public inbox for [email protected]  

help / color / mirror / Atom feed

[PATCH v12 05/11] Add pg_ls_dir_metadata to list a dir with file metadata..
68+ messages / 3 participants
[nested] [flat]

* [PATCH v12 05/11] Add pg_ls_dir_metadata to list a dir with file metadata..
@ 2020-03-08 22:15  Justin Pryzby <[email protected]>
  0 siblings, 0 replies; 68+ messages in thread

From: Justin Pryzby @ 2020-03-08 22:15 UTC (permalink / raw)

Generalize pg_ls_dir_files and retire pg_ls_dir

Need catversion bumped?
---
 doc/src/sgml/func.sgml          |  17 +++
 src/backend/utils/adt/genfile.c | 194 ++++++++++++++++++--------------
 src/include/catalog/pg_proc.dat |   6 +
 3 files changed, 132 insertions(+), 85 deletions(-)

diff --git a/doc/src/sgml/func.sgml b/doc/src/sgml/func.sgml
index 2c6142a0e0..4b966ed847 100644
--- a/doc/src/sgml/func.sgml
+++ b/doc/src/sgml/func.sgml
@@ -21342,6 +21342,15 @@ postgres=# SELECT * FROM pg_walfile_name_offset(pg_stop_backup());
         List the contents of a directory.  Restricted to superusers by default, but other users can be granted EXECUTE to run the function.
        </entry>
       </row>
+      <row>
+       <entry>
+        <literal><function>pg_ls_dir_metadata(<parameter>dirname</parameter> <type>text</type> [, <parameter>missing_ok</parameter> <type>boolean</type>, <parameter>include_dot_dirs</parameter> <type>boolean</type>])</function></literal>
+       </entry>
+       <entry><type>setof text</type></entry>
+       <entry>
+        For each file in a directory, list the file and its metadata.  Restricted to superusers by default, but other users can be granted EXECUTE to run the function.
+       </entry>
+      </row>
       <row>
        <entry>
         <literal><function>pg_ls_logdir()</function></literal>
@@ -21442,6 +21451,14 @@ postgres=# SELECT * FROM pg_walfile_name_offset(pg_stop_backup());
     empty directory from an non-existent directory.
    </para>
 
+   <indexterm>
+    <primary>pg_ls_dir_metadata</primary>
+   </indexterm>
+   <para>
+    <function>pg_ls_dir_metadata</function> lists the files in the specified
+    directory along with the file's metadata.
+   </para>
+
    <indexterm>
     <primary>pg_ls_logdir</primary>
    </indexterm>
diff --git a/src/backend/utils/adt/genfile.c b/src/backend/utils/adt/genfile.c
index bcf9bd1b97..98ab9a2b92 100644
--- a/src/backend/utils/adt/genfile.c
+++ b/src/backend/utils/adt/genfile.c
@@ -36,14 +36,23 @@
 #include "utils/syscache.h"
 #include "utils/timestamp.h"
 
-typedef struct
-{
-	char	   *location;
-	DIR		   *dirdesc;
-	bool		include_dot_dirs;
-} directory_fctx;
+static Datum pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, int flags);
+
+#define	LS_DIR_ISDIR				(1<<0) /* Show column: isdir */
+#define	LS_DIR_METADATA				(1<<1) /* Show columns: mtime, size */
+#define	LS_DIR_MISSING_OK			(1<<2) /* Ignore ENOENT if the toplevel dir is missing */
+#define	LS_DIR_SKIP_DOT_DIRS		(1<<3) /* Do not show . or .. */
+#define	LS_DIR_SKIP_HIDDEN			(1<<4) /* Do not show anything begining with . */
+#define	LS_DIR_SKIP_DIRS			(1<<5) /* Do not show directories */
+#define	LS_DIR_SKIP_SPECIAL			(1<<6) /* Do not show special file types */
 
 
+/*
+ * Shortcut for the historic behavior of the pg_ls_* functions (not including
+ * pg_ls_dir, which skips different files and doesn't show metadata.
+ */
+#define LS_DIR_HISTORIC				(LS_DIR_SKIP_DIRS|LS_DIR_SKIP_HIDDEN|LS_DIR_SKIP_SPECIAL|LS_DIR_METADATA)
+
 /*
  * Convert a "text" filename argument to C string, and check it's allowable.
  *
@@ -447,67 +456,9 @@ pg_stat_file_1arg(PG_FUNCTION_ARGS)
 Datum
 pg_ls_dir(PG_FUNCTION_ARGS)
 {
-	FuncCallContext *funcctx;
-	struct dirent *de;
-	directory_fctx *fctx;
-	MemoryContext oldcontext;
-
-	if (SRF_IS_FIRSTCALL())
-	{
-		bool		missing_ok = false;
-		bool		include_dot_dirs = false;
-
-		/* check the optional arguments */
-		if (PG_NARGS() == 3)
-		{
-			if (!PG_ARGISNULL(1))
-				missing_ok = PG_GETARG_BOOL(1);
-			if (!PG_ARGISNULL(2))
-				include_dot_dirs = PG_GETARG_BOOL(2);
-		}
-
-		funcctx = SRF_FIRSTCALL_INIT();
-		oldcontext = MemoryContextSwitchTo(funcctx->multi_call_memory_ctx);
-
-		fctx = palloc(sizeof(directory_fctx));
-		fctx->location = convert_and_check_filename(PG_GETARG_TEXT_PP(0));
-
-		fctx->include_dot_dirs = include_dot_dirs;
-		fctx->dirdesc = AllocateDir(fctx->location);
-
-		if (!fctx->dirdesc)
-		{
-			if (missing_ok && errno == ENOENT)
-			{
-				MemoryContextSwitchTo(oldcontext);
-				SRF_RETURN_DONE(funcctx);
-			}
-			else
-				ereport(ERROR,
-						(errcode_for_file_access(),
-						 errmsg("could not open directory \"%s\": %m",
-								fctx->location)));
-		}
-		funcctx->user_fctx = fctx;
-		MemoryContextSwitchTo(oldcontext);
-	}
-
-	funcctx = SRF_PERCALL_SETUP();
-	fctx = (directory_fctx *) funcctx->user_fctx;
-
-	while ((de = ReadDir(fctx->dirdesc, fctx->location)) != NULL)
-	{
-		if (!fctx->include_dot_dirs &&
-			(strcmp(de->d_name, ".") == 0 ||
-			 strcmp(de->d_name, "..") == 0))
-			continue;
-
-		SRF_RETURN_NEXT(funcctx, CStringGetTextDatum(de->d_name));
-	}
-
-	FreeDir(fctx->dirdesc);
-
-	SRF_RETURN_DONE(funcctx);
+	text	*filename_t = PG_GETARG_TEXT_PP(0);
+	char	*filename = convert_and_check_filename(filename_t);
+	return pg_ls_dir_files(fcinfo, filename, LS_DIR_SKIP_DOT_DIRS);
 }
 
 /*
@@ -520,7 +471,9 @@ pg_ls_dir(PG_FUNCTION_ARGS)
 Datum
 pg_ls_dir_1arg(PG_FUNCTION_ARGS)
 {
-	return pg_ls_dir(fcinfo);
+	text	*filename_t = PG_GETARG_TEXT_PP(0);
+	char	*filename = convert_and_check_filename(filename_t);
+	return pg_ls_dir_files(fcinfo, filename, LS_DIR_SKIP_DOT_DIRS);
 }
 
 /*
@@ -530,7 +483,7 @@ pg_ls_dir_1arg(PG_FUNCTION_ARGS)
  * Other unreadable-directory cases throw an error.
  */
 static Datum
-pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, bool missing_ok)
+pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, int flags)
 {
 	ReturnSetInfo *rsinfo = (ReturnSetInfo *) fcinfo->resultinfo;
 	bool		randomAccess;
@@ -539,6 +492,32 @@ pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, bool missing_ok)
 	DIR		   *dirdesc;
 	struct dirent *de;
 	MemoryContext oldcontext;
+	TypeFuncClass	tuptype ;
+
+	/* isdir depends on metadata */
+	Assert(!(flags&LS_DIR_ISDIR) || (flags&LS_DIR_METADATA));
+	/* Unreasonable to show isdir and skip dirs */
+	Assert(!(flags&LS_DIR_ISDIR) || !(flags&LS_DIR_SKIP_DIRS));
+
+	/* check the optional arguments */
+	if (PG_NARGS() == 3)
+	{
+		if (!PG_ARGISNULL(1))
+		{
+			if (PG_GETARG_BOOL(1))
+				flags |= LS_DIR_MISSING_OK;
+			else
+				flags &= ~LS_DIR_MISSING_OK;
+		}
+
+		if (!PG_ARGISNULL(2))
+		{
+			if (PG_GETARG_BOOL(2))
+				flags &= ~LS_DIR_SKIP_DOT_DIRS;
+			else
+				flags |= LS_DIR_SKIP_DOT_DIRS;
+		}
+	}
 
 	/* check to see if caller supports us returning a tuplestore */
 	if (rsinfo == NULL || !IsA(rsinfo, ReturnSetInfo))
@@ -554,8 +533,18 @@ pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, bool missing_ok)
 	/* The tupdesc and tuplestore must be created in ecxt_per_query_memory */
 	oldcontext = MemoryContextSwitchTo(rsinfo->econtext->ecxt_per_query_memory);
 
-	if (get_call_result_type(fcinfo, NULL, &tupdesc) != TYPEFUNC_COMPOSITE)
-		elog(ERROR, "return type must be a row type");
+	tuptype = get_call_result_type(fcinfo, NULL, &tupdesc);
+	if (flags & LS_DIR_METADATA)
+	{
+		if (tuptype != TYPEFUNC_COMPOSITE)
+			elog(ERROR, "return type must be a row type");
+	} else {
+		/* pg_ls_dir returns a simple scalar */
+		if (tuptype != TYPEFUNC_SCALAR)
+			elog(ERROR, "return type must be a scalar type");
+		tupdesc = CreateTemplateTupleDesc(1);
+		TupleDescInitEntry(tupdesc, (AttrNumber) 1, "column", TEXTOID, -1, 0);
+	}
 
 	randomAccess = (rsinfo->allowedModes & SFRM_Materialize_Random) != 0;
 	tupstore = tuplestore_begin_heap(randomAccess, false, work_mem);
@@ -574,7 +563,7 @@ pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, bool missing_ok)
 	if (!dirdesc)
 	{
 		/* Return empty tuplestore if appropriate */
-		if (missing_ok && errno == ENOENT)
+		if (flags&LS_DIR_MISSING_OK && errno == ENOENT)
 		{
 			tuplestore_donestoring(tupstore);
 			return (Datum) 0;
@@ -584,13 +573,20 @@ pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, bool missing_ok)
 
 	while ((de = ReadDir(dirdesc, dir)) != NULL)
 	{
-		Datum		values[3];
-		bool		nulls[3];
+		Datum		values[4];
+		bool		nulls[4];
 		char		path[MAXPGPATH * 2];
 		struct stat attrib;
 
-		/* Skip hidden files */
-		if (de->d_name[0] == '.')
+		/* Skip dot dirs? */
+		if (flags & LS_DIR_SKIP_DOT_DIRS &&
+			(strcmp(de->d_name, ".") == 0 ||
+			 strcmp(de->d_name, "..") == 0))
+			continue;
+
+		/* Skip hidden files? */
+		if (flags & LS_DIR_SKIP_HIDDEN &&
+			de->d_name[0] == '.')
 			continue;
 
 		/* Get the file info */
@@ -600,13 +596,27 @@ pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, bool missing_ok)
 					(errcode_for_file_access(),
 					 errmsg("could not stat file \"%s\": %m", path)));
 
-		/* Ignore anything but regular files */
-		if (!S_ISREG(attrib.st_mode))
-			continue;
+		/* Skip dirs or special files? */
+		if (S_ISDIR(attrib.st_mode))
+		{
+			if (flags & LS_DIR_SKIP_DIRS)
+				continue;
+		}
+		else if (!S_ISREG(attrib.st_mode))
+		{
+			if (flags & LS_DIR_SKIP_SPECIAL)
+				continue;
+		}
 
 		values[0] = CStringGetTextDatum(de->d_name);
-		values[1] = Int64GetDatum((int64) attrib.st_size);
-		values[2] = TimestampTzGetDatum(time_t_to_timestamptz(attrib.st_mtime));
+		if (flags & LS_DIR_METADATA)
+		{
+			values[1] = Int64GetDatum((int64) attrib.st_size);
+			values[2] = TimestampTzGetDatum(time_t_to_timestamptz(attrib.st_mtime));
+			if (flags & LS_DIR_ISDIR)
+				values[3] = BoolGetDatum(S_ISDIR(attrib.st_mode));
+		}
+
 		memset(nulls, 0, sizeof(nulls));
 
 		tuplestore_putvalues(tupstore, tupdesc, values, nulls);
@@ -621,14 +631,14 @@ pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, bool missing_ok)
 Datum
 pg_ls_logdir(PG_FUNCTION_ARGS)
 {
-	return pg_ls_dir_files(fcinfo, Log_directory, false);
+	return pg_ls_dir_files(fcinfo, Log_directory, LS_DIR_HISTORIC);
 }
 
 /* Function to return the list of files in the WAL directory */
 Datum
 pg_ls_waldir(PG_FUNCTION_ARGS)
 {
-	return pg_ls_dir_files(fcinfo, XLOGDIR, false);
+	return pg_ls_dir_files(fcinfo, XLOGDIR, LS_DIR_HISTORIC);
 }
 
 /*
@@ -646,7 +656,8 @@ pg_ls_tmpdir(FunctionCallInfo fcinfo, Oid tblspc)
 						tblspc)));
 
 	TempTablespacePath(path, tblspc);
-	return pg_ls_dir_files(fcinfo, path, true);
+	return pg_ls_dir_files(fcinfo, path,
+			LS_DIR_HISTORIC|LS_DIR_MISSING_OK);
 }
 
 /*
@@ -675,5 +686,18 @@ pg_ls_tmpdir_1arg(PG_FUNCTION_ARGS)
 Datum
 pg_ls_archive_statusdir(PG_FUNCTION_ARGS)
 {
-	return pg_ls_dir_files(fcinfo, XLOGDIR "/archive_status", true);
+	return pg_ls_dir_files(fcinfo, XLOGDIR "/archive_status",
+			LS_DIR_HISTORIC|LS_DIR_MISSING_OK);
+}
+
+/*
+ * Function to return the list of files and metadata in an arbitrary directory.
+ */
+Datum
+pg_ls_dir_metadata(PG_FUNCTION_ARGS)
+{
+	char	*dirname = convert_and_check_filename(PG_GETARG_TEXT_PP(0));
+
+	return pg_ls_dir_files(fcinfo, dirname,
+			LS_DIR_METADATA|LS_DIR_SKIP_SPECIAL|LS_DIR_ISDIR);
 }
diff --git a/src/include/catalog/pg_proc.dat b/src/include/catalog/pg_proc.dat
index 7fb574f9dc..0a1859f709 100644
--- a/src/include/catalog/pg_proc.dat
+++ b/src/include/catalog/pg_proc.dat
@@ -10741,6 +10741,12 @@
   proallargtypes => '{oid,text,int8,timestamptz}', proargmodes => '{i,o,o,o}',
   proargnames => '{tablespace,name,size,modification}',
   prosrc => 'pg_ls_tmpdir_1arg' },
+{ oid => '5032', descr => 'list directory with metadata',
+  proname => 'pg_ls_dir_metadata', procost => '10', prorows => '20', proretset => 't',
+  provolatile => 'v', prorettype => 'record', proargtypes => 'text bool bool bool',
+  proallargtypes => '{text,bool,bool,bool,text,int8,timestamptz,bool}', proargmodes => '{i,i,i,i,o,o,o,o}',
+  proargnames => '{dirname,missing_ok,include_dot_dirs,dir_ok,name,size,modification,isdir}',
+  prosrc => 'pg_ls_dir_metadata' },
 
 # hash partitioning constraint function
 { oid => '5028', descr => 'hash partition CHECK constraint',
-- 
2.17.0


--wIc/V6YLA2QdyfT4
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
 filename="v12-0006-Show-links-to-dirs-with-isdir-false.patch"



^ permalink  raw  reply  [nested|flat] 68+ messages in thread

* [PATCH v13 3/8] Add pg_ls_dir_metadata to list a dir with file metadata..
@ 2020-03-08 22:15  Justin Pryzby <[email protected]>
  0 siblings, 0 replies; 68+ messages in thread

From: Justin Pryzby @ 2020-03-08 22:15 UTC (permalink / raw)

Generalize pg_ls_dir_files and retire pg_ls_dir

Change to use lstat() to allow pg_ls_dir_recurse to avoid infinite recursion.
That means:
 - links to dirs are shown with isdir=false;
 - timestamps shown are those of the link;
 - changed pg_stat_file for consistency;

Need catversion bumped?
---
 doc/src/sgml/func.sgml               |  19 ++-
 src/backend/catalog/system_views.sql |   1 +
 src/backend/utils/adt/genfile.c      | 218 +++++++++++++++------------
 src/include/catalog/pg_proc.dat      |   6 +
 4 files changed, 150 insertions(+), 94 deletions(-)

diff --git a/doc/src/sgml/func.sgml b/doc/src/sgml/func.sgml
index 2c6142a0e0..68c7327e1d 100644
--- a/doc/src/sgml/func.sgml
+++ b/doc/src/sgml/func.sgml
@@ -21342,6 +21342,15 @@ postgres=# SELECT * FROM pg_walfile_name_offset(pg_stop_backup());
         List the contents of a directory.  Restricted to superusers by default, but other users can be granted EXECUTE to run the function.
        </entry>
       </row>
+      <row>
+       <entry>
+        <literal><function>pg_ls_dir_metadata(<parameter>dirname</parameter> <type>text</type> [, <parameter>missing_ok</parameter> <type>boolean</type>, <parameter>include_dot_dirs</parameter> <type>boolean</type>])</function></literal>
+       </entry>
+       <entry><type>setof text</type></entry>
+       <entry>
+        For each file in a directory, list the file and its metadata.  Restricted to superusers by default, but other users can be granted EXECUTE to run the function.
+       </entry>
+      </row>
       <row>
        <entry>
         <literal><function>pg_ls_logdir()</function></literal>
@@ -21442,6 +21451,14 @@ postgres=# SELECT * FROM pg_walfile_name_offset(pg_stop_backup());
     empty directory from an non-existent directory.
    </para>
 
+   <indexterm>
+    <primary>pg_ls_dir_metadata</primary>
+   </indexterm>
+   <para>
+    <function>pg_ls_dir_metadata</function> lists the files in the specified
+    directory along with the file's metadata.
+   </para>
+
    <indexterm>
     <primary>pg_ls_logdir</primary>
    </indexterm>
@@ -21528,7 +21545,7 @@ SELECT convert_from(pg_read_binary_file('file_in_utf8.txt'), 'UTF8');
     size, last accessed time stamp, last modified time stamp,
     last file status change time stamp (Unix platforms only),
     file creation time stamp (Windows only), and a <type>boolean</type>
-    indicating if it is a directory (or a symbolic link to a directory).
+    indicating if it is a directory (and not a symbolic link to a directory).
     Typical usages include:
 <programlisting>
 SELECT * FROM pg_stat_file('filename');
diff --git a/src/backend/catalog/system_views.sql b/src/backend/catalog/system_views.sql
index b8a3f46912..05a644a7c9 100644
--- a/src/backend/catalog/system_views.sql
+++ b/src/backend/catalog/system_views.sql
@@ -1435,6 +1435,7 @@ REVOKE EXECUTE ON FUNCTION pg_stat_file(text,boolean) FROM public;
 
 REVOKE EXECUTE ON FUNCTION pg_ls_dir(text) FROM public;
 REVOKE EXECUTE ON FUNCTION pg_ls_dir(text,boolean,boolean) FROM public;
+REVOKE EXECUTE ON FUNCTION pg_ls_dir_metadata(text,boolean,boolean) FROM public;
 
 --
 -- We also set up some things as accessible to standard roles.
diff --git a/src/backend/utils/adt/genfile.c b/src/backend/utils/adt/genfile.c
index 01185f218b..4b70a00a35 100644
--- a/src/backend/utils/adt/genfile.c
+++ b/src/backend/utils/adt/genfile.c
@@ -36,6 +36,21 @@
 #include "utils/syscache.h"
 #include "utils/timestamp.h"
 
+static Datum pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, int flags);
+
+#define	LS_DIR_ISDIR				(1<<0) /* Show column: isdir */
+#define	LS_DIR_METADATA				(1<<1) /* Show columns: mtime, size */
+#define	LS_DIR_MISSING_OK			(1<<2) /* Ignore ENOENT if the toplevel dir is missing */
+#define	LS_DIR_SKIP_DOT_DIRS		(1<<3) /* Do not show . or .. */
+#define	LS_DIR_SKIP_HIDDEN			(1<<4) /* Do not show anything begining with . */
+#define	LS_DIR_SKIP_DIRS			(1<<5) /* Do not show directories */
+#define	LS_DIR_SKIP_SPECIAL			(1<<6) /* Do not show special file types */
+
+/*
+ * Shortcut for the historic behavior of the pg_ls_* functions (not including
+ * pg_ls_dir, which skips different files and doesn't show metadata.
+ */
+#define LS_DIR_HISTORIC				(LS_DIR_SKIP_DIRS|LS_DIR_SKIP_HIDDEN|LS_DIR_SKIP_SPECIAL|LS_DIR_METADATA)
 
 /*
  * Convert a "text" filename argument to C string, and check it's allowable.
@@ -370,7 +385,7 @@ pg_stat_file(PG_FUNCTION_ARGS)
 
 	filename = convert_and_check_filename(filename_t);
 
-	if (stat(filename, &fst) < 0)
+	if (lstat(filename, &fst) < 0)
 	{
 		if (missing_ok && errno == ENOENT)
 			PG_RETURN_NULL();
@@ -413,6 +428,10 @@ pg_stat_file(PG_FUNCTION_ARGS)
 	values[4] = TimestampTzGetDatum(time_t_to_timestamptz(fst.st_ctime));
 #endif
 	values[5] = BoolGetDatum(S_ISDIR(fst.st_mode));
+#ifdef WIN32
+	if (pgwin32_is_junction(path)) /* Links are not directories */
+		values[5] = BoolGetDatum(false);
+#endif
 
 	tuple = heap_form_tuple(tupdesc, values, isnull);
 
@@ -440,79 +459,9 @@ pg_stat_file_1arg(PG_FUNCTION_ARGS)
 Datum
 pg_ls_dir(PG_FUNCTION_ARGS)
 {
-	ReturnSetInfo *rsinfo = (ReturnSetInfo *) fcinfo->resultinfo;
-	char	   *location;
-	bool		missing_ok = false;
-	bool		include_dot_dirs = false;
-	bool		randomAccess;
-	TupleDesc	tupdesc;
-	Tuplestorestate *tupstore;
-	DIR		   *dirdesc;
-	struct dirent *de;
-	MemoryContext oldcontext;
-
-	location = convert_and_check_filename(PG_GETARG_TEXT_PP(0));
-
-	/* check the optional arguments */
-	if (PG_NARGS() == 3)
-	{
-		if (!PG_ARGISNULL(1))
-			missing_ok = PG_GETARG_BOOL(1);
-		if (!PG_ARGISNULL(2))
-			include_dot_dirs = PG_GETARG_BOOL(2);
-	}
-
-	/* check to see if caller supports us returning a tuplestore */
-	if (rsinfo == NULL || !IsA(rsinfo, ReturnSetInfo))
-		ereport(ERROR,
-				(errcode(ERRCODE_FEATURE_NOT_SUPPORTED),
-				 errmsg("set-valued function called in context that cannot accept a set")));
-	if (!(rsinfo->allowedModes & SFRM_Materialize))
-		ereport(ERROR,
-				(errcode(ERRCODE_SYNTAX_ERROR),
-				 errmsg("materialize mode required, but it is not allowed in this context")));
-
-	/* The tupdesc and tuplestore must be created in ecxt_per_query_memory */
-	oldcontext = MemoryContextSwitchTo(rsinfo->econtext->ecxt_per_query_memory);
-
-	tupdesc = CreateTemplateTupleDesc(1);
-	TupleDescInitEntry(tupdesc, (AttrNumber) 1, "pg_ls_dir", TEXTOID, -1, 0);
-
-	randomAccess = (rsinfo->allowedModes & SFRM_Materialize_Random) != 0;
-	tupstore = tuplestore_begin_heap(randomAccess, false, work_mem);
-	rsinfo->returnMode = SFRM_Materialize;
-	rsinfo->setResult = tupstore;
-	rsinfo->setDesc = tupdesc;
-
-	MemoryContextSwitchTo(oldcontext);
-
-	dirdesc = AllocateDir(location);
-	if (!dirdesc)
-	{
-		/* Return empty tuplestore if appropriate */
-		if (missing_ok && errno == ENOENT)
-			return (Datum) 0;
-		/* Otherwise, we can let ReadDir() throw the error */
-	}
-
-	while ((de = ReadDir(dirdesc, location)) != NULL)
-	{
-		Datum		values[1];
-		bool		nulls[1];
-
-		if (!include_dot_dirs &&
-			(strcmp(de->d_name, ".") == 0 ||
-			 strcmp(de->d_name, "..") == 0))
-			continue;
-
-		values[0] = CStringGetTextDatum(de->d_name);
-		nulls[0] = false;
-
-		tuplestore_putvalues(tupstore, tupdesc, values, nulls);
-	}
-
-	FreeDir(dirdesc);
-	return (Datum) 0;
+	text	*filename_t = PG_GETARG_TEXT_PP(0);
+	char	*filename = convert_and_check_filename(filename_t);
+	return pg_ls_dir_files(fcinfo, filename, LS_DIR_SKIP_DOT_DIRS);
 }
 
 /*
@@ -525,7 +474,9 @@ pg_ls_dir(PG_FUNCTION_ARGS)
 Datum
 pg_ls_dir_1arg(PG_FUNCTION_ARGS)
 {
-	return pg_ls_dir(fcinfo);
+	text	*filename_t = PG_GETARG_TEXT_PP(0);
+	char	*filename = convert_and_check_filename(filename_t);
+	return pg_ls_dir_files(fcinfo, filename, LS_DIR_SKIP_DOT_DIRS);
 }
 
 /*
@@ -535,7 +486,7 @@ pg_ls_dir_1arg(PG_FUNCTION_ARGS)
  * Other unreadable-directory cases throw an error.
  */
 static Datum
-pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, bool missing_ok)
+pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, int flags)
 {
 	ReturnSetInfo *rsinfo = (ReturnSetInfo *) fcinfo->resultinfo;
 	bool		randomAccess;
@@ -544,6 +495,32 @@ pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, bool missing_ok)
 	DIR		   *dirdesc;
 	struct dirent *de;
 	MemoryContext oldcontext;
+	TypeFuncClass	tuptype ;
+
+	/* isdir depends on metadata */
+	Assert(!(flags&LS_DIR_ISDIR) || (flags&LS_DIR_METADATA));
+	/* Unreasonable to show isdir and skip dirs */
+	Assert(!(flags&LS_DIR_ISDIR) || !(flags&LS_DIR_SKIP_DIRS));
+
+	/* check the optional arguments */
+	if (PG_NARGS() == 3)
+	{
+		if (!PG_ARGISNULL(1))
+		{
+			if (PG_GETARG_BOOL(1))
+				flags |= LS_DIR_MISSING_OK;
+			else
+				flags &= ~LS_DIR_MISSING_OK;
+		}
+
+		if (!PG_ARGISNULL(2))
+		{
+			if (PG_GETARG_BOOL(2))
+				flags &= ~LS_DIR_SKIP_DOT_DIRS;
+			else
+				flags |= LS_DIR_SKIP_DOT_DIRS;
+		}
+	}
 
 	/* check to see if caller supports us returning a tuplestore */
 	if (rsinfo == NULL || !IsA(rsinfo, ReturnSetInfo))
@@ -558,8 +535,18 @@ pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, bool missing_ok)
 	/* The tupdesc and tuplestore must be created in ecxt_per_query_memory */
 	oldcontext = MemoryContextSwitchTo(rsinfo->econtext->ecxt_per_query_memory);
 
-	if (get_call_result_type(fcinfo, NULL, &tupdesc) != TYPEFUNC_COMPOSITE)
-		elog(ERROR, "return type must be a row type");
+	tuptype = get_call_result_type(fcinfo, NULL, &tupdesc);
+	if (flags & LS_DIR_METADATA)
+	{
+		if (tuptype != TYPEFUNC_COMPOSITE)
+			elog(ERROR, "return type must be a row type");
+	} else {
+		/* pg_ls_dir returns a simple scalar */
+		if (tuptype != TYPEFUNC_SCALAR)
+			elog(ERROR, "return type must be a scalar type");
+		tupdesc = CreateTemplateTupleDesc(1);
+		TupleDescInitEntry(tupdesc, (AttrNumber) 1, "column", TEXTOID, -1, 0);
+	}
 
 	randomAccess = (rsinfo->allowedModes & SFRM_Materialize_Random) != 0;
 	tupstore = tuplestore_begin_heap(randomAccess, false, work_mem);
@@ -578,36 +565,67 @@ pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, bool missing_ok)
 	if (!dirdesc)
 	{
 		/* Return empty tuplestore if appropriate */
-		if (missing_ok && errno == ENOENT)
+		if (flags&LS_DIR_MISSING_OK && errno == ENOENT)
+		{
+			tuplestore_donestoring(tupstore);
 			return (Datum) 0;
+		}
 		/* Otherwise, we can let ReadDir() throw the error */
 	}
 
 	while ((de = ReadDir(dirdesc, dir)) != NULL)
 	{
-		Datum		values[3];
-		bool		nulls[3];
+		Datum		values[4];
+		bool		nulls[4];
 		char		path[MAXPGPATH * 2];
 		struct stat attrib;
 
-		/* Skip hidden files */
-		if (de->d_name[0] == '.')
+		/* Skip dot dirs? */
+		if (flags & LS_DIR_SKIP_DOT_DIRS &&
+			(strcmp(de->d_name, ".") == 0 ||
+			 strcmp(de->d_name, "..") == 0))
+			continue;
+
+		/* Skip hidden files? */
+		if (flags & LS_DIR_SKIP_HIDDEN &&
+			de->d_name[0] == '.')
 			continue;
 
 		/* Get the file info */
 		snprintf(path, sizeof(path), "%s/%s", dir, de->d_name);
-		if (stat(path, &attrib) < 0)
+		if (lstat(path, &attrib) < 0)
 			ereport(ERROR,
 					(errcode_for_file_access(),
 					 errmsg("could not stat file \"%s\": %m", path)));
 
-		/* Ignore anything but regular files */
-		if (!S_ISREG(attrib.st_mode))
-			continue;
+		/* Skip dirs or special files? */
+		if (S_ISDIR(attrib.st_mode))
+		{
+			if (flags & LS_DIR_SKIP_DIRS)
+				continue;
+		}
+		else if (!S_ISREG(attrib.st_mode))
+		{
+			if (flags & LS_DIR_SKIP_SPECIAL)
+				continue;
+		}
 
 		values[0] = CStringGetTextDatum(de->d_name);
-		values[1] = Int64GetDatum((int64) attrib.st_size);
-		values[2] = TimestampTzGetDatum(time_t_to_timestamptz(attrib.st_mtime));
+		if (flags & LS_DIR_METADATA)
+		{
+			values[1] = Int64GetDatum((int64) attrib.st_size);
+			values[2] = TimestampTzGetDatum(time_t_to_timestamptz(attrib.st_mtime));
+			if (flags & LS_DIR_ISDIR)
+			{
+				values[3] = BoolGetDatum(S_ISDIR(attrib.st_mode));
+#ifdef WIN32
+				/* Links are not directories */
+				if (pgwin32_is_junction(path))
+					values[3] = BoolGetDatum(false);
+#endif
+			}
+		}
+
 		memset(nulls, 0, sizeof(nulls));
 
 		tuplestore_putvalues(tupstore, tupdesc, values, nulls);
@@ -621,14 +639,14 @@ pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, bool missing_ok)
 Datum
 pg_ls_logdir(PG_FUNCTION_ARGS)
 {
-	return pg_ls_dir_files(fcinfo, Log_directory, false);
+	return pg_ls_dir_files(fcinfo, Log_directory, LS_DIR_HISTORIC);
 }
 
 /* Function to return the list of files in the WAL directory */
 Datum
 pg_ls_waldir(PG_FUNCTION_ARGS)
 {
-	return pg_ls_dir_files(fcinfo, XLOGDIR, false);
+	return pg_ls_dir_files(fcinfo, XLOGDIR, LS_DIR_HISTORIC);
 }
 
 /*
@@ -646,7 +664,8 @@ pg_ls_tmpdir(FunctionCallInfo fcinfo, Oid tblspc)
 						tblspc)));
 
 	TempTablespacePath(path, tblspc);
-	return pg_ls_dir_files(fcinfo, path, true);
+	return pg_ls_dir_files(fcinfo, path,
+			LS_DIR_HISTORIC|LS_DIR_MISSING_OK);
 }
 
 /*
@@ -675,5 +694,18 @@ pg_ls_tmpdir_1arg(PG_FUNCTION_ARGS)
 Datum
 pg_ls_archive_statusdir(PG_FUNCTION_ARGS)
 {
-	return pg_ls_dir_files(fcinfo, XLOGDIR "/archive_status", true);
+	return pg_ls_dir_files(fcinfo, XLOGDIR "/archive_status",
+			LS_DIR_HISTORIC|LS_DIR_MISSING_OK);
+}
+
+/*
+ * Function to return the list of files and metadata in an arbitrary directory.
+ */
+Datum
+pg_ls_dir_metadata(PG_FUNCTION_ARGS)
+{
+	char	*dirname = convert_and_check_filename(PG_GETARG_TEXT_PP(0));
+
+	return pg_ls_dir_files(fcinfo, dirname,
+			LS_DIR_METADATA|LS_DIR_SKIP_SPECIAL|LS_DIR_ISDIR);
 }
diff --git a/src/include/catalog/pg_proc.dat b/src/include/catalog/pg_proc.dat
index 7fb574f9dc..0a1859f709 100644
--- a/src/include/catalog/pg_proc.dat
+++ b/src/include/catalog/pg_proc.dat
@@ -10741,6 +10741,12 @@
   proallargtypes => '{oid,text,int8,timestamptz}', proargmodes => '{i,o,o,o}',
   proargnames => '{tablespace,name,size,modification}',
   prosrc => 'pg_ls_tmpdir_1arg' },
+{ oid => '5032', descr => 'list directory with metadata',
+  proname => 'pg_ls_dir_metadata', procost => '10', prorows => '20', proretset => 't',
+  provolatile => 'v', prorettype => 'record', proargtypes => 'text bool bool bool',
+  proallargtypes => '{text,bool,bool,bool,text,int8,timestamptz,bool}', proargmodes => '{i,i,i,i,o,o,o,o}',
+  proargnames => '{dirname,missing_ok,include_dot_dirs,dir_ok,name,size,modification,isdir}',
+  prosrc => 'pg_ls_dir_metadata' },
 
 # hash partitioning constraint function
 { oid => '5028', descr => 'hash partition CHECK constraint',
-- 
2.17.0


--E0h0CbphJD8hN+Gf
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
 filename="v13-0004-pg_ls_tmpdir-to-show-isdir-argument.patch"



^ permalink  raw  reply  [nested|flat] 68+ messages in thread

* [PATCH v9 04/11] Add pg_ls_dir_metadata to list a dir with file metadata..
@ 2020-03-08 22:15  Justin Pryzby <[email protected]>
  0 siblings, 0 replies; 68+ messages in thread

From: Justin Pryzby @ 2020-03-08 22:15 UTC (permalink / raw)

Need catversion bumped?
---
 src/backend/utils/adt/genfile.c | 51 ++++++++++++++++++++++++++-------
 src/include/catalog/pg_proc.dat |  6 ++++
 2 files changed, 47 insertions(+), 10 deletions(-)

diff --git a/src/backend/utils/adt/genfile.c b/src/backend/utils/adt/genfile.c
index 897b11a77d..4699aea568 100644
--- a/src/backend/utils/adt/genfile.c
+++ b/src/backend/utils/adt/genfile.c
@@ -524,7 +524,7 @@ pg_ls_dir_1arg(PG_FUNCTION_ARGS)
 
 /* Generic function to return a directory listing of files */
 static Datum
-pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, bool missing_ok)
+pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, bool missing_ok, bool dir_ok)
 {
 	FuncCallContext *funcctx;
 	struct dirent *de;
@@ -540,13 +540,17 @@ pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, bool missing_ok)
 
 		fctx = palloc(sizeof(directory_fctx));
 
-		tupdesc = CreateTemplateTupleDesc(3);
+		tupdesc = CreateTemplateTupleDesc(dir_ok ? 4:3);
 		TupleDescInitEntry(tupdesc, (AttrNumber) 1, "name",
 						   TEXTOID, -1, 0);
 		TupleDescInitEntry(tupdesc, (AttrNumber) 2, "size",
 						   INT8OID, -1, 0);
 		TupleDescInitEntry(tupdesc, (AttrNumber) 3, "modification",
 						   TIMESTAMPTZOID, -1, 0);
+		if (dir_ok)
+			TupleDescInitEntry(tupdesc, (AttrNumber) 4, "isdir",
+					BOOLOID, -1, 0);
+
 		funcctx->tuple_desc = BlessTupleDesc(tupdesc);
 
 		fctx->location = pstrdup(dir);
@@ -575,8 +579,8 @@ pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, bool missing_ok)
 
 	while ((de = ReadDir(fctx->dirdesc, fctx->location)) != NULL)
 	{
-		Datum		values[3];
-		bool		nulls[3];
+		Datum		values[4];
+		bool		nulls[4];
 		char		path[MAXPGPATH * 2];
 		struct stat attrib;
 		HeapTuple	tuple;
@@ -592,13 +596,21 @@ pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, bool missing_ok)
 					(errcode_for_file_access(),
 					 errmsg("could not stat file \"%s\": %m", path)));
 
-		/* Ignore anything but regular files */
-		if (!S_ISREG(attrib.st_mode))
+		if (S_ISDIR(attrib.st_mode))
+		{
+			if (!dir_ok)
+				continue;
+		}
+		else if (!S_ISREG(attrib.st_mode))
+			/* Ignore anything but regular files */
 			continue;
 
 		values[0] = CStringGetTextDatum(de->d_name);
 		values[1] = Int64GetDatum((int64) attrib.st_size);
 		values[2] = TimestampTzGetDatum(time_t_to_timestamptz(attrib.st_mtime));
+		if (dir_ok)
+			values[3] = BoolGetDatum(S_ISDIR(attrib.st_mode));
+
 		memset(nulls, 0, sizeof(nulls));
 
 		tuple = heap_form_tuple(funcctx->tuple_desc, values, nulls);
@@ -613,14 +625,14 @@ pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, bool missing_ok)
 Datum
 pg_ls_logdir(PG_FUNCTION_ARGS)
 {
-	return pg_ls_dir_files(fcinfo, Log_directory, false);
+	return pg_ls_dir_files(fcinfo, Log_directory, false, false);
 }
 
 /* Function to return the list of files in the WAL directory */
 Datum
 pg_ls_waldir(PG_FUNCTION_ARGS)
 {
-	return pg_ls_dir_files(fcinfo, XLOGDIR, false);
+	return pg_ls_dir_files(fcinfo, XLOGDIR, false, false);
 }
 
 /*
@@ -638,7 +650,7 @@ pg_ls_tmpdir(FunctionCallInfo fcinfo, Oid tblspc)
 						tblspc)));
 
 	TempTablespacePath(path, tblspc);
-	return pg_ls_dir_files(fcinfo, path, true);
+	return pg_ls_dir_files(fcinfo, path, true, false);
 }
 
 /*
@@ -667,5 +679,24 @@ pg_ls_tmpdir_1arg(PG_FUNCTION_ARGS)
 Datum
 pg_ls_archive_statusdir(PG_FUNCTION_ARGS)
 {
-	return pg_ls_dir_files(fcinfo, XLOGDIR "/archive_status", true);
+	return pg_ls_dir_files(fcinfo, XLOGDIR "/archive_status", true, false);
+}
+
+/*
+ * Function to return the list of files and metadata in an arbitrary directory.
+ */
+Datum
+pg_ls_dir_metadata(PG_FUNCTION_ARGS)
+{
+	char	*dirname = convert_and_check_filename(PG_GETARG_TEXT_PP(0));
+	bool	missing_ok = false;
+	bool	include_dot_dirs = false;
+
+	if (!PG_ARGISNULL(1))
+		missing_ok = PG_GETARG_BOOL(1);
+	if (!PG_ARGISNULL(2))
+		/* XXX: Not implemented */
+		include_dot_dirs = PG_GETARG_BOOL(2);
+
+	return pg_ls_dir_files(fcinfo, dirname, missing_ok, true);
 }
diff --git a/src/include/catalog/pg_proc.dat b/src/include/catalog/pg_proc.dat
index 7fb574f9dc..0a1859f709 100644
--- a/src/include/catalog/pg_proc.dat
+++ b/src/include/catalog/pg_proc.dat
@@ -10741,6 +10741,12 @@
   proallargtypes => '{oid,text,int8,timestamptz}', proargmodes => '{i,o,o,o}',
   proargnames => '{tablespace,name,size,modification}',
   prosrc => 'pg_ls_tmpdir_1arg' },
+{ oid => '5032', descr => 'list directory with metadata',
+  proname => 'pg_ls_dir_metadata', procost => '10', prorows => '20', proretset => 't',
+  provolatile => 'v', prorettype => 'record', proargtypes => 'text bool bool bool',
+  proallargtypes => '{text,bool,bool,bool,text,int8,timestamptz,bool}', proargmodes => '{i,i,i,i,o,o,o,o}',
+  proargnames => '{dirname,missing_ok,include_dot_dirs,dir_ok,name,size,modification,isdir}',
+  prosrc => 'pg_ls_dir_metadata' },
 
 # hash partitioning constraint function
 { oid => '5028', descr => 'hash partition CHECK constraint',
-- 
2.17.0


--32u276st3Jlj2kUU
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
 filename="v9-0005-Add-tests-exercizing-pg_ls_tmpdir.patch"



^ permalink  raw  reply  [nested|flat] 68+ messages in thread

* [PATCH v10 4/9] Add pg_ls_dir_metadata to list a dir with file metadata..
@ 2020-03-08 22:15  Justin Pryzby <[email protected]>
  0 siblings, 0 replies; 68+ messages in thread

From: Justin Pryzby @ 2020-03-08 22:15 UTC (permalink / raw)

Generalize pg_ls_dir_files and retire pg_ls_dir

Need catversion bumped?
---
 src/backend/utils/adt/genfile.c | 184 +++++++++++++++++---------------
 src/include/catalog/pg_proc.dat |   6 ++
 2 files changed, 106 insertions(+), 84 deletions(-)

diff --git a/src/backend/utils/adt/genfile.c b/src/backend/utils/adt/genfile.c
index bcf9bd1b97..4f437573ab 100644
--- a/src/backend/utils/adt/genfile.c
+++ b/src/backend/utils/adt/genfile.c
@@ -36,13 +36,15 @@
 #include "utils/syscache.h"
 #include "utils/timestamp.h"
 
-typedef struct
-{
-	char	   *location;
-	DIR		   *dirdesc;
-	bool		include_dot_dirs;
-} directory_fctx;
+static Datum pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, int flags);
 
+#define	FLAG_ISDIR				(1<<0) /* Show column: isdir */
+#define	FLAG_METADATA			(1<<1) /* Show columns: mtime, size */
+#define	FLAG_MISSING_OK			(1<<2) /* Ignore ENOENT if the toplevel dir is missing */
+#define	FLAG_SKIP_DOT_DIRS		(1<<3) /* Do not show . or .. */
+#define	FLAG_SKIP_HIDDEN		(1<<4) /* Do not show anything begining with . */
+#define	FLAG_SKIP_DIRS			(1<<5) /* Do not show directories */
+#define	FLAG_SKIP_SPECIAL		(1<<6) /* Do not show special file types */
 
 /*
  * Convert a "text" filename argument to C string, and check it's allowable.
@@ -447,67 +449,9 @@ pg_stat_file_1arg(PG_FUNCTION_ARGS)
 Datum
 pg_ls_dir(PG_FUNCTION_ARGS)
 {
-	FuncCallContext *funcctx;
-	struct dirent *de;
-	directory_fctx *fctx;
-	MemoryContext oldcontext;
-
-	if (SRF_IS_FIRSTCALL())
-	{
-		bool		missing_ok = false;
-		bool		include_dot_dirs = false;
-
-		/* check the optional arguments */
-		if (PG_NARGS() == 3)
-		{
-			if (!PG_ARGISNULL(1))
-				missing_ok = PG_GETARG_BOOL(1);
-			if (!PG_ARGISNULL(2))
-				include_dot_dirs = PG_GETARG_BOOL(2);
-		}
-
-		funcctx = SRF_FIRSTCALL_INIT();
-		oldcontext = MemoryContextSwitchTo(funcctx->multi_call_memory_ctx);
-
-		fctx = palloc(sizeof(directory_fctx));
-		fctx->location = convert_and_check_filename(PG_GETARG_TEXT_PP(0));
-
-		fctx->include_dot_dirs = include_dot_dirs;
-		fctx->dirdesc = AllocateDir(fctx->location);
-
-		if (!fctx->dirdesc)
-		{
-			if (missing_ok && errno == ENOENT)
-			{
-				MemoryContextSwitchTo(oldcontext);
-				SRF_RETURN_DONE(funcctx);
-			}
-			else
-				ereport(ERROR,
-						(errcode_for_file_access(),
-						 errmsg("could not open directory \"%s\": %m",
-								fctx->location)));
-		}
-		funcctx->user_fctx = fctx;
-		MemoryContextSwitchTo(oldcontext);
-	}
-
-	funcctx = SRF_PERCALL_SETUP();
-	fctx = (directory_fctx *) funcctx->user_fctx;
-
-	while ((de = ReadDir(fctx->dirdesc, fctx->location)) != NULL)
-	{
-		if (!fctx->include_dot_dirs &&
-			(strcmp(de->d_name, ".") == 0 ||
-			 strcmp(de->d_name, "..") == 0))
-			continue;
-
-		SRF_RETURN_NEXT(funcctx, CStringGetTextDatum(de->d_name));
-	}
-
-	FreeDir(fctx->dirdesc);
-
-	SRF_RETURN_DONE(funcctx);
+	text	*filename_t = PG_GETARG_TEXT_PP(0);
+	char	*filename = convert_and_check_filename(filename_t);
+	return pg_ls_dir_files(fcinfo, filename, FLAG_SKIP_DOT_DIRS);
 }
 
 /*
@@ -520,7 +464,9 @@ pg_ls_dir(PG_FUNCTION_ARGS)
 Datum
 pg_ls_dir_1arg(PG_FUNCTION_ARGS)
 {
-	return pg_ls_dir(fcinfo);
+	text	*filename_t = PG_GETARG_TEXT_PP(0);
+	char	*filename = convert_and_check_filename(filename_t);
+	return pg_ls_dir_files(fcinfo, filename, FLAG_SKIP_DOT_DIRS);
 }
 
 /*
@@ -530,7 +476,7 @@ pg_ls_dir_1arg(PG_FUNCTION_ARGS)
  * Other unreadable-directory cases throw an error.
  */
 static Datum
-pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, bool missing_ok)
+pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, int flags)
 {
 	ReturnSetInfo *rsinfo = (ReturnSetInfo *) fcinfo->resultinfo;
 	bool		randomAccess;
@@ -539,6 +485,32 @@ pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, bool missing_ok)
 	DIR		   *dirdesc;
 	struct dirent *de;
 	MemoryContext oldcontext;
+	TypeFuncClass	tuptype ;
+
+	/* isdir depends on metadata */
+	Assert(!(flags&FLAG_ISDIR) || (flags&FLAG_METADATA));
+	/* Unreasonable to show isdir and skip dirs */
+	Assert(!(flags&FLAG_ISDIR) || !(flags&FLAG_SKIP_DIRS));
+
+	/* check the optional arguments */
+	if (PG_NARGS() == 3)
+	{
+		if (!PG_ARGISNULL(1))
+		{
+			if (PG_GETARG_BOOL(1))
+				flags |= FLAG_MISSING_OK;
+			else
+				flags &= ~FLAG_MISSING_OK;
+		}
+
+		if (!PG_ARGISNULL(2))
+		{
+			if (PG_GETARG_BOOL(2))
+				flags &= ~FLAG_SKIP_DOT_DIRS;
+			else
+				flags |= FLAG_SKIP_DOT_DIRS;
+		}
+	}
 
 	/* check to see if caller supports us returning a tuplestore */
 	if (rsinfo == NULL || !IsA(rsinfo, ReturnSetInfo))
@@ -554,8 +526,18 @@ pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, bool missing_ok)
 	/* The tupdesc and tuplestore must be created in ecxt_per_query_memory */
 	oldcontext = MemoryContextSwitchTo(rsinfo->econtext->ecxt_per_query_memory);
 
-	if (get_call_result_type(fcinfo, NULL, &tupdesc) != TYPEFUNC_COMPOSITE)
-		elog(ERROR, "return type must be a row type");
+	tuptype = get_call_result_type(fcinfo, NULL, &tupdesc);
+	if (flags&FLAG_METADATA)
+	{
+		if (tuptype != TYPEFUNC_COMPOSITE)
+			elog(ERROR, "return type must be a row type");
+	} else {
+		/* pg_ls_dir returns a simple scalar */
+		if (tuptype != TYPEFUNC_SCALAR)
+			elog(ERROR, "return type must be a scalar type");
+		tupdesc = CreateTemplateTupleDesc(1);
+		TupleDescInitEntry(tupdesc, (AttrNumber) 1, "column", TEXTOID, -1, 0);
+	}
 
 	randomAccess = (rsinfo->allowedModes & SFRM_Materialize_Random) != 0;
 	tupstore = tuplestore_begin_heap(randomAccess, false, work_mem);
@@ -574,7 +556,7 @@ pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, bool missing_ok)
 	if (!dirdesc)
 	{
 		/* Return empty tuplestore if appropriate */
-		if (missing_ok && errno == ENOENT)
+		if (flags&FLAG_MISSING_OK && errno == ENOENT)
 		{
 			tuplestore_donestoring(tupstore);
 			return (Datum) 0;
@@ -584,13 +566,19 @@ pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, bool missing_ok)
 
 	while ((de = ReadDir(dirdesc, dir)) != NULL)
 	{
-		Datum		values[3];
-		bool		nulls[3];
+		Datum		values[4];
+		bool		nulls[4];
 		char		path[MAXPGPATH * 2];
 		struct stat attrib;
 
+		if (flags&FLAG_SKIP_DOT_DIRS &&
+			(strcmp(de->d_name, ".") == 0 ||
+			 strcmp(de->d_name, "..") == 0))
+			continue;
+
 		/* Skip hidden files */
-		if (de->d_name[0] == '.')
+		if (flags&FLAG_SKIP_HIDDEN &&
+			de->d_name[0] == '.')
 			continue;
 
 		/* Get the file info */
@@ -600,13 +588,26 @@ pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, bool missing_ok)
 					(errcode_for_file_access(),
 					 errmsg("could not stat file \"%s\": %m", path)));
 
-		/* Ignore anything but regular files */
-		if (!S_ISREG(attrib.st_mode))
-			continue;
+		if (S_ISDIR(attrib.st_mode))
+		{
+			if (flags&FLAG_SKIP_DIRS)
+				continue;
+		}
+		else if (!S_ISREG(attrib.st_mode))
+		{
+			if (flags&FLAG_SKIP_SPECIAL)
+				continue;
+		}
 
 		values[0] = CStringGetTextDatum(de->d_name);
-		values[1] = Int64GetDatum((int64) attrib.st_size);
-		values[2] = TimestampTzGetDatum(time_t_to_timestamptz(attrib.st_mtime));
+		if (flags & FLAG_METADATA)
+		{
+			values[1] = Int64GetDatum((int64) attrib.st_size);
+			values[2] = TimestampTzGetDatum(time_t_to_timestamptz(attrib.st_mtime));
+			if (flags & FLAG_ISDIR)
+				values[3] = BoolGetDatum(S_ISDIR(attrib.st_mode));
+		}
+
 		memset(nulls, 0, sizeof(nulls));
 
 		tuplestore_putvalues(tupstore, tupdesc, values, nulls);
@@ -621,14 +622,16 @@ pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, bool missing_ok)
 Datum
 pg_ls_logdir(PG_FUNCTION_ARGS)
 {
-	return pg_ls_dir_files(fcinfo, Log_directory, false);
+	return pg_ls_dir_files(fcinfo, Log_directory,
+			FLAG_SKIP_DIRS|FLAG_SKIP_HIDDEN|FLAG_SKIP_SPECIAL|FLAG_METADATA);
 }
 
 /* Function to return the list of files in the WAL directory */
 Datum
 pg_ls_waldir(PG_FUNCTION_ARGS)
 {
-	return pg_ls_dir_files(fcinfo, XLOGDIR, false);
+	return pg_ls_dir_files(fcinfo, XLOGDIR,
+			FLAG_SKIP_DIRS|FLAG_SKIP_HIDDEN|FLAG_SKIP_SPECIAL|FLAG_METADATA);
 }
 
 /*
@@ -646,7 +649,8 @@ pg_ls_tmpdir(FunctionCallInfo fcinfo, Oid tblspc)
 						tblspc)));
 
 	TempTablespacePath(path, tblspc);
-	return pg_ls_dir_files(fcinfo, path, true);
+	return pg_ls_dir_files(fcinfo, path,
+			FLAG_MISSING_OK|FLAG_SKIP_DIRS|FLAG_SKIP_HIDDEN|FLAG_SKIP_SPECIAL|FLAG_METADATA);
 }
 
 /*
@@ -675,5 +679,17 @@ pg_ls_tmpdir_1arg(PG_FUNCTION_ARGS)
 Datum
 pg_ls_archive_statusdir(PG_FUNCTION_ARGS)
 {
-	return pg_ls_dir_files(fcinfo, XLOGDIR "/archive_status", true);
+	return pg_ls_dir_files(fcinfo, XLOGDIR "/archive_status",
+			FLAG_MISSING_OK|FLAG_SKIP_DIRS|FLAG_SKIP_HIDDEN|FLAG_SKIP_SPECIAL|FLAG_METADATA);
+}
+
+/*
+ * Function to return the list of files and metadata in an arbitrary directory.
+ */
+Datum
+pg_ls_dir_metadata(PG_FUNCTION_ARGS)
+{
+	char	*dirname = convert_and_check_filename(PG_GETARG_TEXT_PP(0));
+
+	return pg_ls_dir_files(fcinfo, dirname, FLAG_METADATA|FLAG_SKIP_SPECIAL|FLAG_ISDIR);
 }
diff --git a/src/include/catalog/pg_proc.dat b/src/include/catalog/pg_proc.dat
index 7fb574f9dc..0a1859f709 100644
--- a/src/include/catalog/pg_proc.dat
+++ b/src/include/catalog/pg_proc.dat
@@ -10741,6 +10741,12 @@
   proallargtypes => '{oid,text,int8,timestamptz}', proargmodes => '{i,o,o,o}',
   proargnames => '{tablespace,name,size,modification}',
   prosrc => 'pg_ls_tmpdir_1arg' },
+{ oid => '5032', descr => 'list directory with metadata',
+  proname => 'pg_ls_dir_metadata', procost => '10', prorows => '20', proretset => 't',
+  provolatile => 'v', prorettype => 'record', proargtypes => 'text bool bool bool',
+  proallargtypes => '{text,bool,bool,bool,text,int8,timestamptz,bool}', proargmodes => '{i,i,i,i,o,o,o,o}',
+  proargnames => '{dirname,missing_ok,include_dot_dirs,dir_ok,name,size,modification,isdir}',
+  prosrc => 'pg_ls_dir_metadata' },
 
 # hash partitioning constraint function
 { oid => '5028', descr => 'hash partition CHECK constraint',
-- 
2.17.0


--rCwQ2Y43eQY6RBgR
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
 filename="v10-0005-pg_ls_tmpdir-to-show-isdir-argument.patch"



^ permalink  raw  reply  [nested|flat] 68+ messages in thread

* [PATCH v11 4/9] Add pg_ls_dir_metadata to list a dir with file metadata..
@ 2020-03-08 22:15  Justin Pryzby <[email protected]>
  0 siblings, 0 replies; 68+ messages in thread

From: Justin Pryzby @ 2020-03-08 22:15 UTC (permalink / raw)

Generalize pg_ls_dir_files and retire pg_ls_dir

Need catversion bumped?
---
 doc/src/sgml/func.sgml          |  17 +++
 src/backend/utils/adt/genfile.c | 190 ++++++++++++++++++--------------
 src/include/catalog/pg_proc.dat |   6 +
 3 files changed, 129 insertions(+), 84 deletions(-)

diff --git a/doc/src/sgml/func.sgml b/doc/src/sgml/func.sgml
index fc4d7f0f78..dfea9f8061 100644
--- a/doc/src/sgml/func.sgml
+++ b/doc/src/sgml/func.sgml
@@ -21342,6 +21342,15 @@ postgres=# SELECT * FROM pg_walfile_name_offset(pg_stop_backup());
         List the contents of a directory.  Restricted to superusers by default, but other users can be granted EXECUTE to run the function.
        </entry>
       </row>
+      <row>
+       <entry>
+        <literal><function>pg_ls_dir_metadata(<parameter>dirname</parameter> <type>text</type> [, <parameter>missing_ok</parameter> <type>boolean</type>, <parameter>include_dot_dirs</parameter> <type>boolean</type>])</function></literal>
+       </entry>
+       <entry><type>setof text</type></entry>
+       <entry>
+        For each file in a directory, list the file and its metadata.  Restricted to superusers by default, but other users can be granted EXECUTE to run the function.
+       </entry>
+      </row>
       <row>
        <entry>
         <literal><function>pg_ls_logdir()</function></literal>
@@ -21442,6 +21451,14 @@ postgres=# SELECT * FROM pg_walfile_name_offset(pg_stop_backup());
     empty directory from an non-existent directory.
    </para>
 
+   <indexterm>
+    <primary>pg_ls_dir_metadata</primary>
+   </indexterm>
+   <para>
+    <function>pg_ls_dir_metadata</function> lists the files in the specified
+    directory along with the file's metadata.
+   </para>
+
    <indexterm>
     <primary>pg_ls_logdir</primary>
    </indexterm>
diff --git a/src/backend/utils/adt/genfile.c b/src/backend/utils/adt/genfile.c
index bcf9bd1b97..b105363903 100644
--- a/src/backend/utils/adt/genfile.c
+++ b/src/backend/utils/adt/genfile.c
@@ -36,14 +36,23 @@
 #include "utils/syscache.h"
 #include "utils/timestamp.h"
 
-typedef struct
-{
-	char	   *location;
-	DIR		   *dirdesc;
-	bool		include_dot_dirs;
-} directory_fctx;
+static Datum pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, int flags);
+
+#define	LS_DIR_ISDIR				(1<<0) /* Show column: isdir */
+#define	LS_DIR_METADATA				(1<<1) /* Show columns: mtime, size */
+#define	LS_DIR_MISSING_OK			(1<<2) /* Ignore ENOENT if the toplevel dir is missing */
+#define	LS_DIR_SKIP_DOT_DIRS		(1<<3) /* Do not show . or .. */
+#define	LS_DIR_SKIP_HIDDEN			(1<<4) /* Do not show anything begining with . */
+#define	LS_DIR_SKIP_DIRS			(1<<5) /* Do not show directories */
+#define	LS_DIR_SKIP_SPECIAL			(1<<6) /* Do not show special file types */
 
 
+/*
+ * Shortcut for the historic behavior of the pg_ls_* functions (not including
+ * pg_ls_dir, which skips different files and doesn't show metadata.
+ */
+#define LS_DIR_HISTORIC				(LS_DIR_SKIP_DIRS|LS_DIR_SKIP_HIDDEN|LS_DIR_SKIP_SPECIAL|LS_DIR_METADATA)
+
 /*
  * Convert a "text" filename argument to C string, and check it's allowable.
  *
@@ -447,67 +456,9 @@ pg_stat_file_1arg(PG_FUNCTION_ARGS)
 Datum
 pg_ls_dir(PG_FUNCTION_ARGS)
 {
-	FuncCallContext *funcctx;
-	struct dirent *de;
-	directory_fctx *fctx;
-	MemoryContext oldcontext;
-
-	if (SRF_IS_FIRSTCALL())
-	{
-		bool		missing_ok = false;
-		bool		include_dot_dirs = false;
-
-		/* check the optional arguments */
-		if (PG_NARGS() == 3)
-		{
-			if (!PG_ARGISNULL(1))
-				missing_ok = PG_GETARG_BOOL(1);
-			if (!PG_ARGISNULL(2))
-				include_dot_dirs = PG_GETARG_BOOL(2);
-		}
-
-		funcctx = SRF_FIRSTCALL_INIT();
-		oldcontext = MemoryContextSwitchTo(funcctx->multi_call_memory_ctx);
-
-		fctx = palloc(sizeof(directory_fctx));
-		fctx->location = convert_and_check_filename(PG_GETARG_TEXT_PP(0));
-
-		fctx->include_dot_dirs = include_dot_dirs;
-		fctx->dirdesc = AllocateDir(fctx->location);
-
-		if (!fctx->dirdesc)
-		{
-			if (missing_ok && errno == ENOENT)
-			{
-				MemoryContextSwitchTo(oldcontext);
-				SRF_RETURN_DONE(funcctx);
-			}
-			else
-				ereport(ERROR,
-						(errcode_for_file_access(),
-						 errmsg("could not open directory \"%s\": %m",
-								fctx->location)));
-		}
-		funcctx->user_fctx = fctx;
-		MemoryContextSwitchTo(oldcontext);
-	}
-
-	funcctx = SRF_PERCALL_SETUP();
-	fctx = (directory_fctx *) funcctx->user_fctx;
-
-	while ((de = ReadDir(fctx->dirdesc, fctx->location)) != NULL)
-	{
-		if (!fctx->include_dot_dirs &&
-			(strcmp(de->d_name, ".") == 0 ||
-			 strcmp(de->d_name, "..") == 0))
-			continue;
-
-		SRF_RETURN_NEXT(funcctx, CStringGetTextDatum(de->d_name));
-	}
-
-	FreeDir(fctx->dirdesc);
-
-	SRF_RETURN_DONE(funcctx);
+	text	*filename_t = PG_GETARG_TEXT_PP(0);
+	char	*filename = convert_and_check_filename(filename_t);
+	return pg_ls_dir_files(fcinfo, filename, LS_DIR_SKIP_DOT_DIRS);
 }
 
 /*
@@ -520,7 +471,9 @@ pg_ls_dir(PG_FUNCTION_ARGS)
 Datum
 pg_ls_dir_1arg(PG_FUNCTION_ARGS)
 {
-	return pg_ls_dir(fcinfo);
+	text	*filename_t = PG_GETARG_TEXT_PP(0);
+	char	*filename = convert_and_check_filename(filename_t);
+	return pg_ls_dir_files(fcinfo, filename, LS_DIR_SKIP_DOT_DIRS);
 }
 
 /*
@@ -530,7 +483,7 @@ pg_ls_dir_1arg(PG_FUNCTION_ARGS)
  * Other unreadable-directory cases throw an error.
  */
 static Datum
-pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, bool missing_ok)
+pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, int flags)
 {
 	ReturnSetInfo *rsinfo = (ReturnSetInfo *) fcinfo->resultinfo;
 	bool		randomAccess;
@@ -539,6 +492,32 @@ pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, bool missing_ok)
 	DIR		   *dirdesc;
 	struct dirent *de;
 	MemoryContext oldcontext;
+	TypeFuncClass	tuptype ;
+
+	/* isdir depends on metadata */
+	Assert(!(flags&LS_DIR_ISDIR) || (flags&LS_DIR_METADATA));
+	/* Unreasonable to show isdir and skip dirs */
+	Assert(!(flags&LS_DIR_ISDIR) || !(flags&LS_DIR_SKIP_DIRS));
+
+	/* check the optional arguments */
+	if (PG_NARGS() == 3)
+	{
+		if (!PG_ARGISNULL(1))
+		{
+			if (PG_GETARG_BOOL(1))
+				flags |= LS_DIR_MISSING_OK;
+			else
+				flags &= ~LS_DIR_MISSING_OK;
+		}
+
+		if (!PG_ARGISNULL(2))
+		{
+			if (PG_GETARG_BOOL(2))
+				flags &= ~LS_DIR_SKIP_DOT_DIRS;
+			else
+				flags |= LS_DIR_SKIP_DOT_DIRS;
+		}
+	}
 
 	/* check to see if caller supports us returning a tuplestore */
 	if (rsinfo == NULL || !IsA(rsinfo, ReturnSetInfo))
@@ -554,8 +533,18 @@ pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, bool missing_ok)
 	/* The tupdesc and tuplestore must be created in ecxt_per_query_memory */
 	oldcontext = MemoryContextSwitchTo(rsinfo->econtext->ecxt_per_query_memory);
 
-	if (get_call_result_type(fcinfo, NULL, &tupdesc) != TYPEFUNC_COMPOSITE)
-		elog(ERROR, "return type must be a row type");
+	tuptype = get_call_result_type(fcinfo, NULL, &tupdesc);
+	if (flags & LS_DIR_METADATA)
+	{
+		if (tuptype != TYPEFUNC_COMPOSITE)
+			elog(ERROR, "return type must be a row type");
+	} else {
+		/* pg_ls_dir returns a simple scalar */
+		if (tuptype != TYPEFUNC_SCALAR)
+			elog(ERROR, "return type must be a scalar type");
+		tupdesc = CreateTemplateTupleDesc(1);
+		TupleDescInitEntry(tupdesc, (AttrNumber) 1, "column", TEXTOID, -1, 0);
+	}
 
 	randomAccess = (rsinfo->allowedModes & SFRM_Materialize_Random) != 0;
 	tupstore = tuplestore_begin_heap(randomAccess, false, work_mem);
@@ -574,7 +563,7 @@ pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, bool missing_ok)
 	if (!dirdesc)
 	{
 		/* Return empty tuplestore if appropriate */
-		if (missing_ok && errno == ENOENT)
+		if (flags&LS_DIR_MISSING_OK && errno == ENOENT)
 		{
 			tuplestore_donestoring(tupstore);
 			return (Datum) 0;
@@ -584,13 +573,19 @@ pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, bool missing_ok)
 
 	while ((de = ReadDir(dirdesc, dir)) != NULL)
 	{
-		Datum		values[3];
-		bool		nulls[3];
+		Datum		values[4];
+		bool		nulls[4];
 		char		path[MAXPGPATH * 2];
 		struct stat attrib;
 
+		if (flags & LS_DIR_SKIP_DOT_DIRS &&
+			(strcmp(de->d_name, ".") == 0 ||
+			 strcmp(de->d_name, "..") == 0))
+			continue;
+
 		/* Skip hidden files */
-		if (de->d_name[0] == '.')
+		if (flags & LS_DIR_SKIP_HIDDEN &&
+			de->d_name[0] == '.')
 			continue;
 
 		/* Get the file info */
@@ -600,13 +595,26 @@ pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, bool missing_ok)
 					(errcode_for_file_access(),
 					 errmsg("could not stat file \"%s\": %m", path)));
 
-		/* Ignore anything but regular files */
-		if (!S_ISREG(attrib.st_mode))
-			continue;
+		if (S_ISDIR(attrib.st_mode))
+		{
+			if (flags & LS_DIR_SKIP_DIRS)
+				continue;
+		}
+		else if (!S_ISREG(attrib.st_mode))
+		{
+			if (flags & LS_DIR_SKIP_SPECIAL)
+				continue;
+		}
 
 		values[0] = CStringGetTextDatum(de->d_name);
-		values[1] = Int64GetDatum((int64) attrib.st_size);
-		values[2] = TimestampTzGetDatum(time_t_to_timestamptz(attrib.st_mtime));
+		if (flags & LS_DIR_METADATA)
+		{
+			values[1] = Int64GetDatum((int64) attrib.st_size);
+			values[2] = TimestampTzGetDatum(time_t_to_timestamptz(attrib.st_mtime));
+			if (flags & LS_DIR_ISDIR)
+				values[3] = BoolGetDatum(S_ISDIR(attrib.st_mode));
+		}
+
 		memset(nulls, 0, sizeof(nulls));
 
 		tuplestore_putvalues(tupstore, tupdesc, values, nulls);
@@ -621,14 +629,14 @@ pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, bool missing_ok)
 Datum
 pg_ls_logdir(PG_FUNCTION_ARGS)
 {
-	return pg_ls_dir_files(fcinfo, Log_directory, false);
+	return pg_ls_dir_files(fcinfo, Log_directory, LS_DIR_HISTORIC);
 }
 
 /* Function to return the list of files in the WAL directory */
 Datum
 pg_ls_waldir(PG_FUNCTION_ARGS)
 {
-	return pg_ls_dir_files(fcinfo, XLOGDIR, false);
+	return pg_ls_dir_files(fcinfo, XLOGDIR, LS_DIR_HISTORIC);
 }
 
 /*
@@ -646,7 +654,8 @@ pg_ls_tmpdir(FunctionCallInfo fcinfo, Oid tblspc)
 						tblspc)));
 
 	TempTablespacePath(path, tblspc);
-	return pg_ls_dir_files(fcinfo, path, true);
+	return pg_ls_dir_files(fcinfo, path,
+			LS_DIR_HISTORIC|LS_DIR_MISSING_OK);
 }
 
 /*
@@ -675,5 +684,18 @@ pg_ls_tmpdir_1arg(PG_FUNCTION_ARGS)
 Datum
 pg_ls_archive_statusdir(PG_FUNCTION_ARGS)
 {
-	return pg_ls_dir_files(fcinfo, XLOGDIR "/archive_status", true);
+	return pg_ls_dir_files(fcinfo, XLOGDIR "/archive_status",
+			LS_DIR_HISTORIC|LS_DIR_MISSING_OK);
+}
+
+/*
+ * Function to return the list of files and metadata in an arbitrary directory.
+ */
+Datum
+pg_ls_dir_metadata(PG_FUNCTION_ARGS)
+{
+	char	*dirname = convert_and_check_filename(PG_GETARG_TEXT_PP(0));
+
+	return pg_ls_dir_files(fcinfo, dirname,
+			LS_DIR_METADATA|LS_DIR_SKIP_SPECIAL|LS_DIR_ISDIR);
 }
diff --git a/src/include/catalog/pg_proc.dat b/src/include/catalog/pg_proc.dat
index 7fb574f9dc..0a1859f709 100644
--- a/src/include/catalog/pg_proc.dat
+++ b/src/include/catalog/pg_proc.dat
@@ -10741,6 +10741,12 @@
   proallargtypes => '{oid,text,int8,timestamptz}', proargmodes => '{i,o,o,o}',
   proargnames => '{tablespace,name,size,modification}',
   prosrc => 'pg_ls_tmpdir_1arg' },
+{ oid => '5032', descr => 'list directory with metadata',
+  proname => 'pg_ls_dir_metadata', procost => '10', prorows => '20', proretset => 't',
+  provolatile => 'v', prorettype => 'record', proargtypes => 'text bool bool bool',
+  proallargtypes => '{text,bool,bool,bool,text,int8,timestamptz,bool}', proargmodes => '{i,i,i,i,o,o,o,o}',
+  proargnames => '{dirname,missing_ok,include_dot_dirs,dir_ok,name,size,modification,isdir}',
+  prosrc => 'pg_ls_dir_metadata' },
 
 # hash partitioning constraint function
 { oid => '5028', descr => 'hash partition CHECK constraint',
-- 
2.17.0


--oTHb8nViIGeoXxdp
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
 filename="v11-0005-pg_ls_tmpdir-to-show-isdir-argument.patch"



^ permalink  raw  reply  [nested|flat] 68+ messages in thread

* [PATCH v32 03/11] Add pg_ls_dir_metadata to list a dir with file metadata..
@ 2020-03-10 03:40  Justin Pryzby <[email protected]>
  0 siblings, 0 replies; 68+ messages in thread

From: Justin Pryzby @ 2020-03-10 03:40 UTC (permalink / raw)

Generalize pg_ls_dir_files and retire pg_ls_dir

Need catversion bumped?
---
 doc/src/sgml/func.sgml                       |  21 ++
 src/backend/catalog/system_functions.sql     |   1 +
 src/backend/utils/adt/genfile.c              | 239 +++++++++++--------
 src/include/catalog/pg_proc.dat              |  12 +
 src/test/regress/expected/misc_functions.out |  24 ++
 src/test/regress/expected/tablespace.out     |   8 +
 src/test/regress/sql/misc_functions.sql      |  11 +
 src/test/regress/sql/tablespace.sql          |   5 +
 8 files changed, 225 insertions(+), 96 deletions(-)

diff --git a/doc/src/sgml/func.sgml b/doc/src/sgml/func.sgml
index b5c1befe627..7948e6ec06c 100644
--- a/doc/src/sgml/func.sgml
+++ b/doc/src/sgml/func.sgml
@@ -25945,6 +25945,27 @@ postgres=# SELECT * FROM pg_walfile_name_offset(pg_stop_backup());
        </para></entry>
       </row>
 
+      <row>
+       <entry role="func_table_entry"><para role="func_signature">
+        <indexterm>
+         <primary>pg_ls_dir_metadata</primary>
+        </indexterm>
+        <function>pg_ls_dir_metadata</function> ( <parameter>dirname</parameter> <type>text</type>
+        <optional>, <parameter>missing_ok</parameter> <type>boolean</type>,
+        <parameter>include_dot_dirs</parameter> <type>boolean</type> </optional> )
+        <returnvalue>setof record</returnvalue>
+        ( <parameter>filename</parameter> <type>text</type>,
+        <parameter>size</parameter> <type>bigint</type>,
+        <parameter>modification</parameter> <type>timestamp with time zone</type> )
+       </para>
+       <para>
+        For each file in the specified directory, list the file and its
+        metadata.
+        Restricted to superusers by default, but other users can be granted
+        EXECUTE to run the function.
+       </para></entry>
+      </row>
+
       <row>
        <entry role="func_table_entry"><para role="func_signature">
         <indexterm>
diff --git a/src/backend/catalog/system_functions.sql b/src/backend/catalog/system_functions.sql
index 3a4fa9091b1..5eb6cc8572c 100644
--- a/src/backend/catalog/system_functions.sql
+++ b/src/backend/catalog/system_functions.sql
@@ -702,6 +702,7 @@ REVOKE EXECUTE ON FUNCTION pg_stat_file(text,boolean) FROM public;
 REVOKE EXECUTE ON FUNCTION pg_ls_dir(text) FROM public;
 
 REVOKE EXECUTE ON FUNCTION pg_ls_dir(text,boolean,boolean) FROM public;
+REVOKE EXECUTE ON FUNCTION pg_ls_dir_metadata(text,boolean,boolean) FROM public;
 
 REVOKE EXECUTE ON FUNCTION pg_log_backend_memory_contexts(integer) FROM PUBLIC;
 
diff --git a/src/backend/utils/adt/genfile.c b/src/backend/utils/adt/genfile.c
index 027ed864001..0728547ec2b 100644
--- a/src/backend/utils/adt/genfile.c
+++ b/src/backend/utils/adt/genfile.c
@@ -37,6 +37,21 @@
 #include "utils/syscache.h"
 #include "utils/timestamp.h"
 
+static Datum pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, int flags);
+
+#define	LS_DIR_ISDIR				(1<<0) /* Show column: isdir */
+#define	LS_DIR_METADATA				(1<<1) /* Show columns: mtime, size */
+#define	LS_DIR_MISSING_OK			(1<<2) /* Ignore ENOENT if the toplevel dir is missing */
+#define	LS_DIR_SKIP_DOT_DIRS		(1<<3) /* Do not show . or .. */
+#define	LS_DIR_SKIP_HIDDEN			(1<<4) /* Do not show anything begining with . */
+#define	LS_DIR_SKIP_DIRS			(1<<5) /* Do not show directories */
+#define	LS_DIR_SKIP_SPECIAL			(1<<6) /* Do not show special file types */
+
+/*
+ * Shortcut for the historic behavior of the pg_ls_* functions (not including
+ * pg_ls_dir, which skips different files and doesn't show metadata).
+ */
+#define LS_DIR_HISTORIC				(LS_DIR_SKIP_DIRS | LS_DIR_SKIP_HIDDEN | LS_DIR_SKIP_SPECIAL | LS_DIR_METADATA)
 
 /*
  * Convert a "text" filename argument to C string, and check it's allowable.
@@ -452,6 +467,11 @@ pg_stat_file(PG_FUNCTION_ARGS)
 	values[4] = TimestampTzGetDatum(time_t_to_timestamptz(fst.st_ctime));
 #endif
 	values[5] = BoolGetDatum(S_ISDIR(fst.st_mode));
+#ifdef WIN32
+	/* Links should have isdir=false */
+	if (pgwin32_is_junction(filename))
+		values[5] = BoolGetDatum(false);
+#endif
 
 	tuple = heap_form_tuple(tupdesc, values, isnull);
 
@@ -479,79 +499,9 @@ pg_stat_file_1arg(PG_FUNCTION_ARGS)
 Datum
 pg_ls_dir(PG_FUNCTION_ARGS)
 {
-	ReturnSetInfo *rsinfo = (ReturnSetInfo *) fcinfo->resultinfo;
-	char	   *location;
-	bool		missing_ok = false;
-	bool		include_dot_dirs = false;
-	bool		randomAccess;
-	TupleDesc	tupdesc;
-	Tuplestorestate *tupstore;
-	DIR		   *dirdesc;
-	struct dirent *de;
-	MemoryContext oldcontext;
-
-	location = convert_and_check_filename(PG_GETARG_TEXT_PP(0));
-
-	/* check the optional arguments */
-	if (PG_NARGS() == 3)
-	{
-		if (!PG_ARGISNULL(1))
-			missing_ok = PG_GETARG_BOOL(1);
-		if (!PG_ARGISNULL(2))
-			include_dot_dirs = PG_GETARG_BOOL(2);
-	}
-
-	/* check to see if caller supports us returning a tuplestore */
-	if (rsinfo == NULL || !IsA(rsinfo, ReturnSetInfo))
-		ereport(ERROR,
-				(errcode(ERRCODE_FEATURE_NOT_SUPPORTED),
-				 errmsg("set-valued function called in context that cannot accept a set")));
-	if (!(rsinfo->allowedModes & SFRM_Materialize))
-		ereport(ERROR,
-				(errcode(ERRCODE_SYNTAX_ERROR),
-				 errmsg("materialize mode required, but it is not allowed in this context")));
-
-	/* The tupdesc and tuplestore must be created in ecxt_per_query_memory */
-	oldcontext = MemoryContextSwitchTo(rsinfo->econtext->ecxt_per_query_memory);
-
-	tupdesc = CreateTemplateTupleDesc(1);
-	TupleDescInitEntry(tupdesc, (AttrNumber) 1, "pg_ls_dir", TEXTOID, -1, 0);
-
-	randomAccess = (rsinfo->allowedModes & SFRM_Materialize_Random) != 0;
-	tupstore = tuplestore_begin_heap(randomAccess, false, work_mem);
-	rsinfo->returnMode = SFRM_Materialize;
-	rsinfo->setResult = tupstore;
-	rsinfo->setDesc = tupdesc;
-
-	MemoryContextSwitchTo(oldcontext);
-
-	dirdesc = AllocateDir(location);
-	if (!dirdesc)
-	{
-		/* Return empty tuplestore if appropriate */
-		if (missing_ok && errno == ENOENT)
-			return (Datum) 0;
-		/* Otherwise, we can let ReadDir() throw the error */
-	}
-
-	while ((de = ReadDir(dirdesc, location)) != NULL)
-	{
-		Datum		values[1];
-		bool		nulls[1];
-
-		if (!include_dot_dirs &&
-			(strcmp(de->d_name, ".") == 0 ||
-			 strcmp(de->d_name, "..") == 0))
-			continue;
-
-		values[0] = CStringGetTextDatum(de->d_name);
-		nulls[0] = false;
-
-		tuplestore_putvalues(tupstore, tupdesc, values, nulls);
-	}
-
-	FreeDir(dirdesc);
-	return (Datum) 0;
+	text	*filename_t = PG_GETARG_TEXT_PP(0);
+	char	*filename = convert_and_check_filename(filename_t);
+	return pg_ls_dir_files(fcinfo, filename, LS_DIR_SKIP_DOT_DIRS);
 }
 
 /*
@@ -564,17 +514,19 @@ pg_ls_dir(PG_FUNCTION_ARGS)
 Datum
 pg_ls_dir_1arg(PG_FUNCTION_ARGS)
 {
-	return pg_ls_dir(fcinfo);
+	text	*filename_t = PG_GETARG_TEXT_PP(0);
+	char	*filename = convert_and_check_filename(filename_t);
+	return pg_ls_dir_files(fcinfo, filename, LS_DIR_SKIP_DOT_DIRS);
 }
 
 /*
- * Generic function to return a directory listing of files.
+ * Generic function to return a directory listing of files (and optionally dirs).
  *
- * If the directory isn't there, silently return an empty set if missing_ok.
+ * If the directory isn't there, silently return an empty set if MISSING_OK.
  * Other unreadable-directory cases throw an error.
  */
 static Datum
-pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, bool missing_ok)
+pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, int flags)
 {
 	ReturnSetInfo *rsinfo = (ReturnSetInfo *) fcinfo->resultinfo;
 	bool		randomAccess;
@@ -583,6 +535,32 @@ pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, bool missing_ok)
 	DIR		   *dirdesc;
 	struct dirent *de;
 	MemoryContext oldcontext;
+	TypeFuncClass	tuptype ;
+
+	/* isdir depends on metadata */
+	Assert(!(flags&LS_DIR_ISDIR) || (flags&LS_DIR_METADATA));
+	/* Unreasonable to show isdir and skip dirs */
+	Assert(!(flags&LS_DIR_ISDIR) || !(flags&LS_DIR_SKIP_DIRS));
+
+	/* check the optional arguments */
+	if (PG_NARGS() == 3)
+	{
+		if (!PG_ARGISNULL(1))
+		{
+			if (PG_GETARG_BOOL(1))
+				flags |= LS_DIR_MISSING_OK;
+			else
+				flags &= ~LS_DIR_MISSING_OK;
+		}
+
+		if (!PG_ARGISNULL(2))
+		{
+			if (PG_GETARG_BOOL(2))
+				flags &= ~LS_DIR_SKIP_DOT_DIRS;
+			else
+				flags |= LS_DIR_SKIP_DOT_DIRS;
+		}
+	}
 
 	/* check to see if caller supports us returning a tuplestore */
 	if (rsinfo == NULL || !IsA(rsinfo, ReturnSetInfo))
@@ -597,8 +575,20 @@ pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, bool missing_ok)
 	/* The tupdesc and tuplestore must be created in ecxt_per_query_memory */
 	oldcontext = MemoryContextSwitchTo(rsinfo->econtext->ecxt_per_query_memory);
 
-	if (get_call_result_type(fcinfo, NULL, &tupdesc) != TYPEFUNC_COMPOSITE)
-		elog(ERROR, "return type must be a row type");
+	tuptype = get_call_result_type(fcinfo, NULL, &tupdesc);
+	if (flags & LS_DIR_METADATA)
+	{
+		if (tuptype != TYPEFUNC_COMPOSITE)
+			elog(ERROR, "return type must be a row type");
+	}
+	else
+	{
+		/* pg_ls_dir returns a simple scalar */
+		if (tuptype != TYPEFUNC_SCALAR)
+			elog(ERROR, "return type must be a scalar type");
+		tupdesc = CreateTemplateTupleDesc(1);
+		TupleDescInitEntry(tupdesc, (AttrNumber) 1, "column", TEXTOID, -1, 0);
+	}
 
 	randomAccess = (rsinfo->allowedModes & SFRM_Materialize_Random) != 0;
 	tupstore = tuplestore_begin_heap(randomAccess, false, work_mem);
@@ -617,20 +607,27 @@ pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, bool missing_ok)
 	if (!dirdesc)
 	{
 		/* Return empty tuplestore if appropriate */
-		if (missing_ok && errno == ENOENT)
+		if (flags & LS_DIR_MISSING_OK && errno == ENOENT)
 			return (Datum) 0;
 		/* Otherwise, we can let ReadDir() throw the error */
 	}
 
 	while ((de = ReadDir(dirdesc, dir)) != NULL)
 	{
-		Datum		values[3];
-		bool		nulls[3];
+		Datum		values[4];
+		bool		nulls[4];
 		char		path[MAXPGPATH * 2];
 		struct stat attrib;
 
-		/* Skip hidden files */
-		if (de->d_name[0] == '.')
+		/* Skip dot dirs? */
+		if (flags & LS_DIR_SKIP_DOT_DIRS &&
+			(strcmp(de->d_name, ".") == 0 ||
+			 strcmp(de->d_name, "..") == 0))
+			continue;
+
+		/* Skip hidden files? */
+		if (flags & LS_DIR_SKIP_HIDDEN &&
+			de->d_name[0] == '.')
 			continue;
 
 		/* Get the file info */
@@ -645,13 +642,34 @@ pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, bool missing_ok)
 					 errmsg("could not stat file \"%s\": %m", path)));
 		}
 
-		/* Ignore anything but regular files */
-		if (!S_ISREG(attrib.st_mode))
-			continue;
+		/* Skip dirs or special files? */
+		if (S_ISDIR(attrib.st_mode))
+		{
+			if (flags & LS_DIR_SKIP_DIRS)
+				continue;
+		}
+		else if (!S_ISREG(attrib.st_mode))
+		{
+			if (flags & LS_DIR_SKIP_SPECIAL)
+				continue;
+		}
 
 		values[0] = CStringGetTextDatum(de->d_name);
-		values[1] = Int64GetDatum((int64) attrib.st_size);
-		values[2] = TimestampTzGetDatum(time_t_to_timestamptz(attrib.st_mtime));
+		if (flags & LS_DIR_METADATA)
+		{
+			values[1] = Int64GetDatum((int64) attrib.st_size);
+			values[2] = TimestampTzGetDatum(time_t_to_timestamptz(attrib.st_mtime));
+			if (flags & LS_DIR_ISDIR)
+			{
+				values[3] = BoolGetDatum(S_ISDIR(attrib.st_mode));
+#ifdef WIN32
+				/* Links should have isdir=false */
+				if (pgwin32_is_junction(path))
+					values[3] = BoolGetDatum(false);
+#endif
+			}
+		}
+
 		memset(nulls, 0, sizeof(nulls));
 
 		tuplestore_putvalues(tupstore, tupdesc, values, nulls);
@@ -665,14 +683,14 @@ pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, bool missing_ok)
 Datum
 pg_ls_logdir(PG_FUNCTION_ARGS)
 {
-	return pg_ls_dir_files(fcinfo, Log_directory, false);
+	return pg_ls_dir_files(fcinfo, Log_directory, LS_DIR_HISTORIC);
 }
 
 /* Function to return the list of files in the WAL directory */
 Datum
 pg_ls_waldir(PG_FUNCTION_ARGS)
 {
-	return pg_ls_dir_files(fcinfo, XLOGDIR, false);
+	return pg_ls_dir_files(fcinfo, XLOGDIR, LS_DIR_HISTORIC);
 }
 
 /*
@@ -690,7 +708,8 @@ pg_ls_tmpdir(FunctionCallInfo fcinfo, Oid tblspc)
 						tblspc)));
 
 	TempTablespacePath(path, tblspc);
-	return pg_ls_dir_files(fcinfo, path, true);
+	return pg_ls_dir_files(fcinfo, path,
+			LS_DIR_HISTORIC | LS_DIR_MISSING_OK);
 }
 
 /*
@@ -719,7 +738,35 @@ pg_ls_tmpdir_1arg(PG_FUNCTION_ARGS)
 Datum
 pg_ls_archive_statusdir(PG_FUNCTION_ARGS)
 {
-	return pg_ls_dir_files(fcinfo, XLOGDIR "/archive_status", true);
+	return pg_ls_dir_files(fcinfo, XLOGDIR "/archive_status",
+			LS_DIR_HISTORIC | LS_DIR_MISSING_OK);
+}
+
+/*
+ * Return the list of files and metadata in an arbitrary directory.
+ */
+Datum
+pg_ls_dir_metadata(PG_FUNCTION_ARGS)
+{
+	char	*dirname = convert_and_check_filename(PG_GETARG_TEXT_PP(0));
+
+	return pg_ls_dir_files(fcinfo, dirname,
+			LS_DIR_METADATA | LS_DIR_SKIP_SPECIAL | LS_DIR_ISDIR);
+}
+
+/*
+ * Return the list of files and metadata in an arbitrary directory.
+ * note: this wrapper is necessary to pass the sanity check in opr_sanity,
+ * which checks that all built-in functions that share the implementing C
+ * function take the same number of arguments.
+ */
+Datum
+pg_ls_dir_metadata_1arg(PG_FUNCTION_ARGS)
+{
+	char	*dirname = convert_and_check_filename(PG_GETARG_TEXT_PP(0));
+
+	return pg_ls_dir_files(fcinfo, dirname,
+			LS_DIR_METADATA | LS_DIR_SKIP_SPECIAL | LS_DIR_ISDIR);
 }
 
 /*
@@ -728,7 +775,7 @@ pg_ls_archive_statusdir(PG_FUNCTION_ARGS)
 Datum
 pg_ls_logicalsnapdir(PG_FUNCTION_ARGS)
 {
-	return pg_ls_dir_files(fcinfo, "pg_logical/snapshots", false);
+	return pg_ls_dir_files(fcinfo, "pg_logical/snapshots", LS_DIR_HISTORIC);
 }
 
 /*
@@ -737,7 +784,7 @@ pg_ls_logicalsnapdir(PG_FUNCTION_ARGS)
 Datum
 pg_ls_logicalmapdir(PG_FUNCTION_ARGS)
 {
-	return pg_ls_dir_files(fcinfo, "pg_logical/mappings", false);
+	return pg_ls_dir_files(fcinfo, "pg_logical/mappings", LS_DIR_HISTORIC);
 }
 
 /*
@@ -762,5 +809,5 @@ pg_ls_replslotdir(PG_FUNCTION_ARGS)
 						slotname)));
 
 	snprintf(path, sizeof(path), "pg_replslot/%s", slotname);
-	return pg_ls_dir_files(fcinfo, path, false);
+	return pg_ls_dir_files(fcinfo, path, LS_DIR_HISTORIC);
 }
diff --git a/src/include/catalog/pg_proc.dat b/src/include/catalog/pg_proc.dat
index 4d992dc2241..c99e1633bbf 100644
--- a/src/include/catalog/pg_proc.dat
+++ b/src/include/catalog/pg_proc.dat
@@ -11645,6 +11645,18 @@
   proargmodes => '{i,o,o,o}',
   proargnames => '{slot_name,name,size,modification}',
   prosrc => 'pg_ls_replslotdir' },
+{ oid => '8450', descr => 'list directory with metadata',
+  proname => 'pg_ls_dir_metadata', procost => '10', prorows => '20', proretset => 't',
+  provolatile => 'v', prorettype => 'record', proargtypes => 'text bool bool',
+  proallargtypes => '{text,bool,bool,text,int8,timestamptz,bool}', proargmodes => '{i,i,i,o,o,o,o}',
+  proargnames => '{dirname,missing_ok,include_dot_dirs,filename,size,modification,isdir}',
+  prosrc => 'pg_ls_dir_metadata' },
+{ oid => '8451', descr => 'list directory with metadata',
+  proname => 'pg_ls_dir_metadata', procost => '10', prorows => '20', proretset => 't',
+  provolatile => 'v', prorettype => 'record', proargtypes => 'text',
+  proallargtypes => '{text,text,int8,timestamptz,bool}', proargmodes => '{i,o,o,o,o}',
+  proargnames => '{dirname,filename,size,modification,isdir}',
+  prosrc => 'pg_ls_dir_metadata_1arg' },
 
 # hash partitioning constraint function
 { oid => '5028', descr => 'hash partition CHECK constraint',
diff --git a/src/test/regress/expected/misc_functions.out b/src/test/regress/expected/misc_functions.out
index 830de507e7c..b3a9d11b5c0 100644
--- a/src/test/regress/expected/misc_functions.out
+++ b/src/test/regress/expected/misc_functions.out
@@ -267,6 +267,30 @@ select * from pg_stat_file('.') limit 0;
 ------+--------+--------------+--------+----------+-------
 (0 rows)
 
+-- This tests the missing_ok parameter, which causes pg_ls_tmpdir to succeed even if the tmpdir doesn't exist yet
+-- The name='' condition is never true, so the function runs to completion but returns zero rows.
+select * from pg_ls_tmpdir() where name='Does not exist';
+ name | size | modification 
+------+------+--------------
+(0 rows)
+
+select filename, isdir from pg_ls_dir_metadata('.') where filename='.';
+ filename | isdir 
+----------+-------
+ .        | t
+(1 row)
+
+select filename, isdir from pg_ls_dir_metadata('.', false, false) where filename='.'; -- include_dot_dirs=false
+ filename | isdir 
+----------+-------
+(0 rows)
+
+-- Check that expected columns are present
+select * from pg_ls_dir_metadata('.') limit 0;
+ filename | size | modification | isdir 
+----------+------+--------------+-------
+(0 rows)
+
 --
 -- Test replication slot directory functions
 --
diff --git a/src/test/regress/expected/tablespace.out b/src/test/regress/expected/tablespace.out
index 864f4b6e208..21f7f19b67f 100644
--- a/src/test/regress/expected/tablespace.out
+++ b/src/test/regress/expected/tablespace.out
@@ -16,6 +16,14 @@ SELECT spcoptions FROM pg_tablespace WHERE spcname = 'regress_tblspacewith';
 DROP TABLESPACE regress_tblspacewith;
 -- create a tablespace we can use
 CREATE TABLESPACE regress_tblspace LOCATION :'testtablespace';
+-- This tests the missing_ok parameter, which causes pg_ls_tmpdir to succeed even if the tmpdir doesn't exist yet
+-- The name='' condition is never true, so the function runs to completion but returns zero rows.
+-- The query is written to ERROR if the tablespace doesn't exist, rather than silently failing to call pg_ls_tmpdir()
+SELECT c.* FROM (SELECT oid FROM pg_tablespace b WHERE b.spcname='regress_tblspace' UNION SELECT 0 ORDER BY 1 DESC LIMIT 1) AS b , pg_ls_tmpdir(oid) AS c WHERE c.name='Does not exist';
+ name | size | modification 
+------+------+--------------
+(0 rows)
+
 -- try setting and resetting some properties for the new tablespace
 ALTER TABLESPACE regress_tblspace SET (random_page_cost = 1.0, seq_page_cost = 1.1);
 ALTER TABLESPACE regress_tblspace SET (some_nonexistent_parameter = true);  -- fail
diff --git a/src/test/regress/sql/misc_functions.sql b/src/test/regress/sql/misc_functions.sql
index 422a1369aeb..ca946d08bde 100644
--- a/src/test/regress/sql/misc_functions.sql
+++ b/src/test/regress/sql/misc_functions.sql
@@ -99,6 +99,17 @@ select pg_ls_dir('does not exist'); -- fails with missingok=false
 -- Check that expected columns are present
 select * from pg_stat_file('.') limit 0;
 
+-- This tests the missing_ok parameter, which causes pg_ls_tmpdir to succeed even if the tmpdir doesn't exist yet
+-- The name='' condition is never true, so the function runs to completion but returns zero rows.
+select * from pg_ls_tmpdir() where name='Does not exist';
+
+select filename, isdir from pg_ls_dir_metadata('.') where filename='.';
+
+select filename, isdir from pg_ls_dir_metadata('.', false, false) where filename='.'; -- include_dot_dirs=false
+
+-- Check that expected columns are present
+select * from pg_ls_dir_metadata('.') limit 0;
+
 --
 -- Test replication slot directory functions
 --
diff --git a/src/test/regress/sql/tablespace.sql b/src/test/regress/sql/tablespace.sql
index 92076db9a13..291ca029c14 100644
--- a/src/test/regress/sql/tablespace.sql
+++ b/src/test/regress/sql/tablespace.sql
@@ -16,6 +16,11 @@ DROP TABLESPACE regress_tblspacewith;
 -- create a tablespace we can use
 CREATE TABLESPACE regress_tblspace LOCATION :'testtablespace';
 
+-- This tests the missing_ok parameter, which causes pg_ls_tmpdir to succeed even if the tmpdir doesn't exist yet
+-- The name='' condition is never true, so the function runs to completion but returns zero rows.
+-- The query is written to ERROR if the tablespace doesn't exist, rather than silently failing to call pg_ls_tmpdir()
+SELECT c.* FROM (SELECT oid FROM pg_tablespace b WHERE b.spcname='regress_tblspace' UNION SELECT 0 ORDER BY 1 DESC LIMIT 1) AS b , pg_ls_tmpdir(oid) AS c WHERE c.name='Does not exist';
+
 -- try setting and resetting some properties for the new tablespace
 ALTER TABLESPACE regress_tblspace SET (random_page_cost = 1.0, seq_page_cost = 1.1);
 ALTER TABLESPACE regress_tblspace SET (some_nonexistent_parameter = true);  -- fail
-- 
2.17.0


--Bne5rrxQd65beI7a
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
 filename="v32-0004-pg_ls_tmpdir-to-show-directories-and-isdir-argum.patch"



^ permalink  raw  reply  [nested|flat] 68+ messages in thread

* [PATCH v31 03/11] Add pg_ls_dir_metadata to list a dir with file metadata..
@ 2020-03-10 03:40  Justin Pryzby <[email protected]>
  0 siblings, 0 replies; 68+ messages in thread

From: Justin Pryzby @ 2020-03-10 03:40 UTC (permalink / raw)

Generalize pg_ls_dir_files and retire pg_ls_dir

Need catversion bumped?
---
 doc/src/sgml/func.sgml                       |  21 ++
 src/backend/catalog/system_functions.sql     |   1 +
 src/backend/utils/adt/genfile.c              | 239 +++++++++++--------
 src/include/catalog/pg_proc.dat              |  12 +
 src/test/regress/expected/misc_functions.out |  24 ++
 src/test/regress/input/tablespace.source     |   5 +
 src/test/regress/output/tablespace.source    |   8 +
 src/test/regress/sql/misc_functions.sql      |  11 +
 8 files changed, 225 insertions(+), 96 deletions(-)

diff --git a/doc/src/sgml/func.sgml b/doc/src/sgml/func.sgml
index d36479d86d..e0099e77df 100644
--- a/doc/src/sgml/func.sgml
+++ b/doc/src/sgml/func.sgml
@@ -25860,6 +25860,27 @@ postgres=# SELECT * FROM pg_walfile_name_offset(pg_stop_backup());
        </para></entry>
       </row>
 
+      <row>
+       <entry role="func_table_entry"><para role="func_signature">
+        <indexterm>
+         <primary>pg_ls_dir_metadata</primary>
+        </indexterm>
+        <function>pg_ls_dir_metadata</function> ( <parameter>dirname</parameter> <type>text</type>
+        <optional>, <parameter>missing_ok</parameter> <type>boolean</type>,
+        <parameter>include_dot_dirs</parameter> <type>boolean</type> </optional> )
+        <returnvalue>setof record</returnvalue>
+        ( <parameter>filename</parameter> <type>text</type>,
+        <parameter>size</parameter> <type>bigint</type>,
+        <parameter>modification</parameter> <type>timestamp with time zone</type> )
+       </para>
+       <para>
+        For each file in the specified directory, list the file and its
+        metadata.
+        Restricted to superusers by default, but other users can be granted
+        EXECUTE to run the function.
+       </para></entry>
+      </row>
+
       <row>
        <entry role="func_table_entry"><para role="func_signature">
         <indexterm>
diff --git a/src/backend/catalog/system_functions.sql b/src/backend/catalog/system_functions.sql
index f6789025a5..a45bac89ff 100644
--- a/src/backend/catalog/system_functions.sql
+++ b/src/backend/catalog/system_functions.sql
@@ -698,6 +698,7 @@ REVOKE EXECUTE ON FUNCTION pg_stat_file(text,boolean) FROM public;
 REVOKE EXECUTE ON FUNCTION pg_ls_dir(text) FROM public;
 
 REVOKE EXECUTE ON FUNCTION pg_ls_dir(text,boolean,boolean) FROM public;
+REVOKE EXECUTE ON FUNCTION pg_ls_dir_metadata(text,boolean,boolean) FROM public;
 
 REVOKE EXECUTE ON FUNCTION pg_log_backend_memory_contexts(integer) FROM PUBLIC;
 
diff --git a/src/backend/utils/adt/genfile.c b/src/backend/utils/adt/genfile.c
index 027ed86400..0728547ec2 100644
--- a/src/backend/utils/adt/genfile.c
+++ b/src/backend/utils/adt/genfile.c
@@ -37,6 +37,21 @@
 #include "utils/syscache.h"
 #include "utils/timestamp.h"
 
+static Datum pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, int flags);
+
+#define	LS_DIR_ISDIR				(1<<0) /* Show column: isdir */
+#define	LS_DIR_METADATA				(1<<1) /* Show columns: mtime, size */
+#define	LS_DIR_MISSING_OK			(1<<2) /* Ignore ENOENT if the toplevel dir is missing */
+#define	LS_DIR_SKIP_DOT_DIRS		(1<<3) /* Do not show . or .. */
+#define	LS_DIR_SKIP_HIDDEN			(1<<4) /* Do not show anything begining with . */
+#define	LS_DIR_SKIP_DIRS			(1<<5) /* Do not show directories */
+#define	LS_DIR_SKIP_SPECIAL			(1<<6) /* Do not show special file types */
+
+/*
+ * Shortcut for the historic behavior of the pg_ls_* functions (not including
+ * pg_ls_dir, which skips different files and doesn't show metadata).
+ */
+#define LS_DIR_HISTORIC				(LS_DIR_SKIP_DIRS | LS_DIR_SKIP_HIDDEN | LS_DIR_SKIP_SPECIAL | LS_DIR_METADATA)
 
 /*
  * Convert a "text" filename argument to C string, and check it's allowable.
@@ -452,6 +467,11 @@ pg_stat_file(PG_FUNCTION_ARGS)
 	values[4] = TimestampTzGetDatum(time_t_to_timestamptz(fst.st_ctime));
 #endif
 	values[5] = BoolGetDatum(S_ISDIR(fst.st_mode));
+#ifdef WIN32
+	/* Links should have isdir=false */
+	if (pgwin32_is_junction(filename))
+		values[5] = BoolGetDatum(false);
+#endif
 
 	tuple = heap_form_tuple(tupdesc, values, isnull);
 
@@ -479,79 +499,9 @@ pg_stat_file_1arg(PG_FUNCTION_ARGS)
 Datum
 pg_ls_dir(PG_FUNCTION_ARGS)
 {
-	ReturnSetInfo *rsinfo = (ReturnSetInfo *) fcinfo->resultinfo;
-	char	   *location;
-	bool		missing_ok = false;
-	bool		include_dot_dirs = false;
-	bool		randomAccess;
-	TupleDesc	tupdesc;
-	Tuplestorestate *tupstore;
-	DIR		   *dirdesc;
-	struct dirent *de;
-	MemoryContext oldcontext;
-
-	location = convert_and_check_filename(PG_GETARG_TEXT_PP(0));
-
-	/* check the optional arguments */
-	if (PG_NARGS() == 3)
-	{
-		if (!PG_ARGISNULL(1))
-			missing_ok = PG_GETARG_BOOL(1);
-		if (!PG_ARGISNULL(2))
-			include_dot_dirs = PG_GETARG_BOOL(2);
-	}
-
-	/* check to see if caller supports us returning a tuplestore */
-	if (rsinfo == NULL || !IsA(rsinfo, ReturnSetInfo))
-		ereport(ERROR,
-				(errcode(ERRCODE_FEATURE_NOT_SUPPORTED),
-				 errmsg("set-valued function called in context that cannot accept a set")));
-	if (!(rsinfo->allowedModes & SFRM_Materialize))
-		ereport(ERROR,
-				(errcode(ERRCODE_SYNTAX_ERROR),
-				 errmsg("materialize mode required, but it is not allowed in this context")));
-
-	/* The tupdesc and tuplestore must be created in ecxt_per_query_memory */
-	oldcontext = MemoryContextSwitchTo(rsinfo->econtext->ecxt_per_query_memory);
-
-	tupdesc = CreateTemplateTupleDesc(1);
-	TupleDescInitEntry(tupdesc, (AttrNumber) 1, "pg_ls_dir", TEXTOID, -1, 0);
-
-	randomAccess = (rsinfo->allowedModes & SFRM_Materialize_Random) != 0;
-	tupstore = tuplestore_begin_heap(randomAccess, false, work_mem);
-	rsinfo->returnMode = SFRM_Materialize;
-	rsinfo->setResult = tupstore;
-	rsinfo->setDesc = tupdesc;
-
-	MemoryContextSwitchTo(oldcontext);
-
-	dirdesc = AllocateDir(location);
-	if (!dirdesc)
-	{
-		/* Return empty tuplestore if appropriate */
-		if (missing_ok && errno == ENOENT)
-			return (Datum) 0;
-		/* Otherwise, we can let ReadDir() throw the error */
-	}
-
-	while ((de = ReadDir(dirdesc, location)) != NULL)
-	{
-		Datum		values[1];
-		bool		nulls[1];
-
-		if (!include_dot_dirs &&
-			(strcmp(de->d_name, ".") == 0 ||
-			 strcmp(de->d_name, "..") == 0))
-			continue;
-
-		values[0] = CStringGetTextDatum(de->d_name);
-		nulls[0] = false;
-
-		tuplestore_putvalues(tupstore, tupdesc, values, nulls);
-	}
-
-	FreeDir(dirdesc);
-	return (Datum) 0;
+	text	*filename_t = PG_GETARG_TEXT_PP(0);
+	char	*filename = convert_and_check_filename(filename_t);
+	return pg_ls_dir_files(fcinfo, filename, LS_DIR_SKIP_DOT_DIRS);
 }
 
 /*
@@ -564,17 +514,19 @@ pg_ls_dir(PG_FUNCTION_ARGS)
 Datum
 pg_ls_dir_1arg(PG_FUNCTION_ARGS)
 {
-	return pg_ls_dir(fcinfo);
+	text	*filename_t = PG_GETARG_TEXT_PP(0);
+	char	*filename = convert_and_check_filename(filename_t);
+	return pg_ls_dir_files(fcinfo, filename, LS_DIR_SKIP_DOT_DIRS);
 }
 
 /*
- * Generic function to return a directory listing of files.
+ * Generic function to return a directory listing of files (and optionally dirs).
  *
- * If the directory isn't there, silently return an empty set if missing_ok.
+ * If the directory isn't there, silently return an empty set if MISSING_OK.
  * Other unreadable-directory cases throw an error.
  */
 static Datum
-pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, bool missing_ok)
+pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, int flags)
 {
 	ReturnSetInfo *rsinfo = (ReturnSetInfo *) fcinfo->resultinfo;
 	bool		randomAccess;
@@ -583,6 +535,32 @@ pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, bool missing_ok)
 	DIR		   *dirdesc;
 	struct dirent *de;
 	MemoryContext oldcontext;
+	TypeFuncClass	tuptype ;
+
+	/* isdir depends on metadata */
+	Assert(!(flags&LS_DIR_ISDIR) || (flags&LS_DIR_METADATA));
+	/* Unreasonable to show isdir and skip dirs */
+	Assert(!(flags&LS_DIR_ISDIR) || !(flags&LS_DIR_SKIP_DIRS));
+
+	/* check the optional arguments */
+	if (PG_NARGS() == 3)
+	{
+		if (!PG_ARGISNULL(1))
+		{
+			if (PG_GETARG_BOOL(1))
+				flags |= LS_DIR_MISSING_OK;
+			else
+				flags &= ~LS_DIR_MISSING_OK;
+		}
+
+		if (!PG_ARGISNULL(2))
+		{
+			if (PG_GETARG_BOOL(2))
+				flags &= ~LS_DIR_SKIP_DOT_DIRS;
+			else
+				flags |= LS_DIR_SKIP_DOT_DIRS;
+		}
+	}
 
 	/* check to see if caller supports us returning a tuplestore */
 	if (rsinfo == NULL || !IsA(rsinfo, ReturnSetInfo))
@@ -597,8 +575,20 @@ pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, bool missing_ok)
 	/* The tupdesc and tuplestore must be created in ecxt_per_query_memory */
 	oldcontext = MemoryContextSwitchTo(rsinfo->econtext->ecxt_per_query_memory);
 
-	if (get_call_result_type(fcinfo, NULL, &tupdesc) != TYPEFUNC_COMPOSITE)
-		elog(ERROR, "return type must be a row type");
+	tuptype = get_call_result_type(fcinfo, NULL, &tupdesc);
+	if (flags & LS_DIR_METADATA)
+	{
+		if (tuptype != TYPEFUNC_COMPOSITE)
+			elog(ERROR, "return type must be a row type");
+	}
+	else
+	{
+		/* pg_ls_dir returns a simple scalar */
+		if (tuptype != TYPEFUNC_SCALAR)
+			elog(ERROR, "return type must be a scalar type");
+		tupdesc = CreateTemplateTupleDesc(1);
+		TupleDescInitEntry(tupdesc, (AttrNumber) 1, "column", TEXTOID, -1, 0);
+	}
 
 	randomAccess = (rsinfo->allowedModes & SFRM_Materialize_Random) != 0;
 	tupstore = tuplestore_begin_heap(randomAccess, false, work_mem);
@@ -617,20 +607,27 @@ pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, bool missing_ok)
 	if (!dirdesc)
 	{
 		/* Return empty tuplestore if appropriate */
-		if (missing_ok && errno == ENOENT)
+		if (flags & LS_DIR_MISSING_OK && errno == ENOENT)
 			return (Datum) 0;
 		/* Otherwise, we can let ReadDir() throw the error */
 	}
 
 	while ((de = ReadDir(dirdesc, dir)) != NULL)
 	{
-		Datum		values[3];
-		bool		nulls[3];
+		Datum		values[4];
+		bool		nulls[4];
 		char		path[MAXPGPATH * 2];
 		struct stat attrib;
 
-		/* Skip hidden files */
-		if (de->d_name[0] == '.')
+		/* Skip dot dirs? */
+		if (flags & LS_DIR_SKIP_DOT_DIRS &&
+			(strcmp(de->d_name, ".") == 0 ||
+			 strcmp(de->d_name, "..") == 0))
+			continue;
+
+		/* Skip hidden files? */
+		if (flags & LS_DIR_SKIP_HIDDEN &&
+			de->d_name[0] == '.')
 			continue;
 
 		/* Get the file info */
@@ -645,13 +642,34 @@ pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, bool missing_ok)
 					 errmsg("could not stat file \"%s\": %m", path)));
 		}
 
-		/* Ignore anything but regular files */
-		if (!S_ISREG(attrib.st_mode))
-			continue;
+		/* Skip dirs or special files? */
+		if (S_ISDIR(attrib.st_mode))
+		{
+			if (flags & LS_DIR_SKIP_DIRS)
+				continue;
+		}
+		else if (!S_ISREG(attrib.st_mode))
+		{
+			if (flags & LS_DIR_SKIP_SPECIAL)
+				continue;
+		}
 
 		values[0] = CStringGetTextDatum(de->d_name);
-		values[1] = Int64GetDatum((int64) attrib.st_size);
-		values[2] = TimestampTzGetDatum(time_t_to_timestamptz(attrib.st_mtime));
+		if (flags & LS_DIR_METADATA)
+		{
+			values[1] = Int64GetDatum((int64) attrib.st_size);
+			values[2] = TimestampTzGetDatum(time_t_to_timestamptz(attrib.st_mtime));
+			if (flags & LS_DIR_ISDIR)
+			{
+				values[3] = BoolGetDatum(S_ISDIR(attrib.st_mode));
+#ifdef WIN32
+				/* Links should have isdir=false */
+				if (pgwin32_is_junction(path))
+					values[3] = BoolGetDatum(false);
+#endif
+			}
+		}
+
 		memset(nulls, 0, sizeof(nulls));
 
 		tuplestore_putvalues(tupstore, tupdesc, values, nulls);
@@ -665,14 +683,14 @@ pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, bool missing_ok)
 Datum
 pg_ls_logdir(PG_FUNCTION_ARGS)
 {
-	return pg_ls_dir_files(fcinfo, Log_directory, false);
+	return pg_ls_dir_files(fcinfo, Log_directory, LS_DIR_HISTORIC);
 }
 
 /* Function to return the list of files in the WAL directory */
 Datum
 pg_ls_waldir(PG_FUNCTION_ARGS)
 {
-	return pg_ls_dir_files(fcinfo, XLOGDIR, false);
+	return pg_ls_dir_files(fcinfo, XLOGDIR, LS_DIR_HISTORIC);
 }
 
 /*
@@ -690,7 +708,8 @@ pg_ls_tmpdir(FunctionCallInfo fcinfo, Oid tblspc)
 						tblspc)));
 
 	TempTablespacePath(path, tblspc);
-	return pg_ls_dir_files(fcinfo, path, true);
+	return pg_ls_dir_files(fcinfo, path,
+			LS_DIR_HISTORIC | LS_DIR_MISSING_OK);
 }
 
 /*
@@ -719,7 +738,35 @@ pg_ls_tmpdir_1arg(PG_FUNCTION_ARGS)
 Datum
 pg_ls_archive_statusdir(PG_FUNCTION_ARGS)
 {
-	return pg_ls_dir_files(fcinfo, XLOGDIR "/archive_status", true);
+	return pg_ls_dir_files(fcinfo, XLOGDIR "/archive_status",
+			LS_DIR_HISTORIC | LS_DIR_MISSING_OK);
+}
+
+/*
+ * Return the list of files and metadata in an arbitrary directory.
+ */
+Datum
+pg_ls_dir_metadata(PG_FUNCTION_ARGS)
+{
+	char	*dirname = convert_and_check_filename(PG_GETARG_TEXT_PP(0));
+
+	return pg_ls_dir_files(fcinfo, dirname,
+			LS_DIR_METADATA | LS_DIR_SKIP_SPECIAL | LS_DIR_ISDIR);
+}
+
+/*
+ * Return the list of files and metadata in an arbitrary directory.
+ * note: this wrapper is necessary to pass the sanity check in opr_sanity,
+ * which checks that all built-in functions that share the implementing C
+ * function take the same number of arguments.
+ */
+Datum
+pg_ls_dir_metadata_1arg(PG_FUNCTION_ARGS)
+{
+	char	*dirname = convert_and_check_filename(PG_GETARG_TEXT_PP(0));
+
+	return pg_ls_dir_files(fcinfo, dirname,
+			LS_DIR_METADATA | LS_DIR_SKIP_SPECIAL | LS_DIR_ISDIR);
 }
 
 /*
@@ -728,7 +775,7 @@ pg_ls_archive_statusdir(PG_FUNCTION_ARGS)
 Datum
 pg_ls_logicalsnapdir(PG_FUNCTION_ARGS)
 {
-	return pg_ls_dir_files(fcinfo, "pg_logical/snapshots", false);
+	return pg_ls_dir_files(fcinfo, "pg_logical/snapshots", LS_DIR_HISTORIC);
 }
 
 /*
@@ -737,7 +784,7 @@ pg_ls_logicalsnapdir(PG_FUNCTION_ARGS)
 Datum
 pg_ls_logicalmapdir(PG_FUNCTION_ARGS)
 {
-	return pg_ls_dir_files(fcinfo, "pg_logical/mappings", false);
+	return pg_ls_dir_files(fcinfo, "pg_logical/mappings", LS_DIR_HISTORIC);
 }
 
 /*
@@ -762,5 +809,5 @@ pg_ls_replslotdir(PG_FUNCTION_ARGS)
 						slotname)));
 
 	snprintf(path, sizeof(path), "pg_replslot/%s", slotname);
-	return pg_ls_dir_files(fcinfo, path, false);
+	return pg_ls_dir_files(fcinfo, path, LS_DIR_HISTORIC);
 }
diff --git a/src/include/catalog/pg_proc.dat b/src/include/catalog/pg_proc.dat
index e934361dc3..39fbaacbc9 100644
--- a/src/include/catalog/pg_proc.dat
+++ b/src/include/catalog/pg_proc.dat
@@ -11645,6 +11645,18 @@
   proargmodes => '{i,o,o,o}',
   proargnames => '{slot_name,name,size,modification}',
   prosrc => 'pg_ls_replslotdir' },
+{ oid => '8450', descr => 'list directory with metadata',
+  proname => 'pg_ls_dir_metadata', procost => '10', prorows => '20', proretset => 't',
+  provolatile => 'v', prorettype => 'record', proargtypes => 'text bool bool',
+  proallargtypes => '{text,bool,bool,text,int8,timestamptz,bool}', proargmodes => '{i,i,i,o,o,o,o}',
+  proargnames => '{dirname,missing_ok,include_dot_dirs,filename,size,modification,isdir}',
+  prosrc => 'pg_ls_dir_metadata' },
+{ oid => '8451', descr => 'list directory with metadata',
+  proname => 'pg_ls_dir_metadata', procost => '10', prorows => '20', proretset => 't',
+  provolatile => 'v', prorettype => 'record', proargtypes => 'text',
+  proallargtypes => '{text,text,int8,timestamptz,bool}', proargmodes => '{i,o,o,o,o}',
+  proargnames => '{dirname,filename,size,modification,isdir}',
+  prosrc => 'pg_ls_dir_metadata_1arg' },
 
 # hash partitioning constraint function
 { oid => '5028', descr => 'hash partition CHECK constraint',
diff --git a/src/test/regress/expected/misc_functions.out b/src/test/regress/expected/misc_functions.out
index 830de507e7..b3a9d11b5c 100644
--- a/src/test/regress/expected/misc_functions.out
+++ b/src/test/regress/expected/misc_functions.out
@@ -267,6 +267,30 @@ select * from pg_stat_file('.') limit 0;
 ------+--------+--------------+--------+----------+-------
 (0 rows)
 
+-- This tests the missing_ok parameter, which causes pg_ls_tmpdir to succeed even if the tmpdir doesn't exist yet
+-- The name='' condition is never true, so the function runs to completion but returns zero rows.
+select * from pg_ls_tmpdir() where name='Does not exist';
+ name | size | modification 
+------+------+--------------
+(0 rows)
+
+select filename, isdir from pg_ls_dir_metadata('.') where filename='.';
+ filename | isdir 
+----------+-------
+ .        | t
+(1 row)
+
+select filename, isdir from pg_ls_dir_metadata('.', false, false) where filename='.'; -- include_dot_dirs=false
+ filename | isdir 
+----------+-------
+(0 rows)
+
+-- Check that expected columns are present
+select * from pg_ls_dir_metadata('.') limit 0;
+ filename | size | modification | isdir 
+----------+------+--------------+-------
+(0 rows)
+
 --
 -- Test replication slot directory functions
 --
diff --git a/src/test/regress/input/tablespace.source b/src/test/regress/input/tablespace.source
index cb9774ecc8..e69fa17004 100644
--- a/src/test/regress/input/tablespace.source
+++ b/src/test/regress/input/tablespace.source
@@ -11,6 +11,11 @@ DROP TABLESPACE regress_tblspacewith;
 -- create a tablespace we can use
 CREATE TABLESPACE regress_tblspace LOCATION '@testtablespace@';
 
+-- This tests the missing_ok parameter, which causes pg_ls_tmpdir to succeed even if the tmpdir doesn't exist yet
+-- The name='' condition is never true, so the function runs to completion but returns zero rows.
+-- The query is written to ERROR if the tablespace doesn't exist, rather than silently failing to call pg_ls_tmpdir()
+SELECT c.* FROM (SELECT oid FROM pg_tablespace b WHERE b.spcname='regress_tblspace' UNION SELECT 0 ORDER BY 1 DESC LIMIT 1) AS b , pg_ls_tmpdir(oid) AS c WHERE c.name='Does not exist';
+
 -- try setting and resetting some properties for the new tablespace
 ALTER TABLESPACE regress_tblspace SET (random_page_cost = 1.0, seq_page_cost = 1.1);
 ALTER TABLESPACE regress_tblspace SET (some_nonexistent_parameter = true);  -- fail
diff --git a/src/test/regress/output/tablespace.source b/src/test/regress/output/tablespace.source
index e7629d470e..c2f6d64c5c 100644
--- a/src/test/regress/output/tablespace.source
+++ b/src/test/regress/output/tablespace.source
@@ -13,6 +13,14 @@ SELECT spcoptions FROM pg_tablespace WHERE spcname = 'regress_tblspacewith';
 DROP TABLESPACE regress_tblspacewith;
 -- create a tablespace we can use
 CREATE TABLESPACE regress_tblspace LOCATION '@testtablespace@';
+-- This tests the missing_ok parameter, which causes pg_ls_tmpdir to succeed even if the tmpdir doesn't exist yet
+-- The name='' condition is never true, so the function runs to completion but returns zero rows.
+-- The query is written to ERROR if the tablespace doesn't exist, rather than silently failing to call pg_ls_tmpdir()
+SELECT c.* FROM (SELECT oid FROM pg_tablespace b WHERE b.spcname='regress_tblspace' UNION SELECT 0 ORDER BY 1 DESC LIMIT 1) AS b , pg_ls_tmpdir(oid) AS c WHERE c.name='Does not exist';
+ name | size | modification 
+------+------+--------------
+(0 rows)
+
 -- try setting and resetting some properties for the new tablespace
 ALTER TABLESPACE regress_tblspace SET (random_page_cost = 1.0, seq_page_cost = 1.1);
 ALTER TABLESPACE regress_tblspace SET (some_nonexistent_parameter = true);  -- fail
diff --git a/src/test/regress/sql/misc_functions.sql b/src/test/regress/sql/misc_functions.sql
index 422a1369ae..ca946d08bd 100644
--- a/src/test/regress/sql/misc_functions.sql
+++ b/src/test/regress/sql/misc_functions.sql
@@ -99,6 +99,17 @@ select pg_ls_dir('does not exist'); -- fails with missingok=false
 -- Check that expected columns are present
 select * from pg_stat_file('.') limit 0;
 
+-- This tests the missing_ok parameter, which causes pg_ls_tmpdir to succeed even if the tmpdir doesn't exist yet
+-- The name='' condition is never true, so the function runs to completion but returns zero rows.
+select * from pg_ls_tmpdir() where name='Does not exist';
+
+select filename, isdir from pg_ls_dir_metadata('.') where filename='.';
+
+select filename, isdir from pg_ls_dir_metadata('.', false, false) where filename='.'; -- include_dot_dirs=false
+
+-- Check that expected columns are present
+select * from pg_ls_dir_metadata('.') limit 0;
+
 --
 -- Test replication slot directory functions
 --
-- 
2.17.0


--qZVVwWJgpX9Jzs7f
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
 filename="v31-0004-pg_ls_tmpdir-to-show-directories-and-isdir-argum.patch"



^ permalink  raw  reply  [nested|flat] 68+ messages in thread

* [PATCH v33 03/11] Add pg_ls_dir_metadata to list a dir with file metadata..
@ 2020-03-10 03:40  Justin Pryzby <[email protected]>
  0 siblings, 0 replies; 68+ messages in thread

From: Justin Pryzby @ 2020-03-10 03:40 UTC (permalink / raw)

Generalize pg_ls_dir_files and retire pg_ls_dir

Need catversion bumped?
---
 doc/src/sgml/func.sgml                       |  21 ++
 src/backend/catalog/system_functions.sql     |   1 +
 src/backend/utils/adt/genfile.c              | 240 +++++++++++--------
 src/include/catalog/pg_proc.dat              |  12 +
 src/test/regress/expected/misc_functions.out |  24 ++
 src/test/regress/expected/tablespace.out     |   8 +
 src/test/regress/sql/misc_functions.sql      |  11 +
 src/test/regress/sql/tablespace.sql          |   5 +
 8 files changed, 226 insertions(+), 96 deletions(-)

diff --git a/doc/src/sgml/func.sgml b/doc/src/sgml/func.sgml
index d32b414e04f..14b8fba3123 100644
--- a/doc/src/sgml/func.sgml
+++ b/doc/src/sgml/func.sgml
@@ -25958,6 +25958,27 @@ postgres=# SELECT * FROM pg_walfile_name_offset(pg_stop_backup());
        </para></entry>
       </row>
 
+      <row>
+       <entry role="func_table_entry"><para role="func_signature">
+        <indexterm>
+         <primary>pg_ls_dir_metadata</primary>
+        </indexterm>
+        <function>pg_ls_dir_metadata</function> ( <parameter>dirname</parameter> <type>text</type>
+        <optional>, <parameter>missing_ok</parameter> <type>boolean</type>,
+        <parameter>include_dot_dirs</parameter> <type>boolean</type> </optional> )
+        <returnvalue>setof record</returnvalue>
+        ( <parameter>filename</parameter> <type>text</type>,
+        <parameter>size</parameter> <type>bigint</type>,
+        <parameter>modification</parameter> <type>timestamp with time zone</type> )
+       </para>
+       <para>
+        For each file in the specified directory, list the file and its
+        metadata.
+        Restricted to superusers by default, but other users can be granted
+        EXECUTE to run the function.
+       </para></entry>
+      </row>
+
       <row>
        <entry role="func_table_entry"><para role="func_signature">
         <indexterm>
diff --git a/src/backend/catalog/system_functions.sql b/src/backend/catalog/system_functions.sql
index fd1421788e6..2ffc2f9d1a6 100644
--- a/src/backend/catalog/system_functions.sql
+++ b/src/backend/catalog/system_functions.sql
@@ -702,6 +702,7 @@ REVOKE EXECUTE ON FUNCTION pg_stat_file(text,boolean) FROM public;
 REVOKE EXECUTE ON FUNCTION pg_ls_dir(text) FROM public;
 
 REVOKE EXECUTE ON FUNCTION pg_ls_dir(text,boolean,boolean) FROM public;
+REVOKE EXECUTE ON FUNCTION pg_ls_dir_metadata(text,boolean,boolean) FROM public;
 
 REVOKE EXECUTE ON FUNCTION pg_log_backend_memory_contexts(integer) FROM PUBLIC;
 
diff --git a/src/backend/utils/adt/genfile.c b/src/backend/utils/adt/genfile.c
index 542bbacaa24..3921723fda6 100644
--- a/src/backend/utils/adt/genfile.c
+++ b/src/backend/utils/adt/genfile.c
@@ -37,6 +37,21 @@
 #include "utils/syscache.h"
 #include "utils/timestamp.h"
 
+static Datum pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, int flags);
+
+#define	LS_DIR_ISDIR				(1<<0) /* Show column: isdir */
+#define	LS_DIR_METADATA				(1<<1) /* Show columns: mtime, size */
+#define	LS_DIR_MISSING_OK			(1<<2) /* Ignore ENOENT if the toplevel dir is missing */
+#define	LS_DIR_SKIP_DOT_DIRS		(1<<3) /* Do not show . or .. */
+#define	LS_DIR_SKIP_HIDDEN			(1<<4) /* Do not show anything begining with . */
+#define	LS_DIR_SKIP_DIRS			(1<<5) /* Do not show directories */
+#define	LS_DIR_SKIP_SPECIAL			(1<<6) /* Do not show special file types */
+
+/*
+ * Shortcut for the historic behavior of the pg_ls_* functions (not including
+ * pg_ls_dir, which skips different files and doesn't show metadata).
+ */
+#define LS_DIR_HISTORIC				(LS_DIR_SKIP_DIRS | LS_DIR_SKIP_HIDDEN | LS_DIR_SKIP_SPECIAL | LS_DIR_METADATA)
 
 /*
  * Convert a "text" filename argument to C string, and check it's allowable.
@@ -452,6 +467,11 @@ pg_stat_file(PG_FUNCTION_ARGS)
 	values[4] = TimestampTzGetDatum(time_t_to_timestamptz(fst.st_ctime));
 #endif
 	values[5] = BoolGetDatum(S_ISDIR(fst.st_mode));
+#ifdef WIN32
+	/* Links should have isdir=false */
+	if (pgwin32_is_junction(filename))
+		values[5] = BoolGetDatum(false);
+#endif
 
 	tuple = heap_form_tuple(tupdesc, values, isnull);
 
@@ -479,79 +499,9 @@ pg_stat_file_1arg(PG_FUNCTION_ARGS)
 Datum
 pg_ls_dir(PG_FUNCTION_ARGS)
 {
-	ReturnSetInfo *rsinfo = (ReturnSetInfo *) fcinfo->resultinfo;
-	char	   *location;
-	bool		missing_ok = false;
-	bool		include_dot_dirs = false;
-	bool		randomAccess;
-	TupleDesc	tupdesc;
-	Tuplestorestate *tupstore;
-	DIR		   *dirdesc;
-	struct dirent *de;
-	MemoryContext oldcontext;
-
-	location = convert_and_check_filename(PG_GETARG_TEXT_PP(0));
-
-	/* check the optional arguments */
-	if (PG_NARGS() == 3)
-	{
-		if (!PG_ARGISNULL(1))
-			missing_ok = PG_GETARG_BOOL(1);
-		if (!PG_ARGISNULL(2))
-			include_dot_dirs = PG_GETARG_BOOL(2);
-	}
-
-	/* check to see if caller supports us returning a tuplestore */
-	if (rsinfo == NULL || !IsA(rsinfo, ReturnSetInfo))
-		ereport(ERROR,
-				(errcode(ERRCODE_FEATURE_NOT_SUPPORTED),
-				 errmsg("set-valued function called in context that cannot accept a set")));
-	if (!(rsinfo->allowedModes & SFRM_Materialize))
-		ereport(ERROR,
-				(errcode(ERRCODE_SYNTAX_ERROR),
-				 errmsg("materialize mode required, but it is not allowed in this context")));
-
-	/* The tupdesc and tuplestore must be created in ecxt_per_query_memory */
-	oldcontext = MemoryContextSwitchTo(rsinfo->econtext->ecxt_per_query_memory);
-
-	tupdesc = CreateTemplateTupleDesc(1);
-	TupleDescInitEntry(tupdesc, (AttrNumber) 1, "pg_ls_dir", TEXTOID, -1, 0);
-
-	randomAccess = (rsinfo->allowedModes & SFRM_Materialize_Random) != 0;
-	tupstore = tuplestore_begin_heap(randomAccess, false, work_mem);
-	rsinfo->returnMode = SFRM_Materialize;
-	rsinfo->setResult = tupstore;
-	rsinfo->setDesc = tupdesc;
-
-	MemoryContextSwitchTo(oldcontext);
-
-	dirdesc = AllocateDir(location);
-	if (!dirdesc)
-	{
-		/* Return empty tuplestore if appropriate */
-		if (missing_ok && errno == ENOENT)
-			return (Datum) 0;
-		/* Otherwise, we can let ReadDir() throw the error */
-	}
-
-	while ((de = ReadDir(dirdesc, location)) != NULL)
-	{
-		Datum		values[1];
-		bool		nulls[1];
-
-		if (!include_dot_dirs &&
-			(strcmp(de->d_name, ".") == 0 ||
-			 strcmp(de->d_name, "..") == 0))
-			continue;
-
-		values[0] = CStringGetTextDatum(de->d_name);
-		nulls[0] = false;
-
-		tuplestore_putvalues(tupstore, tupdesc, values, nulls);
-	}
-
-	FreeDir(dirdesc);
-	return (Datum) 0;
+	text	*filename_t = PG_GETARG_TEXT_PP(0);
+	char	*filename = convert_and_check_filename(filename_t);
+	return pg_ls_dir_files(fcinfo, filename, LS_DIR_SKIP_DOT_DIRS);
 }
 
 /*
@@ -564,17 +514,19 @@ pg_ls_dir(PG_FUNCTION_ARGS)
 Datum
 pg_ls_dir_1arg(PG_FUNCTION_ARGS)
 {
-	return pg_ls_dir(fcinfo);
+	text	*filename_t = PG_GETARG_TEXT_PP(0);
+	char	*filename = convert_and_check_filename(filename_t);
+	return pg_ls_dir_files(fcinfo, filename, LS_DIR_SKIP_DOT_DIRS);
 }
 
 /*
- * Generic function to return a directory listing of files.
+ * Generic function to return a directory listing of files (and optionally dirs).
  *
- * If the directory isn't there, silently return an empty set if missing_ok.
+ * If the directory isn't there, silently return an empty set if MISSING_OK.
  * Other unreadable-directory cases throw an error.
  */
 static Datum
-pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, bool missing_ok)
+pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, int flags)
 {
 	ReturnSetInfo *rsinfo = (ReturnSetInfo *) fcinfo->resultinfo;
 	bool		randomAccess;
@@ -583,6 +535,32 @@ pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, bool missing_ok)
 	DIR		   *dirdesc;
 	struct dirent *de;
 	MemoryContext oldcontext;
+	TypeFuncClass	tuptype ;
+
+	/* isdir depends on metadata */
+	Assert(!(flags&LS_DIR_ISDIR) || (flags&LS_DIR_METADATA));
+	/* Unreasonable to show isdir and skip dirs */
+	Assert(!(flags&LS_DIR_ISDIR) || !(flags&LS_DIR_SKIP_DIRS));
+
+	/* check the optional arguments */
+	if (PG_NARGS() == 3)
+	{
+		if (!PG_ARGISNULL(1))
+		{
+			if (PG_GETARG_BOOL(1))
+				flags |= LS_DIR_MISSING_OK;
+			else
+				flags &= ~LS_DIR_MISSING_OK;
+		}
+
+		if (!PG_ARGISNULL(2))
+		{
+			if (PG_GETARG_BOOL(2))
+				flags &= ~LS_DIR_SKIP_DOT_DIRS;
+			else
+				flags |= LS_DIR_SKIP_DOT_DIRS;
+		}
+	}
 
 	/* check to see if caller supports us returning a tuplestore */
 	if (rsinfo == NULL || !IsA(rsinfo, ReturnSetInfo))
@@ -597,8 +575,20 @@ pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, bool missing_ok)
 	/* The tupdesc and tuplestore must be created in ecxt_per_query_memory */
 	oldcontext = MemoryContextSwitchTo(rsinfo->econtext->ecxt_per_query_memory);
 
-	if (get_call_result_type(fcinfo, NULL, &tupdesc) != TYPEFUNC_COMPOSITE)
-		elog(ERROR, "return type must be a row type");
+	tuptype = get_call_result_type(fcinfo, NULL, &tupdesc);
+	if (flags & LS_DIR_METADATA)
+	{
+		if (tuptype != TYPEFUNC_COMPOSITE)
+			elog(ERROR, "return type must be a row type");
+	}
+	else
+	{
+		/* pg_ls_dir returns a simple scalar */
+		if (tuptype != TYPEFUNC_SCALAR)
+			elog(ERROR, "return type must be a scalar type");
+		tupdesc = CreateTemplateTupleDesc(1);
+		TupleDescInitEntry(tupdesc, (AttrNumber) 1, "column", TEXTOID, -1, 0);
+	}
 
 	randomAccess = (rsinfo->allowedModes & SFRM_Materialize_Random) != 0;
 	tupstore = tuplestore_begin_heap(randomAccess, false, work_mem);
@@ -617,20 +607,27 @@ pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, bool missing_ok)
 	if (!dirdesc)
 	{
 		/* Return empty tuplestore if appropriate */
-		if (missing_ok && errno == ENOENT)
+		if (flags & LS_DIR_MISSING_OK && errno == ENOENT)
 			return (Datum) 0;
 		/* Otherwise, we can let ReadDir() throw the error */
 	}
 
 	while ((de = ReadDir(dirdesc, dir)) != NULL)
 	{
-		Datum		values[3];
-		bool		nulls[3];
+		Datum		values[4];
+		bool		nulls[4];
 		char		path[MAXPGPATH * 2];
 		struct stat attrib;
 
-		/* Skip hidden files */
-		if (de->d_name[0] == '.')
+		/* Skip dot dirs? */
+		if (flags & LS_DIR_SKIP_DOT_DIRS &&
+			(strcmp(de->d_name, ".") == 0 ||
+			 strcmp(de->d_name, "..") == 0))
+			continue;
+
+		/* Skip hidden files? */
+		if (flags & LS_DIR_SKIP_HIDDEN &&
+			de->d_name[0] == '.')
 			continue;
 
 		/* Get the file info */
@@ -645,13 +642,35 @@ pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, bool missing_ok)
 					 errmsg("could not stat file \"%s\": %m", path)));
 		}
 
-		/* Ignore anything but regular files */
-		if (!S_ISREG(attrib.st_mode))
-			continue;
+		/* Skip dirs or special files? */
+		if (S_ISDIR(attrib.st_mode))
+		{
+			if (flags & LS_DIR_SKIP_DIRS)
+				continue;
+		}
+		else if (!S_ISREG(attrib.st_mode))
+		{
+			if (flags & LS_DIR_SKIP_SPECIAL)
+				continue;
+		}
 
 		values[0] = CStringGetTextDatum(de->d_name);
-		values[1] = Int64GetDatum((int64) attrib.st_size);
-		values[2] = TimestampTzGetDatum(time_t_to_timestamptz(attrib.st_mtime));
+		if (flags & LS_DIR_METADATA)
+		{
+			values[1] = Int64GetDatum((int64) attrib.st_size);
+			values[2] = TimestampTzGetDatum(time_t_to_timestamptz(attrib.st_mtime));
+			if (flags & LS_DIR_ISDIR)
+			{
+#ifdef WIN32
+				/* Links should have isdir=false */
+				if (pgwin32_is_junction(path))
+					values[3] = BoolGetDatum(false);
+				else
+#endif
+					values[3] = BoolGetDatum(S_ISDIR(attrib.st_mode));
+			}
+		}
+
 		memset(nulls, 0, sizeof(nulls));
 
 		tuplestore_putvalues(tupstore, tupdesc, values, nulls);
@@ -665,14 +684,14 @@ pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, bool missing_ok)
 Datum
 pg_ls_logdir(PG_FUNCTION_ARGS)
 {
-	return pg_ls_dir_files(fcinfo, Log_directory, false);
+	return pg_ls_dir_files(fcinfo, Log_directory, LS_DIR_HISTORIC);
 }
 
 /* Function to return the list of files in the WAL directory */
 Datum
 pg_ls_waldir(PG_FUNCTION_ARGS)
 {
-	return pg_ls_dir_files(fcinfo, XLOGDIR, false);
+	return pg_ls_dir_files(fcinfo, XLOGDIR, LS_DIR_HISTORIC);
 }
 
 /*
@@ -690,7 +709,8 @@ pg_ls_tmpdir(FunctionCallInfo fcinfo, Oid tblspc)
 						tblspc)));
 
 	TempTablespacePath(path, tblspc);
-	return pg_ls_dir_files(fcinfo, path, true);
+	return pg_ls_dir_files(fcinfo, path,
+			LS_DIR_HISTORIC | LS_DIR_MISSING_OK);
 }
 
 /*
@@ -719,7 +739,35 @@ pg_ls_tmpdir_1arg(PG_FUNCTION_ARGS)
 Datum
 pg_ls_archive_statusdir(PG_FUNCTION_ARGS)
 {
-	return pg_ls_dir_files(fcinfo, XLOGDIR "/archive_status", true);
+	return pg_ls_dir_files(fcinfo, XLOGDIR "/archive_status",
+			LS_DIR_HISTORIC | LS_DIR_MISSING_OK);
+}
+
+/*
+ * Return the list of files and metadata in an arbitrary directory.
+ */
+Datum
+pg_ls_dir_metadata(PG_FUNCTION_ARGS)
+{
+	char	*dirname = convert_and_check_filename(PG_GETARG_TEXT_PP(0));
+
+	return pg_ls_dir_files(fcinfo, dirname,
+			LS_DIR_METADATA | LS_DIR_SKIP_SPECIAL | LS_DIR_ISDIR);
+}
+
+/*
+ * Return the list of files and metadata in an arbitrary directory.
+ * note: this wrapper is necessary to pass the sanity check in opr_sanity,
+ * which checks that all built-in functions that share the implementing C
+ * function take the same number of arguments.
+ */
+Datum
+pg_ls_dir_metadata_1arg(PG_FUNCTION_ARGS)
+{
+	char	*dirname = convert_and_check_filename(PG_GETARG_TEXT_PP(0));
+
+	return pg_ls_dir_files(fcinfo, dirname,
+			LS_DIR_METADATA | LS_DIR_SKIP_SPECIAL | LS_DIR_ISDIR);
 }
 
 /*
@@ -728,7 +776,7 @@ pg_ls_archive_statusdir(PG_FUNCTION_ARGS)
 Datum
 pg_ls_logicalsnapdir(PG_FUNCTION_ARGS)
 {
-	return pg_ls_dir_files(fcinfo, "pg_logical/snapshots", false);
+	return pg_ls_dir_files(fcinfo, "pg_logical/snapshots", LS_DIR_HISTORIC);
 }
 
 /*
@@ -737,7 +785,7 @@ pg_ls_logicalsnapdir(PG_FUNCTION_ARGS)
 Datum
 pg_ls_logicalmapdir(PG_FUNCTION_ARGS)
 {
-	return pg_ls_dir_files(fcinfo, "pg_logical/mappings", false);
+	return pg_ls_dir_files(fcinfo, "pg_logical/mappings", LS_DIR_HISTORIC);
 }
 
 /*
@@ -762,5 +810,5 @@ pg_ls_replslotdir(PG_FUNCTION_ARGS)
 						slotname)));
 
 	snprintf(path, sizeof(path), "pg_replslot/%s", slotname);
-	return pg_ls_dir_files(fcinfo, path, false);
+	return pg_ls_dir_files(fcinfo, path, LS_DIR_HISTORIC);
 }
diff --git a/src/include/catalog/pg_proc.dat b/src/include/catalog/pg_proc.dat
index 0859dc81cac..b58987b713c 100644
--- a/src/include/catalog/pg_proc.dat
+++ b/src/include/catalog/pg_proc.dat
@@ -11661,6 +11661,18 @@
   proargmodes => '{i,o,o,o}',
   proargnames => '{slot_name,name,size,modification}',
   prosrc => 'pg_ls_replslotdir' },
+{ oid => '8450', descr => 'list directory with metadata',
+  proname => 'pg_ls_dir_metadata', procost => '10', prorows => '20', proretset => 't',
+  provolatile => 'v', prorettype => 'record', proargtypes => 'text bool bool',
+  proallargtypes => '{text,bool,bool,text,int8,timestamptz,bool}', proargmodes => '{i,i,i,o,o,o,o}',
+  proargnames => '{dirname,missing_ok,include_dot_dirs,filename,size,modification,isdir}',
+  prosrc => 'pg_ls_dir_metadata' },
+{ oid => '8451', descr => 'list directory with metadata',
+  proname => 'pg_ls_dir_metadata', procost => '10', prorows => '20', proretset => 't',
+  provolatile => 'v', prorettype => 'record', proargtypes => 'text',
+  proallargtypes => '{text,text,int8,timestamptz,bool}', proargmodes => '{i,o,o,o,o}',
+  proargnames => '{dirname,filename,size,modification,isdir}',
+  prosrc => 'pg_ls_dir_metadata_1arg' },
 
 # hash partitioning constraint function
 { oid => '5028', descr => 'hash partition CHECK constraint',
diff --git a/src/test/regress/expected/misc_functions.out b/src/test/regress/expected/misc_functions.out
index b7264af3ecb..a627839bc28 100644
--- a/src/test/regress/expected/misc_functions.out
+++ b/src/test/regress/expected/misc_functions.out
@@ -309,6 +309,30 @@ select * from pg_stat_file('.') limit 0;
 ------+--------+--------------+--------+----------+-------
 (0 rows)
 
+-- This tests the missing_ok parameter, which causes pg_ls_tmpdir to succeed even if the tmpdir doesn't exist yet
+-- The name='' condition is never true, so the function runs to completion but returns zero rows.
+select * from pg_ls_tmpdir() where name='Does not exist';
+ name | size | modification 
+------+------+--------------
+(0 rows)
+
+select filename, isdir from pg_ls_dir_metadata('.') where filename='.';
+ filename | isdir 
+----------+-------
+ .        | t
+(1 row)
+
+select filename, isdir from pg_ls_dir_metadata('.', false, false) where filename='.'; -- include_dot_dirs=false
+ filename | isdir 
+----------+-------
+(0 rows)
+
+-- Check that expected columns are present
+select * from pg_ls_dir_metadata('.') limit 0;
+ filename | size | modification | isdir 
+----------+------+--------------+-------
+(0 rows)
+
 --
 -- Test replication slot directory functions
 --
diff --git a/src/test/regress/expected/tablespace.out b/src/test/regress/expected/tablespace.out
index 2dfbcfdebe1..c6b9074ab02 100644
--- a/src/test/regress/expected/tablespace.out
+++ b/src/test/regress/expected/tablespace.out
@@ -24,6 +24,14 @@ SELECT spcoptions FROM pg_tablespace WHERE spcname = 'regress_tblspacewith';
 DROP TABLESPACE regress_tblspacewith;
 -- create a tablespace we can use
 CREATE TABLESPACE regress_tblspace LOCATION '';
+-- This tests the missing_ok parameter, which causes pg_ls_tmpdir to succeed even if the tmpdir doesn't exist yet
+-- The name='' condition is never true, so the function runs to completion but returns zero rows.
+-- The query is written to ERROR if the tablespace doesn't exist, rather than silently failing to call pg_ls_tmpdir()
+SELECT c.* FROM (SELECT oid FROM pg_tablespace b WHERE b.spcname='regress_tblspace' UNION SELECT 0 ORDER BY 1 DESC LIMIT 1) AS b , pg_ls_tmpdir(oid) AS c WHERE c.name='Does not exist';
+ name | size | modification 
+------+------+--------------
+(0 rows)
+
 -- try setting and resetting some properties for the new tablespace
 ALTER TABLESPACE regress_tblspace SET (random_page_cost = 1.0, seq_page_cost = 1.1);
 ALTER TABLESPACE regress_tblspace SET (some_nonexistent_parameter = true);  -- fail
diff --git a/src/test/regress/sql/misc_functions.sql b/src/test/regress/sql/misc_functions.sql
index 169f66416c3..bfa8e309e9a 100644
--- a/src/test/regress/sql/misc_functions.sql
+++ b/src/test/regress/sql/misc_functions.sql
@@ -109,6 +109,17 @@ select * from pg_ls_tmpdir() limit 0;
 select * from pg_ls_waldir() limit 0;
 select * from pg_stat_file('.') limit 0;
 
+-- This tests the missing_ok parameter, which causes pg_ls_tmpdir to succeed even if the tmpdir doesn't exist yet
+-- The name='' condition is never true, so the function runs to completion but returns zero rows.
+select * from pg_ls_tmpdir() where name='Does not exist';
+
+select filename, isdir from pg_ls_dir_metadata('.') where filename='.';
+
+select filename, isdir from pg_ls_dir_metadata('.', false, false) where filename='.'; -- include_dot_dirs=false
+
+-- Check that expected columns are present
+select * from pg_ls_dir_metadata('.') limit 0;
+
 --
 -- Test replication slot directory functions
 --
diff --git a/src/test/regress/sql/tablespace.sql b/src/test/regress/sql/tablespace.sql
index 896f05cea32..3aadac8b611 100644
--- a/src/test/regress/sql/tablespace.sql
+++ b/src/test/regress/sql/tablespace.sql
@@ -23,6 +23,11 @@ DROP TABLESPACE regress_tblspacewith;
 -- create a tablespace we can use
 CREATE TABLESPACE regress_tblspace LOCATION '';
 
+-- This tests the missing_ok parameter, which causes pg_ls_tmpdir to succeed even if the tmpdir doesn't exist yet
+-- The name='' condition is never true, so the function runs to completion but returns zero rows.
+-- The query is written to ERROR if the tablespace doesn't exist, rather than silently failing to call pg_ls_tmpdir()
+SELECT c.* FROM (SELECT oid FROM pg_tablespace b WHERE b.spcname='regress_tblspace' UNION SELECT 0 ORDER BY 1 DESC LIMIT 1) AS b , pg_ls_tmpdir(oid) AS c WHERE c.name='Does not exist';
+
 -- try setting and resetting some properties for the new tablespace
 ALTER TABLESPACE regress_tblspace SET (random_page_cost = 1.0, seq_page_cost = 1.1);
 ALTER TABLESPACE regress_tblspace SET (some_nonexistent_parameter = true);  -- fail
-- 
2.17.1


--9CzcV6dAFIr7O1Ie
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
 filename="v33-0004-pg_ls_tmpdir-to-show-directories-and-isdir-argum.patch"



^ permalink  raw  reply  [nested|flat] 68+ messages in thread

* [PATCH v30 03/11] Add pg_ls_dir_metadata to list a dir with file metadata..
@ 2020-03-10 03:40  Justin Pryzby <[email protected]>
  0 siblings, 0 replies; 68+ messages in thread

From: Justin Pryzby @ 2020-03-10 03:40 UTC (permalink / raw)

Generalize pg_ls_dir_files and retire pg_ls_dir

Need catversion bumped?
---
 doc/src/sgml/func.sgml                       |  21 ++
 src/backend/catalog/system_functions.sql     |   1 +
 src/backend/utils/adt/genfile.c              | 233 +++++++++++--------
 src/include/catalog/pg_proc.dat              |  12 +
 src/test/regress/expected/misc_functions.out |  24 ++
 src/test/regress/input/tablespace.source     |   5 +
 src/test/regress/output/tablespace.source    |   8 +
 src/test/regress/sql/misc_functions.sql      |  11 +
 8 files changed, 222 insertions(+), 93 deletions(-)

diff --git a/doc/src/sgml/func.sgml b/doc/src/sgml/func.sgml
index 7a830f0684..dd9b1afde4 100644
--- a/doc/src/sgml/func.sgml
+++ b/doc/src/sgml/func.sgml
@@ -25848,6 +25848,27 @@ postgres=# SELECT * FROM pg_walfile_name_offset(pg_stop_backup());
        </para></entry>
       </row>
 
+      <row>
+       <entry role="func_table_entry"><para role="func_signature">
+        <indexterm>
+         <primary>pg_ls_dir_metadata</primary>
+        </indexterm>
+        <function>pg_ls_dir_metadata</function> ( <parameter>dirname</parameter> <type>text</type>
+        <optional>, <parameter>missing_ok</parameter> <type>boolean</type>,
+        <parameter>include_dot_dirs</parameter> <type>boolean</type> </optional> )
+        <returnvalue>setof record</returnvalue>
+        ( <parameter>filename</parameter> <type>text</type>,
+        <parameter>size</parameter> <type>bigint</type>,
+        <parameter>modification</parameter> <type>timestamp with time zone</type> )
+       </para>
+       <para>
+        For each file in the specified directory, list the file and its
+        metadata.
+        Restricted to superusers by default, but other users can be granted
+        EXECUTE to run the function.
+       </para></entry>
+      </row>
+
       <row>
        <entry role="func_table_entry"><para role="func_signature">
         <indexterm>
diff --git a/src/backend/catalog/system_functions.sql b/src/backend/catalog/system_functions.sql
index a416e94d37..649a78605a 100644
--- a/src/backend/catalog/system_functions.sql
+++ b/src/backend/catalog/system_functions.sql
@@ -698,6 +698,7 @@ REVOKE EXECUTE ON FUNCTION pg_stat_file(text,boolean) FROM public;
 REVOKE EXECUTE ON FUNCTION pg_ls_dir(text) FROM public;
 
 REVOKE EXECUTE ON FUNCTION pg_ls_dir(text,boolean,boolean) FROM public;
+REVOKE EXECUTE ON FUNCTION pg_ls_dir_metadata(text,boolean,boolean) FROM public;
 
 --
 -- We also set up some things as accessible to standard roles.
diff --git a/src/backend/utils/adt/genfile.c b/src/backend/utils/adt/genfile.c
index c436d9318b..8efb798b2f 100644
--- a/src/backend/utils/adt/genfile.c
+++ b/src/backend/utils/adt/genfile.c
@@ -36,6 +36,21 @@
 #include "utils/syscache.h"
 #include "utils/timestamp.h"
 
+static Datum pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, int flags);
+
+#define	LS_DIR_ISDIR				(1<<0) /* Show column: isdir */
+#define	LS_DIR_METADATA				(1<<1) /* Show columns: mtime, size */
+#define	LS_DIR_MISSING_OK			(1<<2) /* Ignore ENOENT if the toplevel dir is missing */
+#define	LS_DIR_SKIP_DOT_DIRS		(1<<3) /* Do not show . or .. */
+#define	LS_DIR_SKIP_HIDDEN			(1<<4) /* Do not show anything begining with . */
+#define	LS_DIR_SKIP_DIRS			(1<<5) /* Do not show directories */
+#define	LS_DIR_SKIP_SPECIAL			(1<<6) /* Do not show special file types */
+
+/*
+ * Shortcut for the historic behavior of the pg_ls_* functions (not including
+ * pg_ls_dir, which skips different files and doesn't show metadata).
+ */
+#define LS_DIR_HISTORIC				(LS_DIR_SKIP_DIRS | LS_DIR_SKIP_HIDDEN | LS_DIR_SKIP_SPECIAL | LS_DIR_METADATA)
 
 /*
  * Convert a "text" filename argument to C string, and check it's allowable.
@@ -451,6 +466,11 @@ pg_stat_file(PG_FUNCTION_ARGS)
 	values[4] = TimestampTzGetDatum(time_t_to_timestamptz(fst.st_ctime));
 #endif
 	values[5] = BoolGetDatum(S_ISDIR(fst.st_mode));
+#ifdef WIN32
+	/* Links should have isdir=false */
+	if (pgwin32_is_junction(filename))
+		values[5] = BoolGetDatum(false);
+#endif
 
 	tuple = heap_form_tuple(tupdesc, values, isnull);
 
@@ -478,79 +498,9 @@ pg_stat_file_1arg(PG_FUNCTION_ARGS)
 Datum
 pg_ls_dir(PG_FUNCTION_ARGS)
 {
-	ReturnSetInfo *rsinfo = (ReturnSetInfo *) fcinfo->resultinfo;
-	char	   *location;
-	bool		missing_ok = false;
-	bool		include_dot_dirs = false;
-	bool		randomAccess;
-	TupleDesc	tupdesc;
-	Tuplestorestate *tupstore;
-	DIR		   *dirdesc;
-	struct dirent *de;
-	MemoryContext oldcontext;
-
-	location = convert_and_check_filename(PG_GETARG_TEXT_PP(0));
-
-	/* check the optional arguments */
-	if (PG_NARGS() == 3)
-	{
-		if (!PG_ARGISNULL(1))
-			missing_ok = PG_GETARG_BOOL(1);
-		if (!PG_ARGISNULL(2))
-			include_dot_dirs = PG_GETARG_BOOL(2);
-	}
-
-	/* check to see if caller supports us returning a tuplestore */
-	if (rsinfo == NULL || !IsA(rsinfo, ReturnSetInfo))
-		ereport(ERROR,
-				(errcode(ERRCODE_FEATURE_NOT_SUPPORTED),
-				 errmsg("set-valued function called in context that cannot accept a set")));
-	if (!(rsinfo->allowedModes & SFRM_Materialize))
-		ereport(ERROR,
-				(errcode(ERRCODE_SYNTAX_ERROR),
-				 errmsg("materialize mode required, but it is not allowed in this context")));
-
-	/* The tupdesc and tuplestore must be created in ecxt_per_query_memory */
-	oldcontext = MemoryContextSwitchTo(rsinfo->econtext->ecxt_per_query_memory);
-
-	tupdesc = CreateTemplateTupleDesc(1);
-	TupleDescInitEntry(tupdesc, (AttrNumber) 1, "pg_ls_dir", TEXTOID, -1, 0);
-
-	randomAccess = (rsinfo->allowedModes & SFRM_Materialize_Random) != 0;
-	tupstore = tuplestore_begin_heap(randomAccess, false, work_mem);
-	rsinfo->returnMode = SFRM_Materialize;
-	rsinfo->setResult = tupstore;
-	rsinfo->setDesc = tupdesc;
-
-	MemoryContextSwitchTo(oldcontext);
-
-	dirdesc = AllocateDir(location);
-	if (!dirdesc)
-	{
-		/* Return empty tuplestore if appropriate */
-		if (missing_ok && errno == ENOENT)
-			return (Datum) 0;
-		/* Otherwise, we can let ReadDir() throw the error */
-	}
-
-	while ((de = ReadDir(dirdesc, location)) != NULL)
-	{
-		Datum		values[1];
-		bool		nulls[1];
-
-		if (!include_dot_dirs &&
-			(strcmp(de->d_name, ".") == 0 ||
-			 strcmp(de->d_name, "..") == 0))
-			continue;
-
-		values[0] = CStringGetTextDatum(de->d_name);
-		nulls[0] = false;
-
-		tuplestore_putvalues(tupstore, tupdesc, values, nulls);
-	}
-
-	FreeDir(dirdesc);
-	return (Datum) 0;
+	text	*filename_t = PG_GETARG_TEXT_PP(0);
+	char	*filename = convert_and_check_filename(filename_t);
+	return pg_ls_dir_files(fcinfo, filename, LS_DIR_SKIP_DOT_DIRS);
 }
 
 /*
@@ -563,17 +513,19 @@ pg_ls_dir(PG_FUNCTION_ARGS)
 Datum
 pg_ls_dir_1arg(PG_FUNCTION_ARGS)
 {
-	return pg_ls_dir(fcinfo);
+	text	*filename_t = PG_GETARG_TEXT_PP(0);
+	char	*filename = convert_and_check_filename(filename_t);
+	return pg_ls_dir_files(fcinfo, filename, LS_DIR_SKIP_DOT_DIRS);
 }
 
 /*
- * Generic function to return a directory listing of files.
+ * Generic function to return a directory listing of files (and optionally dirs).
  *
- * If the directory isn't there, silently return an empty set if missing_ok.
+ * If the directory isn't there, silently return an empty set if MISSING_OK.
  * Other unreadable-directory cases throw an error.
  */
 static Datum
-pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, bool missing_ok)
+pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, int flags)
 {
 	ReturnSetInfo *rsinfo = (ReturnSetInfo *) fcinfo->resultinfo;
 	bool		randomAccess;
@@ -582,6 +534,32 @@ pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, bool missing_ok)
 	DIR		   *dirdesc;
 	struct dirent *de;
 	MemoryContext oldcontext;
+	TypeFuncClass	tuptype ;
+
+	/* isdir depends on metadata */
+	Assert(!(flags&LS_DIR_ISDIR) || (flags&LS_DIR_METADATA));
+	/* Unreasonable to show isdir and skip dirs */
+	Assert(!(flags&LS_DIR_ISDIR) || !(flags&LS_DIR_SKIP_DIRS));
+
+	/* check the optional arguments */
+	if (PG_NARGS() == 3)
+	{
+		if (!PG_ARGISNULL(1))
+		{
+			if (PG_GETARG_BOOL(1))
+				flags |= LS_DIR_MISSING_OK;
+			else
+				flags &= ~LS_DIR_MISSING_OK;
+		}
+
+		if (!PG_ARGISNULL(2))
+		{
+			if (PG_GETARG_BOOL(2))
+				flags &= ~LS_DIR_SKIP_DOT_DIRS;
+			else
+				flags |= LS_DIR_SKIP_DOT_DIRS;
+		}
+	}
 
 	/* check to see if caller supports us returning a tuplestore */
 	if (rsinfo == NULL || !IsA(rsinfo, ReturnSetInfo))
@@ -596,8 +574,20 @@ pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, bool missing_ok)
 	/* The tupdesc and tuplestore must be created in ecxt_per_query_memory */
 	oldcontext = MemoryContextSwitchTo(rsinfo->econtext->ecxt_per_query_memory);
 
-	if (get_call_result_type(fcinfo, NULL, &tupdesc) != TYPEFUNC_COMPOSITE)
-		elog(ERROR, "return type must be a row type");
+	tuptype = get_call_result_type(fcinfo, NULL, &tupdesc);
+	if (flags & LS_DIR_METADATA)
+	{
+		if (tuptype != TYPEFUNC_COMPOSITE)
+			elog(ERROR, "return type must be a row type");
+	}
+	else
+	{
+		/* pg_ls_dir returns a simple scalar */
+		if (tuptype != TYPEFUNC_SCALAR)
+			elog(ERROR, "return type must be a scalar type");
+		tupdesc = CreateTemplateTupleDesc(1);
+		TupleDescInitEntry(tupdesc, (AttrNumber) 1, "column", TEXTOID, -1, 0);
+	}
 
 	randomAccess = (rsinfo->allowedModes & SFRM_Materialize_Random) != 0;
 	tupstore = tuplestore_begin_heap(randomAccess, false, work_mem);
@@ -616,20 +606,27 @@ pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, bool missing_ok)
 	if (!dirdesc)
 	{
 		/* Return empty tuplestore if appropriate */
-		if (missing_ok && errno == ENOENT)
+		if (flags & LS_DIR_MISSING_OK && errno == ENOENT)
 			return (Datum) 0;
 		/* Otherwise, we can let ReadDir() throw the error */
 	}
 
 	while ((de = ReadDir(dirdesc, dir)) != NULL)
 	{
-		Datum		values[3];
-		bool		nulls[3];
+		Datum		values[4];
+		bool		nulls[4];
 		char		path[MAXPGPATH * 2];
 		struct stat attrib;
 
-		/* Skip hidden files */
-		if (de->d_name[0] == '.')
+		/* Skip dot dirs? */
+		if (flags & LS_DIR_SKIP_DOT_DIRS &&
+			(strcmp(de->d_name, ".") == 0 ||
+			 strcmp(de->d_name, "..") == 0))
+			continue;
+
+		/* Skip hidden files? */
+		if (flags & LS_DIR_SKIP_HIDDEN &&
+			de->d_name[0] == '.')
 			continue;
 
 		/* Get the file info */
@@ -644,13 +641,34 @@ pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, bool missing_ok)
 					 errmsg("could not stat file \"%s\": %m", path)));
 		}
 
-		/* Ignore anything but regular files */
-		if (!S_ISREG(attrib.st_mode))
-			continue;
+		/* Skip dirs or special files? */
+		if (S_ISDIR(attrib.st_mode))
+		{
+			if (flags & LS_DIR_SKIP_DIRS)
+				continue;
+		}
+		else if (!S_ISREG(attrib.st_mode))
+		{
+			if (flags & LS_DIR_SKIP_SPECIAL)
+				continue;
+		}
 
 		values[0] = CStringGetTextDatum(de->d_name);
-		values[1] = Int64GetDatum((int64) attrib.st_size);
-		values[2] = TimestampTzGetDatum(time_t_to_timestamptz(attrib.st_mtime));
+		if (flags & LS_DIR_METADATA)
+		{
+			values[1] = Int64GetDatum((int64) attrib.st_size);
+			values[2] = TimestampTzGetDatum(time_t_to_timestamptz(attrib.st_mtime));
+			if (flags & LS_DIR_ISDIR)
+			{
+				values[3] = BoolGetDatum(S_ISDIR(attrib.st_mode));
+#ifdef WIN32
+				/* Links should have isdir=false */
+				if (pgwin32_is_junction(path))
+					values[3] = BoolGetDatum(false);
+#endif
+			}
+		}
+
 		memset(nulls, 0, sizeof(nulls));
 
 		tuplestore_putvalues(tupstore, tupdesc, values, nulls);
@@ -664,14 +682,14 @@ pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, bool missing_ok)
 Datum
 pg_ls_logdir(PG_FUNCTION_ARGS)
 {
-	return pg_ls_dir_files(fcinfo, Log_directory, false);
+	return pg_ls_dir_files(fcinfo, Log_directory, LS_DIR_HISTORIC);
 }
 
 /* Function to return the list of files in the WAL directory */
 Datum
 pg_ls_waldir(PG_FUNCTION_ARGS)
 {
-	return pg_ls_dir_files(fcinfo, XLOGDIR, false);
+	return pg_ls_dir_files(fcinfo, XLOGDIR, LS_DIR_HISTORIC);
 }
 
 /*
@@ -689,7 +707,8 @@ pg_ls_tmpdir(FunctionCallInfo fcinfo, Oid tblspc)
 						tblspc)));
 
 	TempTablespacePath(path, tblspc);
-	return pg_ls_dir_files(fcinfo, path, true);
+	return pg_ls_dir_files(fcinfo, path,
+			LS_DIR_HISTORIC | LS_DIR_MISSING_OK);
 }
 
 /*
@@ -718,5 +737,33 @@ pg_ls_tmpdir_1arg(PG_FUNCTION_ARGS)
 Datum
 pg_ls_archive_statusdir(PG_FUNCTION_ARGS)
 {
-	return pg_ls_dir_files(fcinfo, XLOGDIR "/archive_status", true);
+	return pg_ls_dir_files(fcinfo, XLOGDIR "/archive_status",
+			LS_DIR_HISTORIC | LS_DIR_MISSING_OK);
+}
+
+/*
+ * Return the list of files and metadata in an arbitrary directory.
+ */
+Datum
+pg_ls_dir_metadata(PG_FUNCTION_ARGS)
+{
+	char	*dirname = convert_and_check_filename(PG_GETARG_TEXT_PP(0));
+
+	return pg_ls_dir_files(fcinfo, dirname,
+			LS_DIR_METADATA | LS_DIR_SKIP_SPECIAL | LS_DIR_ISDIR);
+}
+
+/*
+ * Return the list of files and metadata in an arbitrary directory.
+ * note: this wrapper is necessary to pass the sanity check in opr_sanity,
+ * which checks that all built-in functions that share the implementing C
+ * function take the same number of arguments.
+ */
+Datum
+pg_ls_dir_metadata_1arg(PG_FUNCTION_ARGS)
+{
+	char	*dirname = convert_and_check_filename(PG_GETARG_TEXT_PP(0));
+
+	return pg_ls_dir_files(fcinfo, dirname,
+			LS_DIR_METADATA | LS_DIR_SKIP_SPECIAL | LS_DIR_ISDIR);
 }
diff --git a/src/include/catalog/pg_proc.dat b/src/include/catalog/pg_proc.dat
index fde251fa4f..b405aa341c 100644
--- a/src/include/catalog/pg_proc.dat
+++ b/src/include/catalog/pg_proc.dat
@@ -11547,6 +11547,18 @@
   proallargtypes => '{oid,text,int8,timestamptz}', proargmodes => '{i,o,o,o}',
   proargnames => '{tablespace,name,size,modification}',
   prosrc => 'pg_ls_tmpdir_1arg' },
+{ oid => '8450', descr => 'list directory with metadata',
+  proname => 'pg_ls_dir_metadata', procost => '10', prorows => '20', proretset => 't',
+  provolatile => 'v', prorettype => 'record', proargtypes => 'text bool bool',
+  proallargtypes => '{text,bool,bool,text,int8,timestamptz,bool}', proargmodes => '{i,i,i,o,o,o,o}',
+  proargnames => '{dirname,missing_ok,include_dot_dirs,filename,size,modification,isdir}',
+  prosrc => 'pg_ls_dir_metadata' },
+{ oid => '8451', descr => 'list directory with metadata',
+  proname => 'pg_ls_dir_metadata', procost => '10', prorows => '20', proretset => 't',
+  provolatile => 'v', prorettype => 'record', proargtypes => 'text',
+  proallargtypes => '{text,text,int8,timestamptz,bool}', proargmodes => '{i,o,o,o,o}',
+  proargnames => '{dirname,filename,size,modification,isdir}',
+  prosrc => 'pg_ls_dir_metadata_1arg' },
 
 # hash partitioning constraint function
 { oid => '5028', descr => 'hash partition CHECK constraint',
diff --git a/src/test/regress/expected/misc_functions.out b/src/test/regress/expected/misc_functions.out
index ea0fc48dbd..b4112ec298 100644
--- a/src/test/regress/expected/misc_functions.out
+++ b/src/test/regress/expected/misc_functions.out
@@ -238,6 +238,30 @@ select * from pg_stat_file('.') limit 0;
 ------+--------+--------------+--------+----------+-------
 (0 rows)
 
+-- This tests the missing_ok parameter, which causes pg_ls_tmpdir to succeed even if the tmpdir doesn't exist yet
+-- The name='' condition is never true, so the function runs to completion but returns zero rows.
+select * from pg_ls_tmpdir() where name='Does not exist';
+ name | size | modification 
+------+------+--------------
+(0 rows)
+
+select filename, isdir from pg_ls_dir_metadata('.') where filename='.';
+ filename | isdir 
+----------+-------
+ .        | t
+(1 row)
+
+select filename, isdir from pg_ls_dir_metadata('.', false, false) where filename='.'; -- include_dot_dirs=false
+ filename | isdir 
+----------+-------
+(0 rows)
+
+-- Check that expected columns are present
+select * from pg_ls_dir_metadata('.') limit 0;
+ filename | size | modification | isdir 
+----------+------+--------------+-------
+(0 rows)
+
 --
 -- Test adding a support function to a subject function
 --
diff --git a/src/test/regress/input/tablespace.source b/src/test/regress/input/tablespace.source
index c133e73499..a32212be04 100644
--- a/src/test/regress/input/tablespace.source
+++ b/src/test/regress/input/tablespace.source
@@ -11,6 +11,11 @@ DROP TABLESPACE regress_tblspacewith;
 -- create a tablespace we can use
 CREATE TABLESPACE regress_tblspace LOCATION '@testtablespace@';
 
+-- This tests the missing_ok parameter, which causes pg_ls_tmpdir to succeed even if the tmpdir doesn't exist yet
+-- The name='' condition is never true, so the function runs to completion but returns zero rows.
+-- The query is written to ERROR if the tablespace doesn't exist, rather than silently failing to call pg_ls_tmpdir()
+SELECT c.* FROM (SELECT oid FROM pg_tablespace b WHERE b.spcname='regress_tblspace' UNION SELECT 0 ORDER BY 1 DESC LIMIT 1) AS b , pg_ls_tmpdir(oid) AS c WHERE c.name='Does not exist';
+
 -- try setting and resetting some properties for the new tablespace
 ALTER TABLESPACE regress_tblspace SET (random_page_cost = 1.0, seq_page_cost = 1.1);
 ALTER TABLESPACE regress_tblspace SET (some_nonexistent_parameter = true);  -- fail
diff --git a/src/test/regress/output/tablespace.source b/src/test/regress/output/tablespace.source
index 1bbe7e0323..5f7ce3d09f 100644
--- a/src/test/regress/output/tablespace.source
+++ b/src/test/regress/output/tablespace.source
@@ -13,6 +13,14 @@ SELECT spcoptions FROM pg_tablespace WHERE spcname = 'regress_tblspacewith';
 DROP TABLESPACE regress_tblspacewith;
 -- create a tablespace we can use
 CREATE TABLESPACE regress_tblspace LOCATION '@testtablespace@';
+-- This tests the missing_ok parameter, which causes pg_ls_tmpdir to succeed even if the tmpdir doesn't exist yet
+-- The name='' condition is never true, so the function runs to completion but returns zero rows.
+-- The query is written to ERROR if the tablespace doesn't exist, rather than silently failing to call pg_ls_tmpdir()
+SELECT c.* FROM (SELECT oid FROM pg_tablespace b WHERE b.spcname='regress_tblspace' UNION SELECT 0 ORDER BY 1 DESC LIMIT 1) AS b , pg_ls_tmpdir(oid) AS c WHERE c.name='Does not exist';
+ name | size | modification 
+------+------+--------------
+(0 rows)
+
 -- try setting and resetting some properties for the new tablespace
 ALTER TABLESPACE regress_tblspace SET (random_page_cost = 1.0, seq_page_cost = 1.1);
 ALTER TABLESPACE regress_tblspace SET (some_nonexistent_parameter = true);  -- fail
diff --git a/src/test/regress/sql/misc_functions.sql b/src/test/regress/sql/misc_functions.sql
index eb6ac12ab4..c169e527d9 100644
--- a/src/test/regress/sql/misc_functions.sql
+++ b/src/test/regress/sql/misc_functions.sql
@@ -77,6 +77,17 @@ select pg_ls_dir('does not exist'); -- fails with missingok=false
 -- Check that expected columns are present
 select * from pg_stat_file('.') limit 0;
 
+-- This tests the missing_ok parameter, which causes pg_ls_tmpdir to succeed even if the tmpdir doesn't exist yet
+-- The name='' condition is never true, so the function runs to completion but returns zero rows.
+select * from pg_ls_tmpdir() where name='Does not exist';
+
+select filename, isdir from pg_ls_dir_metadata('.') where filename='.';
+
+select filename, isdir from pg_ls_dir_metadata('.', false, false) where filename='.'; -- include_dot_dirs=false
+
+-- Check that expected columns are present
+select * from pg_ls_dir_metadata('.') limit 0;
+
 --
 -- Test adding a support function to a subject function
 --
-- 
2.17.0


--ZwgA9U+XZDXt4+m+
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
 filename="v30-0004-pg_ls_tmpdir-to-show-directories-and-isdir-argum.patch"



^ permalink  raw  reply  [nested|flat] 68+ messages in thread

* [PATCH v25 04/11] Add pg_ls_dir_metadata to list a dir with file metadata..
@ 2020-03-10 03:40  Justin Pryzby <[email protected]>
  0 siblings, 0 replies; 68+ messages in thread

From: Justin Pryzby @ 2020-03-10 03:40 UTC (permalink / raw)

Generalize pg_ls_dir_files and retire pg_ls_dir

Need catversion bumped?
---
 doc/src/sgml/func.sgml                       |  21 ++
 src/backend/catalog/system_views.sql         |   1 +
 src/backend/utils/adt/genfile.c              | 231 +++++++++++--------
 src/include/catalog/pg_proc.dat              |  12 +
 src/test/regress/expected/misc_functions.out |  24 ++
 src/test/regress/input/tablespace.source     |   5 +
 src/test/regress/output/tablespace.source    |   8 +
 src/test/regress/sql/misc_functions.sql      |  11 +
 8 files changed, 221 insertions(+), 92 deletions(-)

diff --git a/doc/src/sgml/func.sgml b/doc/src/sgml/func.sgml
index df29af6371..226923a134 100644
--- a/doc/src/sgml/func.sgml
+++ b/doc/src/sgml/func.sgml
@@ -25750,6 +25750,27 @@ SELECT pg_size_pretty(sum(pg_relation_size(relid))) AS total_size
        </para></entry>
       </row>
 
+      <row>
+       <entry role="func_table_entry"><para role="func_signature">
+        <indexterm>
+         <primary>pg_ls_dir_metadata</primary>
+        </indexterm>
+        <function>pg_ls_dir_metadata</function> ( <parameter>dirname</parameter> <type>text</type>
+        <optional>, <parameter>missing_ok</parameter> <type>boolean</type>,
+        <parameter>include_dot_dirs</parameter> <type>boolean</type> </optional> )
+        <returnvalue>setof record</returnvalue>
+        ( <parameter>filename</parameter> <type>text</type>,
+        <parameter>size</parameter> <type>bigint</type>,
+        <parameter>modification</parameter> <type>timestamp with time zone</type> )
+       </para>
+       <para>
+        For each file in the specified directory, list the file and its
+        metadata.
+        Restricted to superusers by default, but other users can be granted
+        EXECUTE to run the function.
+       </para></entry>
+      </row>
+
       <row>
        <entry role="func_table_entry"><para role="func_signature">
         <indexterm>
diff --git a/src/backend/catalog/system_views.sql b/src/backend/catalog/system_views.sql
index b140c210bc..93d4f1e4a6 100644
--- a/src/backend/catalog/system_views.sql
+++ b/src/backend/catalog/system_views.sql
@@ -1506,6 +1506,7 @@ REVOKE EXECUTE ON FUNCTION pg_stat_file(text,boolean) FROM public;
 
 REVOKE EXECUTE ON FUNCTION pg_ls_dir(text) FROM public;
 REVOKE EXECUTE ON FUNCTION pg_ls_dir(text,boolean,boolean) FROM public;
+REVOKE EXECUTE ON FUNCTION pg_ls_dir_metadata(text,boolean,boolean) FROM public;
 
 --
 -- We also set up some things as accessible to standard roles.
diff --git a/src/backend/utils/adt/genfile.c b/src/backend/utils/adt/genfile.c
index 9f4927220b..7351615f8a 100644
--- a/src/backend/utils/adt/genfile.c
+++ b/src/backend/utils/adt/genfile.c
@@ -36,6 +36,21 @@
 #include "utils/syscache.h"
 #include "utils/timestamp.h"
 
+static Datum pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, int flags);
+
+#define	LS_DIR_ISDIR				(1<<0) /* Show column: isdir */
+#define	LS_DIR_METADATA				(1<<1) /* Show columns: mtime, size */
+#define	LS_DIR_MISSING_OK			(1<<2) /* Ignore ENOENT if the toplevel dir is missing */
+#define	LS_DIR_SKIP_DOT_DIRS		(1<<3) /* Do not show . or .. */
+#define	LS_DIR_SKIP_HIDDEN			(1<<4) /* Do not show anything begining with . */
+#define	LS_DIR_SKIP_DIRS			(1<<5) /* Do not show directories */
+#define	LS_DIR_SKIP_SPECIAL			(1<<6) /* Do not show special file types */
+
+/*
+ * Shortcut for the historic behavior of the pg_ls_* functions (not including
+ * pg_ls_dir, which skips different files and doesn't show metadata).
+ */
+#define LS_DIR_HISTORIC				(LS_DIR_SKIP_DIRS | LS_DIR_SKIP_HIDDEN | LS_DIR_SKIP_SPECIAL | LS_DIR_METADATA)
 
 /*
  * Convert a "text" filename argument to C string, and check it's allowable.
@@ -449,6 +464,11 @@ pg_stat_file(PG_FUNCTION_ARGS)
 	values[4] = TimestampTzGetDatum(time_t_to_timestamptz(fst.st_ctime));
 #endif
 	values[5] = BoolGetDatum(S_ISDIR(fst.st_mode));
+#ifdef WIN32
+	/* Links should have isdir=false */
+	if (pgwin32_is_junction(filename))
+		values[5] = BoolGetDatum(false);
+#endif
 
 	tuple = heap_form_tuple(tupdesc, values, isnull);
 
@@ -476,79 +496,9 @@ pg_stat_file_1arg(PG_FUNCTION_ARGS)
 Datum
 pg_ls_dir(PG_FUNCTION_ARGS)
 {
-	ReturnSetInfo *rsinfo = (ReturnSetInfo *) fcinfo->resultinfo;
-	char	   *location;
-	bool		missing_ok = false;
-	bool		include_dot_dirs = false;
-	bool		randomAccess;
-	TupleDesc	tupdesc;
-	Tuplestorestate *tupstore;
-	DIR		   *dirdesc;
-	struct dirent *de;
-	MemoryContext oldcontext;
-
-	location = convert_and_check_filename(PG_GETARG_TEXT_PP(0));
-
-	/* check the optional arguments */
-	if (PG_NARGS() == 3)
-	{
-		if (!PG_ARGISNULL(1))
-			missing_ok = PG_GETARG_BOOL(1);
-		if (!PG_ARGISNULL(2))
-			include_dot_dirs = PG_GETARG_BOOL(2);
-	}
-
-	/* check to see if caller supports us returning a tuplestore */
-	if (rsinfo == NULL || !IsA(rsinfo, ReturnSetInfo))
-		ereport(ERROR,
-				(errcode(ERRCODE_FEATURE_NOT_SUPPORTED),
-				 errmsg("set-valued function called in context that cannot accept a set")));
-	if (!(rsinfo->allowedModes & SFRM_Materialize))
-		ereport(ERROR,
-				(errcode(ERRCODE_SYNTAX_ERROR),
-				 errmsg("materialize mode required, but it is not allowed in this context")));
-
-	/* The tupdesc and tuplestore must be created in ecxt_per_query_memory */
-	oldcontext = MemoryContextSwitchTo(rsinfo->econtext->ecxt_per_query_memory);
-
-	tupdesc = CreateTemplateTupleDesc(1);
-	TupleDescInitEntry(tupdesc, (AttrNumber) 1, "pg_ls_dir", TEXTOID, -1, 0);
-
-	randomAccess = (rsinfo->allowedModes & SFRM_Materialize_Random) != 0;
-	tupstore = tuplestore_begin_heap(randomAccess, false, work_mem);
-	rsinfo->returnMode = SFRM_Materialize;
-	rsinfo->setResult = tupstore;
-	rsinfo->setDesc = tupdesc;
-
-	MemoryContextSwitchTo(oldcontext);
-
-	dirdesc = AllocateDir(location);
-	if (!dirdesc)
-	{
-		/* Return empty tuplestore if appropriate */
-		if (missing_ok && errno == ENOENT)
-			return (Datum) 0;
-		/* Otherwise, we can let ReadDir() throw the error */
-	}
-
-	while ((de = ReadDir(dirdesc, location)) != NULL)
-	{
-		Datum		values[1];
-		bool		nulls[1];
-
-		if (!include_dot_dirs &&
-			(strcmp(de->d_name, ".") == 0 ||
-			 strcmp(de->d_name, "..") == 0))
-			continue;
-
-		values[0] = CStringGetTextDatum(de->d_name);
-		nulls[0] = false;
-
-		tuplestore_putvalues(tupstore, tupdesc, values, nulls);
-	}
-
-	FreeDir(dirdesc);
-	return (Datum) 0;
+	text	*filename_t = PG_GETARG_TEXT_PP(0);
+	char	*filename = convert_and_check_filename(filename_t);
+	return pg_ls_dir_files(fcinfo, filename, LS_DIR_SKIP_DOT_DIRS);
 }
 
 /*
@@ -561,17 +511,19 @@ pg_ls_dir(PG_FUNCTION_ARGS)
 Datum
 pg_ls_dir_1arg(PG_FUNCTION_ARGS)
 {
-	return pg_ls_dir(fcinfo);
+	text	*filename_t = PG_GETARG_TEXT_PP(0);
+	char	*filename = convert_and_check_filename(filename_t);
+	return pg_ls_dir_files(fcinfo, filename, LS_DIR_SKIP_DOT_DIRS);
 }
 
 /*
  * Generic function to return a directory listing of files.
  *
- * If the directory isn't there, silently return an empty set if missing_ok.
+ * If the directory isn't there, silently return an empty set if MISSING_OK.
  * Other unreadable-directory cases throw an error.
  */
 static Datum
-pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, bool missing_ok)
+pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, int flags)
 {
 	ReturnSetInfo *rsinfo = (ReturnSetInfo *) fcinfo->resultinfo;
 	bool		randomAccess;
@@ -580,6 +532,32 @@ pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, bool missing_ok)
 	DIR		   *dirdesc;
 	struct dirent *de;
 	MemoryContext oldcontext;
+	TypeFuncClass	tuptype ;
+
+	/* isdir depends on metadata */
+	Assert(!(flags&LS_DIR_ISDIR) || (flags&LS_DIR_METADATA));
+	/* Unreasonable to show isdir and skip dirs */
+	Assert(!(flags&LS_DIR_ISDIR) || !(flags&LS_DIR_SKIP_DIRS));
+
+	/* check the optional arguments */
+	if (PG_NARGS() == 3)
+	{
+		if (!PG_ARGISNULL(1))
+		{
+			if (PG_GETARG_BOOL(1))
+				flags |= LS_DIR_MISSING_OK;
+			else
+				flags &= ~LS_DIR_MISSING_OK;
+		}
+
+		if (!PG_ARGISNULL(2))
+		{
+			if (PG_GETARG_BOOL(2))
+				flags &= ~LS_DIR_SKIP_DOT_DIRS;
+			else
+				flags |= LS_DIR_SKIP_DOT_DIRS;
+		}
+	}
 
 	/* check to see if caller supports us returning a tuplestore */
 	if (rsinfo == NULL || !IsA(rsinfo, ReturnSetInfo))
@@ -594,8 +572,20 @@ pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, bool missing_ok)
 	/* The tupdesc and tuplestore must be created in ecxt_per_query_memory */
 	oldcontext = MemoryContextSwitchTo(rsinfo->econtext->ecxt_per_query_memory);
 
-	if (get_call_result_type(fcinfo, NULL, &tupdesc) != TYPEFUNC_COMPOSITE)
-		elog(ERROR, "return type must be a row type");
+	tuptype = get_call_result_type(fcinfo, NULL, &tupdesc);
+	if (flags & LS_DIR_METADATA)
+	{
+		if (tuptype != TYPEFUNC_COMPOSITE)
+			elog(ERROR, "return type must be a row type");
+	}
+	else
+	{
+		/* pg_ls_dir returns a simple scalar */
+		if (tuptype != TYPEFUNC_SCALAR)
+			elog(ERROR, "return type must be a scalar type");
+		tupdesc = CreateTemplateTupleDesc(1);
+		TupleDescInitEntry(tupdesc, (AttrNumber) 1, "column", TEXTOID, -1, 0);
+	}
 
 	randomAccess = (rsinfo->allowedModes & SFRM_Materialize_Random) != 0;
 	tupstore = tuplestore_begin_heap(randomAccess, false, work_mem);
@@ -614,20 +604,27 @@ pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, bool missing_ok)
 	if (!dirdesc)
 	{
 		/* Return empty tuplestore if appropriate */
-		if (missing_ok && errno == ENOENT)
+		if (flags & LS_DIR_MISSING_OK && errno == ENOENT)
 			return (Datum) 0;
 		/* Otherwise, we can let ReadDir() throw the error */
 	}
 
 	while ((de = ReadDir(dirdesc, dir)) != NULL)
 	{
-		Datum		values[3];
-		bool		nulls[3];
+		Datum		values[4];
+		bool		nulls[4];
 		char		path[MAXPGPATH * 2];
 		struct stat attrib;
 
-		/* Skip hidden files */
-		if (de->d_name[0] == '.')
+		/* Skip dot dirs? */
+		if (flags & LS_DIR_SKIP_DOT_DIRS &&
+			(strcmp(de->d_name, ".") == 0 ||
+			 strcmp(de->d_name, "..") == 0))
+			continue;
+
+		/* Skip hidden files? */
+		if (flags & LS_DIR_SKIP_HIDDEN &&
+			de->d_name[0] == '.')
 			continue;
 
 		/* Get the file info */
@@ -642,13 +639,34 @@ pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, bool missing_ok)
 					 errmsg("could not stat file \"%s\": %m", path)));
 		}
 
-		/* Ignore anything but regular files */
-		if (!S_ISREG(attrib.st_mode))
-			continue;
+		/* Skip dirs or special files? */
+		if (S_ISDIR(attrib.st_mode))
+		{
+			if (flags & LS_DIR_SKIP_DIRS)
+				continue;
+		}
+		else if (!S_ISREG(attrib.st_mode))
+		{
+			if (flags & LS_DIR_SKIP_SPECIAL)
+				continue;
+		}
 
 		values[0] = CStringGetTextDatum(de->d_name);
-		values[1] = Int64GetDatum((int64) attrib.st_size);
-		values[2] = TimestampTzGetDatum(time_t_to_timestamptz(attrib.st_mtime));
+		if (flags & LS_DIR_METADATA)
+		{
+			values[1] = Int64GetDatum((int64) attrib.st_size);
+			values[2] = TimestampTzGetDatum(time_t_to_timestamptz(attrib.st_mtime));
+			if (flags & LS_DIR_ISDIR)
+			{
+				values[3] = BoolGetDatum(S_ISDIR(attrib.st_mode));
+#ifdef WIN32
+				/* Links should have isdir=false */
+				if (pgwin32_is_junction(path))
+					values[3] = BoolGetDatum(false);
+#endif
+			}
+		}
+
 		memset(nulls, 0, sizeof(nulls));
 
 		tuplestore_putvalues(tupstore, tupdesc, values, nulls);
@@ -662,14 +680,14 @@ pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, bool missing_ok)
 Datum
 pg_ls_logdir(PG_FUNCTION_ARGS)
 {
-	return pg_ls_dir_files(fcinfo, Log_directory, false);
+	return pg_ls_dir_files(fcinfo, Log_directory, LS_DIR_HISTORIC);
 }
 
 /* Function to return the list of files in the WAL directory */
 Datum
 pg_ls_waldir(PG_FUNCTION_ARGS)
 {
-	return pg_ls_dir_files(fcinfo, XLOGDIR, false);
+	return pg_ls_dir_files(fcinfo, XLOGDIR, LS_DIR_HISTORIC);
 }
 
 /*
@@ -687,7 +705,8 @@ pg_ls_tmpdir(FunctionCallInfo fcinfo, Oid tblspc)
 						tblspc)));
 
 	TempTablespacePath(path, tblspc);
-	return pg_ls_dir_files(fcinfo, path, true);
+	return pg_ls_dir_files(fcinfo, path,
+			LS_DIR_HISTORIC | LS_DIR_MISSING_OK);
 }
 
 /*
@@ -716,5 +735,33 @@ pg_ls_tmpdir_1arg(PG_FUNCTION_ARGS)
 Datum
 pg_ls_archive_statusdir(PG_FUNCTION_ARGS)
 {
-	return pg_ls_dir_files(fcinfo, XLOGDIR "/archive_status", true);
+	return pg_ls_dir_files(fcinfo, XLOGDIR "/archive_status",
+			LS_DIR_HISTORIC | LS_DIR_MISSING_OK);
+}
+
+/*
+ * Return the list of files and metadata in an arbitrary directory.
+ */
+Datum
+pg_ls_dir_metadata(PG_FUNCTION_ARGS)
+{
+	char	*dirname = convert_and_check_filename(PG_GETARG_TEXT_PP(0));
+
+	return pg_ls_dir_files(fcinfo, dirname,
+			LS_DIR_METADATA | LS_DIR_SKIP_SPECIAL | LS_DIR_ISDIR);
+}
+
+/*
+ * Return the list of files and metadata in an arbitrary directory.
+ * note: this wrapper is necessary to pass the sanity check in opr_sanity,
+ * which checks that all built-in functions that share the implementing C
+ * function take the same number of arguments.
+ */
+Datum
+pg_ls_dir_metadata_1arg(PG_FUNCTION_ARGS)
+{
+	char	*dirname = convert_and_check_filename(PG_GETARG_TEXT_PP(0));
+
+	return pg_ls_dir_files(fcinfo, dirname,
+			LS_DIR_METADATA | LS_DIR_SKIP_SPECIAL | LS_DIR_ISDIR);
 }
diff --git a/src/include/catalog/pg_proc.dat b/src/include/catalog/pg_proc.dat
index fc2202b843..8e88c549d9 100644
--- a/src/include/catalog/pg_proc.dat
+++ b/src/include/catalog/pg_proc.dat
@@ -10976,6 +10976,18 @@
   proallargtypes => '{oid,text,int8,timestamptz}', proargmodes => '{i,o,o,o}',
   proargnames => '{tablespace,name,size,modification}',
   prosrc => 'pg_ls_tmpdir_1arg' },
+{ oid => '9979', descr => 'list directory with metadata',
+  proname => 'pg_ls_dir_metadata', procost => '10', prorows => '20', proretset => 't',
+  provolatile => 'v', prorettype => 'record', proargtypes => 'text bool bool',
+  proallargtypes => '{text,bool,bool,text,int8,timestamptz,bool}', proargmodes => '{i,i,i,o,o,o,o}',
+  proargnames => '{dirname,missing_ok,include_dot_dirs,filename,size,modification,isdir}',
+  prosrc => 'pg_ls_dir_metadata' },
+{ oid => '9980', descr => 'list directory with metadata',
+  proname => 'pg_ls_dir_metadata', procost => '10', prorows => '20', proretset => 't',
+  provolatile => 'v', prorettype => 'record', proargtypes => 'text',
+  proallargtypes => '{text,text,int8,timestamptz,bool}', proargmodes => '{i,o,o,o,o}',
+  proargnames => '{dirname,filename,size,modification,isdir}',
+  prosrc => 'pg_ls_dir_metadata_1arg' },
 
 # hash partitioning constraint function
 { oid => '5028', descr => 'hash partition CHECK constraint',
diff --git a/src/test/regress/expected/misc_functions.out b/src/test/regress/expected/misc_functions.out
index edbfd9abc1..fd7c3c791f 100644
--- a/src/test/regress/expected/misc_functions.out
+++ b/src/test/regress/expected/misc_functions.out
@@ -225,6 +225,30 @@ select * from pg_stat_file('.') limit 0;
 ------+--------+--------------+--------+----------+-------
 (0 rows)
 
+-- This tests the missing_ok parameter, which causes pg_ls_tmpdir to succeed even if the tmpdir doesn't exist yet
+-- The name='' condition is never true, so the function runs to completion but returns zero rows.
+select * from pg_ls_tmpdir() where name='Does not exist';
+ name | size | modification 
+------+------+--------------
+(0 rows)
+
+select filename, isdir from pg_ls_dir_metadata('.') where filename='.';
+ filename | isdir 
+----------+-------
+ .        | t
+(1 row)
+
+select filename, isdir from pg_ls_dir_metadata('.', false, false) where filename='.'; -- include_dot_dirs=false
+ filename | isdir 
+----------+-------
+(0 rows)
+
+-- Check that expected columns are present
+select * from pg_ls_dir_metadata('.') limit 0;
+ filename | size | modification | isdir 
+----------+------+--------------+-------
+(0 rows)
+
 --
 -- Test adding a support function to a subject function
 --
diff --git a/src/test/regress/input/tablespace.source b/src/test/regress/input/tablespace.source
index a5f61a35dc..0b9cfe615e 100644
--- a/src/test/regress/input/tablespace.source
+++ b/src/test/regress/input/tablespace.source
@@ -11,6 +11,11 @@ DROP TABLESPACE regress_tblspacewith;
 -- create a tablespace we can use
 CREATE TABLESPACE regress_tblspace LOCATION '@testtablespace@';
 
+-- This tests the missing_ok parameter, which causes pg_ls_tmpdir to succeed even if the tmpdir doesn't exist yet
+-- The name='' condition is never true, so the function runs to completion but returns zero rows.
+-- The query is written to ERROR if the tablespace doesn't exist, rather than silently failing to call pg_ls_tmpdir()
+SELECT c.* FROM (SELECT oid FROM pg_tablespace b WHERE b.spcname='regress_tblspace' UNION SELECT 0 ORDER BY 1 DESC LIMIT 1) AS b , pg_ls_tmpdir(oid) AS c WHERE c.name='Does not exist';
+
 -- try setting and resetting some properties for the new tablespace
 ALTER TABLESPACE regress_tblspace SET (random_page_cost = 1.0, seq_page_cost = 1.1);
 ALTER TABLESPACE regress_tblspace SET (some_nonexistent_parameter = true);  -- fail
diff --git a/src/test/regress/output/tablespace.source b/src/test/regress/output/tablespace.source
index 162b591b31..a42714bf40 100644
--- a/src/test/regress/output/tablespace.source
+++ b/src/test/regress/output/tablespace.source
@@ -13,6 +13,14 @@ SELECT spcoptions FROM pg_tablespace WHERE spcname = 'regress_tblspacewith';
 DROP TABLESPACE regress_tblspacewith;
 -- create a tablespace we can use
 CREATE TABLESPACE regress_tblspace LOCATION '@testtablespace@';
+-- This tests the missing_ok parameter, which causes pg_ls_tmpdir to succeed even if the tmpdir doesn't exist yet
+-- The name='' condition is never true, so the function runs to completion but returns zero rows.
+-- The query is written to ERROR if the tablespace doesn't exist, rather than silently failing to call pg_ls_tmpdir()
+SELECT c.* FROM (SELECT oid FROM pg_tablespace b WHERE b.spcname='regress_tblspace' UNION SELECT 0 ORDER BY 1 DESC LIMIT 1) AS b , pg_ls_tmpdir(oid) AS c WHERE c.name='Does not exist';
+ name | size | modification 
+------+------+--------------
+(0 rows)
+
 -- try setting and resetting some properties for the new tablespace
 ALTER TABLESPACE regress_tblspace SET (random_page_cost = 1.0, seq_page_cost = 1.1);
 ALTER TABLESPACE regress_tblspace SET (some_nonexistent_parameter = true);  -- fail
diff --git a/src/test/regress/sql/misc_functions.sql b/src/test/regress/sql/misc_functions.sql
index b32a3a4b74..64d9d70874 100644
--- a/src/test/regress/sql/misc_functions.sql
+++ b/src/test/regress/sql/misc_functions.sql
@@ -68,6 +68,17 @@ select pg_ls_dir('does not exist'); -- fails with missingok=false
 -- Check that expected columns are present
 select * from pg_stat_file('.') limit 0;
 
+-- This tests the missing_ok parameter, which causes pg_ls_tmpdir to succeed even if the tmpdir doesn't exist yet
+-- The name='' condition is never true, so the function runs to completion but returns zero rows.
+select * from pg_ls_tmpdir() where name='Does not exist';
+
+select filename, isdir from pg_ls_dir_metadata('.') where filename='.';
+
+select filename, isdir from pg_ls_dir_metadata('.', false, false) where filename='.'; -- include_dot_dirs=false
+
+-- Check that expected columns are present
+select * from pg_ls_dir_metadata('.') limit 0;
+
 --
 -- Test adding a support function to a subject function
 --
-- 
2.17.0


--SBikYMzjhZGK9d4p
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
 filename="v25-0005-pg_ls_tmpdir-to-show-directories-and-isdir-argum.patch"



^ permalink  raw  reply  [nested|flat] 68+ messages in thread

* [PATCH v26 04/11] Add pg_ls_dir_metadata to list a dir with file metadata..
@ 2020-03-10 03:40  Justin Pryzby <[email protected]>
  0 siblings, 0 replies; 68+ messages in thread

From: Justin Pryzby @ 2020-03-10 03:40 UTC (permalink / raw)

Generalize pg_ls_dir_files and retire pg_ls_dir

Need catversion bumped?
---
 doc/src/sgml/func.sgml                       |  21 ++
 src/backend/catalog/system_views.sql         |   1 +
 src/backend/utils/adt/genfile.c              | 233 +++++++++++--------
 src/include/catalog/pg_proc.dat              |  12 +
 src/test/regress/expected/misc_functions.out |  24 ++
 src/test/regress/input/tablespace.source     |   5 +
 src/test/regress/output/tablespace.source    |   8 +
 src/test/regress/sql/misc_functions.sql      |  11 +
 8 files changed, 222 insertions(+), 93 deletions(-)

diff --git a/doc/src/sgml/func.sgml b/doc/src/sgml/func.sgml
index 2707e757ca..79d30a1568 100644
--- a/doc/src/sgml/func.sgml
+++ b/doc/src/sgml/func.sgml
@@ -25799,6 +25799,27 @@ postgres=# SELECT * FROM pg_walfile_name_offset(pg_stop_backup());
        </para></entry>
       </row>
 
+      <row>
+       <entry role="func_table_entry"><para role="func_signature">
+        <indexterm>
+         <primary>pg_ls_dir_metadata</primary>
+        </indexterm>
+        <function>pg_ls_dir_metadata</function> ( <parameter>dirname</parameter> <type>text</type>
+        <optional>, <parameter>missing_ok</parameter> <type>boolean</type>,
+        <parameter>include_dot_dirs</parameter> <type>boolean</type> </optional> )
+        <returnvalue>setof record</returnvalue>
+        ( <parameter>filename</parameter> <type>text</type>,
+        <parameter>size</parameter> <type>bigint</type>,
+        <parameter>modification</parameter> <type>timestamp with time zone</type> )
+       </para>
+       <para>
+        For each file in the specified directory, list the file and its
+        metadata.
+        Restricted to superusers by default, but other users can be granted
+        EXECUTE to run the function.
+       </para></entry>
+      </row>
+
       <row>
        <entry role="func_table_entry"><para role="func_signature">
         <indexterm>
diff --git a/src/backend/catalog/system_views.sql b/src/backend/catalog/system_views.sql
index b140c210bc..93d4f1e4a6 100644
--- a/src/backend/catalog/system_views.sql
+++ b/src/backend/catalog/system_views.sql
@@ -1506,6 +1506,7 @@ REVOKE EXECUTE ON FUNCTION pg_stat_file(text,boolean) FROM public;
 
 REVOKE EXECUTE ON FUNCTION pg_ls_dir(text) FROM public;
 REVOKE EXECUTE ON FUNCTION pg_ls_dir(text,boolean,boolean) FROM public;
+REVOKE EXECUTE ON FUNCTION pg_ls_dir_metadata(text,boolean,boolean) FROM public;
 
 --
 -- We also set up some things as accessible to standard roles.
diff --git a/src/backend/utils/adt/genfile.c b/src/backend/utils/adt/genfile.c
index 9f4927220b..198a99391a 100644
--- a/src/backend/utils/adt/genfile.c
+++ b/src/backend/utils/adt/genfile.c
@@ -36,6 +36,21 @@
 #include "utils/syscache.h"
 #include "utils/timestamp.h"
 
+static Datum pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, int flags);
+
+#define	LS_DIR_ISDIR				(1<<0) /* Show column: isdir */
+#define	LS_DIR_METADATA				(1<<1) /* Show columns: mtime, size */
+#define	LS_DIR_MISSING_OK			(1<<2) /* Ignore ENOENT if the toplevel dir is missing */
+#define	LS_DIR_SKIP_DOT_DIRS		(1<<3) /* Do not show . or .. */
+#define	LS_DIR_SKIP_HIDDEN			(1<<4) /* Do not show anything begining with . */
+#define	LS_DIR_SKIP_DIRS			(1<<5) /* Do not show directories */
+#define	LS_DIR_SKIP_SPECIAL			(1<<6) /* Do not show special file types */
+
+/*
+ * Shortcut for the historic behavior of the pg_ls_* functions (not including
+ * pg_ls_dir, which skips different files and doesn't show metadata).
+ */
+#define LS_DIR_HISTORIC				(LS_DIR_SKIP_DIRS | LS_DIR_SKIP_HIDDEN | LS_DIR_SKIP_SPECIAL | LS_DIR_METADATA)
 
 /*
  * Convert a "text" filename argument to C string, and check it's allowable.
@@ -449,6 +464,11 @@ pg_stat_file(PG_FUNCTION_ARGS)
 	values[4] = TimestampTzGetDatum(time_t_to_timestamptz(fst.st_ctime));
 #endif
 	values[5] = BoolGetDatum(S_ISDIR(fst.st_mode));
+#ifdef WIN32
+	/* Links should have isdir=false */
+	if (pgwin32_is_junction(filename))
+		values[5] = BoolGetDatum(false);
+#endif
 
 	tuple = heap_form_tuple(tupdesc, values, isnull);
 
@@ -476,79 +496,9 @@ pg_stat_file_1arg(PG_FUNCTION_ARGS)
 Datum
 pg_ls_dir(PG_FUNCTION_ARGS)
 {
-	ReturnSetInfo *rsinfo = (ReturnSetInfo *) fcinfo->resultinfo;
-	char	   *location;
-	bool		missing_ok = false;
-	bool		include_dot_dirs = false;
-	bool		randomAccess;
-	TupleDesc	tupdesc;
-	Tuplestorestate *tupstore;
-	DIR		   *dirdesc;
-	struct dirent *de;
-	MemoryContext oldcontext;
-
-	location = convert_and_check_filename(PG_GETARG_TEXT_PP(0));
-
-	/* check the optional arguments */
-	if (PG_NARGS() == 3)
-	{
-		if (!PG_ARGISNULL(1))
-			missing_ok = PG_GETARG_BOOL(1);
-		if (!PG_ARGISNULL(2))
-			include_dot_dirs = PG_GETARG_BOOL(2);
-	}
-
-	/* check to see if caller supports us returning a tuplestore */
-	if (rsinfo == NULL || !IsA(rsinfo, ReturnSetInfo))
-		ereport(ERROR,
-				(errcode(ERRCODE_FEATURE_NOT_SUPPORTED),
-				 errmsg("set-valued function called in context that cannot accept a set")));
-	if (!(rsinfo->allowedModes & SFRM_Materialize))
-		ereport(ERROR,
-				(errcode(ERRCODE_SYNTAX_ERROR),
-				 errmsg("materialize mode required, but it is not allowed in this context")));
-
-	/* The tupdesc and tuplestore must be created in ecxt_per_query_memory */
-	oldcontext = MemoryContextSwitchTo(rsinfo->econtext->ecxt_per_query_memory);
-
-	tupdesc = CreateTemplateTupleDesc(1);
-	TupleDescInitEntry(tupdesc, (AttrNumber) 1, "pg_ls_dir", TEXTOID, -1, 0);
-
-	randomAccess = (rsinfo->allowedModes & SFRM_Materialize_Random) != 0;
-	tupstore = tuplestore_begin_heap(randomAccess, false, work_mem);
-	rsinfo->returnMode = SFRM_Materialize;
-	rsinfo->setResult = tupstore;
-	rsinfo->setDesc = tupdesc;
-
-	MemoryContextSwitchTo(oldcontext);
-
-	dirdesc = AllocateDir(location);
-	if (!dirdesc)
-	{
-		/* Return empty tuplestore if appropriate */
-		if (missing_ok && errno == ENOENT)
-			return (Datum) 0;
-		/* Otherwise, we can let ReadDir() throw the error */
-	}
-
-	while ((de = ReadDir(dirdesc, location)) != NULL)
-	{
-		Datum		values[1];
-		bool		nulls[1];
-
-		if (!include_dot_dirs &&
-			(strcmp(de->d_name, ".") == 0 ||
-			 strcmp(de->d_name, "..") == 0))
-			continue;
-
-		values[0] = CStringGetTextDatum(de->d_name);
-		nulls[0] = false;
-
-		tuplestore_putvalues(tupstore, tupdesc, values, nulls);
-	}
-
-	FreeDir(dirdesc);
-	return (Datum) 0;
+	text	*filename_t = PG_GETARG_TEXT_PP(0);
+	char	*filename = convert_and_check_filename(filename_t);
+	return pg_ls_dir_files(fcinfo, filename, LS_DIR_SKIP_DOT_DIRS);
 }
 
 /*
@@ -561,17 +511,19 @@ pg_ls_dir(PG_FUNCTION_ARGS)
 Datum
 pg_ls_dir_1arg(PG_FUNCTION_ARGS)
 {
-	return pg_ls_dir(fcinfo);
+	text	*filename_t = PG_GETARG_TEXT_PP(0);
+	char	*filename = convert_and_check_filename(filename_t);
+	return pg_ls_dir_files(fcinfo, filename, LS_DIR_SKIP_DOT_DIRS);
 }
 
 /*
- * Generic function to return a directory listing of files.
+ * Generic function to return a directory listing of files (and optionally dirs).
  *
- * If the directory isn't there, silently return an empty set if missing_ok.
+ * If the directory isn't there, silently return an empty set if MISSING_OK.
  * Other unreadable-directory cases throw an error.
  */
 static Datum
-pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, bool missing_ok)
+pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, int flags)
 {
 	ReturnSetInfo *rsinfo = (ReturnSetInfo *) fcinfo->resultinfo;
 	bool		randomAccess;
@@ -580,6 +532,32 @@ pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, bool missing_ok)
 	DIR		   *dirdesc;
 	struct dirent *de;
 	MemoryContext oldcontext;
+	TypeFuncClass	tuptype ;
+
+	/* isdir depends on metadata */
+	Assert(!(flags&LS_DIR_ISDIR) || (flags&LS_DIR_METADATA));
+	/* Unreasonable to show isdir and skip dirs */
+	Assert(!(flags&LS_DIR_ISDIR) || !(flags&LS_DIR_SKIP_DIRS));
+
+	/* check the optional arguments */
+	if (PG_NARGS() == 3)
+	{
+		if (!PG_ARGISNULL(1))
+		{
+			if (PG_GETARG_BOOL(1))
+				flags |= LS_DIR_MISSING_OK;
+			else
+				flags &= ~LS_DIR_MISSING_OK;
+		}
+
+		if (!PG_ARGISNULL(2))
+		{
+			if (PG_GETARG_BOOL(2))
+				flags &= ~LS_DIR_SKIP_DOT_DIRS;
+			else
+				flags |= LS_DIR_SKIP_DOT_DIRS;
+		}
+	}
 
 	/* check to see if caller supports us returning a tuplestore */
 	if (rsinfo == NULL || !IsA(rsinfo, ReturnSetInfo))
@@ -594,8 +572,20 @@ pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, bool missing_ok)
 	/* The tupdesc and tuplestore must be created in ecxt_per_query_memory */
 	oldcontext = MemoryContextSwitchTo(rsinfo->econtext->ecxt_per_query_memory);
 
-	if (get_call_result_type(fcinfo, NULL, &tupdesc) != TYPEFUNC_COMPOSITE)
-		elog(ERROR, "return type must be a row type");
+	tuptype = get_call_result_type(fcinfo, NULL, &tupdesc);
+	if (flags & LS_DIR_METADATA)
+	{
+		if (tuptype != TYPEFUNC_COMPOSITE)
+			elog(ERROR, "return type must be a row type");
+	}
+	else
+	{
+		/* pg_ls_dir returns a simple scalar */
+		if (tuptype != TYPEFUNC_SCALAR)
+			elog(ERROR, "return type must be a scalar type");
+		tupdesc = CreateTemplateTupleDesc(1);
+		TupleDescInitEntry(tupdesc, (AttrNumber) 1, "column", TEXTOID, -1, 0);
+	}
 
 	randomAccess = (rsinfo->allowedModes & SFRM_Materialize_Random) != 0;
 	tupstore = tuplestore_begin_heap(randomAccess, false, work_mem);
@@ -614,20 +604,27 @@ pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, bool missing_ok)
 	if (!dirdesc)
 	{
 		/* Return empty tuplestore if appropriate */
-		if (missing_ok && errno == ENOENT)
+		if (flags & LS_DIR_MISSING_OK && errno == ENOENT)
 			return (Datum) 0;
 		/* Otherwise, we can let ReadDir() throw the error */
 	}
 
 	while ((de = ReadDir(dirdesc, dir)) != NULL)
 	{
-		Datum		values[3];
-		bool		nulls[3];
+		Datum		values[4];
+		bool		nulls[4];
 		char		path[MAXPGPATH * 2];
 		struct stat attrib;
 
-		/* Skip hidden files */
-		if (de->d_name[0] == '.')
+		/* Skip dot dirs? */
+		if (flags & LS_DIR_SKIP_DOT_DIRS &&
+			(strcmp(de->d_name, ".") == 0 ||
+			 strcmp(de->d_name, "..") == 0))
+			continue;
+
+		/* Skip hidden files? */
+		if (flags & LS_DIR_SKIP_HIDDEN &&
+			de->d_name[0] == '.')
 			continue;
 
 		/* Get the file info */
@@ -642,13 +639,34 @@ pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, bool missing_ok)
 					 errmsg("could not stat file \"%s\": %m", path)));
 		}
 
-		/* Ignore anything but regular files */
-		if (!S_ISREG(attrib.st_mode))
-			continue;
+		/* Skip dirs or special files? */
+		if (S_ISDIR(attrib.st_mode))
+		{
+			if (flags & LS_DIR_SKIP_DIRS)
+				continue;
+		}
+		else if (!S_ISREG(attrib.st_mode))
+		{
+			if (flags & LS_DIR_SKIP_SPECIAL)
+				continue;
+		}
 
 		values[0] = CStringGetTextDatum(de->d_name);
-		values[1] = Int64GetDatum((int64) attrib.st_size);
-		values[2] = TimestampTzGetDatum(time_t_to_timestamptz(attrib.st_mtime));
+		if (flags & LS_DIR_METADATA)
+		{
+			values[1] = Int64GetDatum((int64) attrib.st_size);
+			values[2] = TimestampTzGetDatum(time_t_to_timestamptz(attrib.st_mtime));
+			if (flags & LS_DIR_ISDIR)
+			{
+				values[3] = BoolGetDatum(S_ISDIR(attrib.st_mode));
+#ifdef WIN32
+				/* Links should have isdir=false */
+				if (pgwin32_is_junction(path))
+					values[3] = BoolGetDatum(false);
+#endif
+			}
+		}
+
 		memset(nulls, 0, sizeof(nulls));
 
 		tuplestore_putvalues(tupstore, tupdesc, values, nulls);
@@ -662,14 +680,14 @@ pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, bool missing_ok)
 Datum
 pg_ls_logdir(PG_FUNCTION_ARGS)
 {
-	return pg_ls_dir_files(fcinfo, Log_directory, false);
+	return pg_ls_dir_files(fcinfo, Log_directory, LS_DIR_HISTORIC);
 }
 
 /* Function to return the list of files in the WAL directory */
 Datum
 pg_ls_waldir(PG_FUNCTION_ARGS)
 {
-	return pg_ls_dir_files(fcinfo, XLOGDIR, false);
+	return pg_ls_dir_files(fcinfo, XLOGDIR, LS_DIR_HISTORIC);
 }
 
 /*
@@ -687,7 +705,8 @@ pg_ls_tmpdir(FunctionCallInfo fcinfo, Oid tblspc)
 						tblspc)));
 
 	TempTablespacePath(path, tblspc);
-	return pg_ls_dir_files(fcinfo, path, true);
+	return pg_ls_dir_files(fcinfo, path,
+			LS_DIR_HISTORIC | LS_DIR_MISSING_OK);
 }
 
 /*
@@ -716,5 +735,33 @@ pg_ls_tmpdir_1arg(PG_FUNCTION_ARGS)
 Datum
 pg_ls_archive_statusdir(PG_FUNCTION_ARGS)
 {
-	return pg_ls_dir_files(fcinfo, XLOGDIR "/archive_status", true);
+	return pg_ls_dir_files(fcinfo, XLOGDIR "/archive_status",
+			LS_DIR_HISTORIC | LS_DIR_MISSING_OK);
+}
+
+/*
+ * Return the list of files and metadata in an arbitrary directory.
+ */
+Datum
+pg_ls_dir_metadata(PG_FUNCTION_ARGS)
+{
+	char	*dirname = convert_and_check_filename(PG_GETARG_TEXT_PP(0));
+
+	return pg_ls_dir_files(fcinfo, dirname,
+			LS_DIR_METADATA | LS_DIR_SKIP_SPECIAL | LS_DIR_ISDIR);
+}
+
+/*
+ * Return the list of files and metadata in an arbitrary directory.
+ * note: this wrapper is necessary to pass the sanity check in opr_sanity,
+ * which checks that all built-in functions that share the implementing C
+ * function take the same number of arguments.
+ */
+Datum
+pg_ls_dir_metadata_1arg(PG_FUNCTION_ARGS)
+{
+	char	*dirname = convert_and_check_filename(PG_GETARG_TEXT_PP(0));
+
+	return pg_ls_dir_files(fcinfo, dirname,
+			LS_DIR_METADATA | LS_DIR_SKIP_SPECIAL | LS_DIR_ISDIR);
 }
diff --git a/src/include/catalog/pg_proc.dat b/src/include/catalog/pg_proc.dat
index 22970f46cd..17b03d1abd 100644
--- a/src/include/catalog/pg_proc.dat
+++ b/src/include/catalog/pg_proc.dat
@@ -11266,6 +11266,18 @@
   proallargtypes => '{oid,text,int8,timestamptz}', proargmodes => '{i,o,o,o}',
   proargnames => '{tablespace,name,size,modification}',
   prosrc => 'pg_ls_tmpdir_1arg' },
+{ oid => '9979', descr => 'list directory with metadata',
+  proname => 'pg_ls_dir_metadata', procost => '10', prorows => '20', proretset => 't',
+  provolatile => 'v', prorettype => 'record', proargtypes => 'text bool bool',
+  proallargtypes => '{text,bool,bool,text,int8,timestamptz,bool}', proargmodes => '{i,i,i,o,o,o,o}',
+  proargnames => '{dirname,missing_ok,include_dot_dirs,filename,size,modification,isdir}',
+  prosrc => 'pg_ls_dir_metadata' },
+{ oid => '9980', descr => 'list directory with metadata',
+  proname => 'pg_ls_dir_metadata', procost => '10', prorows => '20', proretset => 't',
+  provolatile => 'v', prorettype => 'record', proargtypes => 'text',
+  proallargtypes => '{text,text,int8,timestamptz,bool}', proargmodes => '{i,o,o,o,o}',
+  proargnames => '{dirname,filename,size,modification,isdir}',
+  prosrc => 'pg_ls_dir_metadata_1arg' },
 
 # hash partitioning constraint function
 { oid => '5028', descr => 'hash partition CHECK constraint',
diff --git a/src/test/regress/expected/misc_functions.out b/src/test/regress/expected/misc_functions.out
index edbfd9abc1..fd7c3c791f 100644
--- a/src/test/regress/expected/misc_functions.out
+++ b/src/test/regress/expected/misc_functions.out
@@ -225,6 +225,30 @@ select * from pg_stat_file('.') limit 0;
 ------+--------+--------------+--------+----------+-------
 (0 rows)
 
+-- This tests the missing_ok parameter, which causes pg_ls_tmpdir to succeed even if the tmpdir doesn't exist yet
+-- The name='' condition is never true, so the function runs to completion but returns zero rows.
+select * from pg_ls_tmpdir() where name='Does not exist';
+ name | size | modification 
+------+------+--------------
+(0 rows)
+
+select filename, isdir from pg_ls_dir_metadata('.') where filename='.';
+ filename | isdir 
+----------+-------
+ .        | t
+(1 row)
+
+select filename, isdir from pg_ls_dir_metadata('.', false, false) where filename='.'; -- include_dot_dirs=false
+ filename | isdir 
+----------+-------
+(0 rows)
+
+-- Check that expected columns are present
+select * from pg_ls_dir_metadata('.') limit 0;
+ filename | size | modification | isdir 
+----------+------+--------------+-------
+(0 rows)
+
 --
 -- Test adding a support function to a subject function
 --
diff --git a/src/test/regress/input/tablespace.source b/src/test/regress/input/tablespace.source
index a5f61a35dc..0b9cfe615e 100644
--- a/src/test/regress/input/tablespace.source
+++ b/src/test/regress/input/tablespace.source
@@ -11,6 +11,11 @@ DROP TABLESPACE regress_tblspacewith;
 -- create a tablespace we can use
 CREATE TABLESPACE regress_tblspace LOCATION '@testtablespace@';
 
+-- This tests the missing_ok parameter, which causes pg_ls_tmpdir to succeed even if the tmpdir doesn't exist yet
+-- The name='' condition is never true, so the function runs to completion but returns zero rows.
+-- The query is written to ERROR if the tablespace doesn't exist, rather than silently failing to call pg_ls_tmpdir()
+SELECT c.* FROM (SELECT oid FROM pg_tablespace b WHERE b.spcname='regress_tblspace' UNION SELECT 0 ORDER BY 1 DESC LIMIT 1) AS b , pg_ls_tmpdir(oid) AS c WHERE c.name='Does not exist';
+
 -- try setting and resetting some properties for the new tablespace
 ALTER TABLESPACE regress_tblspace SET (random_page_cost = 1.0, seq_page_cost = 1.1);
 ALTER TABLESPACE regress_tblspace SET (some_nonexistent_parameter = true);  -- fail
diff --git a/src/test/regress/output/tablespace.source b/src/test/regress/output/tablespace.source
index 162b591b31..a42714bf40 100644
--- a/src/test/regress/output/tablespace.source
+++ b/src/test/regress/output/tablespace.source
@@ -13,6 +13,14 @@ SELECT spcoptions FROM pg_tablespace WHERE spcname = 'regress_tblspacewith';
 DROP TABLESPACE regress_tblspacewith;
 -- create a tablespace we can use
 CREATE TABLESPACE regress_tblspace LOCATION '@testtablespace@';
+-- This tests the missing_ok parameter, which causes pg_ls_tmpdir to succeed even if the tmpdir doesn't exist yet
+-- The name='' condition is never true, so the function runs to completion but returns zero rows.
+-- The query is written to ERROR if the tablespace doesn't exist, rather than silently failing to call pg_ls_tmpdir()
+SELECT c.* FROM (SELECT oid FROM pg_tablespace b WHERE b.spcname='regress_tblspace' UNION SELECT 0 ORDER BY 1 DESC LIMIT 1) AS b , pg_ls_tmpdir(oid) AS c WHERE c.name='Does not exist';
+ name | size | modification 
+------+------+--------------
+(0 rows)
+
 -- try setting and resetting some properties for the new tablespace
 ALTER TABLESPACE regress_tblspace SET (random_page_cost = 1.0, seq_page_cost = 1.1);
 ALTER TABLESPACE regress_tblspace SET (some_nonexistent_parameter = true);  -- fail
diff --git a/src/test/regress/sql/misc_functions.sql b/src/test/regress/sql/misc_functions.sql
index b32a3a4b74..64d9d70874 100644
--- a/src/test/regress/sql/misc_functions.sql
+++ b/src/test/regress/sql/misc_functions.sql
@@ -68,6 +68,17 @@ select pg_ls_dir('does not exist'); -- fails with missingok=false
 -- Check that expected columns are present
 select * from pg_stat_file('.') limit 0;
 
+-- This tests the missing_ok parameter, which causes pg_ls_tmpdir to succeed even if the tmpdir doesn't exist yet
+-- The name='' condition is never true, so the function runs to completion but returns zero rows.
+select * from pg_ls_tmpdir() where name='Does not exist';
+
+select filename, isdir from pg_ls_dir_metadata('.') where filename='.';
+
+select filename, isdir from pg_ls_dir_metadata('.', false, false) where filename='.'; -- include_dot_dirs=false
+
+-- Check that expected columns are present
+select * from pg_ls_dir_metadata('.') limit 0;
+
 --
 -- Test adding a support function to a subject function
 --
-- 
2.17.0


--yKkOmjQZXRsvHRX8
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
 filename="v26-0005-pg_ls_tmpdir-to-show-directories-and-isdir-argum.patch"



^ permalink  raw  reply  [nested|flat] 68+ messages in thread

* [PATCH v27 03/11] Add pg_ls_dir_metadata to list a dir with file metadata..
@ 2020-03-10 03:40  Justin Pryzby <[email protected]>
  0 siblings, 0 replies; 68+ messages in thread

From: Justin Pryzby @ 2020-03-10 03:40 UTC (permalink / raw)

Generalize pg_ls_dir_files and retire pg_ls_dir

Need catversion bumped?
---
 doc/src/sgml/func.sgml                       |  21 ++
 src/backend/catalog/system_views.sql         |   1 +
 src/backend/utils/adt/genfile.c              | 233 +++++++++++--------
 src/include/catalog/pg_proc.dat              |  12 +
 src/test/regress/expected/misc_functions.out |  24 ++
 src/test/regress/input/tablespace.source     |   5 +
 src/test/regress/output/tablespace.source    |   8 +
 src/test/regress/sql/misc_functions.sql      |  11 +
 8 files changed, 222 insertions(+), 93 deletions(-)

diff --git a/doc/src/sgml/func.sgml b/doc/src/sgml/func.sgml
index 977b75a531..76ae2f2dd8 100644
--- a/doc/src/sgml/func.sgml
+++ b/doc/src/sgml/func.sgml
@@ -25793,6 +25793,27 @@ postgres=# SELECT * FROM pg_walfile_name_offset(pg_stop_backup());
        </para></entry>
       </row>
 
+      <row>
+       <entry role="func_table_entry"><para role="func_signature">
+        <indexterm>
+         <primary>pg_ls_dir_metadata</primary>
+        </indexterm>
+        <function>pg_ls_dir_metadata</function> ( <parameter>dirname</parameter> <type>text</type>
+        <optional>, <parameter>missing_ok</parameter> <type>boolean</type>,
+        <parameter>include_dot_dirs</parameter> <type>boolean</type> </optional> )
+        <returnvalue>setof record</returnvalue>
+        ( <parameter>filename</parameter> <type>text</type>,
+        <parameter>size</parameter> <type>bigint</type>,
+        <parameter>modification</parameter> <type>timestamp with time zone</type> )
+       </para>
+       <para>
+        For each file in the specified directory, list the file and its
+        metadata.
+        Restricted to superusers by default, but other users can be granted
+        EXECUTE to run the function.
+       </para></entry>
+      </row>
+
       <row>
        <entry role="func_table_entry"><para role="func_signature">
         <indexterm>
diff --git a/src/backend/catalog/system_views.sql b/src/backend/catalog/system_views.sql
index 5f2541d316..fcd050b752 100644
--- a/src/backend/catalog/system_views.sql
+++ b/src/backend/catalog/system_views.sql
@@ -1607,6 +1607,7 @@ REVOKE EXECUTE ON FUNCTION pg_stat_file(text,boolean) FROM public;
 
 REVOKE EXECUTE ON FUNCTION pg_ls_dir(text) FROM public;
 REVOKE EXECUTE ON FUNCTION pg_ls_dir(text,boolean,boolean) FROM public;
+REVOKE EXECUTE ON FUNCTION pg_ls_dir_metadata(text,boolean,boolean) FROM public;
 
 --
 -- We also set up some things as accessible to standard roles.
diff --git a/src/backend/utils/adt/genfile.c b/src/backend/utils/adt/genfile.c
index 322152ebd9..f59f853983 100644
--- a/src/backend/utils/adt/genfile.c
+++ b/src/backend/utils/adt/genfile.c
@@ -36,6 +36,21 @@
 #include "utils/syscache.h"
 #include "utils/timestamp.h"
 
+static Datum pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, int flags);
+
+#define	LS_DIR_ISDIR				(1<<0) /* Show column: isdir */
+#define	LS_DIR_METADATA				(1<<1) /* Show columns: mtime, size */
+#define	LS_DIR_MISSING_OK			(1<<2) /* Ignore ENOENT if the toplevel dir is missing */
+#define	LS_DIR_SKIP_DOT_DIRS		(1<<3) /* Do not show . or .. */
+#define	LS_DIR_SKIP_HIDDEN			(1<<4) /* Do not show anything begining with . */
+#define	LS_DIR_SKIP_DIRS			(1<<5) /* Do not show directories */
+#define	LS_DIR_SKIP_SPECIAL			(1<<6) /* Do not show special file types */
+
+/*
+ * Shortcut for the historic behavior of the pg_ls_* functions (not including
+ * pg_ls_dir, which skips different files and doesn't show metadata).
+ */
+#define LS_DIR_HISTORIC				(LS_DIR_SKIP_DIRS | LS_DIR_SKIP_HIDDEN | LS_DIR_SKIP_SPECIAL | LS_DIR_METADATA)
 
 /*
  * Convert a "text" filename argument to C string, and check it's allowable.
@@ -452,6 +467,11 @@ pg_stat_file(PG_FUNCTION_ARGS)
 	values[4] = TimestampTzGetDatum(time_t_to_timestamptz(fst.st_ctime));
 #endif
 	values[5] = BoolGetDatum(S_ISDIR(fst.st_mode));
+#ifdef WIN32
+	/* Links should have isdir=false */
+	if (pgwin32_is_junction(filename))
+		values[5] = BoolGetDatum(false);
+#endif
 
 	tuple = heap_form_tuple(tupdesc, values, isnull);
 
@@ -479,79 +499,9 @@ pg_stat_file_1arg(PG_FUNCTION_ARGS)
 Datum
 pg_ls_dir(PG_FUNCTION_ARGS)
 {
-	ReturnSetInfo *rsinfo = (ReturnSetInfo *) fcinfo->resultinfo;
-	char	   *location;
-	bool		missing_ok = false;
-	bool		include_dot_dirs = false;
-	bool		randomAccess;
-	TupleDesc	tupdesc;
-	Tuplestorestate *tupstore;
-	DIR		   *dirdesc;
-	struct dirent *de;
-	MemoryContext oldcontext;
-
-	location = convert_and_check_filename(PG_GETARG_TEXT_PP(0));
-
-	/* check the optional arguments */
-	if (PG_NARGS() == 3)
-	{
-		if (!PG_ARGISNULL(1))
-			missing_ok = PG_GETARG_BOOL(1);
-		if (!PG_ARGISNULL(2))
-			include_dot_dirs = PG_GETARG_BOOL(2);
-	}
-
-	/* check to see if caller supports us returning a tuplestore */
-	if (rsinfo == NULL || !IsA(rsinfo, ReturnSetInfo))
-		ereport(ERROR,
-				(errcode(ERRCODE_FEATURE_NOT_SUPPORTED),
-				 errmsg("set-valued function called in context that cannot accept a set")));
-	if (!(rsinfo->allowedModes & SFRM_Materialize))
-		ereport(ERROR,
-				(errcode(ERRCODE_SYNTAX_ERROR),
-				 errmsg("materialize mode required, but it is not allowed in this context")));
-
-	/* The tupdesc and tuplestore must be created in ecxt_per_query_memory */
-	oldcontext = MemoryContextSwitchTo(rsinfo->econtext->ecxt_per_query_memory);
-
-	tupdesc = CreateTemplateTupleDesc(1);
-	TupleDescInitEntry(tupdesc, (AttrNumber) 1, "pg_ls_dir", TEXTOID, -1, 0);
-
-	randomAccess = (rsinfo->allowedModes & SFRM_Materialize_Random) != 0;
-	tupstore = tuplestore_begin_heap(randomAccess, false, work_mem);
-	rsinfo->returnMode = SFRM_Materialize;
-	rsinfo->setResult = tupstore;
-	rsinfo->setDesc = tupdesc;
-
-	MemoryContextSwitchTo(oldcontext);
-
-	dirdesc = AllocateDir(location);
-	if (!dirdesc)
-	{
-		/* Return empty tuplestore if appropriate */
-		if (missing_ok && errno == ENOENT)
-			return (Datum) 0;
-		/* Otherwise, we can let ReadDir() throw the error */
-	}
-
-	while ((de = ReadDir(dirdesc, location)) != NULL)
-	{
-		Datum		values[1];
-		bool		nulls[1];
-
-		if (!include_dot_dirs &&
-			(strcmp(de->d_name, ".") == 0 ||
-			 strcmp(de->d_name, "..") == 0))
-			continue;
-
-		values[0] = CStringGetTextDatum(de->d_name);
-		nulls[0] = false;
-
-		tuplestore_putvalues(tupstore, tupdesc, values, nulls);
-	}
-
-	FreeDir(dirdesc);
-	return (Datum) 0;
+	text	*filename_t = PG_GETARG_TEXT_PP(0);
+	char	*filename = convert_and_check_filename(filename_t);
+	return pg_ls_dir_files(fcinfo, filename, LS_DIR_SKIP_DOT_DIRS);
 }
 
 /*
@@ -564,17 +514,19 @@ pg_ls_dir(PG_FUNCTION_ARGS)
 Datum
 pg_ls_dir_1arg(PG_FUNCTION_ARGS)
 {
-	return pg_ls_dir(fcinfo);
+	text	*filename_t = PG_GETARG_TEXT_PP(0);
+	char	*filename = convert_and_check_filename(filename_t);
+	return pg_ls_dir_files(fcinfo, filename, LS_DIR_SKIP_DOT_DIRS);
 }
 
 /*
- * Generic function to return a directory listing of files.
+ * Generic function to return a directory listing of files (and optionally dirs).
  *
- * If the directory isn't there, silently return an empty set if missing_ok.
+ * If the directory isn't there, silently return an empty set if MISSING_OK.
  * Other unreadable-directory cases throw an error.
  */
 static Datum
-pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, bool missing_ok)
+pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, int flags)
 {
 	ReturnSetInfo *rsinfo = (ReturnSetInfo *) fcinfo->resultinfo;
 	bool		randomAccess;
@@ -583,6 +535,32 @@ pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, bool missing_ok)
 	DIR		   *dirdesc;
 	struct dirent *de;
 	MemoryContext oldcontext;
+	TypeFuncClass	tuptype ;
+
+	/* isdir depends on metadata */
+	Assert(!(flags&LS_DIR_ISDIR) || (flags&LS_DIR_METADATA));
+	/* Unreasonable to show isdir and skip dirs */
+	Assert(!(flags&LS_DIR_ISDIR) || !(flags&LS_DIR_SKIP_DIRS));
+
+	/* check the optional arguments */
+	if (PG_NARGS() == 3)
+	{
+		if (!PG_ARGISNULL(1))
+		{
+			if (PG_GETARG_BOOL(1))
+				flags |= LS_DIR_MISSING_OK;
+			else
+				flags &= ~LS_DIR_MISSING_OK;
+		}
+
+		if (!PG_ARGISNULL(2))
+		{
+			if (PG_GETARG_BOOL(2))
+				flags &= ~LS_DIR_SKIP_DOT_DIRS;
+			else
+				flags |= LS_DIR_SKIP_DOT_DIRS;
+		}
+	}
 
 	/* check to see if caller supports us returning a tuplestore */
 	if (rsinfo == NULL || !IsA(rsinfo, ReturnSetInfo))
@@ -597,8 +575,20 @@ pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, bool missing_ok)
 	/* The tupdesc and tuplestore must be created in ecxt_per_query_memory */
 	oldcontext = MemoryContextSwitchTo(rsinfo->econtext->ecxt_per_query_memory);
 
-	if (get_call_result_type(fcinfo, NULL, &tupdesc) != TYPEFUNC_COMPOSITE)
-		elog(ERROR, "return type must be a row type");
+	tuptype = get_call_result_type(fcinfo, NULL, &tupdesc);
+	if (flags & LS_DIR_METADATA)
+	{
+		if (tuptype != TYPEFUNC_COMPOSITE)
+			elog(ERROR, "return type must be a row type");
+	}
+	else
+	{
+		/* pg_ls_dir returns a simple scalar */
+		if (tuptype != TYPEFUNC_SCALAR)
+			elog(ERROR, "return type must be a scalar type");
+		tupdesc = CreateTemplateTupleDesc(1);
+		TupleDescInitEntry(tupdesc, (AttrNumber) 1, "column", TEXTOID, -1, 0);
+	}
 
 	randomAccess = (rsinfo->allowedModes & SFRM_Materialize_Random) != 0;
 	tupstore = tuplestore_begin_heap(randomAccess, false, work_mem);
@@ -617,20 +607,27 @@ pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, bool missing_ok)
 	if (!dirdesc)
 	{
 		/* Return empty tuplestore if appropriate */
-		if (missing_ok && errno == ENOENT)
+		if (flags & LS_DIR_MISSING_OK && errno == ENOENT)
 			return (Datum) 0;
 		/* Otherwise, we can let ReadDir() throw the error */
 	}
 
 	while ((de = ReadDir(dirdesc, dir)) != NULL)
 	{
-		Datum		values[3];
-		bool		nulls[3];
+		Datum		values[4];
+		bool		nulls[4];
 		char		path[MAXPGPATH * 2];
 		struct stat attrib;
 
-		/* Skip hidden files */
-		if (de->d_name[0] == '.')
+		/* Skip dot dirs? */
+		if (flags & LS_DIR_SKIP_DOT_DIRS &&
+			(strcmp(de->d_name, ".") == 0 ||
+			 strcmp(de->d_name, "..") == 0))
+			continue;
+
+		/* Skip hidden files? */
+		if (flags & LS_DIR_SKIP_HIDDEN &&
+			de->d_name[0] == '.')
 			continue;
 
 		/* Get the file info */
@@ -645,13 +642,34 @@ pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, bool missing_ok)
 					 errmsg("could not stat file \"%s\": %m", path)));
 		}
 
-		/* Ignore anything but regular files */
-		if (!S_ISREG(attrib.st_mode))
-			continue;
+		/* Skip dirs or special files? */
+		if (S_ISDIR(attrib.st_mode))
+		{
+			if (flags & LS_DIR_SKIP_DIRS)
+				continue;
+		}
+		else if (!S_ISREG(attrib.st_mode))
+		{
+			if (flags & LS_DIR_SKIP_SPECIAL)
+				continue;
+		}
 
 		values[0] = CStringGetTextDatum(de->d_name);
-		values[1] = Int64GetDatum((int64) attrib.st_size);
-		values[2] = TimestampTzGetDatum(time_t_to_timestamptz(attrib.st_mtime));
+		if (flags & LS_DIR_METADATA)
+		{
+			values[1] = Int64GetDatum((int64) attrib.st_size);
+			values[2] = TimestampTzGetDatum(time_t_to_timestamptz(attrib.st_mtime));
+			if (flags & LS_DIR_ISDIR)
+			{
+				values[3] = BoolGetDatum(S_ISDIR(attrib.st_mode));
+#ifdef WIN32
+				/* Links should have isdir=false */
+				if (pgwin32_is_junction(path))
+					values[3] = BoolGetDatum(false);
+#endif
+			}
+		}
+
 		memset(nulls, 0, sizeof(nulls));
 
 		tuplestore_putvalues(tupstore, tupdesc, values, nulls);
@@ -665,14 +683,14 @@ pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, bool missing_ok)
 Datum
 pg_ls_logdir(PG_FUNCTION_ARGS)
 {
-	return pg_ls_dir_files(fcinfo, Log_directory, false);
+	return pg_ls_dir_files(fcinfo, Log_directory, LS_DIR_HISTORIC);
 }
 
 /* Function to return the list of files in the WAL directory */
 Datum
 pg_ls_waldir(PG_FUNCTION_ARGS)
 {
-	return pg_ls_dir_files(fcinfo, XLOGDIR, false);
+	return pg_ls_dir_files(fcinfo, XLOGDIR, LS_DIR_HISTORIC);
 }
 
 /*
@@ -690,7 +708,8 @@ pg_ls_tmpdir(FunctionCallInfo fcinfo, Oid tblspc)
 						tblspc)));
 
 	TempTablespacePath(path, tblspc);
-	return pg_ls_dir_files(fcinfo, path, true);
+	return pg_ls_dir_files(fcinfo, path,
+			LS_DIR_HISTORIC | LS_DIR_MISSING_OK);
 }
 
 /*
@@ -719,5 +738,33 @@ pg_ls_tmpdir_1arg(PG_FUNCTION_ARGS)
 Datum
 pg_ls_archive_statusdir(PG_FUNCTION_ARGS)
 {
-	return pg_ls_dir_files(fcinfo, XLOGDIR "/archive_status", true);
+	return pg_ls_dir_files(fcinfo, XLOGDIR "/archive_status",
+			LS_DIR_HISTORIC | LS_DIR_MISSING_OK);
+}
+
+/*
+ * Return the list of files and metadata in an arbitrary directory.
+ */
+Datum
+pg_ls_dir_metadata(PG_FUNCTION_ARGS)
+{
+	char	*dirname = convert_and_check_filename(PG_GETARG_TEXT_PP(0));
+
+	return pg_ls_dir_files(fcinfo, dirname,
+			LS_DIR_METADATA | LS_DIR_SKIP_SPECIAL | LS_DIR_ISDIR);
+}
+
+/*
+ * Return the list of files and metadata in an arbitrary directory.
+ * note: this wrapper is necessary to pass the sanity check in opr_sanity,
+ * which checks that all built-in functions that share the implementing C
+ * function take the same number of arguments.
+ */
+Datum
+pg_ls_dir_metadata_1arg(PG_FUNCTION_ARGS)
+{
+	char	*dirname = convert_and_check_filename(PG_GETARG_TEXT_PP(0));
+
+	return pg_ls_dir_files(fcinfo, dirname,
+			LS_DIR_METADATA | LS_DIR_SKIP_SPECIAL | LS_DIR_ISDIR);
 }
diff --git a/src/include/catalog/pg_proc.dat b/src/include/catalog/pg_proc.dat
index 4309fa40dd..b85bee2d1e 100644
--- a/src/include/catalog/pg_proc.dat
+++ b/src/include/catalog/pg_proc.dat
@@ -11517,6 +11517,18 @@
   proallargtypes => '{oid,text,int8,timestamptz}', proargmodes => '{i,o,o,o}',
   proargnames => '{tablespace,name,size,modification}',
   prosrc => 'pg_ls_tmpdir_1arg' },
+{ oid => '8450', descr => 'list directory with metadata',
+  proname => 'pg_ls_dir_metadata', procost => '10', prorows => '20', proretset => 't',
+  provolatile => 'v', prorettype => 'record', proargtypes => 'text bool bool',
+  proallargtypes => '{text,bool,bool,text,int8,timestamptz,bool}', proargmodes => '{i,i,i,o,o,o,o}',
+  proargnames => '{dirname,missing_ok,include_dot_dirs,filename,size,modification,isdir}',
+  prosrc => 'pg_ls_dir_metadata' },
+{ oid => '8451', descr => 'list directory with metadata',
+  proname => 'pg_ls_dir_metadata', procost => '10', prorows => '20', proretset => 't',
+  provolatile => 'v', prorettype => 'record', proargtypes => 'text',
+  proallargtypes => '{text,text,int8,timestamptz,bool}', proargmodes => '{i,o,o,o,o}',
+  proargnames => '{dirname,filename,size,modification,isdir}',
+  prosrc => 'pg_ls_dir_metadata_1arg' },
 
 # hash partitioning constraint function
 { oid => '5028', descr => 'hash partition CHECK constraint',
diff --git a/src/test/regress/expected/misc_functions.out b/src/test/regress/expected/misc_functions.out
index ea0fc48dbd..b4112ec298 100644
--- a/src/test/regress/expected/misc_functions.out
+++ b/src/test/regress/expected/misc_functions.out
@@ -238,6 +238,30 @@ select * from pg_stat_file('.') limit 0;
 ------+--------+--------------+--------+----------+-------
 (0 rows)
 
+-- This tests the missing_ok parameter, which causes pg_ls_tmpdir to succeed even if the tmpdir doesn't exist yet
+-- The name='' condition is never true, so the function runs to completion but returns zero rows.
+select * from pg_ls_tmpdir() where name='Does not exist';
+ name | size | modification 
+------+------+--------------
+(0 rows)
+
+select filename, isdir from pg_ls_dir_metadata('.') where filename='.';
+ filename | isdir 
+----------+-------
+ .        | t
+(1 row)
+
+select filename, isdir from pg_ls_dir_metadata('.', false, false) where filename='.'; -- include_dot_dirs=false
+ filename | isdir 
+----------+-------
+(0 rows)
+
+-- Check that expected columns are present
+select * from pg_ls_dir_metadata('.') limit 0;
+ filename | size | modification | isdir 
+----------+------+--------------+-------
+(0 rows)
+
 --
 -- Test adding a support function to a subject function
 --
diff --git a/src/test/regress/input/tablespace.source b/src/test/regress/input/tablespace.source
index c133e73499..a32212be04 100644
--- a/src/test/regress/input/tablespace.source
+++ b/src/test/regress/input/tablespace.source
@@ -11,6 +11,11 @@ DROP TABLESPACE regress_tblspacewith;
 -- create a tablespace we can use
 CREATE TABLESPACE regress_tblspace LOCATION '@testtablespace@';
 
+-- This tests the missing_ok parameter, which causes pg_ls_tmpdir to succeed even if the tmpdir doesn't exist yet
+-- The name='' condition is never true, so the function runs to completion but returns zero rows.
+-- The query is written to ERROR if the tablespace doesn't exist, rather than silently failing to call pg_ls_tmpdir()
+SELECT c.* FROM (SELECT oid FROM pg_tablespace b WHERE b.spcname='regress_tblspace' UNION SELECT 0 ORDER BY 1 DESC LIMIT 1) AS b , pg_ls_tmpdir(oid) AS c WHERE c.name='Does not exist';
+
 -- try setting and resetting some properties for the new tablespace
 ALTER TABLESPACE regress_tblspace SET (random_page_cost = 1.0, seq_page_cost = 1.1);
 ALTER TABLESPACE regress_tblspace SET (some_nonexistent_parameter = true);  -- fail
diff --git a/src/test/regress/output/tablespace.source b/src/test/regress/output/tablespace.source
index 1bbe7e0323..5f7ce3d09f 100644
--- a/src/test/regress/output/tablespace.source
+++ b/src/test/regress/output/tablespace.source
@@ -13,6 +13,14 @@ SELECT spcoptions FROM pg_tablespace WHERE spcname = 'regress_tblspacewith';
 DROP TABLESPACE regress_tblspacewith;
 -- create a tablespace we can use
 CREATE TABLESPACE regress_tblspace LOCATION '@testtablespace@';
+-- This tests the missing_ok parameter, which causes pg_ls_tmpdir to succeed even if the tmpdir doesn't exist yet
+-- The name='' condition is never true, so the function runs to completion but returns zero rows.
+-- The query is written to ERROR if the tablespace doesn't exist, rather than silently failing to call pg_ls_tmpdir()
+SELECT c.* FROM (SELECT oid FROM pg_tablespace b WHERE b.spcname='regress_tblspace' UNION SELECT 0 ORDER BY 1 DESC LIMIT 1) AS b , pg_ls_tmpdir(oid) AS c WHERE c.name='Does not exist';
+ name | size | modification 
+------+------+--------------
+(0 rows)
+
 -- try setting and resetting some properties for the new tablespace
 ALTER TABLESPACE regress_tblspace SET (random_page_cost = 1.0, seq_page_cost = 1.1);
 ALTER TABLESPACE regress_tblspace SET (some_nonexistent_parameter = true);  -- fail
diff --git a/src/test/regress/sql/misc_functions.sql b/src/test/regress/sql/misc_functions.sql
index eb6ac12ab4..c169e527d9 100644
--- a/src/test/regress/sql/misc_functions.sql
+++ b/src/test/regress/sql/misc_functions.sql
@@ -77,6 +77,17 @@ select pg_ls_dir('does not exist'); -- fails with missingok=false
 -- Check that expected columns are present
 select * from pg_stat_file('.') limit 0;
 
+-- This tests the missing_ok parameter, which causes pg_ls_tmpdir to succeed even if the tmpdir doesn't exist yet
+-- The name='' condition is never true, so the function runs to completion but returns zero rows.
+select * from pg_ls_tmpdir() where name='Does not exist';
+
+select filename, isdir from pg_ls_dir_metadata('.') where filename='.';
+
+select filename, isdir from pg_ls_dir_metadata('.', false, false) where filename='.'; -- include_dot_dirs=false
+
+-- Check that expected columns are present
+select * from pg_ls_dir_metadata('.') limit 0;
+
 --
 -- Test adding a support function to a subject function
 --
-- 
2.17.0


--19uQFt6ulqmgNgg1
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
 filename="v27-0004-pg_ls_tmpdir-to-show-directories-and-isdir-argum.patch"



^ permalink  raw  reply  [nested|flat] 68+ messages in thread

* [PATCH v28 03/11] Add pg_ls_dir_metadata to list a dir with file metadata..
@ 2020-03-10 03:40  Justin Pryzby <[email protected]>
  0 siblings, 0 replies; 68+ messages in thread

From: Justin Pryzby @ 2020-03-10 03:40 UTC (permalink / raw)

Generalize pg_ls_dir_files and retire pg_ls_dir

Need catversion bumped?
---
 doc/src/sgml/func.sgml                       |  21 ++
 src/backend/catalog/system_views.sql         |   1 +
 src/backend/utils/adt/genfile.c              | 233 +++++++++++--------
 src/include/catalog/pg_proc.dat              |  12 +
 src/test/regress/expected/misc_functions.out |  24 ++
 src/test/regress/input/tablespace.source     |   5 +
 src/test/regress/output/tablespace.source    |   8 +
 src/test/regress/sql/misc_functions.sql      |  11 +
 8 files changed, 222 insertions(+), 93 deletions(-)

diff --git a/doc/src/sgml/func.sgml b/doc/src/sgml/func.sgml
index aa0dcde886..507a6d73f8 100644
--- a/doc/src/sgml/func.sgml
+++ b/doc/src/sgml/func.sgml
@@ -25821,6 +25821,27 @@ postgres=# SELECT * FROM pg_walfile_name_offset(pg_stop_backup());
        </para></entry>
       </row>
 
+      <row>
+       <entry role="func_table_entry"><para role="func_signature">
+        <indexterm>
+         <primary>pg_ls_dir_metadata</primary>
+        </indexterm>
+        <function>pg_ls_dir_metadata</function> ( <parameter>dirname</parameter> <type>text</type>
+        <optional>, <parameter>missing_ok</parameter> <type>boolean</type>,
+        <parameter>include_dot_dirs</parameter> <type>boolean</type> </optional> )
+        <returnvalue>setof record</returnvalue>
+        ( <parameter>filename</parameter> <type>text</type>,
+        <parameter>size</parameter> <type>bigint</type>,
+        <parameter>modification</parameter> <type>timestamp with time zone</type> )
+       </para>
+       <para>
+        For each file in the specified directory, list the file and its
+        metadata.
+        Restricted to superusers by default, but other users can be granted
+        EXECUTE to run the function.
+       </para></entry>
+      </row>
+
       <row>
        <entry role="func_table_entry"><para role="func_signature">
         <indexterm>
diff --git a/src/backend/catalog/system_views.sql b/src/backend/catalog/system_views.sql
index 451db2ee0a..dd84b0f11a 100644
--- a/src/backend/catalog/system_views.sql
+++ b/src/backend/catalog/system_views.sql
@@ -1632,6 +1632,7 @@ REVOKE EXECUTE ON FUNCTION pg_stat_file(text,boolean) FROM public;
 
 REVOKE EXECUTE ON FUNCTION pg_ls_dir(text) FROM public;
 REVOKE EXECUTE ON FUNCTION pg_ls_dir(text,boolean,boolean) FROM public;
+REVOKE EXECUTE ON FUNCTION pg_ls_dir_metadata(text,boolean,boolean) FROM public;
 
 --
 -- We also set up some things as accessible to standard roles.
diff --git a/src/backend/utils/adt/genfile.c b/src/backend/utils/adt/genfile.c
index 322152ebd9..f59f853983 100644
--- a/src/backend/utils/adt/genfile.c
+++ b/src/backend/utils/adt/genfile.c
@@ -36,6 +36,21 @@
 #include "utils/syscache.h"
 #include "utils/timestamp.h"
 
+static Datum pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, int flags);
+
+#define	LS_DIR_ISDIR				(1<<0) /* Show column: isdir */
+#define	LS_DIR_METADATA				(1<<1) /* Show columns: mtime, size */
+#define	LS_DIR_MISSING_OK			(1<<2) /* Ignore ENOENT if the toplevel dir is missing */
+#define	LS_DIR_SKIP_DOT_DIRS		(1<<3) /* Do not show . or .. */
+#define	LS_DIR_SKIP_HIDDEN			(1<<4) /* Do not show anything begining with . */
+#define	LS_DIR_SKIP_DIRS			(1<<5) /* Do not show directories */
+#define	LS_DIR_SKIP_SPECIAL			(1<<6) /* Do not show special file types */
+
+/*
+ * Shortcut for the historic behavior of the pg_ls_* functions (not including
+ * pg_ls_dir, which skips different files and doesn't show metadata).
+ */
+#define LS_DIR_HISTORIC				(LS_DIR_SKIP_DIRS | LS_DIR_SKIP_HIDDEN | LS_DIR_SKIP_SPECIAL | LS_DIR_METADATA)
 
 /*
  * Convert a "text" filename argument to C string, and check it's allowable.
@@ -452,6 +467,11 @@ pg_stat_file(PG_FUNCTION_ARGS)
 	values[4] = TimestampTzGetDatum(time_t_to_timestamptz(fst.st_ctime));
 #endif
 	values[5] = BoolGetDatum(S_ISDIR(fst.st_mode));
+#ifdef WIN32
+	/* Links should have isdir=false */
+	if (pgwin32_is_junction(filename))
+		values[5] = BoolGetDatum(false);
+#endif
 
 	tuple = heap_form_tuple(tupdesc, values, isnull);
 
@@ -479,79 +499,9 @@ pg_stat_file_1arg(PG_FUNCTION_ARGS)
 Datum
 pg_ls_dir(PG_FUNCTION_ARGS)
 {
-	ReturnSetInfo *rsinfo = (ReturnSetInfo *) fcinfo->resultinfo;
-	char	   *location;
-	bool		missing_ok = false;
-	bool		include_dot_dirs = false;
-	bool		randomAccess;
-	TupleDesc	tupdesc;
-	Tuplestorestate *tupstore;
-	DIR		   *dirdesc;
-	struct dirent *de;
-	MemoryContext oldcontext;
-
-	location = convert_and_check_filename(PG_GETARG_TEXT_PP(0));
-
-	/* check the optional arguments */
-	if (PG_NARGS() == 3)
-	{
-		if (!PG_ARGISNULL(1))
-			missing_ok = PG_GETARG_BOOL(1);
-		if (!PG_ARGISNULL(2))
-			include_dot_dirs = PG_GETARG_BOOL(2);
-	}
-
-	/* check to see if caller supports us returning a tuplestore */
-	if (rsinfo == NULL || !IsA(rsinfo, ReturnSetInfo))
-		ereport(ERROR,
-				(errcode(ERRCODE_FEATURE_NOT_SUPPORTED),
-				 errmsg("set-valued function called in context that cannot accept a set")));
-	if (!(rsinfo->allowedModes & SFRM_Materialize))
-		ereport(ERROR,
-				(errcode(ERRCODE_SYNTAX_ERROR),
-				 errmsg("materialize mode required, but it is not allowed in this context")));
-
-	/* The tupdesc and tuplestore must be created in ecxt_per_query_memory */
-	oldcontext = MemoryContextSwitchTo(rsinfo->econtext->ecxt_per_query_memory);
-
-	tupdesc = CreateTemplateTupleDesc(1);
-	TupleDescInitEntry(tupdesc, (AttrNumber) 1, "pg_ls_dir", TEXTOID, -1, 0);
-
-	randomAccess = (rsinfo->allowedModes & SFRM_Materialize_Random) != 0;
-	tupstore = tuplestore_begin_heap(randomAccess, false, work_mem);
-	rsinfo->returnMode = SFRM_Materialize;
-	rsinfo->setResult = tupstore;
-	rsinfo->setDesc = tupdesc;
-
-	MemoryContextSwitchTo(oldcontext);
-
-	dirdesc = AllocateDir(location);
-	if (!dirdesc)
-	{
-		/* Return empty tuplestore if appropriate */
-		if (missing_ok && errno == ENOENT)
-			return (Datum) 0;
-		/* Otherwise, we can let ReadDir() throw the error */
-	}
-
-	while ((de = ReadDir(dirdesc, location)) != NULL)
-	{
-		Datum		values[1];
-		bool		nulls[1];
-
-		if (!include_dot_dirs &&
-			(strcmp(de->d_name, ".") == 0 ||
-			 strcmp(de->d_name, "..") == 0))
-			continue;
-
-		values[0] = CStringGetTextDatum(de->d_name);
-		nulls[0] = false;
-
-		tuplestore_putvalues(tupstore, tupdesc, values, nulls);
-	}
-
-	FreeDir(dirdesc);
-	return (Datum) 0;
+	text	*filename_t = PG_GETARG_TEXT_PP(0);
+	char	*filename = convert_and_check_filename(filename_t);
+	return pg_ls_dir_files(fcinfo, filename, LS_DIR_SKIP_DOT_DIRS);
 }
 
 /*
@@ -564,17 +514,19 @@ pg_ls_dir(PG_FUNCTION_ARGS)
 Datum
 pg_ls_dir_1arg(PG_FUNCTION_ARGS)
 {
-	return pg_ls_dir(fcinfo);
+	text	*filename_t = PG_GETARG_TEXT_PP(0);
+	char	*filename = convert_and_check_filename(filename_t);
+	return pg_ls_dir_files(fcinfo, filename, LS_DIR_SKIP_DOT_DIRS);
 }
 
 /*
- * Generic function to return a directory listing of files.
+ * Generic function to return a directory listing of files (and optionally dirs).
  *
- * If the directory isn't there, silently return an empty set if missing_ok.
+ * If the directory isn't there, silently return an empty set if MISSING_OK.
  * Other unreadable-directory cases throw an error.
  */
 static Datum
-pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, bool missing_ok)
+pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, int flags)
 {
 	ReturnSetInfo *rsinfo = (ReturnSetInfo *) fcinfo->resultinfo;
 	bool		randomAccess;
@@ -583,6 +535,32 @@ pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, bool missing_ok)
 	DIR		   *dirdesc;
 	struct dirent *de;
 	MemoryContext oldcontext;
+	TypeFuncClass	tuptype ;
+
+	/* isdir depends on metadata */
+	Assert(!(flags&LS_DIR_ISDIR) || (flags&LS_DIR_METADATA));
+	/* Unreasonable to show isdir and skip dirs */
+	Assert(!(flags&LS_DIR_ISDIR) || !(flags&LS_DIR_SKIP_DIRS));
+
+	/* check the optional arguments */
+	if (PG_NARGS() == 3)
+	{
+		if (!PG_ARGISNULL(1))
+		{
+			if (PG_GETARG_BOOL(1))
+				flags |= LS_DIR_MISSING_OK;
+			else
+				flags &= ~LS_DIR_MISSING_OK;
+		}
+
+		if (!PG_ARGISNULL(2))
+		{
+			if (PG_GETARG_BOOL(2))
+				flags &= ~LS_DIR_SKIP_DOT_DIRS;
+			else
+				flags |= LS_DIR_SKIP_DOT_DIRS;
+		}
+	}
 
 	/* check to see if caller supports us returning a tuplestore */
 	if (rsinfo == NULL || !IsA(rsinfo, ReturnSetInfo))
@@ -597,8 +575,20 @@ pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, bool missing_ok)
 	/* The tupdesc and tuplestore must be created in ecxt_per_query_memory */
 	oldcontext = MemoryContextSwitchTo(rsinfo->econtext->ecxt_per_query_memory);
 
-	if (get_call_result_type(fcinfo, NULL, &tupdesc) != TYPEFUNC_COMPOSITE)
-		elog(ERROR, "return type must be a row type");
+	tuptype = get_call_result_type(fcinfo, NULL, &tupdesc);
+	if (flags & LS_DIR_METADATA)
+	{
+		if (tuptype != TYPEFUNC_COMPOSITE)
+			elog(ERROR, "return type must be a row type");
+	}
+	else
+	{
+		/* pg_ls_dir returns a simple scalar */
+		if (tuptype != TYPEFUNC_SCALAR)
+			elog(ERROR, "return type must be a scalar type");
+		tupdesc = CreateTemplateTupleDesc(1);
+		TupleDescInitEntry(tupdesc, (AttrNumber) 1, "column", TEXTOID, -1, 0);
+	}
 
 	randomAccess = (rsinfo->allowedModes & SFRM_Materialize_Random) != 0;
 	tupstore = tuplestore_begin_heap(randomAccess, false, work_mem);
@@ -617,20 +607,27 @@ pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, bool missing_ok)
 	if (!dirdesc)
 	{
 		/* Return empty tuplestore if appropriate */
-		if (missing_ok && errno == ENOENT)
+		if (flags & LS_DIR_MISSING_OK && errno == ENOENT)
 			return (Datum) 0;
 		/* Otherwise, we can let ReadDir() throw the error */
 	}
 
 	while ((de = ReadDir(dirdesc, dir)) != NULL)
 	{
-		Datum		values[3];
-		bool		nulls[3];
+		Datum		values[4];
+		bool		nulls[4];
 		char		path[MAXPGPATH * 2];
 		struct stat attrib;
 
-		/* Skip hidden files */
-		if (de->d_name[0] == '.')
+		/* Skip dot dirs? */
+		if (flags & LS_DIR_SKIP_DOT_DIRS &&
+			(strcmp(de->d_name, ".") == 0 ||
+			 strcmp(de->d_name, "..") == 0))
+			continue;
+
+		/* Skip hidden files? */
+		if (flags & LS_DIR_SKIP_HIDDEN &&
+			de->d_name[0] == '.')
 			continue;
 
 		/* Get the file info */
@@ -645,13 +642,34 @@ pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, bool missing_ok)
 					 errmsg("could not stat file \"%s\": %m", path)));
 		}
 
-		/* Ignore anything but regular files */
-		if (!S_ISREG(attrib.st_mode))
-			continue;
+		/* Skip dirs or special files? */
+		if (S_ISDIR(attrib.st_mode))
+		{
+			if (flags & LS_DIR_SKIP_DIRS)
+				continue;
+		}
+		else if (!S_ISREG(attrib.st_mode))
+		{
+			if (flags & LS_DIR_SKIP_SPECIAL)
+				continue;
+		}
 
 		values[0] = CStringGetTextDatum(de->d_name);
-		values[1] = Int64GetDatum((int64) attrib.st_size);
-		values[2] = TimestampTzGetDatum(time_t_to_timestamptz(attrib.st_mtime));
+		if (flags & LS_DIR_METADATA)
+		{
+			values[1] = Int64GetDatum((int64) attrib.st_size);
+			values[2] = TimestampTzGetDatum(time_t_to_timestamptz(attrib.st_mtime));
+			if (flags & LS_DIR_ISDIR)
+			{
+				values[3] = BoolGetDatum(S_ISDIR(attrib.st_mode));
+#ifdef WIN32
+				/* Links should have isdir=false */
+				if (pgwin32_is_junction(path))
+					values[3] = BoolGetDatum(false);
+#endif
+			}
+		}
+
 		memset(nulls, 0, sizeof(nulls));
 
 		tuplestore_putvalues(tupstore, tupdesc, values, nulls);
@@ -665,14 +683,14 @@ pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, bool missing_ok)
 Datum
 pg_ls_logdir(PG_FUNCTION_ARGS)
 {
-	return pg_ls_dir_files(fcinfo, Log_directory, false);
+	return pg_ls_dir_files(fcinfo, Log_directory, LS_DIR_HISTORIC);
 }
 
 /* Function to return the list of files in the WAL directory */
 Datum
 pg_ls_waldir(PG_FUNCTION_ARGS)
 {
-	return pg_ls_dir_files(fcinfo, XLOGDIR, false);
+	return pg_ls_dir_files(fcinfo, XLOGDIR, LS_DIR_HISTORIC);
 }
 
 /*
@@ -690,7 +708,8 @@ pg_ls_tmpdir(FunctionCallInfo fcinfo, Oid tblspc)
 						tblspc)));
 
 	TempTablespacePath(path, tblspc);
-	return pg_ls_dir_files(fcinfo, path, true);
+	return pg_ls_dir_files(fcinfo, path,
+			LS_DIR_HISTORIC | LS_DIR_MISSING_OK);
 }
 
 /*
@@ -719,5 +738,33 @@ pg_ls_tmpdir_1arg(PG_FUNCTION_ARGS)
 Datum
 pg_ls_archive_statusdir(PG_FUNCTION_ARGS)
 {
-	return pg_ls_dir_files(fcinfo, XLOGDIR "/archive_status", true);
+	return pg_ls_dir_files(fcinfo, XLOGDIR "/archive_status",
+			LS_DIR_HISTORIC | LS_DIR_MISSING_OK);
+}
+
+/*
+ * Return the list of files and metadata in an arbitrary directory.
+ */
+Datum
+pg_ls_dir_metadata(PG_FUNCTION_ARGS)
+{
+	char	*dirname = convert_and_check_filename(PG_GETARG_TEXT_PP(0));
+
+	return pg_ls_dir_files(fcinfo, dirname,
+			LS_DIR_METADATA | LS_DIR_SKIP_SPECIAL | LS_DIR_ISDIR);
+}
+
+/*
+ * Return the list of files and metadata in an arbitrary directory.
+ * note: this wrapper is necessary to pass the sanity check in opr_sanity,
+ * which checks that all built-in functions that share the implementing C
+ * function take the same number of arguments.
+ */
+Datum
+pg_ls_dir_metadata_1arg(PG_FUNCTION_ARGS)
+{
+	char	*dirname = convert_and_check_filename(PG_GETARG_TEXT_PP(0));
+
+	return pg_ls_dir_files(fcinfo, dirname,
+			LS_DIR_METADATA | LS_DIR_SKIP_SPECIAL | LS_DIR_ISDIR);
 }
diff --git a/src/include/catalog/pg_proc.dat b/src/include/catalog/pg_proc.dat
index f4957653ae..550fbf734a 100644
--- a/src/include/catalog/pg_proc.dat
+++ b/src/include/catalog/pg_proc.dat
@@ -11534,6 +11534,18 @@
   proallargtypes => '{oid,text,int8,timestamptz}', proargmodes => '{i,o,o,o}',
   proargnames => '{tablespace,name,size,modification}',
   prosrc => 'pg_ls_tmpdir_1arg' },
+{ oid => '8450', descr => 'list directory with metadata',
+  proname => 'pg_ls_dir_metadata', procost => '10', prorows => '20', proretset => 't',
+  provolatile => 'v', prorettype => 'record', proargtypes => 'text bool bool',
+  proallargtypes => '{text,bool,bool,text,int8,timestamptz,bool}', proargmodes => '{i,i,i,o,o,o,o}',
+  proargnames => '{dirname,missing_ok,include_dot_dirs,filename,size,modification,isdir}',
+  prosrc => 'pg_ls_dir_metadata' },
+{ oid => '8451', descr => 'list directory with metadata',
+  proname => 'pg_ls_dir_metadata', procost => '10', prorows => '20', proretset => 't',
+  provolatile => 'v', prorettype => 'record', proargtypes => 'text',
+  proallargtypes => '{text,text,int8,timestamptz,bool}', proargmodes => '{i,o,o,o,o}',
+  proargnames => '{dirname,filename,size,modification,isdir}',
+  prosrc => 'pg_ls_dir_metadata_1arg' },
 
 # hash partitioning constraint function
 { oid => '5028', descr => 'hash partition CHECK constraint',
diff --git a/src/test/regress/expected/misc_functions.out b/src/test/regress/expected/misc_functions.out
index ea0fc48dbd..b4112ec298 100644
--- a/src/test/regress/expected/misc_functions.out
+++ b/src/test/regress/expected/misc_functions.out
@@ -238,6 +238,30 @@ select * from pg_stat_file('.') limit 0;
 ------+--------+--------------+--------+----------+-------
 (0 rows)
 
+-- This tests the missing_ok parameter, which causes pg_ls_tmpdir to succeed even if the tmpdir doesn't exist yet
+-- The name='' condition is never true, so the function runs to completion but returns zero rows.
+select * from pg_ls_tmpdir() where name='Does not exist';
+ name | size | modification 
+------+------+--------------
+(0 rows)
+
+select filename, isdir from pg_ls_dir_metadata('.') where filename='.';
+ filename | isdir 
+----------+-------
+ .        | t
+(1 row)
+
+select filename, isdir from pg_ls_dir_metadata('.', false, false) where filename='.'; -- include_dot_dirs=false
+ filename | isdir 
+----------+-------
+(0 rows)
+
+-- Check that expected columns are present
+select * from pg_ls_dir_metadata('.') limit 0;
+ filename | size | modification | isdir 
+----------+------+--------------+-------
+(0 rows)
+
 --
 -- Test adding a support function to a subject function
 --
diff --git a/src/test/regress/input/tablespace.source b/src/test/regress/input/tablespace.source
index c133e73499..a32212be04 100644
--- a/src/test/regress/input/tablespace.source
+++ b/src/test/regress/input/tablespace.source
@@ -11,6 +11,11 @@ DROP TABLESPACE regress_tblspacewith;
 -- create a tablespace we can use
 CREATE TABLESPACE regress_tblspace LOCATION '@testtablespace@';
 
+-- This tests the missing_ok parameter, which causes pg_ls_tmpdir to succeed even if the tmpdir doesn't exist yet
+-- The name='' condition is never true, so the function runs to completion but returns zero rows.
+-- The query is written to ERROR if the tablespace doesn't exist, rather than silently failing to call pg_ls_tmpdir()
+SELECT c.* FROM (SELECT oid FROM pg_tablespace b WHERE b.spcname='regress_tblspace' UNION SELECT 0 ORDER BY 1 DESC LIMIT 1) AS b , pg_ls_tmpdir(oid) AS c WHERE c.name='Does not exist';
+
 -- try setting and resetting some properties for the new tablespace
 ALTER TABLESPACE regress_tblspace SET (random_page_cost = 1.0, seq_page_cost = 1.1);
 ALTER TABLESPACE regress_tblspace SET (some_nonexistent_parameter = true);  -- fail
diff --git a/src/test/regress/output/tablespace.source b/src/test/regress/output/tablespace.source
index 1bbe7e0323..5f7ce3d09f 100644
--- a/src/test/regress/output/tablespace.source
+++ b/src/test/regress/output/tablespace.source
@@ -13,6 +13,14 @@ SELECT spcoptions FROM pg_tablespace WHERE spcname = 'regress_tblspacewith';
 DROP TABLESPACE regress_tblspacewith;
 -- create a tablespace we can use
 CREATE TABLESPACE regress_tblspace LOCATION '@testtablespace@';
+-- This tests the missing_ok parameter, which causes pg_ls_tmpdir to succeed even if the tmpdir doesn't exist yet
+-- The name='' condition is never true, so the function runs to completion but returns zero rows.
+-- The query is written to ERROR if the tablespace doesn't exist, rather than silently failing to call pg_ls_tmpdir()
+SELECT c.* FROM (SELECT oid FROM pg_tablespace b WHERE b.spcname='regress_tblspace' UNION SELECT 0 ORDER BY 1 DESC LIMIT 1) AS b , pg_ls_tmpdir(oid) AS c WHERE c.name='Does not exist';
+ name | size | modification 
+------+------+--------------
+(0 rows)
+
 -- try setting and resetting some properties for the new tablespace
 ALTER TABLESPACE regress_tblspace SET (random_page_cost = 1.0, seq_page_cost = 1.1);
 ALTER TABLESPACE regress_tblspace SET (some_nonexistent_parameter = true);  -- fail
diff --git a/src/test/regress/sql/misc_functions.sql b/src/test/regress/sql/misc_functions.sql
index eb6ac12ab4..c169e527d9 100644
--- a/src/test/regress/sql/misc_functions.sql
+++ b/src/test/regress/sql/misc_functions.sql
@@ -77,6 +77,17 @@ select pg_ls_dir('does not exist'); -- fails with missingok=false
 -- Check that expected columns are present
 select * from pg_stat_file('.') limit 0;
 
+-- This tests the missing_ok parameter, which causes pg_ls_tmpdir to succeed even if the tmpdir doesn't exist yet
+-- The name='' condition is never true, so the function runs to completion but returns zero rows.
+select * from pg_ls_tmpdir() where name='Does not exist';
+
+select filename, isdir from pg_ls_dir_metadata('.') where filename='.';
+
+select filename, isdir from pg_ls_dir_metadata('.', false, false) where filename='.'; -- include_dot_dirs=false
+
+-- Check that expected columns are present
+select * from pg_ls_dir_metadata('.') limit 0;
+
 --
 -- Test adding a support function to a subject function
 --
-- 
2.17.0


--vk/v8fjDPiDepTtA
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
 filename="v28-0004-pg_ls_tmpdir-to-show-directories-and-isdir-argum.patch"



^ permalink  raw  reply  [nested|flat] 68+ messages in thread

* [PATCH v22 04/10] Add pg_ls_dir_metadata to list a dir with file metadata..
@ 2020-03-10 03:40  Justin Pryzby <[email protected]>
  0 siblings, 0 replies; 68+ messages in thread

From: Justin Pryzby @ 2020-03-10 03:40 UTC (permalink / raw)

Generalize pg_ls_dir_files and retire pg_ls_dir

Need catversion bumped?
---
 doc/src/sgml/func.sgml                       |  21 ++
 src/backend/catalog/system_views.sql         |   1 +
 src/backend/utils/adt/genfile.c              | 229 +++++++++++--------
 src/include/catalog/pg_proc.dat              |  12 +
 src/test/regress/expected/misc_functions.out |  24 ++
 src/test/regress/input/tablespace.source     |   5 +
 src/test/regress/output/tablespace.source    |   8 +
 src/test/regress/sql/misc_functions.sql      |  11 +
 8 files changed, 220 insertions(+), 91 deletions(-)

diff --git a/doc/src/sgml/func.sgml b/doc/src/sgml/func.sgml
index 7ef2ec9972..34c92903d8 100644
--- a/doc/src/sgml/func.sgml
+++ b/doc/src/sgml/func.sgml
@@ -25747,6 +25747,27 @@ SELECT pg_size_pretty(sum(pg_relation_size(relid))) AS total_size
        </para></entry>
       </row>
 
+      <row>
+       <entry role="func_table_entry"><para role="func_signature">
+        <indexterm>
+         <primary>pg_ls_dir_metadata</primary>
+        </indexterm>
+        <function>pg_ls_dir_metadata</function> ( <parameter>dirname</parameter> <type>text</type>
+        <optional>, <parameter>missing_ok</parameter> <type>boolean</type>,
+        <parameter>include_dot_dirs</parameter> <type>boolean</type>] </optional> )
+        <returnvalue>setof record</returnvalue>
+        ( <parameter>name</parameter> <type>text</type>,
+        <parameter>size</parameter> <type>bigint</type>,
+        <parameter>modification</parameter> <type>timestamp with time zone</type> )
+       </para>
+       <para>
+        For each file in the specified directory, list the file and its
+        metadata.
+        Restricted to superusers by default, but other users can be granted
+        EXECUTE to run the function.
+       </para></entry>
+      </row>
+
       <row>
        <entry role="func_table_entry"><para role="func_signature">
         <indexterm>
diff --git a/src/backend/catalog/system_views.sql b/src/backend/catalog/system_views.sql
index c6dd084fbc..169602c1b6 100644
--- a/src/backend/catalog/system_views.sql
+++ b/src/backend/catalog/system_views.sql
@@ -1509,6 +1509,7 @@ REVOKE EXECUTE ON FUNCTION pg_stat_file(text,boolean) FROM public;
 
 REVOKE EXECUTE ON FUNCTION pg_ls_dir(text) FROM public;
 REVOKE EXECUTE ON FUNCTION pg_ls_dir(text,boolean,boolean) FROM public;
+REVOKE EXECUTE ON FUNCTION pg_ls_dir_metadata(text,boolean,boolean) FROM public;
 
 --
 -- We also set up some things as accessible to standard roles.
diff --git a/src/backend/utils/adt/genfile.c b/src/backend/utils/adt/genfile.c
index 9f4927220b..054b8d4b1a 100644
--- a/src/backend/utils/adt/genfile.c
+++ b/src/backend/utils/adt/genfile.c
@@ -36,6 +36,21 @@
 #include "utils/syscache.h"
 #include "utils/timestamp.h"
 
+static Datum pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, int flags);
+
+#define	LS_DIR_ISDIR				(1<<0) /* Show column: isdir */
+#define	LS_DIR_METADATA				(1<<1) /* Show columns: mtime, size */
+#define	LS_DIR_MISSING_OK			(1<<2) /* Ignore ENOENT if the toplevel dir is missing */
+#define	LS_DIR_SKIP_DOT_DIRS		(1<<3) /* Do not show . or .. */
+#define	LS_DIR_SKIP_HIDDEN			(1<<4) /* Do not show anything begining with . */
+#define	LS_DIR_SKIP_DIRS			(1<<5) /* Do not show directories */
+#define	LS_DIR_SKIP_SPECIAL			(1<<6) /* Do not show special file types */
+
+/*
+ * Shortcut for the historic behavior of the pg_ls_* functions (not including
+ * pg_ls_dir, which skips different files and doesn't show metadata.
+ */
+#define LS_DIR_HISTORIC				(LS_DIR_SKIP_DIRS | LS_DIR_SKIP_HIDDEN | LS_DIR_SKIP_SPECIAL | LS_DIR_METADATA)
 
 /*
  * Convert a "text" filename argument to C string, and check it's allowable.
@@ -449,6 +464,11 @@ pg_stat_file(PG_FUNCTION_ARGS)
 	values[4] = TimestampTzGetDatum(time_t_to_timestamptz(fst.st_ctime));
 #endif
 	values[5] = BoolGetDatum(S_ISDIR(fst.st_mode));
+#ifdef WIN32
+	/* Links should have isdir=false */
+	if (pgwin32_is_junction(filename))
+		values[5] = BoolGetDatum(false);
+#endif
 
 	tuple = heap_form_tuple(tupdesc, values, isnull);
 
@@ -476,79 +496,9 @@ pg_stat_file_1arg(PG_FUNCTION_ARGS)
 Datum
 pg_ls_dir(PG_FUNCTION_ARGS)
 {
-	ReturnSetInfo *rsinfo = (ReturnSetInfo *) fcinfo->resultinfo;
-	char	   *location;
-	bool		missing_ok = false;
-	bool		include_dot_dirs = false;
-	bool		randomAccess;
-	TupleDesc	tupdesc;
-	Tuplestorestate *tupstore;
-	DIR		   *dirdesc;
-	struct dirent *de;
-	MemoryContext oldcontext;
-
-	location = convert_and_check_filename(PG_GETARG_TEXT_PP(0));
-
-	/* check the optional arguments */
-	if (PG_NARGS() == 3)
-	{
-		if (!PG_ARGISNULL(1))
-			missing_ok = PG_GETARG_BOOL(1);
-		if (!PG_ARGISNULL(2))
-			include_dot_dirs = PG_GETARG_BOOL(2);
-	}
-
-	/* check to see if caller supports us returning a tuplestore */
-	if (rsinfo == NULL || !IsA(rsinfo, ReturnSetInfo))
-		ereport(ERROR,
-				(errcode(ERRCODE_FEATURE_NOT_SUPPORTED),
-				 errmsg("set-valued function called in context that cannot accept a set")));
-	if (!(rsinfo->allowedModes & SFRM_Materialize))
-		ereport(ERROR,
-				(errcode(ERRCODE_SYNTAX_ERROR),
-				 errmsg("materialize mode required, but it is not allowed in this context")));
-
-	/* The tupdesc and tuplestore must be created in ecxt_per_query_memory */
-	oldcontext = MemoryContextSwitchTo(rsinfo->econtext->ecxt_per_query_memory);
-
-	tupdesc = CreateTemplateTupleDesc(1);
-	TupleDescInitEntry(tupdesc, (AttrNumber) 1, "pg_ls_dir", TEXTOID, -1, 0);
-
-	randomAccess = (rsinfo->allowedModes & SFRM_Materialize_Random) != 0;
-	tupstore = tuplestore_begin_heap(randomAccess, false, work_mem);
-	rsinfo->returnMode = SFRM_Materialize;
-	rsinfo->setResult = tupstore;
-	rsinfo->setDesc = tupdesc;
-
-	MemoryContextSwitchTo(oldcontext);
-
-	dirdesc = AllocateDir(location);
-	if (!dirdesc)
-	{
-		/* Return empty tuplestore if appropriate */
-		if (missing_ok && errno == ENOENT)
-			return (Datum) 0;
-		/* Otherwise, we can let ReadDir() throw the error */
-	}
-
-	while ((de = ReadDir(dirdesc, location)) != NULL)
-	{
-		Datum		values[1];
-		bool		nulls[1];
-
-		if (!include_dot_dirs &&
-			(strcmp(de->d_name, ".") == 0 ||
-			 strcmp(de->d_name, "..") == 0))
-			continue;
-
-		values[0] = CStringGetTextDatum(de->d_name);
-		nulls[0] = false;
-
-		tuplestore_putvalues(tupstore, tupdesc, values, nulls);
-	}
-
-	FreeDir(dirdesc);
-	return (Datum) 0;
+	text	*filename_t = PG_GETARG_TEXT_PP(0);
+	char	*filename = convert_and_check_filename(filename_t);
+	return pg_ls_dir_files(fcinfo, filename, LS_DIR_SKIP_DOT_DIRS);
 }
 
 /*
@@ -561,7 +511,9 @@ pg_ls_dir(PG_FUNCTION_ARGS)
 Datum
 pg_ls_dir_1arg(PG_FUNCTION_ARGS)
 {
-	return pg_ls_dir(fcinfo);
+	text	*filename_t = PG_GETARG_TEXT_PP(0);
+	char	*filename = convert_and_check_filename(filename_t);
+	return pg_ls_dir_files(fcinfo, filename, LS_DIR_SKIP_DOT_DIRS);
 }
 
 /*
@@ -571,7 +523,7 @@ pg_ls_dir_1arg(PG_FUNCTION_ARGS)
  * Other unreadable-directory cases throw an error.
  */
 static Datum
-pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, bool missing_ok)
+pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, int flags)
 {
 	ReturnSetInfo *rsinfo = (ReturnSetInfo *) fcinfo->resultinfo;
 	bool		randomAccess;
@@ -580,6 +532,32 @@ pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, bool missing_ok)
 	DIR		   *dirdesc;
 	struct dirent *de;
 	MemoryContext oldcontext;
+	TypeFuncClass	tuptype ;
+
+	/* isdir depends on metadata */
+	Assert(!(flags&LS_DIR_ISDIR) || (flags&LS_DIR_METADATA));
+	/* Unreasonable to show isdir and skip dirs */
+	Assert(!(flags&LS_DIR_ISDIR) || !(flags&LS_DIR_SKIP_DIRS));
+
+	/* check the optional arguments */
+	if (PG_NARGS() == 3)
+	{
+		if (!PG_ARGISNULL(1))
+		{
+			if (PG_GETARG_BOOL(1))
+				flags |= LS_DIR_MISSING_OK;
+			else
+				flags &= ~LS_DIR_MISSING_OK;
+		}
+
+		if (!PG_ARGISNULL(2))
+		{
+			if (PG_GETARG_BOOL(2))
+				flags &= ~LS_DIR_SKIP_DOT_DIRS;
+			else
+				flags |= LS_DIR_SKIP_DOT_DIRS;
+		}
+	}
 
 	/* check to see if caller supports us returning a tuplestore */
 	if (rsinfo == NULL || !IsA(rsinfo, ReturnSetInfo))
@@ -594,8 +572,20 @@ pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, bool missing_ok)
 	/* The tupdesc and tuplestore must be created in ecxt_per_query_memory */
 	oldcontext = MemoryContextSwitchTo(rsinfo->econtext->ecxt_per_query_memory);
 
-	if (get_call_result_type(fcinfo, NULL, &tupdesc) != TYPEFUNC_COMPOSITE)
-		elog(ERROR, "return type must be a row type");
+	tuptype = get_call_result_type(fcinfo, NULL, &tupdesc);
+	if (flags & LS_DIR_METADATA)
+	{
+		if (tuptype != TYPEFUNC_COMPOSITE)
+			elog(ERROR, "return type must be a row type");
+	}
+	else
+	{
+		/* pg_ls_dir returns a simple scalar */
+		if (tuptype != TYPEFUNC_SCALAR)
+			elog(ERROR, "return type must be a scalar type");
+		tupdesc = CreateTemplateTupleDesc(1);
+		TupleDescInitEntry(tupdesc, (AttrNumber) 1, "column", TEXTOID, -1, 0);
+	}
 
 	randomAccess = (rsinfo->allowedModes & SFRM_Materialize_Random) != 0;
 	tupstore = tuplestore_begin_heap(randomAccess, false, work_mem);
@@ -614,20 +604,27 @@ pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, bool missing_ok)
 	if (!dirdesc)
 	{
 		/* Return empty tuplestore if appropriate */
-		if (missing_ok && errno == ENOENT)
+		if (flags & LS_DIR_MISSING_OK && errno == ENOENT)
 			return (Datum) 0;
 		/* Otherwise, we can let ReadDir() throw the error */
 	}
 
 	while ((de = ReadDir(dirdesc, dir)) != NULL)
 	{
-		Datum		values[3];
-		bool		nulls[3];
+		Datum		values[4];
+		bool		nulls[4];
 		char		path[MAXPGPATH * 2];
 		struct stat attrib;
 
-		/* Skip hidden files */
-		if (de->d_name[0] == '.')
+		/* Skip dot dirs? */
+		if (flags & LS_DIR_SKIP_DOT_DIRS &&
+			(strcmp(de->d_name, ".") == 0 ||
+			 strcmp(de->d_name, "..") == 0))
+			continue;
+
+		/* Skip hidden files? */
+		if (flags & LS_DIR_SKIP_HIDDEN &&
+			de->d_name[0] == '.')
 			continue;
 
 		/* Get the file info */
@@ -642,13 +639,34 @@ pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, bool missing_ok)
 					 errmsg("could not stat file \"%s\": %m", path)));
 		}
 
-		/* Ignore anything but regular files */
-		if (!S_ISREG(attrib.st_mode))
-			continue;
+		/* Skip dirs or special files? */
+		if (S_ISDIR(attrib.st_mode))
+		{
+			if (flags & LS_DIR_SKIP_DIRS)
+				continue;
+		}
+		else if (!S_ISREG(attrib.st_mode))
+		{
+			if (flags & LS_DIR_SKIP_SPECIAL)
+				continue;
+		}
 
 		values[0] = CStringGetTextDatum(de->d_name);
-		values[1] = Int64GetDatum((int64) attrib.st_size);
-		values[2] = TimestampTzGetDatum(time_t_to_timestamptz(attrib.st_mtime));
+		if (flags & LS_DIR_METADATA)
+		{
+			values[1] = Int64GetDatum((int64) attrib.st_size);
+			values[2] = TimestampTzGetDatum(time_t_to_timestamptz(attrib.st_mtime));
+			if (flags & LS_DIR_ISDIR)
+			{
+				values[3] = BoolGetDatum(S_ISDIR(attrib.st_mode));
+#ifdef WIN32
+				/* Links should have isdir=false */
+				if (pgwin32_is_junction(path))
+					values[3] = BoolGetDatum(false);
+#endif
+			}
+		}
+
 		memset(nulls, 0, sizeof(nulls));
 
 		tuplestore_putvalues(tupstore, tupdesc, values, nulls);
@@ -662,14 +680,14 @@ pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, bool missing_ok)
 Datum
 pg_ls_logdir(PG_FUNCTION_ARGS)
 {
-	return pg_ls_dir_files(fcinfo, Log_directory, false);
+	return pg_ls_dir_files(fcinfo, Log_directory, LS_DIR_HISTORIC);
 }
 
 /* Function to return the list of files in the WAL directory */
 Datum
 pg_ls_waldir(PG_FUNCTION_ARGS)
 {
-	return pg_ls_dir_files(fcinfo, XLOGDIR, false);
+	return pg_ls_dir_files(fcinfo, XLOGDIR, LS_DIR_HISTORIC);
 }
 
 /*
@@ -687,7 +705,8 @@ pg_ls_tmpdir(FunctionCallInfo fcinfo, Oid tblspc)
 						tblspc)));
 
 	TempTablespacePath(path, tblspc);
-	return pg_ls_dir_files(fcinfo, path, true);
+	return pg_ls_dir_files(fcinfo, path,
+			LS_DIR_HISTORIC | LS_DIR_MISSING_OK);
 }
 
 /*
@@ -716,5 +735,33 @@ pg_ls_tmpdir_1arg(PG_FUNCTION_ARGS)
 Datum
 pg_ls_archive_statusdir(PG_FUNCTION_ARGS)
 {
-	return pg_ls_dir_files(fcinfo, XLOGDIR "/archive_status", true);
+	return pg_ls_dir_files(fcinfo, XLOGDIR "/archive_status",
+			LS_DIR_HISTORIC | LS_DIR_MISSING_OK);
+}
+
+/*
+ * Function to return the list of files and metadata in an arbitrary directory.
+ */
+Datum
+pg_ls_dir_metadata(PG_FUNCTION_ARGS)
+{
+	char	*dirname = convert_and_check_filename(PG_GETARG_TEXT_PP(0));
+
+	return pg_ls_dir_files(fcinfo, dirname,
+			LS_DIR_METADATA | LS_DIR_SKIP_SPECIAL | LS_DIR_ISDIR);
+}
+
+/*
+ * Function to return the list of files and metadata in an arbitrary directory.
+ * note: this wrapper is necessary to pass the sanity check in opr_sanity,
+ * which checks that all built-in functions that share the implementing C
+ * function take the same number of arguments.
+ */
+Datum
+pg_ls_dir_metadata_1arg(PG_FUNCTION_ARGS)
+{
+	char	*dirname = convert_and_check_filename(PG_GETARG_TEXT_PP(0));
+
+	return pg_ls_dir_files(fcinfo, dirname,
+			LS_DIR_METADATA | LS_DIR_SKIP_SPECIAL | LS_DIR_ISDIR);
 }
diff --git a/src/include/catalog/pg_proc.dat b/src/include/catalog/pg_proc.dat
index 24ec2cfed6..02b1436d83 100644
--- a/src/include/catalog/pg_proc.dat
+++ b/src/include/catalog/pg_proc.dat
@@ -10964,6 +10964,18 @@
   proallargtypes => '{regclass,text,text,int8}', proargmodes => '{i,i,o,o}',
   proargnames => '{relation,fork,path,failed_block_num}',
   prosrc => 'pg_relation_check_pages' },
+{ oid => '9979', descr => 'list directory with metadata',
+  proname => 'pg_ls_dir_metadata', procost => '10', prorows => '20', proretset => 't',
+  provolatile => 'v', prorettype => 'record', proargtypes => 'text bool bool',
+  proallargtypes => '{text,bool,bool,text,int8,timestamptz,bool}', proargmodes => '{i,i,i,o,o,o,o}',
+  proargnames => '{dirname,missing_ok,include_dot_dirs,filename,size,modification,isdir}',
+  prosrc => 'pg_ls_dir_metadata' },
+{ oid => '9980', descr => 'list directory with metadata',
+  proname => 'pg_ls_dir_metadata', procost => '10', prorows => '20', proretset => 't',
+  provolatile => 'v', prorettype => 'record', proargtypes => 'text',
+  proallargtypes => '{text,text,int8,timestamptz,bool}', proargmodes => '{i,o,o,o,o}',
+  proargnames => '{dirname,filename,size,modification,isdir}',
+  prosrc => 'pg_ls_dir_metadata_1arg' },
 
 # hash partitioning constraint function
 { oid => '5028', descr => 'hash partition CHECK constraint',
diff --git a/src/test/regress/expected/misc_functions.out b/src/test/regress/expected/misc_functions.out
index 2e87c548eb..20bfe71822 100644
--- a/src/test/regress/expected/misc_functions.out
+++ b/src/test/regress/expected/misc_functions.out
@@ -219,6 +219,30 @@ select pg_ls_dir('does not exist', true, false); -- ok with missingok=true
 
 select pg_ls_dir('does not exist'); -- fails with missingok=false
 ERROR:  could not open directory "does not exist": No such file or directory
+-- This tests the missing_ok parameter, which causes pg_ls_tmpdir to succeed even if the tmpdir doesn't exist yet
+-- The name='' condition is never true, so the function runs to completion but returns zero rows.
+select * from pg_ls_tmpdir() where name='Does not exist';
+ name | size | modification 
+------+------+--------------
+(0 rows)
+
+select filename, isdir from pg_ls_dir_metadata('.') where filename='.';
+ filename | isdir 
+----------+-------
+ .        | t
+(1 row)
+
+select filename, isdir from pg_ls_dir_metadata('.', false, false) where filename='.'; -- include_dot_dirs=false
+ filename | isdir 
+----------+-------
+(0 rows)
+
+-- Check that expected columns are present
+select * from pg_ls_dir_metadata('.') limit 0;
+ filename | size | modification | isdir 
+----------+------+--------------+-------
+(0 rows)
+
 --
 -- Test adding a support function to a subject function
 --
diff --git a/src/test/regress/input/tablespace.source b/src/test/regress/input/tablespace.source
index a5f61a35dc..0b9cfe615e 100644
--- a/src/test/regress/input/tablespace.source
+++ b/src/test/regress/input/tablespace.source
@@ -11,6 +11,11 @@ DROP TABLESPACE regress_tblspacewith;
 -- create a tablespace we can use
 CREATE TABLESPACE regress_tblspace LOCATION '@testtablespace@';
 
+-- This tests the missing_ok parameter, which causes pg_ls_tmpdir to succeed even if the tmpdir doesn't exist yet
+-- The name='' condition is never true, so the function runs to completion but returns zero rows.
+-- The query is written to ERROR if the tablespace doesn't exist, rather than silently failing to call pg_ls_tmpdir()
+SELECT c.* FROM (SELECT oid FROM pg_tablespace b WHERE b.spcname='regress_tblspace' UNION SELECT 0 ORDER BY 1 DESC LIMIT 1) AS b , pg_ls_tmpdir(oid) AS c WHERE c.name='Does not exist';
+
 -- try setting and resetting some properties for the new tablespace
 ALTER TABLESPACE regress_tblspace SET (random_page_cost = 1.0, seq_page_cost = 1.1);
 ALTER TABLESPACE regress_tblspace SET (some_nonexistent_parameter = true);  -- fail
diff --git a/src/test/regress/output/tablespace.source b/src/test/regress/output/tablespace.source
index 162b591b31..a42714bf40 100644
--- a/src/test/regress/output/tablespace.source
+++ b/src/test/regress/output/tablespace.source
@@ -13,6 +13,14 @@ SELECT spcoptions FROM pg_tablespace WHERE spcname = 'regress_tblspacewith';
 DROP TABLESPACE regress_tblspacewith;
 -- create a tablespace we can use
 CREATE TABLESPACE regress_tblspace LOCATION '@testtablespace@';
+-- This tests the missing_ok parameter, which causes pg_ls_tmpdir to succeed even if the tmpdir doesn't exist yet
+-- The name='' condition is never true, so the function runs to completion but returns zero rows.
+-- The query is written to ERROR if the tablespace doesn't exist, rather than silently failing to call pg_ls_tmpdir()
+SELECT c.* FROM (SELECT oid FROM pg_tablespace b WHERE b.spcname='regress_tblspace' UNION SELECT 0 ORDER BY 1 DESC LIMIT 1) AS b , pg_ls_tmpdir(oid) AS c WHERE c.name='Does not exist';
+ name | size | modification 
+------+------+--------------
+(0 rows)
+
 -- try setting and resetting some properties for the new tablespace
 ALTER TABLESPACE regress_tblspace SET (random_page_cost = 1.0, seq_page_cost = 1.1);
 ALTER TABLESPACE regress_tblspace SET (some_nonexistent_parameter = true);  -- fail
diff --git a/src/test/regress/sql/misc_functions.sql b/src/test/regress/sql/misc_functions.sql
index f6857ad177..68e2bc6586 100644
--- a/src/test/regress/sql/misc_functions.sql
+++ b/src/test/regress/sql/misc_functions.sql
@@ -65,6 +65,17 @@ select * from (select pg_ls_dir('.', false, false) as name) as ls where ls.name=
 select pg_ls_dir('does not exist', true, false); -- ok with missingok=true
 select pg_ls_dir('does not exist'); -- fails with missingok=false
 
+-- This tests the missing_ok parameter, which causes pg_ls_tmpdir to succeed even if the tmpdir doesn't exist yet
+-- The name='' condition is never true, so the function runs to completion but returns zero rows.
+select * from pg_ls_tmpdir() where name='Does not exist';
+
+select filename, isdir from pg_ls_dir_metadata('.') where filename='.';
+
+select filename, isdir from pg_ls_dir_metadata('.', false, false) where filename='.'; -- include_dot_dirs=false
+
+-- Check that expected columns are present
+select * from pg_ls_dir_metadata('.') limit 0;
+
 --
 -- Test adding a support function to a subject function
 --
-- 
2.17.0


--d6Gm4EdcadzBjdND
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
 filename="v22-0005-pg_ls_tmpdir-to-show-directories-and-isdir-argum.patch"



^ permalink  raw  reply  [nested|flat] 68+ messages in thread

* [PATCH v23 04/10] Add pg_ls_dir_metadata to list a dir with file metadata..
@ 2020-03-10 03:40  Justin Pryzby <[email protected]>
  0 siblings, 0 replies; 68+ messages in thread

From: Justin Pryzby @ 2020-03-10 03:40 UTC (permalink / raw)

Generalize pg_ls_dir_files and retire pg_ls_dir

Need catversion bumped?
---
 doc/src/sgml/func.sgml                       |  21 ++
 src/backend/catalog/system_views.sql         |   1 +
 src/backend/utils/adt/genfile.c              | 229 +++++++++++--------
 src/include/catalog/pg_proc.dat              |  12 +
 src/test/regress/expected/misc_functions.out |  24 ++
 src/test/regress/input/tablespace.source     |   5 +
 src/test/regress/output/tablespace.source    |   8 +
 src/test/regress/sql/misc_functions.sql      |  11 +
 8 files changed, 220 insertions(+), 91 deletions(-)

diff --git a/doc/src/sgml/func.sgml b/doc/src/sgml/func.sgml
index 7b1dc264f6..0a924fdcc3 100644
--- a/doc/src/sgml/func.sgml
+++ b/doc/src/sgml/func.sgml
@@ -25738,6 +25738,27 @@ SELECT pg_size_pretty(sum(pg_relation_size(relid))) AS total_size
        </para></entry>
       </row>
 
+      <row>
+       <entry role="func_table_entry"><para role="func_signature">
+        <indexterm>
+         <primary>pg_ls_dir_metadata</primary>
+        </indexterm>
+        <function>pg_ls_dir_metadata</function> ( <parameter>dirname</parameter> <type>text</type>
+        <optional>, <parameter>missing_ok</parameter> <type>boolean</type>,
+        <parameter>include_dot_dirs</parameter> <type>boolean</type>] </optional> )
+        <returnvalue>setof record</returnvalue>
+        ( <parameter>name</parameter> <type>text</type>,
+        <parameter>size</parameter> <type>bigint</type>,
+        <parameter>modification</parameter> <type>timestamp with time zone</type> )
+       </para>
+       <para>
+        For each file in the specified directory, list the file and its
+        metadata.
+        Restricted to superusers by default, but other users can be granted
+        EXECUTE to run the function.
+       </para></entry>
+      </row>
+
       <row>
        <entry role="func_table_entry"><para role="func_signature">
         <indexterm>
diff --git a/src/backend/catalog/system_views.sql b/src/backend/catalog/system_views.sql
index 2e4aa1c4b6..e7647787cf 100644
--- a/src/backend/catalog/system_views.sql
+++ b/src/backend/catalog/system_views.sql
@@ -1503,6 +1503,7 @@ REVOKE EXECUTE ON FUNCTION pg_stat_file(text,boolean) FROM public;
 
 REVOKE EXECUTE ON FUNCTION pg_ls_dir(text) FROM public;
 REVOKE EXECUTE ON FUNCTION pg_ls_dir(text,boolean,boolean) FROM public;
+REVOKE EXECUTE ON FUNCTION pg_ls_dir_metadata(text,boolean,boolean) FROM public;
 
 --
 -- We also set up some things as accessible to standard roles.
diff --git a/src/backend/utils/adt/genfile.c b/src/backend/utils/adt/genfile.c
index 9f4927220b..6f003bbf67 100644
--- a/src/backend/utils/adt/genfile.c
+++ b/src/backend/utils/adt/genfile.c
@@ -36,6 +36,21 @@
 #include "utils/syscache.h"
 #include "utils/timestamp.h"
 
+static Datum pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, int flags);
+
+#define	LS_DIR_ISDIR				(1<<0) /* Show column: isdir */
+#define	LS_DIR_METADATA				(1<<1) /* Show columns: mtime, size */
+#define	LS_DIR_MISSING_OK			(1<<2) /* Ignore ENOENT if the toplevel dir is missing */
+#define	LS_DIR_SKIP_DOT_DIRS		(1<<3) /* Do not show . or .. */
+#define	LS_DIR_SKIP_HIDDEN			(1<<4) /* Do not show anything begining with . */
+#define	LS_DIR_SKIP_DIRS			(1<<5) /* Do not show directories */
+#define	LS_DIR_SKIP_SPECIAL			(1<<6) /* Do not show special file types */
+
+/*
+ * Shortcut for the historic behavior of the pg_ls_* functions (not including
+ * pg_ls_dir, which skips different files and doesn't show metadata).
+ */
+#define LS_DIR_HISTORIC				(LS_DIR_SKIP_DIRS | LS_DIR_SKIP_HIDDEN | LS_DIR_SKIP_SPECIAL | LS_DIR_METADATA)
 
 /*
  * Convert a "text" filename argument to C string, and check it's allowable.
@@ -449,6 +464,11 @@ pg_stat_file(PG_FUNCTION_ARGS)
 	values[4] = TimestampTzGetDatum(time_t_to_timestamptz(fst.st_ctime));
 #endif
 	values[5] = BoolGetDatum(S_ISDIR(fst.st_mode));
+#ifdef WIN32
+	/* Links should have isdir=false */
+	if (pgwin32_is_junction(filename))
+		values[5] = BoolGetDatum(false);
+#endif
 
 	tuple = heap_form_tuple(tupdesc, values, isnull);
 
@@ -476,79 +496,9 @@ pg_stat_file_1arg(PG_FUNCTION_ARGS)
 Datum
 pg_ls_dir(PG_FUNCTION_ARGS)
 {
-	ReturnSetInfo *rsinfo = (ReturnSetInfo *) fcinfo->resultinfo;
-	char	   *location;
-	bool		missing_ok = false;
-	bool		include_dot_dirs = false;
-	bool		randomAccess;
-	TupleDesc	tupdesc;
-	Tuplestorestate *tupstore;
-	DIR		   *dirdesc;
-	struct dirent *de;
-	MemoryContext oldcontext;
-
-	location = convert_and_check_filename(PG_GETARG_TEXT_PP(0));
-
-	/* check the optional arguments */
-	if (PG_NARGS() == 3)
-	{
-		if (!PG_ARGISNULL(1))
-			missing_ok = PG_GETARG_BOOL(1);
-		if (!PG_ARGISNULL(2))
-			include_dot_dirs = PG_GETARG_BOOL(2);
-	}
-
-	/* check to see if caller supports us returning a tuplestore */
-	if (rsinfo == NULL || !IsA(rsinfo, ReturnSetInfo))
-		ereport(ERROR,
-				(errcode(ERRCODE_FEATURE_NOT_SUPPORTED),
-				 errmsg("set-valued function called in context that cannot accept a set")));
-	if (!(rsinfo->allowedModes & SFRM_Materialize))
-		ereport(ERROR,
-				(errcode(ERRCODE_SYNTAX_ERROR),
-				 errmsg("materialize mode required, but it is not allowed in this context")));
-
-	/* The tupdesc and tuplestore must be created in ecxt_per_query_memory */
-	oldcontext = MemoryContextSwitchTo(rsinfo->econtext->ecxt_per_query_memory);
-
-	tupdesc = CreateTemplateTupleDesc(1);
-	TupleDescInitEntry(tupdesc, (AttrNumber) 1, "pg_ls_dir", TEXTOID, -1, 0);
-
-	randomAccess = (rsinfo->allowedModes & SFRM_Materialize_Random) != 0;
-	tupstore = tuplestore_begin_heap(randomAccess, false, work_mem);
-	rsinfo->returnMode = SFRM_Materialize;
-	rsinfo->setResult = tupstore;
-	rsinfo->setDesc = tupdesc;
-
-	MemoryContextSwitchTo(oldcontext);
-
-	dirdesc = AllocateDir(location);
-	if (!dirdesc)
-	{
-		/* Return empty tuplestore if appropriate */
-		if (missing_ok && errno == ENOENT)
-			return (Datum) 0;
-		/* Otherwise, we can let ReadDir() throw the error */
-	}
-
-	while ((de = ReadDir(dirdesc, location)) != NULL)
-	{
-		Datum		values[1];
-		bool		nulls[1];
-
-		if (!include_dot_dirs &&
-			(strcmp(de->d_name, ".") == 0 ||
-			 strcmp(de->d_name, "..") == 0))
-			continue;
-
-		values[0] = CStringGetTextDatum(de->d_name);
-		nulls[0] = false;
-
-		tuplestore_putvalues(tupstore, tupdesc, values, nulls);
-	}
-
-	FreeDir(dirdesc);
-	return (Datum) 0;
+	text	*filename_t = PG_GETARG_TEXT_PP(0);
+	char	*filename = convert_and_check_filename(filename_t);
+	return pg_ls_dir_files(fcinfo, filename, LS_DIR_SKIP_DOT_DIRS);
 }
 
 /*
@@ -561,7 +511,9 @@ pg_ls_dir(PG_FUNCTION_ARGS)
 Datum
 pg_ls_dir_1arg(PG_FUNCTION_ARGS)
 {
-	return pg_ls_dir(fcinfo);
+	text	*filename_t = PG_GETARG_TEXT_PP(0);
+	char	*filename = convert_and_check_filename(filename_t);
+	return pg_ls_dir_files(fcinfo, filename, LS_DIR_SKIP_DOT_DIRS);
 }
 
 /*
@@ -571,7 +523,7 @@ pg_ls_dir_1arg(PG_FUNCTION_ARGS)
  * Other unreadable-directory cases throw an error.
  */
 static Datum
-pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, bool missing_ok)
+pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, int flags)
 {
 	ReturnSetInfo *rsinfo = (ReturnSetInfo *) fcinfo->resultinfo;
 	bool		randomAccess;
@@ -580,6 +532,32 @@ pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, bool missing_ok)
 	DIR		   *dirdesc;
 	struct dirent *de;
 	MemoryContext oldcontext;
+	TypeFuncClass	tuptype ;
+
+	/* isdir depends on metadata */
+	Assert(!(flags&LS_DIR_ISDIR) || (flags&LS_DIR_METADATA));
+	/* Unreasonable to show isdir and skip dirs */
+	Assert(!(flags&LS_DIR_ISDIR) || !(flags&LS_DIR_SKIP_DIRS));
+
+	/* check the optional arguments */
+	if (PG_NARGS() == 3)
+	{
+		if (!PG_ARGISNULL(1))
+		{
+			if (PG_GETARG_BOOL(1))
+				flags |= LS_DIR_MISSING_OK;
+			else
+				flags &= ~LS_DIR_MISSING_OK;
+		}
+
+		if (!PG_ARGISNULL(2))
+		{
+			if (PG_GETARG_BOOL(2))
+				flags &= ~LS_DIR_SKIP_DOT_DIRS;
+			else
+				flags |= LS_DIR_SKIP_DOT_DIRS;
+		}
+	}
 
 	/* check to see if caller supports us returning a tuplestore */
 	if (rsinfo == NULL || !IsA(rsinfo, ReturnSetInfo))
@@ -594,8 +572,20 @@ pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, bool missing_ok)
 	/* The tupdesc and tuplestore must be created in ecxt_per_query_memory */
 	oldcontext = MemoryContextSwitchTo(rsinfo->econtext->ecxt_per_query_memory);
 
-	if (get_call_result_type(fcinfo, NULL, &tupdesc) != TYPEFUNC_COMPOSITE)
-		elog(ERROR, "return type must be a row type");
+	tuptype = get_call_result_type(fcinfo, NULL, &tupdesc);
+	if (flags & LS_DIR_METADATA)
+	{
+		if (tuptype != TYPEFUNC_COMPOSITE)
+			elog(ERROR, "return type must be a row type");
+	}
+	else
+	{
+		/* pg_ls_dir returns a simple scalar */
+		if (tuptype != TYPEFUNC_SCALAR)
+			elog(ERROR, "return type must be a scalar type");
+		tupdesc = CreateTemplateTupleDesc(1);
+		TupleDescInitEntry(tupdesc, (AttrNumber) 1, "column", TEXTOID, -1, 0);
+	}
 
 	randomAccess = (rsinfo->allowedModes & SFRM_Materialize_Random) != 0;
 	tupstore = tuplestore_begin_heap(randomAccess, false, work_mem);
@@ -614,20 +604,27 @@ pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, bool missing_ok)
 	if (!dirdesc)
 	{
 		/* Return empty tuplestore if appropriate */
-		if (missing_ok && errno == ENOENT)
+		if (flags & LS_DIR_MISSING_OK && errno == ENOENT)
 			return (Datum) 0;
 		/* Otherwise, we can let ReadDir() throw the error */
 	}
 
 	while ((de = ReadDir(dirdesc, dir)) != NULL)
 	{
-		Datum		values[3];
-		bool		nulls[3];
+		Datum		values[4];
+		bool		nulls[4];
 		char		path[MAXPGPATH * 2];
 		struct stat attrib;
 
-		/* Skip hidden files */
-		if (de->d_name[0] == '.')
+		/* Skip dot dirs? */
+		if (flags & LS_DIR_SKIP_DOT_DIRS &&
+			(strcmp(de->d_name, ".") == 0 ||
+			 strcmp(de->d_name, "..") == 0))
+			continue;
+
+		/* Skip hidden files? */
+		if (flags & LS_DIR_SKIP_HIDDEN &&
+			de->d_name[0] == '.')
 			continue;
 
 		/* Get the file info */
@@ -642,13 +639,34 @@ pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, bool missing_ok)
 					 errmsg("could not stat file \"%s\": %m", path)));
 		}
 
-		/* Ignore anything but regular files */
-		if (!S_ISREG(attrib.st_mode))
-			continue;
+		/* Skip dirs or special files? */
+		if (S_ISDIR(attrib.st_mode))
+		{
+			if (flags & LS_DIR_SKIP_DIRS)
+				continue;
+		}
+		else if (!S_ISREG(attrib.st_mode))
+		{
+			if (flags & LS_DIR_SKIP_SPECIAL)
+				continue;
+		}
 
 		values[0] = CStringGetTextDatum(de->d_name);
-		values[1] = Int64GetDatum((int64) attrib.st_size);
-		values[2] = TimestampTzGetDatum(time_t_to_timestamptz(attrib.st_mtime));
+		if (flags & LS_DIR_METADATA)
+		{
+			values[1] = Int64GetDatum((int64) attrib.st_size);
+			values[2] = TimestampTzGetDatum(time_t_to_timestamptz(attrib.st_mtime));
+			if (flags & LS_DIR_ISDIR)
+			{
+				values[3] = BoolGetDatum(S_ISDIR(attrib.st_mode));
+#ifdef WIN32
+				/* Links should have isdir=false */
+				if (pgwin32_is_junction(path))
+					values[3] = BoolGetDatum(false);
+#endif
+			}
+		}
+
 		memset(nulls, 0, sizeof(nulls));
 
 		tuplestore_putvalues(tupstore, tupdesc, values, nulls);
@@ -662,14 +680,14 @@ pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, bool missing_ok)
 Datum
 pg_ls_logdir(PG_FUNCTION_ARGS)
 {
-	return pg_ls_dir_files(fcinfo, Log_directory, false);
+	return pg_ls_dir_files(fcinfo, Log_directory, LS_DIR_HISTORIC);
 }
 
 /* Function to return the list of files in the WAL directory */
 Datum
 pg_ls_waldir(PG_FUNCTION_ARGS)
 {
-	return pg_ls_dir_files(fcinfo, XLOGDIR, false);
+	return pg_ls_dir_files(fcinfo, XLOGDIR, LS_DIR_HISTORIC);
 }
 
 /*
@@ -687,7 +705,8 @@ pg_ls_tmpdir(FunctionCallInfo fcinfo, Oid tblspc)
 						tblspc)));
 
 	TempTablespacePath(path, tblspc);
-	return pg_ls_dir_files(fcinfo, path, true);
+	return pg_ls_dir_files(fcinfo, path,
+			LS_DIR_HISTORIC | LS_DIR_MISSING_OK);
 }
 
 /*
@@ -716,5 +735,33 @@ pg_ls_tmpdir_1arg(PG_FUNCTION_ARGS)
 Datum
 pg_ls_archive_statusdir(PG_FUNCTION_ARGS)
 {
-	return pg_ls_dir_files(fcinfo, XLOGDIR "/archive_status", true);
+	return pg_ls_dir_files(fcinfo, XLOGDIR "/archive_status",
+			LS_DIR_HISTORIC | LS_DIR_MISSING_OK);
+}
+
+/*
+ * Return the list of files and metadata in an arbitrary directory.
+ */
+Datum
+pg_ls_dir_metadata(PG_FUNCTION_ARGS)
+{
+	char	*dirname = convert_and_check_filename(PG_GETARG_TEXT_PP(0));
+
+	return pg_ls_dir_files(fcinfo, dirname,
+			LS_DIR_METADATA | LS_DIR_SKIP_SPECIAL | LS_DIR_ISDIR);
+}
+
+/*
+ * Return the list of files and metadata in an arbitrary directory.
+ * note: this wrapper is necessary to pass the sanity check in opr_sanity,
+ * which checks that all built-in functions that share the implementing C
+ * function take the same number of arguments.
+ */
+Datum
+pg_ls_dir_metadata_1arg(PG_FUNCTION_ARGS)
+{
+	char	*dirname = convert_and_check_filename(PG_GETARG_TEXT_PP(0));
+
+	return pg_ls_dir_files(fcinfo, dirname,
+			LS_DIR_METADATA | LS_DIR_SKIP_SPECIAL | LS_DIR_ISDIR);
 }
diff --git a/src/include/catalog/pg_proc.dat b/src/include/catalog/pg_proc.dat
index c01da4bf01..add3c7dc62 100644
--- a/src/include/catalog/pg_proc.dat
+++ b/src/include/catalog/pg_proc.dat
@@ -10971,6 +10971,18 @@
   proallargtypes => '{oid,text,int8,timestamptz}', proargmodes => '{i,o,o,o}',
   proargnames => '{tablespace,name,size,modification}',
   prosrc => 'pg_ls_tmpdir_1arg' },
+{ oid => '9979', descr => 'list directory with metadata',
+  proname => 'pg_ls_dir_metadata', procost => '10', prorows => '20', proretset => 't',
+  provolatile => 'v', prorettype => 'record', proargtypes => 'text bool bool',
+  proallargtypes => '{text,bool,bool,text,int8,timestamptz,bool}', proargmodes => '{i,i,i,o,o,o,o}',
+  proargnames => '{dirname,missing_ok,include_dot_dirs,filename,size,modification,isdir}',
+  prosrc => 'pg_ls_dir_metadata' },
+{ oid => '9980', descr => 'list directory with metadata',
+  proname => 'pg_ls_dir_metadata', procost => '10', prorows => '20', proretset => 't',
+  provolatile => 'v', prorettype => 'record', proargtypes => 'text',
+  proallargtypes => '{text,text,int8,timestamptz,bool}', proargmodes => '{i,o,o,o,o}',
+  proargnames => '{dirname,filename,size,modification,isdir}',
+  prosrc => 'pg_ls_dir_metadata_1arg' },
 
 # hash partitioning constraint function
 { oid => '5028', descr => 'hash partition CHECK constraint',
diff --git a/src/test/regress/expected/misc_functions.out b/src/test/regress/expected/misc_functions.out
index 2e87c548eb..20bfe71822 100644
--- a/src/test/regress/expected/misc_functions.out
+++ b/src/test/regress/expected/misc_functions.out
@@ -219,6 +219,30 @@ select pg_ls_dir('does not exist', true, false); -- ok with missingok=true
 
 select pg_ls_dir('does not exist'); -- fails with missingok=false
 ERROR:  could not open directory "does not exist": No such file or directory
+-- This tests the missing_ok parameter, which causes pg_ls_tmpdir to succeed even if the tmpdir doesn't exist yet
+-- The name='' condition is never true, so the function runs to completion but returns zero rows.
+select * from pg_ls_tmpdir() where name='Does not exist';
+ name | size | modification 
+------+------+--------------
+(0 rows)
+
+select filename, isdir from pg_ls_dir_metadata('.') where filename='.';
+ filename | isdir 
+----------+-------
+ .        | t
+(1 row)
+
+select filename, isdir from pg_ls_dir_metadata('.', false, false) where filename='.'; -- include_dot_dirs=false
+ filename | isdir 
+----------+-------
+(0 rows)
+
+-- Check that expected columns are present
+select * from pg_ls_dir_metadata('.') limit 0;
+ filename | size | modification | isdir 
+----------+------+--------------+-------
+(0 rows)
+
 --
 -- Test adding a support function to a subject function
 --
diff --git a/src/test/regress/input/tablespace.source b/src/test/regress/input/tablespace.source
index a5f61a35dc..0b9cfe615e 100644
--- a/src/test/regress/input/tablespace.source
+++ b/src/test/regress/input/tablespace.source
@@ -11,6 +11,11 @@ DROP TABLESPACE regress_tblspacewith;
 -- create a tablespace we can use
 CREATE TABLESPACE regress_tblspace LOCATION '@testtablespace@';
 
+-- This tests the missing_ok parameter, which causes pg_ls_tmpdir to succeed even if the tmpdir doesn't exist yet
+-- The name='' condition is never true, so the function runs to completion but returns zero rows.
+-- The query is written to ERROR if the tablespace doesn't exist, rather than silently failing to call pg_ls_tmpdir()
+SELECT c.* FROM (SELECT oid FROM pg_tablespace b WHERE b.spcname='regress_tblspace' UNION SELECT 0 ORDER BY 1 DESC LIMIT 1) AS b , pg_ls_tmpdir(oid) AS c WHERE c.name='Does not exist';
+
 -- try setting and resetting some properties for the new tablespace
 ALTER TABLESPACE regress_tblspace SET (random_page_cost = 1.0, seq_page_cost = 1.1);
 ALTER TABLESPACE regress_tblspace SET (some_nonexistent_parameter = true);  -- fail
diff --git a/src/test/regress/output/tablespace.source b/src/test/regress/output/tablespace.source
index 162b591b31..a42714bf40 100644
--- a/src/test/regress/output/tablespace.source
+++ b/src/test/regress/output/tablespace.source
@@ -13,6 +13,14 @@ SELECT spcoptions FROM pg_tablespace WHERE spcname = 'regress_tblspacewith';
 DROP TABLESPACE regress_tblspacewith;
 -- create a tablespace we can use
 CREATE TABLESPACE regress_tblspace LOCATION '@testtablespace@';
+-- This tests the missing_ok parameter, which causes pg_ls_tmpdir to succeed even if the tmpdir doesn't exist yet
+-- The name='' condition is never true, so the function runs to completion but returns zero rows.
+-- The query is written to ERROR if the tablespace doesn't exist, rather than silently failing to call pg_ls_tmpdir()
+SELECT c.* FROM (SELECT oid FROM pg_tablespace b WHERE b.spcname='regress_tblspace' UNION SELECT 0 ORDER BY 1 DESC LIMIT 1) AS b , pg_ls_tmpdir(oid) AS c WHERE c.name='Does not exist';
+ name | size | modification 
+------+------+--------------
+(0 rows)
+
 -- try setting and resetting some properties for the new tablespace
 ALTER TABLESPACE regress_tblspace SET (random_page_cost = 1.0, seq_page_cost = 1.1);
 ALTER TABLESPACE regress_tblspace SET (some_nonexistent_parameter = true);  -- fail
diff --git a/src/test/regress/sql/misc_functions.sql b/src/test/regress/sql/misc_functions.sql
index f6857ad177..68e2bc6586 100644
--- a/src/test/regress/sql/misc_functions.sql
+++ b/src/test/regress/sql/misc_functions.sql
@@ -65,6 +65,17 @@ select * from (select pg_ls_dir('.', false, false) as name) as ls where ls.name=
 select pg_ls_dir('does not exist', true, false); -- ok with missingok=true
 select pg_ls_dir('does not exist'); -- fails with missingok=false
 
+-- This tests the missing_ok parameter, which causes pg_ls_tmpdir to succeed even if the tmpdir doesn't exist yet
+-- The name='' condition is never true, so the function runs to completion but returns zero rows.
+select * from pg_ls_tmpdir() where name='Does not exist';
+
+select filename, isdir from pg_ls_dir_metadata('.') where filename='.';
+
+select filename, isdir from pg_ls_dir_metadata('.', false, false) where filename='.'; -- include_dot_dirs=false
+
+-- Check that expected columns are present
+select * from pg_ls_dir_metadata('.') limit 0;
+
 --
 -- Test adding a support function to a subject function
 --
-- 
2.17.0


--mhjHhnbe5PrRcwjY
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
 filename="v23-0005-pg_ls_tmpdir-to-show-directories-and-isdir-argum.patch"



^ permalink  raw  reply  [nested|flat] 68+ messages in thread

* [PATCH v24 04/11] Add pg_ls_dir_metadata to list a dir with file metadata..
@ 2020-03-10 03:40  Justin Pryzby <[email protected]>
  0 siblings, 0 replies; 68+ messages in thread

From: Justin Pryzby @ 2020-03-10 03:40 UTC (permalink / raw)

Generalize pg_ls_dir_files and retire pg_ls_dir

Need catversion bumped?
---
 doc/src/sgml/func.sgml                       |  21 ++
 src/backend/catalog/system_views.sql         |   1 +
 src/backend/utils/adt/genfile.c              | 231 +++++++++++--------
 src/include/catalog/pg_proc.dat              |  12 +
 src/test/regress/expected/misc_functions.out |  24 ++
 src/test/regress/input/tablespace.source     |   5 +
 src/test/regress/output/tablespace.source    |   8 +
 src/test/regress/sql/misc_functions.sql      |  11 +
 8 files changed, 221 insertions(+), 92 deletions(-)

diff --git a/doc/src/sgml/func.sgml b/doc/src/sgml/func.sgml
index 507bc1a668..5e267c4f8d 100644
--- a/doc/src/sgml/func.sgml
+++ b/doc/src/sgml/func.sgml
@@ -25757,6 +25757,27 @@ SELECT pg_size_pretty(sum(pg_relation_size(relid))) AS total_size
        </para></entry>
       </row>
 
+      <row>
+       <entry role="func_table_entry"><para role="func_signature">
+        <indexterm>
+         <primary>pg_ls_dir_metadata</primary>
+        </indexterm>
+        <function>pg_ls_dir_metadata</function> ( <parameter>dirname</parameter> <type>text</type>
+        <optional>, <parameter>missing_ok</parameter> <type>boolean</type>,
+        <parameter>include_dot_dirs</parameter> <type>boolean</type> </optional> )
+        <returnvalue>setof record</returnvalue>
+        ( <parameter>filename</parameter> <type>text</type>,
+        <parameter>size</parameter> <type>bigint</type>,
+        <parameter>modification</parameter> <type>timestamp with time zone</type> )
+       </para>
+       <para>
+        For each file in the specified directory, list the file and its
+        metadata.
+        Restricted to superusers by default, but other users can be granted
+        EXECUTE to run the function.
+       </para></entry>
+      </row>
+
       <row>
        <entry role="func_table_entry"><para role="func_signature">
         <indexterm>
diff --git a/src/backend/catalog/system_views.sql b/src/backend/catalog/system_views.sql
index 2e4aa1c4b6..e7647787cf 100644
--- a/src/backend/catalog/system_views.sql
+++ b/src/backend/catalog/system_views.sql
@@ -1503,6 +1503,7 @@ REVOKE EXECUTE ON FUNCTION pg_stat_file(text,boolean) FROM public;
 
 REVOKE EXECUTE ON FUNCTION pg_ls_dir(text) FROM public;
 REVOKE EXECUTE ON FUNCTION pg_ls_dir(text,boolean,boolean) FROM public;
+REVOKE EXECUTE ON FUNCTION pg_ls_dir_metadata(text,boolean,boolean) FROM public;
 
 --
 -- We also set up some things as accessible to standard roles.
diff --git a/src/backend/utils/adt/genfile.c b/src/backend/utils/adt/genfile.c
index 9f4927220b..7351615f8a 100644
--- a/src/backend/utils/adt/genfile.c
+++ b/src/backend/utils/adt/genfile.c
@@ -36,6 +36,21 @@
 #include "utils/syscache.h"
 #include "utils/timestamp.h"
 
+static Datum pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, int flags);
+
+#define	LS_DIR_ISDIR				(1<<0) /* Show column: isdir */
+#define	LS_DIR_METADATA				(1<<1) /* Show columns: mtime, size */
+#define	LS_DIR_MISSING_OK			(1<<2) /* Ignore ENOENT if the toplevel dir is missing */
+#define	LS_DIR_SKIP_DOT_DIRS		(1<<3) /* Do not show . or .. */
+#define	LS_DIR_SKIP_HIDDEN			(1<<4) /* Do not show anything begining with . */
+#define	LS_DIR_SKIP_DIRS			(1<<5) /* Do not show directories */
+#define	LS_DIR_SKIP_SPECIAL			(1<<6) /* Do not show special file types */
+
+/*
+ * Shortcut for the historic behavior of the pg_ls_* functions (not including
+ * pg_ls_dir, which skips different files and doesn't show metadata).
+ */
+#define LS_DIR_HISTORIC				(LS_DIR_SKIP_DIRS | LS_DIR_SKIP_HIDDEN | LS_DIR_SKIP_SPECIAL | LS_DIR_METADATA)
 
 /*
  * Convert a "text" filename argument to C string, and check it's allowable.
@@ -449,6 +464,11 @@ pg_stat_file(PG_FUNCTION_ARGS)
 	values[4] = TimestampTzGetDatum(time_t_to_timestamptz(fst.st_ctime));
 #endif
 	values[5] = BoolGetDatum(S_ISDIR(fst.st_mode));
+#ifdef WIN32
+	/* Links should have isdir=false */
+	if (pgwin32_is_junction(filename))
+		values[5] = BoolGetDatum(false);
+#endif
 
 	tuple = heap_form_tuple(tupdesc, values, isnull);
 
@@ -476,79 +496,9 @@ pg_stat_file_1arg(PG_FUNCTION_ARGS)
 Datum
 pg_ls_dir(PG_FUNCTION_ARGS)
 {
-	ReturnSetInfo *rsinfo = (ReturnSetInfo *) fcinfo->resultinfo;
-	char	   *location;
-	bool		missing_ok = false;
-	bool		include_dot_dirs = false;
-	bool		randomAccess;
-	TupleDesc	tupdesc;
-	Tuplestorestate *tupstore;
-	DIR		   *dirdesc;
-	struct dirent *de;
-	MemoryContext oldcontext;
-
-	location = convert_and_check_filename(PG_GETARG_TEXT_PP(0));
-
-	/* check the optional arguments */
-	if (PG_NARGS() == 3)
-	{
-		if (!PG_ARGISNULL(1))
-			missing_ok = PG_GETARG_BOOL(1);
-		if (!PG_ARGISNULL(2))
-			include_dot_dirs = PG_GETARG_BOOL(2);
-	}
-
-	/* check to see if caller supports us returning a tuplestore */
-	if (rsinfo == NULL || !IsA(rsinfo, ReturnSetInfo))
-		ereport(ERROR,
-				(errcode(ERRCODE_FEATURE_NOT_SUPPORTED),
-				 errmsg("set-valued function called in context that cannot accept a set")));
-	if (!(rsinfo->allowedModes & SFRM_Materialize))
-		ereport(ERROR,
-				(errcode(ERRCODE_SYNTAX_ERROR),
-				 errmsg("materialize mode required, but it is not allowed in this context")));
-
-	/* The tupdesc and tuplestore must be created in ecxt_per_query_memory */
-	oldcontext = MemoryContextSwitchTo(rsinfo->econtext->ecxt_per_query_memory);
-
-	tupdesc = CreateTemplateTupleDesc(1);
-	TupleDescInitEntry(tupdesc, (AttrNumber) 1, "pg_ls_dir", TEXTOID, -1, 0);
-
-	randomAccess = (rsinfo->allowedModes & SFRM_Materialize_Random) != 0;
-	tupstore = tuplestore_begin_heap(randomAccess, false, work_mem);
-	rsinfo->returnMode = SFRM_Materialize;
-	rsinfo->setResult = tupstore;
-	rsinfo->setDesc = tupdesc;
-
-	MemoryContextSwitchTo(oldcontext);
-
-	dirdesc = AllocateDir(location);
-	if (!dirdesc)
-	{
-		/* Return empty tuplestore if appropriate */
-		if (missing_ok && errno == ENOENT)
-			return (Datum) 0;
-		/* Otherwise, we can let ReadDir() throw the error */
-	}
-
-	while ((de = ReadDir(dirdesc, location)) != NULL)
-	{
-		Datum		values[1];
-		bool		nulls[1];
-
-		if (!include_dot_dirs &&
-			(strcmp(de->d_name, ".") == 0 ||
-			 strcmp(de->d_name, "..") == 0))
-			continue;
-
-		values[0] = CStringGetTextDatum(de->d_name);
-		nulls[0] = false;
-
-		tuplestore_putvalues(tupstore, tupdesc, values, nulls);
-	}
-
-	FreeDir(dirdesc);
-	return (Datum) 0;
+	text	*filename_t = PG_GETARG_TEXT_PP(0);
+	char	*filename = convert_and_check_filename(filename_t);
+	return pg_ls_dir_files(fcinfo, filename, LS_DIR_SKIP_DOT_DIRS);
 }
 
 /*
@@ -561,17 +511,19 @@ pg_ls_dir(PG_FUNCTION_ARGS)
 Datum
 pg_ls_dir_1arg(PG_FUNCTION_ARGS)
 {
-	return pg_ls_dir(fcinfo);
+	text	*filename_t = PG_GETARG_TEXT_PP(0);
+	char	*filename = convert_and_check_filename(filename_t);
+	return pg_ls_dir_files(fcinfo, filename, LS_DIR_SKIP_DOT_DIRS);
 }
 
 /*
  * Generic function to return a directory listing of files.
  *
- * If the directory isn't there, silently return an empty set if missing_ok.
+ * If the directory isn't there, silently return an empty set if MISSING_OK.
  * Other unreadable-directory cases throw an error.
  */
 static Datum
-pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, bool missing_ok)
+pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, int flags)
 {
 	ReturnSetInfo *rsinfo = (ReturnSetInfo *) fcinfo->resultinfo;
 	bool		randomAccess;
@@ -580,6 +532,32 @@ pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, bool missing_ok)
 	DIR		   *dirdesc;
 	struct dirent *de;
 	MemoryContext oldcontext;
+	TypeFuncClass	tuptype ;
+
+	/* isdir depends on metadata */
+	Assert(!(flags&LS_DIR_ISDIR) || (flags&LS_DIR_METADATA));
+	/* Unreasonable to show isdir and skip dirs */
+	Assert(!(flags&LS_DIR_ISDIR) || !(flags&LS_DIR_SKIP_DIRS));
+
+	/* check the optional arguments */
+	if (PG_NARGS() == 3)
+	{
+		if (!PG_ARGISNULL(1))
+		{
+			if (PG_GETARG_BOOL(1))
+				flags |= LS_DIR_MISSING_OK;
+			else
+				flags &= ~LS_DIR_MISSING_OK;
+		}
+
+		if (!PG_ARGISNULL(2))
+		{
+			if (PG_GETARG_BOOL(2))
+				flags &= ~LS_DIR_SKIP_DOT_DIRS;
+			else
+				flags |= LS_DIR_SKIP_DOT_DIRS;
+		}
+	}
 
 	/* check to see if caller supports us returning a tuplestore */
 	if (rsinfo == NULL || !IsA(rsinfo, ReturnSetInfo))
@@ -594,8 +572,20 @@ pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, bool missing_ok)
 	/* The tupdesc and tuplestore must be created in ecxt_per_query_memory */
 	oldcontext = MemoryContextSwitchTo(rsinfo->econtext->ecxt_per_query_memory);
 
-	if (get_call_result_type(fcinfo, NULL, &tupdesc) != TYPEFUNC_COMPOSITE)
-		elog(ERROR, "return type must be a row type");
+	tuptype = get_call_result_type(fcinfo, NULL, &tupdesc);
+	if (flags & LS_DIR_METADATA)
+	{
+		if (tuptype != TYPEFUNC_COMPOSITE)
+			elog(ERROR, "return type must be a row type");
+	}
+	else
+	{
+		/* pg_ls_dir returns a simple scalar */
+		if (tuptype != TYPEFUNC_SCALAR)
+			elog(ERROR, "return type must be a scalar type");
+		tupdesc = CreateTemplateTupleDesc(1);
+		TupleDescInitEntry(tupdesc, (AttrNumber) 1, "column", TEXTOID, -1, 0);
+	}
 
 	randomAccess = (rsinfo->allowedModes & SFRM_Materialize_Random) != 0;
 	tupstore = tuplestore_begin_heap(randomAccess, false, work_mem);
@@ -614,20 +604,27 @@ pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, bool missing_ok)
 	if (!dirdesc)
 	{
 		/* Return empty tuplestore if appropriate */
-		if (missing_ok && errno == ENOENT)
+		if (flags & LS_DIR_MISSING_OK && errno == ENOENT)
 			return (Datum) 0;
 		/* Otherwise, we can let ReadDir() throw the error */
 	}
 
 	while ((de = ReadDir(dirdesc, dir)) != NULL)
 	{
-		Datum		values[3];
-		bool		nulls[3];
+		Datum		values[4];
+		bool		nulls[4];
 		char		path[MAXPGPATH * 2];
 		struct stat attrib;
 
-		/* Skip hidden files */
-		if (de->d_name[0] == '.')
+		/* Skip dot dirs? */
+		if (flags & LS_DIR_SKIP_DOT_DIRS &&
+			(strcmp(de->d_name, ".") == 0 ||
+			 strcmp(de->d_name, "..") == 0))
+			continue;
+
+		/* Skip hidden files? */
+		if (flags & LS_DIR_SKIP_HIDDEN &&
+			de->d_name[0] == '.')
 			continue;
 
 		/* Get the file info */
@@ -642,13 +639,34 @@ pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, bool missing_ok)
 					 errmsg("could not stat file \"%s\": %m", path)));
 		}
 
-		/* Ignore anything but regular files */
-		if (!S_ISREG(attrib.st_mode))
-			continue;
+		/* Skip dirs or special files? */
+		if (S_ISDIR(attrib.st_mode))
+		{
+			if (flags & LS_DIR_SKIP_DIRS)
+				continue;
+		}
+		else if (!S_ISREG(attrib.st_mode))
+		{
+			if (flags & LS_DIR_SKIP_SPECIAL)
+				continue;
+		}
 
 		values[0] = CStringGetTextDatum(de->d_name);
-		values[1] = Int64GetDatum((int64) attrib.st_size);
-		values[2] = TimestampTzGetDatum(time_t_to_timestamptz(attrib.st_mtime));
+		if (flags & LS_DIR_METADATA)
+		{
+			values[1] = Int64GetDatum((int64) attrib.st_size);
+			values[2] = TimestampTzGetDatum(time_t_to_timestamptz(attrib.st_mtime));
+			if (flags & LS_DIR_ISDIR)
+			{
+				values[3] = BoolGetDatum(S_ISDIR(attrib.st_mode));
+#ifdef WIN32
+				/* Links should have isdir=false */
+				if (pgwin32_is_junction(path))
+					values[3] = BoolGetDatum(false);
+#endif
+			}
+		}
+
 		memset(nulls, 0, sizeof(nulls));
 
 		tuplestore_putvalues(tupstore, tupdesc, values, nulls);
@@ -662,14 +680,14 @@ pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, bool missing_ok)
 Datum
 pg_ls_logdir(PG_FUNCTION_ARGS)
 {
-	return pg_ls_dir_files(fcinfo, Log_directory, false);
+	return pg_ls_dir_files(fcinfo, Log_directory, LS_DIR_HISTORIC);
 }
 
 /* Function to return the list of files in the WAL directory */
 Datum
 pg_ls_waldir(PG_FUNCTION_ARGS)
 {
-	return pg_ls_dir_files(fcinfo, XLOGDIR, false);
+	return pg_ls_dir_files(fcinfo, XLOGDIR, LS_DIR_HISTORIC);
 }
 
 /*
@@ -687,7 +705,8 @@ pg_ls_tmpdir(FunctionCallInfo fcinfo, Oid tblspc)
 						tblspc)));
 
 	TempTablespacePath(path, tblspc);
-	return pg_ls_dir_files(fcinfo, path, true);
+	return pg_ls_dir_files(fcinfo, path,
+			LS_DIR_HISTORIC | LS_DIR_MISSING_OK);
 }
 
 /*
@@ -716,5 +735,33 @@ pg_ls_tmpdir_1arg(PG_FUNCTION_ARGS)
 Datum
 pg_ls_archive_statusdir(PG_FUNCTION_ARGS)
 {
-	return pg_ls_dir_files(fcinfo, XLOGDIR "/archive_status", true);
+	return pg_ls_dir_files(fcinfo, XLOGDIR "/archive_status",
+			LS_DIR_HISTORIC | LS_DIR_MISSING_OK);
+}
+
+/*
+ * Return the list of files and metadata in an arbitrary directory.
+ */
+Datum
+pg_ls_dir_metadata(PG_FUNCTION_ARGS)
+{
+	char	*dirname = convert_and_check_filename(PG_GETARG_TEXT_PP(0));
+
+	return pg_ls_dir_files(fcinfo, dirname,
+			LS_DIR_METADATA | LS_DIR_SKIP_SPECIAL | LS_DIR_ISDIR);
+}
+
+/*
+ * Return the list of files and metadata in an arbitrary directory.
+ * note: this wrapper is necessary to pass the sanity check in opr_sanity,
+ * which checks that all built-in functions that share the implementing C
+ * function take the same number of arguments.
+ */
+Datum
+pg_ls_dir_metadata_1arg(PG_FUNCTION_ARGS)
+{
+	char	*dirname = convert_and_check_filename(PG_GETARG_TEXT_PP(0));
+
+	return pg_ls_dir_files(fcinfo, dirname,
+			LS_DIR_METADATA | LS_DIR_SKIP_SPECIAL | LS_DIR_ISDIR);
 }
diff --git a/src/include/catalog/pg_proc.dat b/src/include/catalog/pg_proc.dat
index e7fbda9f81..f35a684dcf 100644
--- a/src/include/catalog/pg_proc.dat
+++ b/src/include/catalog/pg_proc.dat
@@ -10975,6 +10975,18 @@
   proallargtypes => '{oid,text,int8,timestamptz}', proargmodes => '{i,o,o,o}',
   proargnames => '{tablespace,name,size,modification}',
   prosrc => 'pg_ls_tmpdir_1arg' },
+{ oid => '9979', descr => 'list directory with metadata',
+  proname => 'pg_ls_dir_metadata', procost => '10', prorows => '20', proretset => 't',
+  provolatile => 'v', prorettype => 'record', proargtypes => 'text bool bool',
+  proallargtypes => '{text,bool,bool,text,int8,timestamptz,bool}', proargmodes => '{i,i,i,o,o,o,o}',
+  proargnames => '{dirname,missing_ok,include_dot_dirs,filename,size,modification,isdir}',
+  prosrc => 'pg_ls_dir_metadata' },
+{ oid => '9980', descr => 'list directory with metadata',
+  proname => 'pg_ls_dir_metadata', procost => '10', prorows => '20', proretset => 't',
+  provolatile => 'v', prorettype => 'record', proargtypes => 'text',
+  proallargtypes => '{text,text,int8,timestamptz,bool}', proargmodes => '{i,o,o,o,o}',
+  proargnames => '{dirname,filename,size,modification,isdir}',
+  prosrc => 'pg_ls_dir_metadata_1arg' },
 
 # hash partitioning constraint function
 { oid => '5028', descr => 'hash partition CHECK constraint',
diff --git a/src/test/regress/expected/misc_functions.out b/src/test/regress/expected/misc_functions.out
index edbfd9abc1..fd7c3c791f 100644
--- a/src/test/regress/expected/misc_functions.out
+++ b/src/test/regress/expected/misc_functions.out
@@ -225,6 +225,30 @@ select * from pg_stat_file('.') limit 0;
 ------+--------+--------------+--------+----------+-------
 (0 rows)
 
+-- This tests the missing_ok parameter, which causes pg_ls_tmpdir to succeed even if the tmpdir doesn't exist yet
+-- The name='' condition is never true, so the function runs to completion but returns zero rows.
+select * from pg_ls_tmpdir() where name='Does not exist';
+ name | size | modification 
+------+------+--------------
+(0 rows)
+
+select filename, isdir from pg_ls_dir_metadata('.') where filename='.';
+ filename | isdir 
+----------+-------
+ .        | t
+(1 row)
+
+select filename, isdir from pg_ls_dir_metadata('.', false, false) where filename='.'; -- include_dot_dirs=false
+ filename | isdir 
+----------+-------
+(0 rows)
+
+-- Check that expected columns are present
+select * from pg_ls_dir_metadata('.') limit 0;
+ filename | size | modification | isdir 
+----------+------+--------------+-------
+(0 rows)
+
 --
 -- Test adding a support function to a subject function
 --
diff --git a/src/test/regress/input/tablespace.source b/src/test/regress/input/tablespace.source
index a5f61a35dc..0b9cfe615e 100644
--- a/src/test/regress/input/tablespace.source
+++ b/src/test/regress/input/tablespace.source
@@ -11,6 +11,11 @@ DROP TABLESPACE regress_tblspacewith;
 -- create a tablespace we can use
 CREATE TABLESPACE regress_tblspace LOCATION '@testtablespace@';
 
+-- This tests the missing_ok parameter, which causes pg_ls_tmpdir to succeed even if the tmpdir doesn't exist yet
+-- The name='' condition is never true, so the function runs to completion but returns zero rows.
+-- The query is written to ERROR if the tablespace doesn't exist, rather than silently failing to call pg_ls_tmpdir()
+SELECT c.* FROM (SELECT oid FROM pg_tablespace b WHERE b.spcname='regress_tblspace' UNION SELECT 0 ORDER BY 1 DESC LIMIT 1) AS b , pg_ls_tmpdir(oid) AS c WHERE c.name='Does not exist';
+
 -- try setting and resetting some properties for the new tablespace
 ALTER TABLESPACE regress_tblspace SET (random_page_cost = 1.0, seq_page_cost = 1.1);
 ALTER TABLESPACE regress_tblspace SET (some_nonexistent_parameter = true);  -- fail
diff --git a/src/test/regress/output/tablespace.source b/src/test/regress/output/tablespace.source
index 162b591b31..a42714bf40 100644
--- a/src/test/regress/output/tablespace.source
+++ b/src/test/regress/output/tablespace.source
@@ -13,6 +13,14 @@ SELECT spcoptions FROM pg_tablespace WHERE spcname = 'regress_tblspacewith';
 DROP TABLESPACE regress_tblspacewith;
 -- create a tablespace we can use
 CREATE TABLESPACE regress_tblspace LOCATION '@testtablespace@';
+-- This tests the missing_ok parameter, which causes pg_ls_tmpdir to succeed even if the tmpdir doesn't exist yet
+-- The name='' condition is never true, so the function runs to completion but returns zero rows.
+-- The query is written to ERROR if the tablespace doesn't exist, rather than silently failing to call pg_ls_tmpdir()
+SELECT c.* FROM (SELECT oid FROM pg_tablespace b WHERE b.spcname='regress_tblspace' UNION SELECT 0 ORDER BY 1 DESC LIMIT 1) AS b , pg_ls_tmpdir(oid) AS c WHERE c.name='Does not exist';
+ name | size | modification 
+------+------+--------------
+(0 rows)
+
 -- try setting and resetting some properties for the new tablespace
 ALTER TABLESPACE regress_tblspace SET (random_page_cost = 1.0, seq_page_cost = 1.1);
 ALTER TABLESPACE regress_tblspace SET (some_nonexistent_parameter = true);  -- fail
diff --git a/src/test/regress/sql/misc_functions.sql b/src/test/regress/sql/misc_functions.sql
index b32a3a4b74..64d9d70874 100644
--- a/src/test/regress/sql/misc_functions.sql
+++ b/src/test/regress/sql/misc_functions.sql
@@ -68,6 +68,17 @@ select pg_ls_dir('does not exist'); -- fails with missingok=false
 -- Check that expected columns are present
 select * from pg_stat_file('.') limit 0;
 
+-- This tests the missing_ok parameter, which causes pg_ls_tmpdir to succeed even if the tmpdir doesn't exist yet
+-- The name='' condition is never true, so the function runs to completion but returns zero rows.
+select * from pg_ls_tmpdir() where name='Does not exist';
+
+select filename, isdir from pg_ls_dir_metadata('.') where filename='.';
+
+select filename, isdir from pg_ls_dir_metadata('.', false, false) where filename='.'; -- include_dot_dirs=false
+
+-- Check that expected columns are present
+select * from pg_ls_dir_metadata('.') limit 0;
+
 --
 -- Test adding a support function to a subject function
 --
-- 
2.17.0


--mPTHnM80CEnHQ2WJ
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
 filename="v24-0005-pg_ls_tmpdir-to-show-directories-and-isdir-argum.patch"



^ permalink  raw  reply  [nested|flat] 68+ messages in thread

* [PATCH v19 04/10] Add pg_ls_dir_metadata to list a dir with file metadata..
@ 2020-03-10 03:40  Justin Pryzby <[email protected]>
  0 siblings, 0 replies; 68+ messages in thread

From: Justin Pryzby @ 2020-03-10 03:40 UTC (permalink / raw)

Generalize pg_ls_dir_files and retire pg_ls_dir

Need catversion bumped?
---
 doc/src/sgml/func.sgml                       |  21 ++
 src/backend/catalog/system_views.sql         |   1 +
 src/backend/utils/adt/genfile.c              | 229 +++++++++++--------
 src/include/catalog/pg_proc.dat              |  12 +
 src/test/regress/expected/misc_functions.out |  24 ++
 src/test/regress/input/tablespace.source     |   5 +
 src/test/regress/output/tablespace.source    |   8 +
 src/test/regress/sql/misc_functions.sql      |  11 +
 8 files changed, 220 insertions(+), 91 deletions(-)

diff --git a/doc/src/sgml/func.sgml b/doc/src/sgml/func.sgml
index b7c450ea29..2931f9d7d5 100644
--- a/doc/src/sgml/func.sgml
+++ b/doc/src/sgml/func.sgml
@@ -25711,6 +25711,27 @@ SELECT pg_size_pretty(sum(pg_relation_size(relid))) AS total_size
        </para></entry>
       </row>
 
+      <row>
+       <entry role="func_table_entry"><para role="func_signature">
+        <indexterm>
+         <primary>pg_ls_dir_metadata</primary>
+        </indexterm>
+        <function>pg_ls_dir_metadata</function> ( <parameter>dirname</parameter> <type>text</type>
+        <optional>, <parameter>missing_ok</parameter> <type>boolean</type>,
+        <parameter>include_dot_dirs</parameter> <type>boolean</type>] </optional> )
+        <returnvalue>setof record</returnvalue>
+        ( <parameter>name</parameter> <type>text</type>,
+        <parameter>size</parameter> <type>bigint</type>,
+        <parameter>modification</parameter> <type>timestamp with time zone</type> )
+       </para>
+       <para>
+        For each file in the specified directory, list the file and its
+        metadata.
+        Restricted to superusers by default, but other users can be granted
+        EXECUTE to run the function.
+       </para></entry>
+      </row>
+
       <row>
        <entry role="func_table_entry"><para role="func_signature">
         <indexterm>
diff --git a/src/backend/catalog/system_views.sql b/src/backend/catalog/system_views.sql
index 5314e9348f..f5217a0d93 100644
--- a/src/backend/catalog/system_views.sql
+++ b/src/backend/catalog/system_views.sql
@@ -1481,6 +1481,7 @@ REVOKE EXECUTE ON FUNCTION pg_stat_file(text,boolean) FROM public;
 
 REVOKE EXECUTE ON FUNCTION pg_ls_dir(text) FROM public;
 REVOKE EXECUTE ON FUNCTION pg_ls_dir(text,boolean,boolean) FROM public;
+REVOKE EXECUTE ON FUNCTION pg_ls_dir_metadata(text,boolean,boolean) FROM public;
 
 --
 -- We also set up some things as accessible to standard roles.
diff --git a/src/backend/utils/adt/genfile.c b/src/backend/utils/adt/genfile.c
index 219ac160f8..4824a55480 100644
--- a/src/backend/utils/adt/genfile.c
+++ b/src/backend/utils/adt/genfile.c
@@ -36,6 +36,21 @@
 #include "utils/syscache.h"
 #include "utils/timestamp.h"
 
+static Datum pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, int flags);
+
+#define	LS_DIR_ISDIR				(1<<0) /* Show column: isdir */
+#define	LS_DIR_METADATA				(1<<1) /* Show columns: mtime, size */
+#define	LS_DIR_MISSING_OK			(1<<2) /* Ignore ENOENT if the toplevel dir is missing */
+#define	LS_DIR_SKIP_DOT_DIRS		(1<<3) /* Do not show . or .. */
+#define	LS_DIR_SKIP_HIDDEN			(1<<4) /* Do not show anything begining with . */
+#define	LS_DIR_SKIP_DIRS			(1<<5) /* Do not show directories */
+#define	LS_DIR_SKIP_SPECIAL			(1<<6) /* Do not show special file types */
+
+/*
+ * Shortcut for the historic behavior of the pg_ls_* functions (not including
+ * pg_ls_dir, which skips different files and doesn't show metadata.
+ */
+#define LS_DIR_HISTORIC				(LS_DIR_SKIP_DIRS|LS_DIR_SKIP_HIDDEN|LS_DIR_SKIP_SPECIAL|LS_DIR_METADATA)
 
 /*
  * Convert a "text" filename argument to C string, and check it's allowable.
@@ -413,6 +428,11 @@ pg_stat_file(PG_FUNCTION_ARGS)
 	values[4] = TimestampTzGetDatum(time_t_to_timestamptz(fst.st_ctime));
 #endif
 	values[5] = BoolGetDatum(S_ISDIR(fst.st_mode));
+#ifdef WIN32
+	/* Links should have isdir=false */
+	if (pgwin32_is_junction(filename))
+		values[5] = BoolGetDatum(false);
+#endif
 
 	tuple = heap_form_tuple(tupdesc, values, isnull);
 
@@ -440,79 +460,9 @@ pg_stat_file_1arg(PG_FUNCTION_ARGS)
 Datum
 pg_ls_dir(PG_FUNCTION_ARGS)
 {
-	ReturnSetInfo *rsinfo = (ReturnSetInfo *) fcinfo->resultinfo;
-	char	   *location;
-	bool		missing_ok = false;
-	bool		include_dot_dirs = false;
-	bool		randomAccess;
-	TupleDesc	tupdesc;
-	Tuplestorestate *tupstore;
-	DIR		   *dirdesc;
-	struct dirent *de;
-	MemoryContext oldcontext;
-
-	location = convert_and_check_filename(PG_GETARG_TEXT_PP(0));
-
-	/* check the optional arguments */
-	if (PG_NARGS() == 3)
-	{
-		if (!PG_ARGISNULL(1))
-			missing_ok = PG_GETARG_BOOL(1);
-		if (!PG_ARGISNULL(2))
-			include_dot_dirs = PG_GETARG_BOOL(2);
-	}
-
-	/* check to see if caller supports us returning a tuplestore */
-	if (rsinfo == NULL || !IsA(rsinfo, ReturnSetInfo))
-		ereport(ERROR,
-				(errcode(ERRCODE_FEATURE_NOT_SUPPORTED),
-				 errmsg("set-valued function called in context that cannot accept a set")));
-	if (!(rsinfo->allowedModes & SFRM_Materialize))
-		ereport(ERROR,
-				(errcode(ERRCODE_SYNTAX_ERROR),
-				 errmsg("materialize mode required, but it is not allowed in this context")));
-
-	/* The tupdesc and tuplestore must be created in ecxt_per_query_memory */
-	oldcontext = MemoryContextSwitchTo(rsinfo->econtext->ecxt_per_query_memory);
-
-	tupdesc = CreateTemplateTupleDesc(1);
-	TupleDescInitEntry(tupdesc, (AttrNumber) 1, "pg_ls_dir", TEXTOID, -1, 0);
-
-	randomAccess = (rsinfo->allowedModes & SFRM_Materialize_Random) != 0;
-	tupstore = tuplestore_begin_heap(randomAccess, false, work_mem);
-	rsinfo->returnMode = SFRM_Materialize;
-	rsinfo->setResult = tupstore;
-	rsinfo->setDesc = tupdesc;
-
-	MemoryContextSwitchTo(oldcontext);
-
-	dirdesc = AllocateDir(location);
-	if (!dirdesc)
-	{
-		/* Return empty tuplestore if appropriate */
-		if (missing_ok && errno == ENOENT)
-			return (Datum) 0;
-		/* Otherwise, we can let ReadDir() throw the error */
-	}
-
-	while ((de = ReadDir(dirdesc, location)) != NULL)
-	{
-		Datum		values[1];
-		bool		nulls[1];
-
-		if (!include_dot_dirs &&
-			(strcmp(de->d_name, ".") == 0 ||
-			 strcmp(de->d_name, "..") == 0))
-			continue;
-
-		values[0] = CStringGetTextDatum(de->d_name);
-		nulls[0] = false;
-
-		tuplestore_putvalues(tupstore, tupdesc, values, nulls);
-	}
-
-	FreeDir(dirdesc);
-	return (Datum) 0;
+	text	*filename_t = PG_GETARG_TEXT_PP(0);
+	char	*filename = convert_and_check_filename(filename_t);
+	return pg_ls_dir_files(fcinfo, filename, LS_DIR_SKIP_DOT_DIRS);
 }
 
 /*
@@ -525,7 +475,9 @@ pg_ls_dir(PG_FUNCTION_ARGS)
 Datum
 pg_ls_dir_1arg(PG_FUNCTION_ARGS)
 {
-	return pg_ls_dir(fcinfo);
+	text	*filename_t = PG_GETARG_TEXT_PP(0);
+	char	*filename = convert_and_check_filename(filename_t);
+	return pg_ls_dir_files(fcinfo, filename, LS_DIR_SKIP_DOT_DIRS);
 }
 
 /*
@@ -535,7 +487,7 @@ pg_ls_dir_1arg(PG_FUNCTION_ARGS)
  * Other unreadable-directory cases throw an error.
  */
 static Datum
-pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, bool missing_ok)
+pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, int flags)
 {
 	ReturnSetInfo *rsinfo = (ReturnSetInfo *) fcinfo->resultinfo;
 	bool		randomAccess;
@@ -544,6 +496,32 @@ pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, bool missing_ok)
 	DIR		   *dirdesc;
 	struct dirent *de;
 	MemoryContext oldcontext;
+	TypeFuncClass	tuptype ;
+
+	/* isdir depends on metadata */
+	Assert(!(flags&LS_DIR_ISDIR) || (flags&LS_DIR_METADATA));
+	/* Unreasonable to show isdir and skip dirs */
+	Assert(!(flags&LS_DIR_ISDIR) || !(flags&LS_DIR_SKIP_DIRS));
+
+	/* check the optional arguments */
+	if (PG_NARGS() == 3)
+	{
+		if (!PG_ARGISNULL(1))
+		{
+			if (PG_GETARG_BOOL(1))
+				flags |= LS_DIR_MISSING_OK;
+			else
+				flags &= ~LS_DIR_MISSING_OK;
+		}
+
+		if (!PG_ARGISNULL(2))
+		{
+			if (PG_GETARG_BOOL(2))
+				flags &= ~LS_DIR_SKIP_DOT_DIRS;
+			else
+				flags |= LS_DIR_SKIP_DOT_DIRS;
+		}
+	}
 
 	/* check to see if caller supports us returning a tuplestore */
 	if (rsinfo == NULL || !IsA(rsinfo, ReturnSetInfo))
@@ -558,8 +536,20 @@ pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, bool missing_ok)
 	/* The tupdesc and tuplestore must be created in ecxt_per_query_memory */
 	oldcontext = MemoryContextSwitchTo(rsinfo->econtext->ecxt_per_query_memory);
 
-	if (get_call_result_type(fcinfo, NULL, &tupdesc) != TYPEFUNC_COMPOSITE)
-		elog(ERROR, "return type must be a row type");
+	tuptype = get_call_result_type(fcinfo, NULL, &tupdesc);
+	if (flags & LS_DIR_METADATA)
+	{
+		if (tuptype != TYPEFUNC_COMPOSITE)
+			elog(ERROR, "return type must be a row type");
+	}
+	else
+	{
+		/* pg_ls_dir returns a simple scalar */
+		if (tuptype != TYPEFUNC_SCALAR)
+			elog(ERROR, "return type must be a scalar type");
+		tupdesc = CreateTemplateTupleDesc(1);
+		TupleDescInitEntry(tupdesc, (AttrNumber) 1, "column", TEXTOID, -1, 0);
+	}
 
 	randomAccess = (rsinfo->allowedModes & SFRM_Materialize_Random) != 0;
 	tupstore = tuplestore_begin_heap(randomAccess, false, work_mem);
@@ -578,20 +568,27 @@ pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, bool missing_ok)
 	if (!dirdesc)
 	{
 		/* Return empty tuplestore if appropriate */
-		if (missing_ok && errno == ENOENT)
+		if (flags & LS_DIR_MISSING_OK && errno == ENOENT)
 			return (Datum) 0;
 		/* Otherwise, we can let ReadDir() throw the error */
 	}
 
 	while ((de = ReadDir(dirdesc, dir)) != NULL)
 	{
-		Datum		values[3];
-		bool		nulls[3];
+		Datum		values[4];
+		bool		nulls[4];
 		char		path[MAXPGPATH * 2];
 		struct stat attrib;
 
-		/* Skip hidden files */
-		if (de->d_name[0] == '.')
+		/* Skip dot dirs? */
+		if (flags & LS_DIR_SKIP_DOT_DIRS &&
+			(strcmp(de->d_name, ".") == 0 ||
+			 strcmp(de->d_name, "..") == 0))
+			continue;
+
+		/* Skip hidden files? */
+		if (flags & LS_DIR_SKIP_HIDDEN &&
+			de->d_name[0] == '.')
 			continue;
 
 		/* Get the file info */
@@ -606,13 +603,34 @@ pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, bool missing_ok)
 					 errmsg("could not stat file \"%s\": %m", path)));
 		}
 
-		/* Ignore anything but regular files */
-		if (!S_ISREG(attrib.st_mode))
-			continue;
+		/* Skip dirs or special files? */
+		if (S_ISDIR(attrib.st_mode))
+		{
+			if (flags & LS_DIR_SKIP_DIRS)
+				continue;
+		}
+		else if (!S_ISREG(attrib.st_mode))
+		{
+			if (flags & LS_DIR_SKIP_SPECIAL)
+				continue;
+		}
 
 		values[0] = CStringGetTextDatum(de->d_name);
-		values[1] = Int64GetDatum((int64) attrib.st_size);
-		values[2] = TimestampTzGetDatum(time_t_to_timestamptz(attrib.st_mtime));
+		if (flags & LS_DIR_METADATA)
+		{
+			values[1] = Int64GetDatum((int64) attrib.st_size);
+			values[2] = TimestampTzGetDatum(time_t_to_timestamptz(attrib.st_mtime));
+			if (flags & LS_DIR_ISDIR)
+			{
+				values[3] = BoolGetDatum(S_ISDIR(attrib.st_mode));
+#ifdef WIN32
+				/* Links should have isdir=false */
+				if (pgwin32_is_junction(path))
+					values[3] = BoolGetDatum(false);
+#endif
+			}
+		}
+
 		memset(nulls, 0, sizeof(nulls));
 
 		tuplestore_putvalues(tupstore, tupdesc, values, nulls);
@@ -626,14 +644,14 @@ pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, bool missing_ok)
 Datum
 pg_ls_logdir(PG_FUNCTION_ARGS)
 {
-	return pg_ls_dir_files(fcinfo, Log_directory, false);
+	return pg_ls_dir_files(fcinfo, Log_directory, LS_DIR_HISTORIC);
 }
 
 /* Function to return the list of files in the WAL directory */
 Datum
 pg_ls_waldir(PG_FUNCTION_ARGS)
 {
-	return pg_ls_dir_files(fcinfo, XLOGDIR, false);
+	return pg_ls_dir_files(fcinfo, XLOGDIR, LS_DIR_HISTORIC);
 }
 
 /*
@@ -651,7 +669,8 @@ pg_ls_tmpdir(FunctionCallInfo fcinfo, Oid tblspc)
 						tblspc)));
 
 	TempTablespacePath(path, tblspc);
-	return pg_ls_dir_files(fcinfo, path, true);
+	return pg_ls_dir_files(fcinfo, path,
+			LS_DIR_HISTORIC | LS_DIR_MISSING_OK);
 }
 
 /*
@@ -680,5 +699,33 @@ pg_ls_tmpdir_1arg(PG_FUNCTION_ARGS)
 Datum
 pg_ls_archive_statusdir(PG_FUNCTION_ARGS)
 {
-	return pg_ls_dir_files(fcinfo, XLOGDIR "/archive_status", true);
+	return pg_ls_dir_files(fcinfo, XLOGDIR "/archive_status",
+			LS_DIR_HISTORIC | LS_DIR_MISSING_OK);
+}
+
+/*
+ * Function to return the list of files and metadata in an arbitrary directory.
+ */
+Datum
+pg_ls_dir_metadata(PG_FUNCTION_ARGS)
+{
+	char	*dirname = convert_and_check_filename(PG_GETARG_TEXT_PP(0));
+
+	return pg_ls_dir_files(fcinfo, dirname,
+			LS_DIR_METADATA | LS_DIR_SKIP_SPECIAL | LS_DIR_ISDIR);
+}
+
+/*
+ * Function to return the list of files and metadata in an arbitrary directory.
+ * note: this wrapper is necessary to pass the sanity check in opr_sanity,
+ * which checks that all built-in functions that share the implementing C
+ * function take the same number of arguments.
+ */
+Datum
+pg_ls_dir_metadata_1arg(PG_FUNCTION_ARGS)
+{
+	char	*dirname = convert_and_check_filename(PG_GETARG_TEXT_PP(0));
+
+	return pg_ls_dir_files(fcinfo, dirname,
+			LS_DIR_METADATA | LS_DIR_SKIP_SPECIAL | LS_DIR_ISDIR);
 }
diff --git a/src/include/catalog/pg_proc.dat b/src/include/catalog/pg_proc.dat
index 61f2c2f5b4..5b0147ee2d 100644
--- a/src/include/catalog/pg_proc.dat
+++ b/src/include/catalog/pg_proc.dat
@@ -10902,6 +10902,18 @@
   proallargtypes => '{oid,text,int8,timestamptz}', proargmodes => '{i,o,o,o}',
   proargnames => '{tablespace,name,size,modification}',
   prosrc => 'pg_ls_tmpdir_1arg' },
+{ oid => '9979', descr => 'list directory with metadata',
+  proname => 'pg_ls_dir_metadata', procost => '10', prorows => '20', proretset => 't',
+  provolatile => 'v', prorettype => 'record', proargtypes => 'text bool bool',
+  proallargtypes => '{text,bool,bool,text,int8,timestamptz,bool}', proargmodes => '{i,i,i,o,o,o,o}',
+  proargnames => '{dirname,missing_ok,include_dot_dirs,name,size,modification,isdir}',
+  prosrc => 'pg_ls_dir_metadata' },
+{ oid => '9980', descr => 'list directory with metadata',
+  proname => 'pg_ls_dir_metadata', procost => '10', prorows => '20', proretset => 't',
+  provolatile => 'v', prorettype => 'record', proargtypes => 'text',
+  proallargtypes => '{text,text,int8,timestamptz,bool}', proargmodes => '{i,o,o,o,o}',
+  proargnames => '{dirname,name,size,modification,isdir}',
+  prosrc => 'pg_ls_dir_metadata_1arg' },
 
 # hash partitioning constraint function
 { oid => '5028', descr => 'hash partition CHECK constraint',
diff --git a/src/test/regress/expected/misc_functions.out b/src/test/regress/expected/misc_functions.out
index 2e87c548eb..7930909f02 100644
--- a/src/test/regress/expected/misc_functions.out
+++ b/src/test/regress/expected/misc_functions.out
@@ -219,6 +219,30 @@ select pg_ls_dir('does not exist', true, false); -- ok with missingok=true
 
 select pg_ls_dir('does not exist'); -- fails with missingok=false
 ERROR:  could not open directory "does not exist": No such file or directory
+-- This tests the missing_ok parameter, which causes pg_ls_tmpdir to succeed even if the tmpdir doesn't exist yet
+-- The name='' condition is never true, so the function runs to completion but returns zero rows.
+select * from pg_ls_tmpdir() where name='Does not exist';
+ name | size | modification 
+------+------+--------------
+(0 rows)
+
+select name, isdir from pg_ls_dir_metadata('.') where name='.';
+ name | isdir 
+------+-------
+ .    | t
+(1 row)
+
+select name, isdir from pg_ls_dir_metadata('.', false, false) where name='.'; -- include_dot_dirs=false
+ name | isdir 
+------+-------
+(0 rows)
+
+-- Check that expected columns are present
+select * from pg_ls_dir_metadata('.') limit 0;
+ name | size | modification | isdir 
+------+------+--------------+-------
+(0 rows)
+
 --
 -- Test adding a support function to a subject function
 --
diff --git a/src/test/regress/input/tablespace.source b/src/test/regress/input/tablespace.source
index a5f61a35dc..0b9cfe615e 100644
--- a/src/test/regress/input/tablespace.source
+++ b/src/test/regress/input/tablespace.source
@@ -11,6 +11,11 @@ DROP TABLESPACE regress_tblspacewith;
 -- create a tablespace we can use
 CREATE TABLESPACE regress_tblspace LOCATION '@testtablespace@';
 
+-- This tests the missing_ok parameter, which causes pg_ls_tmpdir to succeed even if the tmpdir doesn't exist yet
+-- The name='' condition is never true, so the function runs to completion but returns zero rows.
+-- The query is written to ERROR if the tablespace doesn't exist, rather than silently failing to call pg_ls_tmpdir()
+SELECT c.* FROM (SELECT oid FROM pg_tablespace b WHERE b.spcname='regress_tblspace' UNION SELECT 0 ORDER BY 1 DESC LIMIT 1) AS b , pg_ls_tmpdir(oid) AS c WHERE c.name='Does not exist';
+
 -- try setting and resetting some properties for the new tablespace
 ALTER TABLESPACE regress_tblspace SET (random_page_cost = 1.0, seq_page_cost = 1.1);
 ALTER TABLESPACE regress_tblspace SET (some_nonexistent_parameter = true);  -- fail
diff --git a/src/test/regress/output/tablespace.source b/src/test/regress/output/tablespace.source
index 162b591b31..a42714bf40 100644
--- a/src/test/regress/output/tablespace.source
+++ b/src/test/regress/output/tablespace.source
@@ -13,6 +13,14 @@ SELECT spcoptions FROM pg_tablespace WHERE spcname = 'regress_tblspacewith';
 DROP TABLESPACE regress_tblspacewith;
 -- create a tablespace we can use
 CREATE TABLESPACE regress_tblspace LOCATION '@testtablespace@';
+-- This tests the missing_ok parameter, which causes pg_ls_tmpdir to succeed even if the tmpdir doesn't exist yet
+-- The name='' condition is never true, so the function runs to completion but returns zero rows.
+-- The query is written to ERROR if the tablespace doesn't exist, rather than silently failing to call pg_ls_tmpdir()
+SELECT c.* FROM (SELECT oid FROM pg_tablespace b WHERE b.spcname='regress_tblspace' UNION SELECT 0 ORDER BY 1 DESC LIMIT 1) AS b , pg_ls_tmpdir(oid) AS c WHERE c.name='Does not exist';
+ name | size | modification 
+------+------+--------------
+(0 rows)
+
 -- try setting and resetting some properties for the new tablespace
 ALTER TABLESPACE regress_tblspace SET (random_page_cost = 1.0, seq_page_cost = 1.1);
 ALTER TABLESPACE regress_tblspace SET (some_nonexistent_parameter = true);  -- fail
diff --git a/src/test/regress/sql/misc_functions.sql b/src/test/regress/sql/misc_functions.sql
index f6857ad177..372345720d 100644
--- a/src/test/regress/sql/misc_functions.sql
+++ b/src/test/regress/sql/misc_functions.sql
@@ -65,6 +65,17 @@ select * from (select pg_ls_dir('.', false, false) as name) as ls where ls.name=
 select pg_ls_dir('does not exist', true, false); -- ok with missingok=true
 select pg_ls_dir('does not exist'); -- fails with missingok=false
 
+-- This tests the missing_ok parameter, which causes pg_ls_tmpdir to succeed even if the tmpdir doesn't exist yet
+-- The name='' condition is never true, so the function runs to completion but returns zero rows.
+select * from pg_ls_tmpdir() where name='Does not exist';
+
+select name, isdir from pg_ls_dir_metadata('.') where name='.';
+
+select name, isdir from pg_ls_dir_metadata('.', false, false) where name='.'; -- include_dot_dirs=false
+
+-- Check that expected columns are present
+select * from pg_ls_dir_metadata('.') limit 0;
+
 --
 -- Test adding a support function to a subject function
 --
-- 
2.17.0


--S0GG+JvAI2G0KxBG
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
 filename="v19-0005-pg_ls_tmpdir-to-show-directories-and-isdir-argum.patch"



^ permalink  raw  reply  [nested|flat] 68+ messages in thread

* [PATCH v16 04/10] Add pg_ls_dir_metadata to list a dir with file metadata..
@ 2020-03-10 03:40  Justin Pryzby <[email protected]>
  0 siblings, 0 replies; 68+ messages in thread

From: Justin Pryzby @ 2020-03-10 03:40 UTC (permalink / raw)

Generalize pg_ls_dir_files and retire pg_ls_dir

Need catversion bumped?
---
 doc/src/sgml/func.sgml                       |  19 ++
 src/backend/catalog/system_views.sql         |   1 +
 src/backend/utils/adt/genfile.c              | 229 +++++++++++--------
 src/include/catalog/pg_proc.dat              |  12 +
 src/test/regress/expected/misc_functions.out |  24 ++
 src/test/regress/input/tablespace.source     |   5 +
 src/test/regress/output/tablespace.source    |   8 +
 src/test/regress/sql/misc_functions.sql      |  11 +
 8 files changed, 218 insertions(+), 91 deletions(-)

diff --git a/doc/src/sgml/func.sgml b/doc/src/sgml/func.sgml
index 96b08d0500..9d617f95b9 100644
--- a/doc/src/sgml/func.sgml
+++ b/doc/src/sgml/func.sgml
@@ -25300,6 +25300,17 @@ postgres=# SELECT * FROM pg_walfile_name_offset(pg_stop_backup());
         List the contents of a directory.  Restricted to superusers by default, but other users can be granted EXECUTE to run the function.
        </entry>
       </row>
+      <row>
+       <entry>
+        <literal><function>pg_ls_dir_metadata(<parameter>dirname</parameter> <type>text</type> [, <parameter>missing_ok</parameter> <type>boolean</type>, <parameter>include_dot_dirs</parameter> <type>boolean</type>])</function></literal>
+       </entry>
+       <entry><type>setof text</type></entry>
+       <entry>
+        For each file in a directory, list the file and its metadata.
+        Restricted to superusers by default, but other users can be granted
+        EXECUTE to run the function.
+       </entry>
+      </row>
       <row>
        <entry>
         <literal><function>pg_ls_logdir()</function></literal>
@@ -25400,6 +25411,14 @@ postgres=# SELECT * FROM pg_walfile_name_offset(pg_stop_backup());
     empty directory from an non-existent directory.
    </para>
 
+   <indexterm>
+    <primary>pg_ls_dir_metadata</primary>
+   </indexterm>
+   <para>
+    <function>pg_ls_dir_metadata</function> lists the files in the specified
+    directory along with the file's metadata.
+   </para>
+
    <indexterm>
     <primary>pg_ls_logdir</primary>
    </indexterm>
diff --git a/src/backend/catalog/system_views.sql b/src/backend/catalog/system_views.sql
index 2bd5f5ea14..1c77430f0c 100644
--- a/src/backend/catalog/system_views.sql
+++ b/src/backend/catalog/system_views.sql
@@ -1467,6 +1467,7 @@ REVOKE EXECUTE ON FUNCTION pg_stat_file(text,boolean) FROM public;
 
 REVOKE EXECUTE ON FUNCTION pg_ls_dir(text) FROM public;
 REVOKE EXECUTE ON FUNCTION pg_ls_dir(text,boolean,boolean) FROM public;
+REVOKE EXECUTE ON FUNCTION pg_ls_dir_metadata(text,boolean,boolean) FROM public;
 
 --
 -- We also set up some things as accessible to standard roles.
diff --git a/src/backend/utils/adt/genfile.c b/src/backend/utils/adt/genfile.c
index 219ac160f8..4824a55480 100644
--- a/src/backend/utils/adt/genfile.c
+++ b/src/backend/utils/adt/genfile.c
@@ -36,6 +36,21 @@
 #include "utils/syscache.h"
 #include "utils/timestamp.h"
 
+static Datum pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, int flags);
+
+#define	LS_DIR_ISDIR				(1<<0) /* Show column: isdir */
+#define	LS_DIR_METADATA				(1<<1) /* Show columns: mtime, size */
+#define	LS_DIR_MISSING_OK			(1<<2) /* Ignore ENOENT if the toplevel dir is missing */
+#define	LS_DIR_SKIP_DOT_DIRS		(1<<3) /* Do not show . or .. */
+#define	LS_DIR_SKIP_HIDDEN			(1<<4) /* Do not show anything begining with . */
+#define	LS_DIR_SKIP_DIRS			(1<<5) /* Do not show directories */
+#define	LS_DIR_SKIP_SPECIAL			(1<<6) /* Do not show special file types */
+
+/*
+ * Shortcut for the historic behavior of the pg_ls_* functions (not including
+ * pg_ls_dir, which skips different files and doesn't show metadata.
+ */
+#define LS_DIR_HISTORIC				(LS_DIR_SKIP_DIRS|LS_DIR_SKIP_HIDDEN|LS_DIR_SKIP_SPECIAL|LS_DIR_METADATA)
 
 /*
  * Convert a "text" filename argument to C string, and check it's allowable.
@@ -413,6 +428,11 @@ pg_stat_file(PG_FUNCTION_ARGS)
 	values[4] = TimestampTzGetDatum(time_t_to_timestamptz(fst.st_ctime));
 #endif
 	values[5] = BoolGetDatum(S_ISDIR(fst.st_mode));
+#ifdef WIN32
+	/* Links should have isdir=false */
+	if (pgwin32_is_junction(filename))
+		values[5] = BoolGetDatum(false);
+#endif
 
 	tuple = heap_form_tuple(tupdesc, values, isnull);
 
@@ -440,79 +460,9 @@ pg_stat_file_1arg(PG_FUNCTION_ARGS)
 Datum
 pg_ls_dir(PG_FUNCTION_ARGS)
 {
-	ReturnSetInfo *rsinfo = (ReturnSetInfo *) fcinfo->resultinfo;
-	char	   *location;
-	bool		missing_ok = false;
-	bool		include_dot_dirs = false;
-	bool		randomAccess;
-	TupleDesc	tupdesc;
-	Tuplestorestate *tupstore;
-	DIR		   *dirdesc;
-	struct dirent *de;
-	MemoryContext oldcontext;
-
-	location = convert_and_check_filename(PG_GETARG_TEXT_PP(0));
-
-	/* check the optional arguments */
-	if (PG_NARGS() == 3)
-	{
-		if (!PG_ARGISNULL(1))
-			missing_ok = PG_GETARG_BOOL(1);
-		if (!PG_ARGISNULL(2))
-			include_dot_dirs = PG_GETARG_BOOL(2);
-	}
-
-	/* check to see if caller supports us returning a tuplestore */
-	if (rsinfo == NULL || !IsA(rsinfo, ReturnSetInfo))
-		ereport(ERROR,
-				(errcode(ERRCODE_FEATURE_NOT_SUPPORTED),
-				 errmsg("set-valued function called in context that cannot accept a set")));
-	if (!(rsinfo->allowedModes & SFRM_Materialize))
-		ereport(ERROR,
-				(errcode(ERRCODE_SYNTAX_ERROR),
-				 errmsg("materialize mode required, but it is not allowed in this context")));
-
-	/* The tupdesc and tuplestore must be created in ecxt_per_query_memory */
-	oldcontext = MemoryContextSwitchTo(rsinfo->econtext->ecxt_per_query_memory);
-
-	tupdesc = CreateTemplateTupleDesc(1);
-	TupleDescInitEntry(tupdesc, (AttrNumber) 1, "pg_ls_dir", TEXTOID, -1, 0);
-
-	randomAccess = (rsinfo->allowedModes & SFRM_Materialize_Random) != 0;
-	tupstore = tuplestore_begin_heap(randomAccess, false, work_mem);
-	rsinfo->returnMode = SFRM_Materialize;
-	rsinfo->setResult = tupstore;
-	rsinfo->setDesc = tupdesc;
-
-	MemoryContextSwitchTo(oldcontext);
-
-	dirdesc = AllocateDir(location);
-	if (!dirdesc)
-	{
-		/* Return empty tuplestore if appropriate */
-		if (missing_ok && errno == ENOENT)
-			return (Datum) 0;
-		/* Otherwise, we can let ReadDir() throw the error */
-	}
-
-	while ((de = ReadDir(dirdesc, location)) != NULL)
-	{
-		Datum		values[1];
-		bool		nulls[1];
-
-		if (!include_dot_dirs &&
-			(strcmp(de->d_name, ".") == 0 ||
-			 strcmp(de->d_name, "..") == 0))
-			continue;
-
-		values[0] = CStringGetTextDatum(de->d_name);
-		nulls[0] = false;
-
-		tuplestore_putvalues(tupstore, tupdesc, values, nulls);
-	}
-
-	FreeDir(dirdesc);
-	return (Datum) 0;
+	text	*filename_t = PG_GETARG_TEXT_PP(0);
+	char	*filename = convert_and_check_filename(filename_t);
+	return pg_ls_dir_files(fcinfo, filename, LS_DIR_SKIP_DOT_DIRS);
 }
 
 /*
@@ -525,7 +475,9 @@ pg_ls_dir(PG_FUNCTION_ARGS)
 Datum
 pg_ls_dir_1arg(PG_FUNCTION_ARGS)
 {
-	return pg_ls_dir(fcinfo);
+	text	*filename_t = PG_GETARG_TEXT_PP(0);
+	char	*filename = convert_and_check_filename(filename_t);
+	return pg_ls_dir_files(fcinfo, filename, LS_DIR_SKIP_DOT_DIRS);
 }
 
 /*
@@ -535,7 +487,7 @@ pg_ls_dir_1arg(PG_FUNCTION_ARGS)
  * Other unreadable-directory cases throw an error.
  */
 static Datum
-pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, bool missing_ok)
+pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, int flags)
 {
 	ReturnSetInfo *rsinfo = (ReturnSetInfo *) fcinfo->resultinfo;
 	bool		randomAccess;
@@ -544,6 +496,32 @@ pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, bool missing_ok)
 	DIR		   *dirdesc;
 	struct dirent *de;
 	MemoryContext oldcontext;
+	TypeFuncClass	tuptype ;
+
+	/* isdir depends on metadata */
+	Assert(!(flags&LS_DIR_ISDIR) || (flags&LS_DIR_METADATA));
+	/* Unreasonable to show isdir and skip dirs */
+	Assert(!(flags&LS_DIR_ISDIR) || !(flags&LS_DIR_SKIP_DIRS));
+
+	/* check the optional arguments */
+	if (PG_NARGS() == 3)
+	{
+		if (!PG_ARGISNULL(1))
+		{
+			if (PG_GETARG_BOOL(1))
+				flags |= LS_DIR_MISSING_OK;
+			else
+				flags &= ~LS_DIR_MISSING_OK;
+		}
+
+		if (!PG_ARGISNULL(2))
+		{
+			if (PG_GETARG_BOOL(2))
+				flags &= ~LS_DIR_SKIP_DOT_DIRS;
+			else
+				flags |= LS_DIR_SKIP_DOT_DIRS;
+		}
+	}
 
 	/* check to see if caller supports us returning a tuplestore */
 	if (rsinfo == NULL || !IsA(rsinfo, ReturnSetInfo))
@@ -558,8 +536,20 @@ pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, bool missing_ok)
 	/* The tupdesc and tuplestore must be created in ecxt_per_query_memory */
 	oldcontext = MemoryContextSwitchTo(rsinfo->econtext->ecxt_per_query_memory);
 
-	if (get_call_result_type(fcinfo, NULL, &tupdesc) != TYPEFUNC_COMPOSITE)
-		elog(ERROR, "return type must be a row type");
+	tuptype = get_call_result_type(fcinfo, NULL, &tupdesc);
+	if (flags & LS_DIR_METADATA)
+	{
+		if (tuptype != TYPEFUNC_COMPOSITE)
+			elog(ERROR, "return type must be a row type");
+	}
+	else
+	{
+		/* pg_ls_dir returns a simple scalar */
+		if (tuptype != TYPEFUNC_SCALAR)
+			elog(ERROR, "return type must be a scalar type");
+		tupdesc = CreateTemplateTupleDesc(1);
+		TupleDescInitEntry(tupdesc, (AttrNumber) 1, "column", TEXTOID, -1, 0);
+	}
 
 	randomAccess = (rsinfo->allowedModes & SFRM_Materialize_Random) != 0;
 	tupstore = tuplestore_begin_heap(randomAccess, false, work_mem);
@@ -578,20 +568,27 @@ pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, bool missing_ok)
 	if (!dirdesc)
 	{
 		/* Return empty tuplestore if appropriate */
-		if (missing_ok && errno == ENOENT)
+		if (flags & LS_DIR_MISSING_OK && errno == ENOENT)
 			return (Datum) 0;
 		/* Otherwise, we can let ReadDir() throw the error */
 	}
 
 	while ((de = ReadDir(dirdesc, dir)) != NULL)
 	{
-		Datum		values[3];
-		bool		nulls[3];
+		Datum		values[4];
+		bool		nulls[4];
 		char		path[MAXPGPATH * 2];
 		struct stat attrib;
 
-		/* Skip hidden files */
-		if (de->d_name[0] == '.')
+		/* Skip dot dirs? */
+		if (flags & LS_DIR_SKIP_DOT_DIRS &&
+			(strcmp(de->d_name, ".") == 0 ||
+			 strcmp(de->d_name, "..") == 0))
+			continue;
+
+		/* Skip hidden files? */
+		if (flags & LS_DIR_SKIP_HIDDEN &&
+			de->d_name[0] == '.')
 			continue;
 
 		/* Get the file info */
@@ -606,13 +603,34 @@ pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, bool missing_ok)
 					 errmsg("could not stat file \"%s\": %m", path)));
 		}
 
-		/* Ignore anything but regular files */
-		if (!S_ISREG(attrib.st_mode))
-			continue;
+		/* Skip dirs or special files? */
+		if (S_ISDIR(attrib.st_mode))
+		{
+			if (flags & LS_DIR_SKIP_DIRS)
+				continue;
+		}
+		else if (!S_ISREG(attrib.st_mode))
+		{
+			if (flags & LS_DIR_SKIP_SPECIAL)
+				continue;
+		}
 
 		values[0] = CStringGetTextDatum(de->d_name);
-		values[1] = Int64GetDatum((int64) attrib.st_size);
-		values[2] = TimestampTzGetDatum(time_t_to_timestamptz(attrib.st_mtime));
+		if (flags & LS_DIR_METADATA)
+		{
+			values[1] = Int64GetDatum((int64) attrib.st_size);
+			values[2] = TimestampTzGetDatum(time_t_to_timestamptz(attrib.st_mtime));
+			if (flags & LS_DIR_ISDIR)
+			{
+				values[3] = BoolGetDatum(S_ISDIR(attrib.st_mode));
+#ifdef WIN32
+				/* Links should have isdir=false */
+				if (pgwin32_is_junction(path))
+					values[3] = BoolGetDatum(false);
+#endif
+			}
+		}
+
 		memset(nulls, 0, sizeof(nulls));
 
 		tuplestore_putvalues(tupstore, tupdesc, values, nulls);
@@ -626,14 +644,14 @@ pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, bool missing_ok)
 Datum
 pg_ls_logdir(PG_FUNCTION_ARGS)
 {
-	return pg_ls_dir_files(fcinfo, Log_directory, false);
+	return pg_ls_dir_files(fcinfo, Log_directory, LS_DIR_HISTORIC);
 }
 
 /* Function to return the list of files in the WAL directory */
 Datum
 pg_ls_waldir(PG_FUNCTION_ARGS)
 {
-	return pg_ls_dir_files(fcinfo, XLOGDIR, false);
+	return pg_ls_dir_files(fcinfo, XLOGDIR, LS_DIR_HISTORIC);
 }
 
 /*
@@ -651,7 +669,8 @@ pg_ls_tmpdir(FunctionCallInfo fcinfo, Oid tblspc)
 						tblspc)));
 
 	TempTablespacePath(path, tblspc);
-	return pg_ls_dir_files(fcinfo, path, true);
+	return pg_ls_dir_files(fcinfo, path,
+			LS_DIR_HISTORIC | LS_DIR_MISSING_OK);
 }
 
 /*
@@ -680,5 +699,33 @@ pg_ls_tmpdir_1arg(PG_FUNCTION_ARGS)
 Datum
 pg_ls_archive_statusdir(PG_FUNCTION_ARGS)
 {
-	return pg_ls_dir_files(fcinfo, XLOGDIR "/archive_status", true);
+	return pg_ls_dir_files(fcinfo, XLOGDIR "/archive_status",
+			LS_DIR_HISTORIC | LS_DIR_MISSING_OK);
+}
+
+/*
+ * Function to return the list of files and metadata in an arbitrary directory.
+ */
+Datum
+pg_ls_dir_metadata(PG_FUNCTION_ARGS)
+{
+	char	*dirname = convert_and_check_filename(PG_GETARG_TEXT_PP(0));
+
+	return pg_ls_dir_files(fcinfo, dirname,
+			LS_DIR_METADATA | LS_DIR_SKIP_SPECIAL | LS_DIR_ISDIR);
+}
+
+/*
+ * Function to return the list of files and metadata in an arbitrary directory.
+ * note: this wrapper is necessary to pass the sanity check in opr_sanity,
+ * which checks that all built-in functions that share the implementing C
+ * function take the same number of arguments.
+ */
+Datum
+pg_ls_dir_metadata_1arg(PG_FUNCTION_ARGS)
+{
+	char	*dirname = convert_and_check_filename(PG_GETARG_TEXT_PP(0));
+
+	return pg_ls_dir_files(fcinfo, dirname,
+			LS_DIR_METADATA | LS_DIR_SKIP_SPECIAL | LS_DIR_ISDIR);
 }
diff --git a/src/include/catalog/pg_proc.dat b/src/include/catalog/pg_proc.dat
index 4bce3ad8de..9f46cba5ed 100644
--- a/src/include/catalog/pg_proc.dat
+++ b/src/include/catalog/pg_proc.dat
@@ -10904,6 +10904,18 @@
   proallargtypes => '{oid,text,int8,timestamptz}', proargmodes => '{i,o,o,o}',
   proargnames => '{tablespace,name,size,modification}',
   prosrc => 'pg_ls_tmpdir_1arg' },
+{ oid => '5032', descr => 'list directory with metadata',
+  proname => 'pg_ls_dir_metadata', procost => '10', prorows => '20', proretset => 't',
+  provolatile => 'v', prorettype => 'record', proargtypes => 'text bool bool',
+  proallargtypes => '{text,bool,bool,text,int8,timestamptz,bool}', proargmodes => '{i,i,i,o,o,o,o}',
+  proargnames => '{dirname,missing_ok,include_dot_dirs,name,size,modification,isdir}',
+  prosrc => 'pg_ls_dir_metadata' },
+{ oid => '5033', descr => 'list directory with metadata',
+  proname => 'pg_ls_dir_metadata', procost => '10', prorows => '20', proretset => 't',
+  provolatile => 'v', prorettype => 'record', proargtypes => 'text',
+  proallargtypes => '{text,text,int8,timestamptz,bool}', proargmodes => '{i,o,o,o,o}',
+  proargnames => '{dirname,name,size,modification,isdir}',
+  prosrc => 'pg_ls_dir_metadata_1arg' },
 
 # hash partitioning constraint function
 { oid => '5028', descr => 'hash partition CHECK constraint',
diff --git a/src/test/regress/expected/misc_functions.out b/src/test/regress/expected/misc_functions.out
index 2e87c548eb..7930909f02 100644
--- a/src/test/regress/expected/misc_functions.out
+++ b/src/test/regress/expected/misc_functions.out
@@ -219,6 +219,30 @@ select pg_ls_dir('does not exist', true, false); -- ok with missingok=true
 
 select pg_ls_dir('does not exist'); -- fails with missingok=false
 ERROR:  could not open directory "does not exist": No such file or directory
+-- This tests the missing_ok parameter, which causes pg_ls_tmpdir to succeed even if the tmpdir doesn't exist yet
+-- The name='' condition is never true, so the function runs to completion but returns zero rows.
+select * from pg_ls_tmpdir() where name='Does not exist';
+ name | size | modification 
+------+------+--------------
+(0 rows)
+
+select name, isdir from pg_ls_dir_metadata('.') where name='.';
+ name | isdir 
+------+-------
+ .    | t
+(1 row)
+
+select name, isdir from pg_ls_dir_metadata('.', false, false) where name='.'; -- include_dot_dirs=false
+ name | isdir 
+------+-------
+(0 rows)
+
+-- Check that expected columns are present
+select * from pg_ls_dir_metadata('.') limit 0;
+ name | size | modification | isdir 
+------+------+--------------+-------
+(0 rows)
+
 --
 -- Test adding a support function to a subject function
 --
diff --git a/src/test/regress/input/tablespace.source b/src/test/regress/input/tablespace.source
index a5f61a35dc..0b9cfe615e 100644
--- a/src/test/regress/input/tablespace.source
+++ b/src/test/regress/input/tablespace.source
@@ -11,6 +11,11 @@ DROP TABLESPACE regress_tblspacewith;
 -- create a tablespace we can use
 CREATE TABLESPACE regress_tblspace LOCATION '@testtablespace@';
 
+-- This tests the missing_ok parameter, which causes pg_ls_tmpdir to succeed even if the tmpdir doesn't exist yet
+-- The name='' condition is never true, so the function runs to completion but returns zero rows.
+-- The query is written to ERROR if the tablespace doesn't exist, rather than silently failing to call pg_ls_tmpdir()
+SELECT c.* FROM (SELECT oid FROM pg_tablespace b WHERE b.spcname='regress_tblspace' UNION SELECT 0 ORDER BY 1 DESC LIMIT 1) AS b , pg_ls_tmpdir(oid) AS c WHERE c.name='Does not exist';
+
 -- try setting and resetting some properties for the new tablespace
 ALTER TABLESPACE regress_tblspace SET (random_page_cost = 1.0, seq_page_cost = 1.1);
 ALTER TABLESPACE regress_tblspace SET (some_nonexistent_parameter = true);  -- fail
diff --git a/src/test/regress/output/tablespace.source b/src/test/regress/output/tablespace.source
index 162b591b31..a42714bf40 100644
--- a/src/test/regress/output/tablespace.source
+++ b/src/test/regress/output/tablespace.source
@@ -13,6 +13,14 @@ SELECT spcoptions FROM pg_tablespace WHERE spcname = 'regress_tblspacewith';
 DROP TABLESPACE regress_tblspacewith;
 -- create a tablespace we can use
 CREATE TABLESPACE regress_tblspace LOCATION '@testtablespace@';
+-- This tests the missing_ok parameter, which causes pg_ls_tmpdir to succeed even if the tmpdir doesn't exist yet
+-- The name='' condition is never true, so the function runs to completion but returns zero rows.
+-- The query is written to ERROR if the tablespace doesn't exist, rather than silently failing to call pg_ls_tmpdir()
+SELECT c.* FROM (SELECT oid FROM pg_tablespace b WHERE b.spcname='regress_tblspace' UNION SELECT 0 ORDER BY 1 DESC LIMIT 1) AS b , pg_ls_tmpdir(oid) AS c WHERE c.name='Does not exist';
+ name | size | modification 
+------+------+--------------
+(0 rows)
+
 -- try setting and resetting some properties for the new tablespace
 ALTER TABLESPACE regress_tblspace SET (random_page_cost = 1.0, seq_page_cost = 1.1);
 ALTER TABLESPACE regress_tblspace SET (some_nonexistent_parameter = true);  -- fail
diff --git a/src/test/regress/sql/misc_functions.sql b/src/test/regress/sql/misc_functions.sql
index f6857ad177..372345720d 100644
--- a/src/test/regress/sql/misc_functions.sql
+++ b/src/test/regress/sql/misc_functions.sql
@@ -65,6 +65,17 @@ select * from (select pg_ls_dir('.', false, false) as name) as ls where ls.name=
 select pg_ls_dir('does not exist', true, false); -- ok with missingok=true
 select pg_ls_dir('does not exist'); -- fails with missingok=false
 
+-- This tests the missing_ok parameter, which causes pg_ls_tmpdir to succeed even if the tmpdir doesn't exist yet
+-- The name='' condition is never true, so the function runs to completion but returns zero rows.
+select * from pg_ls_tmpdir() where name='Does not exist';
+
+select name, isdir from pg_ls_dir_metadata('.') where name='.';
+
+select name, isdir from pg_ls_dir_metadata('.', false, false) where name='.'; -- include_dot_dirs=false
+
+-- Check that expected columns are present
+select * from pg_ls_dir_metadata('.') limit 0;
+
 --
 -- Test adding a support function to a subject function
 --
-- 
2.17.0


--2FkSFaIQeDFoAt0B
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
 filename="v16-0005-pg_ls_tmpdir-to-show-directories-and-isdir-argum.patch"



^ permalink  raw  reply  [nested|flat] 68+ messages in thread

* [PATCH v17 04/10] Add pg_ls_dir_metadata to list a dir with file metadata..
@ 2020-03-10 03:40  Justin Pryzby <[email protected]>
  0 siblings, 0 replies; 68+ messages in thread

From: Justin Pryzby @ 2020-03-10 03:40 UTC (permalink / raw)

Generalize pg_ls_dir_files and retire pg_ls_dir

Need catversion bumped?
---
 doc/src/sgml/func.sgml                       |  21 ++
 src/backend/catalog/system_views.sql         |   1 +
 src/backend/utils/adt/genfile.c              | 229 +++++++++++--------
 src/include/catalog/pg_proc.dat              |  12 +
 src/test/regress/expected/misc_functions.out |  24 ++
 src/test/regress/input/tablespace.source     |   5 +
 src/test/regress/output/tablespace.source    |   8 +
 src/test/regress/sql/misc_functions.sql      |  11 +
 8 files changed, 220 insertions(+), 91 deletions(-)

diff --git a/doc/src/sgml/func.sgml b/doc/src/sgml/func.sgml
index d9b3598977..fc1b4ac98c 100644
--- a/doc/src/sgml/func.sgml
+++ b/doc/src/sgml/func.sgml
@@ -25714,6 +25714,27 @@ SELECT pg_size_pretty(sum(pg_relation_size(relid))) AS total_size
        </para></entry>
       </row>
 
+      <row>
+       <entry role="func_table_entry"><para role="func_signature">
+        <indexterm>
+         <primary>pg_ls_dir_metadata</primary>
+        </indexterm>
+        <function>pg_ls_dir_metadata</function> ( <parameter>dirname</parameter> <type>text</type>
+        <optional>, <parameter>missing_ok</parameter> <type>boolean</type>,
+        <parameter>include_dot_dirs</parameter> <type>boolean</type>] </optional> )
+        <returnvalue>setof record</returnvalue>
+        ( <parameter>name</parameter> <type>text</type>,
+        <parameter>size</parameter> <type>bigint</type>,
+        <parameter>modification</parameter> <type>timestamp with time zone</type> )
+       </para>
+       <para>
+        For each file in the specified directory, list the file and its
+        metadata.
+        Restricted to superusers by default, but other users can be granted
+        EXECUTE to run the function.
+       </para></entry>
+      </row>
+
       <row>
        <entry role="func_table_entry"><para role="func_signature">
         <indexterm>
diff --git a/src/backend/catalog/system_views.sql b/src/backend/catalog/system_views.sql
index 2bd5f5ea14..1c77430f0c 100644
--- a/src/backend/catalog/system_views.sql
+++ b/src/backend/catalog/system_views.sql
@@ -1467,6 +1467,7 @@ REVOKE EXECUTE ON FUNCTION pg_stat_file(text,boolean) FROM public;
 
 REVOKE EXECUTE ON FUNCTION pg_ls_dir(text) FROM public;
 REVOKE EXECUTE ON FUNCTION pg_ls_dir(text,boolean,boolean) FROM public;
+REVOKE EXECUTE ON FUNCTION pg_ls_dir_metadata(text,boolean,boolean) FROM public;
 
 --
 -- We also set up some things as accessible to standard roles.
diff --git a/src/backend/utils/adt/genfile.c b/src/backend/utils/adt/genfile.c
index 219ac160f8..4824a55480 100644
--- a/src/backend/utils/adt/genfile.c
+++ b/src/backend/utils/adt/genfile.c
@@ -36,6 +36,21 @@
 #include "utils/syscache.h"
 #include "utils/timestamp.h"
 
+static Datum pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, int flags);
+
+#define	LS_DIR_ISDIR				(1<<0) /* Show column: isdir */
+#define	LS_DIR_METADATA				(1<<1) /* Show columns: mtime, size */
+#define	LS_DIR_MISSING_OK			(1<<2) /* Ignore ENOENT if the toplevel dir is missing */
+#define	LS_DIR_SKIP_DOT_DIRS		(1<<3) /* Do not show . or .. */
+#define	LS_DIR_SKIP_HIDDEN			(1<<4) /* Do not show anything begining with . */
+#define	LS_DIR_SKIP_DIRS			(1<<5) /* Do not show directories */
+#define	LS_DIR_SKIP_SPECIAL			(1<<6) /* Do not show special file types */
+
+/*
+ * Shortcut for the historic behavior of the pg_ls_* functions (not including
+ * pg_ls_dir, which skips different files and doesn't show metadata.
+ */
+#define LS_DIR_HISTORIC				(LS_DIR_SKIP_DIRS|LS_DIR_SKIP_HIDDEN|LS_DIR_SKIP_SPECIAL|LS_DIR_METADATA)
 
 /*
  * Convert a "text" filename argument to C string, and check it's allowable.
@@ -413,6 +428,11 @@ pg_stat_file(PG_FUNCTION_ARGS)
 	values[4] = TimestampTzGetDatum(time_t_to_timestamptz(fst.st_ctime));
 #endif
 	values[5] = BoolGetDatum(S_ISDIR(fst.st_mode));
+#ifdef WIN32
+	/* Links should have isdir=false */
+	if (pgwin32_is_junction(filename))
+		values[5] = BoolGetDatum(false);
+#endif
 
 	tuple = heap_form_tuple(tupdesc, values, isnull);
 
@@ -440,79 +460,9 @@ pg_stat_file_1arg(PG_FUNCTION_ARGS)
 Datum
 pg_ls_dir(PG_FUNCTION_ARGS)
 {
-	ReturnSetInfo *rsinfo = (ReturnSetInfo *) fcinfo->resultinfo;
-	char	   *location;
-	bool		missing_ok = false;
-	bool		include_dot_dirs = false;
-	bool		randomAccess;
-	TupleDesc	tupdesc;
-	Tuplestorestate *tupstore;
-	DIR		   *dirdesc;
-	struct dirent *de;
-	MemoryContext oldcontext;
-
-	location = convert_and_check_filename(PG_GETARG_TEXT_PP(0));
-
-	/* check the optional arguments */
-	if (PG_NARGS() == 3)
-	{
-		if (!PG_ARGISNULL(1))
-			missing_ok = PG_GETARG_BOOL(1);
-		if (!PG_ARGISNULL(2))
-			include_dot_dirs = PG_GETARG_BOOL(2);
-	}
-
-	/* check to see if caller supports us returning a tuplestore */
-	if (rsinfo == NULL || !IsA(rsinfo, ReturnSetInfo))
-		ereport(ERROR,
-				(errcode(ERRCODE_FEATURE_NOT_SUPPORTED),
-				 errmsg("set-valued function called in context that cannot accept a set")));
-	if (!(rsinfo->allowedModes & SFRM_Materialize))
-		ereport(ERROR,
-				(errcode(ERRCODE_SYNTAX_ERROR),
-				 errmsg("materialize mode required, but it is not allowed in this context")));
-
-	/* The tupdesc and tuplestore must be created in ecxt_per_query_memory */
-	oldcontext = MemoryContextSwitchTo(rsinfo->econtext->ecxt_per_query_memory);
-
-	tupdesc = CreateTemplateTupleDesc(1);
-	TupleDescInitEntry(tupdesc, (AttrNumber) 1, "pg_ls_dir", TEXTOID, -1, 0);
-
-	randomAccess = (rsinfo->allowedModes & SFRM_Materialize_Random) != 0;
-	tupstore = tuplestore_begin_heap(randomAccess, false, work_mem);
-	rsinfo->returnMode = SFRM_Materialize;
-	rsinfo->setResult = tupstore;
-	rsinfo->setDesc = tupdesc;
-
-	MemoryContextSwitchTo(oldcontext);
-
-	dirdesc = AllocateDir(location);
-	if (!dirdesc)
-	{
-		/* Return empty tuplestore if appropriate */
-		if (missing_ok && errno == ENOENT)
-			return (Datum) 0;
-		/* Otherwise, we can let ReadDir() throw the error */
-	}
-
-	while ((de = ReadDir(dirdesc, location)) != NULL)
-	{
-		Datum		values[1];
-		bool		nulls[1];
-
-		if (!include_dot_dirs &&
-			(strcmp(de->d_name, ".") == 0 ||
-			 strcmp(de->d_name, "..") == 0))
-			continue;
-
-		values[0] = CStringGetTextDatum(de->d_name);
-		nulls[0] = false;
-
-		tuplestore_putvalues(tupstore, tupdesc, values, nulls);
-	}
-
-	FreeDir(dirdesc);
-	return (Datum) 0;
+	text	*filename_t = PG_GETARG_TEXT_PP(0);
+	char	*filename = convert_and_check_filename(filename_t);
+	return pg_ls_dir_files(fcinfo, filename, LS_DIR_SKIP_DOT_DIRS);
 }
 
 /*
@@ -525,7 +475,9 @@ pg_ls_dir(PG_FUNCTION_ARGS)
 Datum
 pg_ls_dir_1arg(PG_FUNCTION_ARGS)
 {
-	return pg_ls_dir(fcinfo);
+	text	*filename_t = PG_GETARG_TEXT_PP(0);
+	char	*filename = convert_and_check_filename(filename_t);
+	return pg_ls_dir_files(fcinfo, filename, LS_DIR_SKIP_DOT_DIRS);
 }
 
 /*
@@ -535,7 +487,7 @@ pg_ls_dir_1arg(PG_FUNCTION_ARGS)
  * Other unreadable-directory cases throw an error.
  */
 static Datum
-pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, bool missing_ok)
+pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, int flags)
 {
 	ReturnSetInfo *rsinfo = (ReturnSetInfo *) fcinfo->resultinfo;
 	bool		randomAccess;
@@ -544,6 +496,32 @@ pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, bool missing_ok)
 	DIR		   *dirdesc;
 	struct dirent *de;
 	MemoryContext oldcontext;
+	TypeFuncClass	tuptype ;
+
+	/* isdir depends on metadata */
+	Assert(!(flags&LS_DIR_ISDIR) || (flags&LS_DIR_METADATA));
+	/* Unreasonable to show isdir and skip dirs */
+	Assert(!(flags&LS_DIR_ISDIR) || !(flags&LS_DIR_SKIP_DIRS));
+
+	/* check the optional arguments */
+	if (PG_NARGS() == 3)
+	{
+		if (!PG_ARGISNULL(1))
+		{
+			if (PG_GETARG_BOOL(1))
+				flags |= LS_DIR_MISSING_OK;
+			else
+				flags &= ~LS_DIR_MISSING_OK;
+		}
+
+		if (!PG_ARGISNULL(2))
+		{
+			if (PG_GETARG_BOOL(2))
+				flags &= ~LS_DIR_SKIP_DOT_DIRS;
+			else
+				flags |= LS_DIR_SKIP_DOT_DIRS;
+		}
+	}
 
 	/* check to see if caller supports us returning a tuplestore */
 	if (rsinfo == NULL || !IsA(rsinfo, ReturnSetInfo))
@@ -558,8 +536,20 @@ pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, bool missing_ok)
 	/* The tupdesc and tuplestore must be created in ecxt_per_query_memory */
 	oldcontext = MemoryContextSwitchTo(rsinfo->econtext->ecxt_per_query_memory);
 
-	if (get_call_result_type(fcinfo, NULL, &tupdesc) != TYPEFUNC_COMPOSITE)
-		elog(ERROR, "return type must be a row type");
+	tuptype = get_call_result_type(fcinfo, NULL, &tupdesc);
+	if (flags & LS_DIR_METADATA)
+	{
+		if (tuptype != TYPEFUNC_COMPOSITE)
+			elog(ERROR, "return type must be a row type");
+	}
+	else
+	{
+		/* pg_ls_dir returns a simple scalar */
+		if (tuptype != TYPEFUNC_SCALAR)
+			elog(ERROR, "return type must be a scalar type");
+		tupdesc = CreateTemplateTupleDesc(1);
+		TupleDescInitEntry(tupdesc, (AttrNumber) 1, "column", TEXTOID, -1, 0);
+	}
 
 	randomAccess = (rsinfo->allowedModes & SFRM_Materialize_Random) != 0;
 	tupstore = tuplestore_begin_heap(randomAccess, false, work_mem);
@@ -578,20 +568,27 @@ pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, bool missing_ok)
 	if (!dirdesc)
 	{
 		/* Return empty tuplestore if appropriate */
-		if (missing_ok && errno == ENOENT)
+		if (flags & LS_DIR_MISSING_OK && errno == ENOENT)
 			return (Datum) 0;
 		/* Otherwise, we can let ReadDir() throw the error */
 	}
 
 	while ((de = ReadDir(dirdesc, dir)) != NULL)
 	{
-		Datum		values[3];
-		bool		nulls[3];
+		Datum		values[4];
+		bool		nulls[4];
 		char		path[MAXPGPATH * 2];
 		struct stat attrib;
 
-		/* Skip hidden files */
-		if (de->d_name[0] == '.')
+		/* Skip dot dirs? */
+		if (flags & LS_DIR_SKIP_DOT_DIRS &&
+			(strcmp(de->d_name, ".") == 0 ||
+			 strcmp(de->d_name, "..") == 0))
+			continue;
+
+		/* Skip hidden files? */
+		if (flags & LS_DIR_SKIP_HIDDEN &&
+			de->d_name[0] == '.')
 			continue;
 
 		/* Get the file info */
@@ -606,13 +603,34 @@ pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, bool missing_ok)
 					 errmsg("could not stat file \"%s\": %m", path)));
 		}
 
-		/* Ignore anything but regular files */
-		if (!S_ISREG(attrib.st_mode))
-			continue;
+		/* Skip dirs or special files? */
+		if (S_ISDIR(attrib.st_mode))
+		{
+			if (flags & LS_DIR_SKIP_DIRS)
+				continue;
+		}
+		else if (!S_ISREG(attrib.st_mode))
+		{
+			if (flags & LS_DIR_SKIP_SPECIAL)
+				continue;
+		}
 
 		values[0] = CStringGetTextDatum(de->d_name);
-		values[1] = Int64GetDatum((int64) attrib.st_size);
-		values[2] = TimestampTzGetDatum(time_t_to_timestamptz(attrib.st_mtime));
+		if (flags & LS_DIR_METADATA)
+		{
+			values[1] = Int64GetDatum((int64) attrib.st_size);
+			values[2] = TimestampTzGetDatum(time_t_to_timestamptz(attrib.st_mtime));
+			if (flags & LS_DIR_ISDIR)
+			{
+				values[3] = BoolGetDatum(S_ISDIR(attrib.st_mode));
+#ifdef WIN32
+				/* Links should have isdir=false */
+				if (pgwin32_is_junction(path))
+					values[3] = BoolGetDatum(false);
+#endif
+			}
+		}
+
 		memset(nulls, 0, sizeof(nulls));
 
 		tuplestore_putvalues(tupstore, tupdesc, values, nulls);
@@ -626,14 +644,14 @@ pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, bool missing_ok)
 Datum
 pg_ls_logdir(PG_FUNCTION_ARGS)
 {
-	return pg_ls_dir_files(fcinfo, Log_directory, false);
+	return pg_ls_dir_files(fcinfo, Log_directory, LS_DIR_HISTORIC);
 }
 
 /* Function to return the list of files in the WAL directory */
 Datum
 pg_ls_waldir(PG_FUNCTION_ARGS)
 {
-	return pg_ls_dir_files(fcinfo, XLOGDIR, false);
+	return pg_ls_dir_files(fcinfo, XLOGDIR, LS_DIR_HISTORIC);
 }
 
 /*
@@ -651,7 +669,8 @@ pg_ls_tmpdir(FunctionCallInfo fcinfo, Oid tblspc)
 						tblspc)));
 
 	TempTablespacePath(path, tblspc);
-	return pg_ls_dir_files(fcinfo, path, true);
+	return pg_ls_dir_files(fcinfo, path,
+			LS_DIR_HISTORIC | LS_DIR_MISSING_OK);
 }
 
 /*
@@ -680,5 +699,33 @@ pg_ls_tmpdir_1arg(PG_FUNCTION_ARGS)
 Datum
 pg_ls_archive_statusdir(PG_FUNCTION_ARGS)
 {
-	return pg_ls_dir_files(fcinfo, XLOGDIR "/archive_status", true);
+	return pg_ls_dir_files(fcinfo, XLOGDIR "/archive_status",
+			LS_DIR_HISTORIC | LS_DIR_MISSING_OK);
+}
+
+/*
+ * Function to return the list of files and metadata in an arbitrary directory.
+ */
+Datum
+pg_ls_dir_metadata(PG_FUNCTION_ARGS)
+{
+	char	*dirname = convert_and_check_filename(PG_GETARG_TEXT_PP(0));
+
+	return pg_ls_dir_files(fcinfo, dirname,
+			LS_DIR_METADATA | LS_DIR_SKIP_SPECIAL | LS_DIR_ISDIR);
+}
+
+/*
+ * Function to return the list of files and metadata in an arbitrary directory.
+ * note: this wrapper is necessary to pass the sanity check in opr_sanity,
+ * which checks that all built-in functions that share the implementing C
+ * function take the same number of arguments.
+ */
+Datum
+pg_ls_dir_metadata_1arg(PG_FUNCTION_ARGS)
+{
+	char	*dirname = convert_and_check_filename(PG_GETARG_TEXT_PP(0));
+
+	return pg_ls_dir_files(fcinfo, dirname,
+			LS_DIR_METADATA | LS_DIR_SKIP_SPECIAL | LS_DIR_ISDIR);
 }
diff --git a/src/include/catalog/pg_proc.dat b/src/include/catalog/pg_proc.dat
index 4bce3ad8de..9f46cba5ed 100644
--- a/src/include/catalog/pg_proc.dat
+++ b/src/include/catalog/pg_proc.dat
@@ -10904,6 +10904,18 @@
   proallargtypes => '{oid,text,int8,timestamptz}', proargmodes => '{i,o,o,o}',
   proargnames => '{tablespace,name,size,modification}',
   prosrc => 'pg_ls_tmpdir_1arg' },
+{ oid => '5032', descr => 'list directory with metadata',
+  proname => 'pg_ls_dir_metadata', procost => '10', prorows => '20', proretset => 't',
+  provolatile => 'v', prorettype => 'record', proargtypes => 'text bool bool',
+  proallargtypes => '{text,bool,bool,text,int8,timestamptz,bool}', proargmodes => '{i,i,i,o,o,o,o}',
+  proargnames => '{dirname,missing_ok,include_dot_dirs,name,size,modification,isdir}',
+  prosrc => 'pg_ls_dir_metadata' },
+{ oid => '5033', descr => 'list directory with metadata',
+  proname => 'pg_ls_dir_metadata', procost => '10', prorows => '20', proretset => 't',
+  provolatile => 'v', prorettype => 'record', proargtypes => 'text',
+  proallargtypes => '{text,text,int8,timestamptz,bool}', proargmodes => '{i,o,o,o,o}',
+  proargnames => '{dirname,name,size,modification,isdir}',
+  prosrc => 'pg_ls_dir_metadata_1arg' },
 
 # hash partitioning constraint function
 { oid => '5028', descr => 'hash partition CHECK constraint',
diff --git a/src/test/regress/expected/misc_functions.out b/src/test/regress/expected/misc_functions.out
index 2e87c548eb..7930909f02 100644
--- a/src/test/regress/expected/misc_functions.out
+++ b/src/test/regress/expected/misc_functions.out
@@ -219,6 +219,30 @@ select pg_ls_dir('does not exist', true, false); -- ok with missingok=true
 
 select pg_ls_dir('does not exist'); -- fails with missingok=false
 ERROR:  could not open directory "does not exist": No such file or directory
+-- This tests the missing_ok parameter, which causes pg_ls_tmpdir to succeed even if the tmpdir doesn't exist yet
+-- The name='' condition is never true, so the function runs to completion but returns zero rows.
+select * from pg_ls_tmpdir() where name='Does not exist';
+ name | size | modification 
+------+------+--------------
+(0 rows)
+
+select name, isdir from pg_ls_dir_metadata('.') where name='.';
+ name | isdir 
+------+-------
+ .    | t
+(1 row)
+
+select name, isdir from pg_ls_dir_metadata('.', false, false) where name='.'; -- include_dot_dirs=false
+ name | isdir 
+------+-------
+(0 rows)
+
+-- Check that expected columns are present
+select * from pg_ls_dir_metadata('.') limit 0;
+ name | size | modification | isdir 
+------+------+--------------+-------
+(0 rows)
+
 --
 -- Test adding a support function to a subject function
 --
diff --git a/src/test/regress/input/tablespace.source b/src/test/regress/input/tablespace.source
index a5f61a35dc..0b9cfe615e 100644
--- a/src/test/regress/input/tablespace.source
+++ b/src/test/regress/input/tablespace.source
@@ -11,6 +11,11 @@ DROP TABLESPACE regress_tblspacewith;
 -- create a tablespace we can use
 CREATE TABLESPACE regress_tblspace LOCATION '@testtablespace@';
 
+-- This tests the missing_ok parameter, which causes pg_ls_tmpdir to succeed even if the tmpdir doesn't exist yet
+-- The name='' condition is never true, so the function runs to completion but returns zero rows.
+-- The query is written to ERROR if the tablespace doesn't exist, rather than silently failing to call pg_ls_tmpdir()
+SELECT c.* FROM (SELECT oid FROM pg_tablespace b WHERE b.spcname='regress_tblspace' UNION SELECT 0 ORDER BY 1 DESC LIMIT 1) AS b , pg_ls_tmpdir(oid) AS c WHERE c.name='Does not exist';
+
 -- try setting and resetting some properties for the new tablespace
 ALTER TABLESPACE regress_tblspace SET (random_page_cost = 1.0, seq_page_cost = 1.1);
 ALTER TABLESPACE regress_tblspace SET (some_nonexistent_parameter = true);  -- fail
diff --git a/src/test/regress/output/tablespace.source b/src/test/regress/output/tablespace.source
index 162b591b31..a42714bf40 100644
--- a/src/test/regress/output/tablespace.source
+++ b/src/test/regress/output/tablespace.source
@@ -13,6 +13,14 @@ SELECT spcoptions FROM pg_tablespace WHERE spcname = 'regress_tblspacewith';
 DROP TABLESPACE regress_tblspacewith;
 -- create a tablespace we can use
 CREATE TABLESPACE regress_tblspace LOCATION '@testtablespace@';
+-- This tests the missing_ok parameter, which causes pg_ls_tmpdir to succeed even if the tmpdir doesn't exist yet
+-- The name='' condition is never true, so the function runs to completion but returns zero rows.
+-- The query is written to ERROR if the tablespace doesn't exist, rather than silently failing to call pg_ls_tmpdir()
+SELECT c.* FROM (SELECT oid FROM pg_tablespace b WHERE b.spcname='regress_tblspace' UNION SELECT 0 ORDER BY 1 DESC LIMIT 1) AS b , pg_ls_tmpdir(oid) AS c WHERE c.name='Does not exist';
+ name | size | modification 
+------+------+--------------
+(0 rows)
+
 -- try setting and resetting some properties for the new tablespace
 ALTER TABLESPACE regress_tblspace SET (random_page_cost = 1.0, seq_page_cost = 1.1);
 ALTER TABLESPACE regress_tblspace SET (some_nonexistent_parameter = true);  -- fail
diff --git a/src/test/regress/sql/misc_functions.sql b/src/test/regress/sql/misc_functions.sql
index f6857ad177..372345720d 100644
--- a/src/test/regress/sql/misc_functions.sql
+++ b/src/test/regress/sql/misc_functions.sql
@@ -65,6 +65,17 @@ select * from (select pg_ls_dir('.', false, false) as name) as ls where ls.name=
 select pg_ls_dir('does not exist', true, false); -- ok with missingok=true
 select pg_ls_dir('does not exist'); -- fails with missingok=false
 
+-- This tests the missing_ok parameter, which causes pg_ls_tmpdir to succeed even if the tmpdir doesn't exist yet
+-- The name='' condition is never true, so the function runs to completion but returns zero rows.
+select * from pg_ls_tmpdir() where name='Does not exist';
+
+select name, isdir from pg_ls_dir_metadata('.') where name='.';
+
+select name, isdir from pg_ls_dir_metadata('.', false, false) where name='.'; -- include_dot_dirs=false
+
+-- Check that expected columns are present
+select * from pg_ls_dir_metadata('.') limit 0;
+
 --
 -- Test adding a support function to a subject function
 --
-- 
2.17.0


--4LFBTxd4L5NLO6ly
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
 filename="v17-0005-pg_ls_tmpdir-to-show-directories-and-isdir-argum.patch"



^ permalink  raw  reply  [nested|flat] 68+ messages in thread

* [PATCH v16 04/10] Add pg_ls_dir_metadata to list a dir with file metadata..
@ 2020-03-10 03:40  Justin Pryzby <[email protected]>
  0 siblings, 0 replies; 68+ messages in thread

From: Justin Pryzby @ 2020-03-10 03:40 UTC (permalink / raw)

Generalize pg_ls_dir_files and retire pg_ls_dir

Need catversion bumped?
---
 doc/src/sgml/func.sgml                       |  19 ++
 src/backend/catalog/system_views.sql         |   1 +
 src/backend/utils/adt/genfile.c              | 229 +++++++++++--------
 src/include/catalog/pg_proc.dat              |  12 +
 src/test/regress/expected/misc_functions.out |  24 ++
 src/test/regress/input/tablespace.source     |   5 +
 src/test/regress/output/tablespace.source    |   8 +
 src/test/regress/sql/misc_functions.sql      |  11 +
 8 files changed, 218 insertions(+), 91 deletions(-)

diff --git a/doc/src/sgml/func.sgml b/doc/src/sgml/func.sgml
index 96b08d0500..9d617f95b9 100644
--- a/doc/src/sgml/func.sgml
+++ b/doc/src/sgml/func.sgml
@@ -25300,6 +25300,17 @@ postgres=# SELECT * FROM pg_walfile_name_offset(pg_stop_backup());
         List the contents of a directory.  Restricted to superusers by default, but other users can be granted EXECUTE to run the function.
        </entry>
       </row>
+      <row>
+       <entry>
+        <literal><function>pg_ls_dir_metadata(<parameter>dirname</parameter> <type>text</type> [, <parameter>missing_ok</parameter> <type>boolean</type>, <parameter>include_dot_dirs</parameter> <type>boolean</type>])</function></literal>
+       </entry>
+       <entry><type>setof text</type></entry>
+       <entry>
+        For each file in a directory, list the file and its metadata.
+        Restricted to superusers by default, but other users can be granted
+        EXECUTE to run the function.
+       </entry>
+      </row>
       <row>
        <entry>
         <literal><function>pg_ls_logdir()</function></literal>
@@ -25400,6 +25411,14 @@ postgres=# SELECT * FROM pg_walfile_name_offset(pg_stop_backup());
     empty directory from an non-existent directory.
    </para>
 
+   <indexterm>
+    <primary>pg_ls_dir_metadata</primary>
+   </indexterm>
+   <para>
+    <function>pg_ls_dir_metadata</function> lists the files in the specified
+    directory along with the file's metadata.
+   </para>
+
    <indexterm>
     <primary>pg_ls_logdir</primary>
    </indexterm>
diff --git a/src/backend/catalog/system_views.sql b/src/backend/catalog/system_views.sql
index 2bd5f5ea14..1c77430f0c 100644
--- a/src/backend/catalog/system_views.sql
+++ b/src/backend/catalog/system_views.sql
@@ -1467,6 +1467,7 @@ REVOKE EXECUTE ON FUNCTION pg_stat_file(text,boolean) FROM public;
 
 REVOKE EXECUTE ON FUNCTION pg_ls_dir(text) FROM public;
 REVOKE EXECUTE ON FUNCTION pg_ls_dir(text,boolean,boolean) FROM public;
+REVOKE EXECUTE ON FUNCTION pg_ls_dir_metadata(text,boolean,boolean) FROM public;
 
 --
 -- We also set up some things as accessible to standard roles.
diff --git a/src/backend/utils/adt/genfile.c b/src/backend/utils/adt/genfile.c
index 219ac160f8..4824a55480 100644
--- a/src/backend/utils/adt/genfile.c
+++ b/src/backend/utils/adt/genfile.c
@@ -36,6 +36,21 @@
 #include "utils/syscache.h"
 #include "utils/timestamp.h"
 
+static Datum pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, int flags);
+
+#define	LS_DIR_ISDIR				(1<<0) /* Show column: isdir */
+#define	LS_DIR_METADATA				(1<<1) /* Show columns: mtime, size */
+#define	LS_DIR_MISSING_OK			(1<<2) /* Ignore ENOENT if the toplevel dir is missing */
+#define	LS_DIR_SKIP_DOT_DIRS		(1<<3) /* Do not show . or .. */
+#define	LS_DIR_SKIP_HIDDEN			(1<<4) /* Do not show anything begining with . */
+#define	LS_DIR_SKIP_DIRS			(1<<5) /* Do not show directories */
+#define	LS_DIR_SKIP_SPECIAL			(1<<6) /* Do not show special file types */
+
+/*
+ * Shortcut for the historic behavior of the pg_ls_* functions (not including
+ * pg_ls_dir, which skips different files and doesn't show metadata.
+ */
+#define LS_DIR_HISTORIC				(LS_DIR_SKIP_DIRS|LS_DIR_SKIP_HIDDEN|LS_DIR_SKIP_SPECIAL|LS_DIR_METADATA)
 
 /*
  * Convert a "text" filename argument to C string, and check it's allowable.
@@ -413,6 +428,11 @@ pg_stat_file(PG_FUNCTION_ARGS)
 	values[4] = TimestampTzGetDatum(time_t_to_timestamptz(fst.st_ctime));
 #endif
 	values[5] = BoolGetDatum(S_ISDIR(fst.st_mode));
+#ifdef WIN32
+	/* Links should have isdir=false */
+	if (pgwin32_is_junction(filename))
+		values[5] = BoolGetDatum(false);
+#endif
 
 	tuple = heap_form_tuple(tupdesc, values, isnull);
 
@@ -440,79 +460,9 @@ pg_stat_file_1arg(PG_FUNCTION_ARGS)
 Datum
 pg_ls_dir(PG_FUNCTION_ARGS)
 {
-	ReturnSetInfo *rsinfo = (ReturnSetInfo *) fcinfo->resultinfo;
-	char	   *location;
-	bool		missing_ok = false;
-	bool		include_dot_dirs = false;
-	bool		randomAccess;
-	TupleDesc	tupdesc;
-	Tuplestorestate *tupstore;
-	DIR		   *dirdesc;
-	struct dirent *de;
-	MemoryContext oldcontext;
-
-	location = convert_and_check_filename(PG_GETARG_TEXT_PP(0));
-
-	/* check the optional arguments */
-	if (PG_NARGS() == 3)
-	{
-		if (!PG_ARGISNULL(1))
-			missing_ok = PG_GETARG_BOOL(1);
-		if (!PG_ARGISNULL(2))
-			include_dot_dirs = PG_GETARG_BOOL(2);
-	}
-
-	/* check to see if caller supports us returning a tuplestore */
-	if (rsinfo == NULL || !IsA(rsinfo, ReturnSetInfo))
-		ereport(ERROR,
-				(errcode(ERRCODE_FEATURE_NOT_SUPPORTED),
-				 errmsg("set-valued function called in context that cannot accept a set")));
-	if (!(rsinfo->allowedModes & SFRM_Materialize))
-		ereport(ERROR,
-				(errcode(ERRCODE_SYNTAX_ERROR),
-				 errmsg("materialize mode required, but it is not allowed in this context")));
-
-	/* The tupdesc and tuplestore must be created in ecxt_per_query_memory */
-	oldcontext = MemoryContextSwitchTo(rsinfo->econtext->ecxt_per_query_memory);
-
-	tupdesc = CreateTemplateTupleDesc(1);
-	TupleDescInitEntry(tupdesc, (AttrNumber) 1, "pg_ls_dir", TEXTOID, -1, 0);
-
-	randomAccess = (rsinfo->allowedModes & SFRM_Materialize_Random) != 0;
-	tupstore = tuplestore_begin_heap(randomAccess, false, work_mem);
-	rsinfo->returnMode = SFRM_Materialize;
-	rsinfo->setResult = tupstore;
-	rsinfo->setDesc = tupdesc;
-
-	MemoryContextSwitchTo(oldcontext);
-
-	dirdesc = AllocateDir(location);
-	if (!dirdesc)
-	{
-		/* Return empty tuplestore if appropriate */
-		if (missing_ok && errno == ENOENT)
-			return (Datum) 0;
-		/* Otherwise, we can let ReadDir() throw the error */
-	}
-
-	while ((de = ReadDir(dirdesc, location)) != NULL)
-	{
-		Datum		values[1];
-		bool		nulls[1];
-
-		if (!include_dot_dirs &&
-			(strcmp(de->d_name, ".") == 0 ||
-			 strcmp(de->d_name, "..") == 0))
-			continue;
-
-		values[0] = CStringGetTextDatum(de->d_name);
-		nulls[0] = false;
-
-		tuplestore_putvalues(tupstore, tupdesc, values, nulls);
-	}
-
-	FreeDir(dirdesc);
-	return (Datum) 0;
+	text	*filename_t = PG_GETARG_TEXT_PP(0);
+	char	*filename = convert_and_check_filename(filename_t);
+	return pg_ls_dir_files(fcinfo, filename, LS_DIR_SKIP_DOT_DIRS);
 }
 
 /*
@@ -525,7 +475,9 @@ pg_ls_dir(PG_FUNCTION_ARGS)
 Datum
 pg_ls_dir_1arg(PG_FUNCTION_ARGS)
 {
-	return pg_ls_dir(fcinfo);
+	text	*filename_t = PG_GETARG_TEXT_PP(0);
+	char	*filename = convert_and_check_filename(filename_t);
+	return pg_ls_dir_files(fcinfo, filename, LS_DIR_SKIP_DOT_DIRS);
 }
 
 /*
@@ -535,7 +487,7 @@ pg_ls_dir_1arg(PG_FUNCTION_ARGS)
  * Other unreadable-directory cases throw an error.
  */
 static Datum
-pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, bool missing_ok)
+pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, int flags)
 {
 	ReturnSetInfo *rsinfo = (ReturnSetInfo *) fcinfo->resultinfo;
 	bool		randomAccess;
@@ -544,6 +496,32 @@ pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, bool missing_ok)
 	DIR		   *dirdesc;
 	struct dirent *de;
 	MemoryContext oldcontext;
+	TypeFuncClass	tuptype ;
+
+	/* isdir depends on metadata */
+	Assert(!(flags&LS_DIR_ISDIR) || (flags&LS_DIR_METADATA));
+	/* Unreasonable to show isdir and skip dirs */
+	Assert(!(flags&LS_DIR_ISDIR) || !(flags&LS_DIR_SKIP_DIRS));
+
+	/* check the optional arguments */
+	if (PG_NARGS() == 3)
+	{
+		if (!PG_ARGISNULL(1))
+		{
+			if (PG_GETARG_BOOL(1))
+				flags |= LS_DIR_MISSING_OK;
+			else
+				flags &= ~LS_DIR_MISSING_OK;
+		}
+
+		if (!PG_ARGISNULL(2))
+		{
+			if (PG_GETARG_BOOL(2))
+				flags &= ~LS_DIR_SKIP_DOT_DIRS;
+			else
+				flags |= LS_DIR_SKIP_DOT_DIRS;
+		}
+	}
 
 	/* check to see if caller supports us returning a tuplestore */
 	if (rsinfo == NULL || !IsA(rsinfo, ReturnSetInfo))
@@ -558,8 +536,20 @@ pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, bool missing_ok)
 	/* The tupdesc and tuplestore must be created in ecxt_per_query_memory */
 	oldcontext = MemoryContextSwitchTo(rsinfo->econtext->ecxt_per_query_memory);
 
-	if (get_call_result_type(fcinfo, NULL, &tupdesc) != TYPEFUNC_COMPOSITE)
-		elog(ERROR, "return type must be a row type");
+	tuptype = get_call_result_type(fcinfo, NULL, &tupdesc);
+	if (flags & LS_DIR_METADATA)
+	{
+		if (tuptype != TYPEFUNC_COMPOSITE)
+			elog(ERROR, "return type must be a row type");
+	}
+	else
+	{
+		/* pg_ls_dir returns a simple scalar */
+		if (tuptype != TYPEFUNC_SCALAR)
+			elog(ERROR, "return type must be a scalar type");
+		tupdesc = CreateTemplateTupleDesc(1);
+		TupleDescInitEntry(tupdesc, (AttrNumber) 1, "column", TEXTOID, -1, 0);
+	}
 
 	randomAccess = (rsinfo->allowedModes & SFRM_Materialize_Random) != 0;
 	tupstore = tuplestore_begin_heap(randomAccess, false, work_mem);
@@ -578,20 +568,27 @@ pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, bool missing_ok)
 	if (!dirdesc)
 	{
 		/* Return empty tuplestore if appropriate */
-		if (missing_ok && errno == ENOENT)
+		if (flags & LS_DIR_MISSING_OK && errno == ENOENT)
 			return (Datum) 0;
 		/* Otherwise, we can let ReadDir() throw the error */
 	}
 
 	while ((de = ReadDir(dirdesc, dir)) != NULL)
 	{
-		Datum		values[3];
-		bool		nulls[3];
+		Datum		values[4];
+		bool		nulls[4];
 		char		path[MAXPGPATH * 2];
 		struct stat attrib;
 
-		/* Skip hidden files */
-		if (de->d_name[0] == '.')
+		/* Skip dot dirs? */
+		if (flags & LS_DIR_SKIP_DOT_DIRS &&
+			(strcmp(de->d_name, ".") == 0 ||
+			 strcmp(de->d_name, "..") == 0))
+			continue;
+
+		/* Skip hidden files? */
+		if (flags & LS_DIR_SKIP_HIDDEN &&
+			de->d_name[0] == '.')
 			continue;
 
 		/* Get the file info */
@@ -606,13 +603,34 @@ pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, bool missing_ok)
 					 errmsg("could not stat file \"%s\": %m", path)));
 		}
 
-		/* Ignore anything but regular files */
-		if (!S_ISREG(attrib.st_mode))
-			continue;
+		/* Skip dirs or special files? */
+		if (S_ISDIR(attrib.st_mode))
+		{
+			if (flags & LS_DIR_SKIP_DIRS)
+				continue;
+		}
+		else if (!S_ISREG(attrib.st_mode))
+		{
+			if (flags & LS_DIR_SKIP_SPECIAL)
+				continue;
+		}
 
 		values[0] = CStringGetTextDatum(de->d_name);
-		values[1] = Int64GetDatum((int64) attrib.st_size);
-		values[2] = TimestampTzGetDatum(time_t_to_timestamptz(attrib.st_mtime));
+		if (flags & LS_DIR_METADATA)
+		{
+			values[1] = Int64GetDatum((int64) attrib.st_size);
+			values[2] = TimestampTzGetDatum(time_t_to_timestamptz(attrib.st_mtime));
+			if (flags & LS_DIR_ISDIR)
+			{
+				values[3] = BoolGetDatum(S_ISDIR(attrib.st_mode));
+#ifdef WIN32
+				/* Links should have isdir=false */
+				if (pgwin32_is_junction(path))
+					values[3] = BoolGetDatum(false);
+#endif
+			}
+		}
+
 		memset(nulls, 0, sizeof(nulls));
 
 		tuplestore_putvalues(tupstore, tupdesc, values, nulls);
@@ -626,14 +644,14 @@ pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, bool missing_ok)
 Datum
 pg_ls_logdir(PG_FUNCTION_ARGS)
 {
-	return pg_ls_dir_files(fcinfo, Log_directory, false);
+	return pg_ls_dir_files(fcinfo, Log_directory, LS_DIR_HISTORIC);
 }
 
 /* Function to return the list of files in the WAL directory */
 Datum
 pg_ls_waldir(PG_FUNCTION_ARGS)
 {
-	return pg_ls_dir_files(fcinfo, XLOGDIR, false);
+	return pg_ls_dir_files(fcinfo, XLOGDIR, LS_DIR_HISTORIC);
 }
 
 /*
@@ -651,7 +669,8 @@ pg_ls_tmpdir(FunctionCallInfo fcinfo, Oid tblspc)
 						tblspc)));
 
 	TempTablespacePath(path, tblspc);
-	return pg_ls_dir_files(fcinfo, path, true);
+	return pg_ls_dir_files(fcinfo, path,
+			LS_DIR_HISTORIC | LS_DIR_MISSING_OK);
 }
 
 /*
@@ -680,5 +699,33 @@ pg_ls_tmpdir_1arg(PG_FUNCTION_ARGS)
 Datum
 pg_ls_archive_statusdir(PG_FUNCTION_ARGS)
 {
-	return pg_ls_dir_files(fcinfo, XLOGDIR "/archive_status", true);
+	return pg_ls_dir_files(fcinfo, XLOGDIR "/archive_status",
+			LS_DIR_HISTORIC | LS_DIR_MISSING_OK);
+}
+
+/*
+ * Function to return the list of files and metadata in an arbitrary directory.
+ */
+Datum
+pg_ls_dir_metadata(PG_FUNCTION_ARGS)
+{
+	char	*dirname = convert_and_check_filename(PG_GETARG_TEXT_PP(0));
+
+	return pg_ls_dir_files(fcinfo, dirname,
+			LS_DIR_METADATA | LS_DIR_SKIP_SPECIAL | LS_DIR_ISDIR);
+}
+
+/*
+ * Function to return the list of files and metadata in an arbitrary directory.
+ * note: this wrapper is necessary to pass the sanity check in opr_sanity,
+ * which checks that all built-in functions that share the implementing C
+ * function take the same number of arguments.
+ */
+Datum
+pg_ls_dir_metadata_1arg(PG_FUNCTION_ARGS)
+{
+	char	*dirname = convert_and_check_filename(PG_GETARG_TEXT_PP(0));
+
+	return pg_ls_dir_files(fcinfo, dirname,
+			LS_DIR_METADATA | LS_DIR_SKIP_SPECIAL | LS_DIR_ISDIR);
 }
diff --git a/src/include/catalog/pg_proc.dat b/src/include/catalog/pg_proc.dat
index 4bce3ad8de..9f46cba5ed 100644
--- a/src/include/catalog/pg_proc.dat
+++ b/src/include/catalog/pg_proc.dat
@@ -10904,6 +10904,18 @@
   proallargtypes => '{oid,text,int8,timestamptz}', proargmodes => '{i,o,o,o}',
   proargnames => '{tablespace,name,size,modification}',
   prosrc => 'pg_ls_tmpdir_1arg' },
+{ oid => '5032', descr => 'list directory with metadata',
+  proname => 'pg_ls_dir_metadata', procost => '10', prorows => '20', proretset => 't',
+  provolatile => 'v', prorettype => 'record', proargtypes => 'text bool bool',
+  proallargtypes => '{text,bool,bool,text,int8,timestamptz,bool}', proargmodes => '{i,i,i,o,o,o,o}',
+  proargnames => '{dirname,missing_ok,include_dot_dirs,name,size,modification,isdir}',
+  prosrc => 'pg_ls_dir_metadata' },
+{ oid => '5033', descr => 'list directory with metadata',
+  proname => 'pg_ls_dir_metadata', procost => '10', prorows => '20', proretset => 't',
+  provolatile => 'v', prorettype => 'record', proargtypes => 'text',
+  proallargtypes => '{text,text,int8,timestamptz,bool}', proargmodes => '{i,o,o,o,o}',
+  proargnames => '{dirname,name,size,modification,isdir}',
+  prosrc => 'pg_ls_dir_metadata_1arg' },
 
 # hash partitioning constraint function
 { oid => '5028', descr => 'hash partition CHECK constraint',
diff --git a/src/test/regress/expected/misc_functions.out b/src/test/regress/expected/misc_functions.out
index 2e87c548eb..7930909f02 100644
--- a/src/test/regress/expected/misc_functions.out
+++ b/src/test/regress/expected/misc_functions.out
@@ -219,6 +219,30 @@ select pg_ls_dir('does not exist', true, false); -- ok with missingok=true
 
 select pg_ls_dir('does not exist'); -- fails with missingok=false
 ERROR:  could not open directory "does not exist": No such file or directory
+-- This tests the missing_ok parameter, which causes pg_ls_tmpdir to succeed even if the tmpdir doesn't exist yet
+-- The name='' condition is never true, so the function runs to completion but returns zero rows.
+select * from pg_ls_tmpdir() where name='Does not exist';
+ name | size | modification 
+------+------+--------------
+(0 rows)
+
+select name, isdir from pg_ls_dir_metadata('.') where name='.';
+ name | isdir 
+------+-------
+ .    | t
+(1 row)
+
+select name, isdir from pg_ls_dir_metadata('.', false, false) where name='.'; -- include_dot_dirs=false
+ name | isdir 
+------+-------
+(0 rows)
+
+-- Check that expected columns are present
+select * from pg_ls_dir_metadata('.') limit 0;
+ name | size | modification | isdir 
+------+------+--------------+-------
+(0 rows)
+
 --
 -- Test adding a support function to a subject function
 --
diff --git a/src/test/regress/input/tablespace.source b/src/test/regress/input/tablespace.source
index a5f61a35dc..0b9cfe615e 100644
--- a/src/test/regress/input/tablespace.source
+++ b/src/test/regress/input/tablespace.source
@@ -11,6 +11,11 @@ DROP TABLESPACE regress_tblspacewith;
 -- create a tablespace we can use
 CREATE TABLESPACE regress_tblspace LOCATION '@testtablespace@';
 
+-- This tests the missing_ok parameter, which causes pg_ls_tmpdir to succeed even if the tmpdir doesn't exist yet
+-- The name='' condition is never true, so the function runs to completion but returns zero rows.
+-- The query is written to ERROR if the tablespace doesn't exist, rather than silently failing to call pg_ls_tmpdir()
+SELECT c.* FROM (SELECT oid FROM pg_tablespace b WHERE b.spcname='regress_tblspace' UNION SELECT 0 ORDER BY 1 DESC LIMIT 1) AS b , pg_ls_tmpdir(oid) AS c WHERE c.name='Does not exist';
+
 -- try setting and resetting some properties for the new tablespace
 ALTER TABLESPACE regress_tblspace SET (random_page_cost = 1.0, seq_page_cost = 1.1);
 ALTER TABLESPACE regress_tblspace SET (some_nonexistent_parameter = true);  -- fail
diff --git a/src/test/regress/output/tablespace.source b/src/test/regress/output/tablespace.source
index 162b591b31..a42714bf40 100644
--- a/src/test/regress/output/tablespace.source
+++ b/src/test/regress/output/tablespace.source
@@ -13,6 +13,14 @@ SELECT spcoptions FROM pg_tablespace WHERE spcname = 'regress_tblspacewith';
 DROP TABLESPACE regress_tblspacewith;
 -- create a tablespace we can use
 CREATE TABLESPACE regress_tblspace LOCATION '@testtablespace@';
+-- This tests the missing_ok parameter, which causes pg_ls_tmpdir to succeed even if the tmpdir doesn't exist yet
+-- The name='' condition is never true, so the function runs to completion but returns zero rows.
+-- The query is written to ERROR if the tablespace doesn't exist, rather than silently failing to call pg_ls_tmpdir()
+SELECT c.* FROM (SELECT oid FROM pg_tablespace b WHERE b.spcname='regress_tblspace' UNION SELECT 0 ORDER BY 1 DESC LIMIT 1) AS b , pg_ls_tmpdir(oid) AS c WHERE c.name='Does not exist';
+ name | size | modification 
+------+------+--------------
+(0 rows)
+
 -- try setting and resetting some properties for the new tablespace
 ALTER TABLESPACE regress_tblspace SET (random_page_cost = 1.0, seq_page_cost = 1.1);
 ALTER TABLESPACE regress_tblspace SET (some_nonexistent_parameter = true);  -- fail
diff --git a/src/test/regress/sql/misc_functions.sql b/src/test/regress/sql/misc_functions.sql
index f6857ad177..372345720d 100644
--- a/src/test/regress/sql/misc_functions.sql
+++ b/src/test/regress/sql/misc_functions.sql
@@ -65,6 +65,17 @@ select * from (select pg_ls_dir('.', false, false) as name) as ls where ls.name=
 select pg_ls_dir('does not exist', true, false); -- ok with missingok=true
 select pg_ls_dir('does not exist'); -- fails with missingok=false
 
+-- This tests the missing_ok parameter, which causes pg_ls_tmpdir to succeed even if the tmpdir doesn't exist yet
+-- The name='' condition is never true, so the function runs to completion but returns zero rows.
+select * from pg_ls_tmpdir() where name='Does not exist';
+
+select name, isdir from pg_ls_dir_metadata('.') where name='.';
+
+select name, isdir from pg_ls_dir_metadata('.', false, false) where name='.'; -- include_dot_dirs=false
+
+-- Check that expected columns are present
+select * from pg_ls_dir_metadata('.') limit 0;
+
 --
 -- Test adding a support function to a subject function
 --
-- 
2.17.0


--2FkSFaIQeDFoAt0B
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
 filename="v16-0005-pg_ls_tmpdir-to-show-directories-and-isdir-argum.patch"



^ permalink  raw  reply  [nested|flat] 68+ messages in thread

* [PATCH v17 04/10] Add pg_ls_dir_metadata to list a dir with file metadata..
@ 2020-03-10 03:40  Justin Pryzby <[email protected]>
  0 siblings, 0 replies; 68+ messages in thread

From: Justin Pryzby @ 2020-03-10 03:40 UTC (permalink / raw)

Generalize pg_ls_dir_files and retire pg_ls_dir

Need catversion bumped?
---
 doc/src/sgml/func.sgml                       |  21 ++
 src/backend/catalog/system_views.sql         |   1 +
 src/backend/utils/adt/genfile.c              | 229 +++++++++++--------
 src/include/catalog/pg_proc.dat              |  12 +
 src/test/regress/expected/misc_functions.out |  24 ++
 src/test/regress/input/tablespace.source     |   5 +
 src/test/regress/output/tablespace.source    |   8 +
 src/test/regress/sql/misc_functions.sql      |  11 +
 8 files changed, 220 insertions(+), 91 deletions(-)

diff --git a/doc/src/sgml/func.sgml b/doc/src/sgml/func.sgml
index d9b3598977..fc1b4ac98c 100644
--- a/doc/src/sgml/func.sgml
+++ b/doc/src/sgml/func.sgml
@@ -25714,6 +25714,27 @@ SELECT pg_size_pretty(sum(pg_relation_size(relid))) AS total_size
        </para></entry>
       </row>
 
+      <row>
+       <entry role="func_table_entry"><para role="func_signature">
+        <indexterm>
+         <primary>pg_ls_dir_metadata</primary>
+        </indexterm>
+        <function>pg_ls_dir_metadata</function> ( <parameter>dirname</parameter> <type>text</type>
+        <optional>, <parameter>missing_ok</parameter> <type>boolean</type>,
+        <parameter>include_dot_dirs</parameter> <type>boolean</type>] </optional> )
+        <returnvalue>setof record</returnvalue>
+        ( <parameter>name</parameter> <type>text</type>,
+        <parameter>size</parameter> <type>bigint</type>,
+        <parameter>modification</parameter> <type>timestamp with time zone</type> )
+       </para>
+       <para>
+        For each file in the specified directory, list the file and its
+        metadata.
+        Restricted to superusers by default, but other users can be granted
+        EXECUTE to run the function.
+       </para></entry>
+      </row>
+
       <row>
        <entry role="func_table_entry"><para role="func_signature">
         <indexterm>
diff --git a/src/backend/catalog/system_views.sql b/src/backend/catalog/system_views.sql
index 2bd5f5ea14..1c77430f0c 100644
--- a/src/backend/catalog/system_views.sql
+++ b/src/backend/catalog/system_views.sql
@@ -1467,6 +1467,7 @@ REVOKE EXECUTE ON FUNCTION pg_stat_file(text,boolean) FROM public;
 
 REVOKE EXECUTE ON FUNCTION pg_ls_dir(text) FROM public;
 REVOKE EXECUTE ON FUNCTION pg_ls_dir(text,boolean,boolean) FROM public;
+REVOKE EXECUTE ON FUNCTION pg_ls_dir_metadata(text,boolean,boolean) FROM public;
 
 --
 -- We also set up some things as accessible to standard roles.
diff --git a/src/backend/utils/adt/genfile.c b/src/backend/utils/adt/genfile.c
index 219ac160f8..4824a55480 100644
--- a/src/backend/utils/adt/genfile.c
+++ b/src/backend/utils/adt/genfile.c
@@ -36,6 +36,21 @@
 #include "utils/syscache.h"
 #include "utils/timestamp.h"
 
+static Datum pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, int flags);
+
+#define	LS_DIR_ISDIR				(1<<0) /* Show column: isdir */
+#define	LS_DIR_METADATA				(1<<1) /* Show columns: mtime, size */
+#define	LS_DIR_MISSING_OK			(1<<2) /* Ignore ENOENT if the toplevel dir is missing */
+#define	LS_DIR_SKIP_DOT_DIRS		(1<<3) /* Do not show . or .. */
+#define	LS_DIR_SKIP_HIDDEN			(1<<4) /* Do not show anything begining with . */
+#define	LS_DIR_SKIP_DIRS			(1<<5) /* Do not show directories */
+#define	LS_DIR_SKIP_SPECIAL			(1<<6) /* Do not show special file types */
+
+/*
+ * Shortcut for the historic behavior of the pg_ls_* functions (not including
+ * pg_ls_dir, which skips different files and doesn't show metadata.
+ */
+#define LS_DIR_HISTORIC				(LS_DIR_SKIP_DIRS|LS_DIR_SKIP_HIDDEN|LS_DIR_SKIP_SPECIAL|LS_DIR_METADATA)
 
 /*
  * Convert a "text" filename argument to C string, and check it's allowable.
@@ -413,6 +428,11 @@ pg_stat_file(PG_FUNCTION_ARGS)
 	values[4] = TimestampTzGetDatum(time_t_to_timestamptz(fst.st_ctime));
 #endif
 	values[5] = BoolGetDatum(S_ISDIR(fst.st_mode));
+#ifdef WIN32
+	/* Links should have isdir=false */
+	if (pgwin32_is_junction(filename))
+		values[5] = BoolGetDatum(false);
+#endif
 
 	tuple = heap_form_tuple(tupdesc, values, isnull);
 
@@ -440,79 +460,9 @@ pg_stat_file_1arg(PG_FUNCTION_ARGS)
 Datum
 pg_ls_dir(PG_FUNCTION_ARGS)
 {
-	ReturnSetInfo *rsinfo = (ReturnSetInfo *) fcinfo->resultinfo;
-	char	   *location;
-	bool		missing_ok = false;
-	bool		include_dot_dirs = false;
-	bool		randomAccess;
-	TupleDesc	tupdesc;
-	Tuplestorestate *tupstore;
-	DIR		   *dirdesc;
-	struct dirent *de;
-	MemoryContext oldcontext;
-
-	location = convert_and_check_filename(PG_GETARG_TEXT_PP(0));
-
-	/* check the optional arguments */
-	if (PG_NARGS() == 3)
-	{
-		if (!PG_ARGISNULL(1))
-			missing_ok = PG_GETARG_BOOL(1);
-		if (!PG_ARGISNULL(2))
-			include_dot_dirs = PG_GETARG_BOOL(2);
-	}
-
-	/* check to see if caller supports us returning a tuplestore */
-	if (rsinfo == NULL || !IsA(rsinfo, ReturnSetInfo))
-		ereport(ERROR,
-				(errcode(ERRCODE_FEATURE_NOT_SUPPORTED),
-				 errmsg("set-valued function called in context that cannot accept a set")));
-	if (!(rsinfo->allowedModes & SFRM_Materialize))
-		ereport(ERROR,
-				(errcode(ERRCODE_SYNTAX_ERROR),
-				 errmsg("materialize mode required, but it is not allowed in this context")));
-
-	/* The tupdesc and tuplestore must be created in ecxt_per_query_memory */
-	oldcontext = MemoryContextSwitchTo(rsinfo->econtext->ecxt_per_query_memory);
-
-	tupdesc = CreateTemplateTupleDesc(1);
-	TupleDescInitEntry(tupdesc, (AttrNumber) 1, "pg_ls_dir", TEXTOID, -1, 0);
-
-	randomAccess = (rsinfo->allowedModes & SFRM_Materialize_Random) != 0;
-	tupstore = tuplestore_begin_heap(randomAccess, false, work_mem);
-	rsinfo->returnMode = SFRM_Materialize;
-	rsinfo->setResult = tupstore;
-	rsinfo->setDesc = tupdesc;
-
-	MemoryContextSwitchTo(oldcontext);
-
-	dirdesc = AllocateDir(location);
-	if (!dirdesc)
-	{
-		/* Return empty tuplestore if appropriate */
-		if (missing_ok && errno == ENOENT)
-			return (Datum) 0;
-		/* Otherwise, we can let ReadDir() throw the error */
-	}
-
-	while ((de = ReadDir(dirdesc, location)) != NULL)
-	{
-		Datum		values[1];
-		bool		nulls[1];
-
-		if (!include_dot_dirs &&
-			(strcmp(de->d_name, ".") == 0 ||
-			 strcmp(de->d_name, "..") == 0))
-			continue;
-
-		values[0] = CStringGetTextDatum(de->d_name);
-		nulls[0] = false;
-
-		tuplestore_putvalues(tupstore, tupdesc, values, nulls);
-	}
-
-	FreeDir(dirdesc);
-	return (Datum) 0;
+	text	*filename_t = PG_GETARG_TEXT_PP(0);
+	char	*filename = convert_and_check_filename(filename_t);
+	return pg_ls_dir_files(fcinfo, filename, LS_DIR_SKIP_DOT_DIRS);
 }
 
 /*
@@ -525,7 +475,9 @@ pg_ls_dir(PG_FUNCTION_ARGS)
 Datum
 pg_ls_dir_1arg(PG_FUNCTION_ARGS)
 {
-	return pg_ls_dir(fcinfo);
+	text	*filename_t = PG_GETARG_TEXT_PP(0);
+	char	*filename = convert_and_check_filename(filename_t);
+	return pg_ls_dir_files(fcinfo, filename, LS_DIR_SKIP_DOT_DIRS);
 }
 
 /*
@@ -535,7 +487,7 @@ pg_ls_dir_1arg(PG_FUNCTION_ARGS)
  * Other unreadable-directory cases throw an error.
  */
 static Datum
-pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, bool missing_ok)
+pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, int flags)
 {
 	ReturnSetInfo *rsinfo = (ReturnSetInfo *) fcinfo->resultinfo;
 	bool		randomAccess;
@@ -544,6 +496,32 @@ pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, bool missing_ok)
 	DIR		   *dirdesc;
 	struct dirent *de;
 	MemoryContext oldcontext;
+	TypeFuncClass	tuptype ;
+
+	/* isdir depends on metadata */
+	Assert(!(flags&LS_DIR_ISDIR) || (flags&LS_DIR_METADATA));
+	/* Unreasonable to show isdir and skip dirs */
+	Assert(!(flags&LS_DIR_ISDIR) || !(flags&LS_DIR_SKIP_DIRS));
+
+	/* check the optional arguments */
+	if (PG_NARGS() == 3)
+	{
+		if (!PG_ARGISNULL(1))
+		{
+			if (PG_GETARG_BOOL(1))
+				flags |= LS_DIR_MISSING_OK;
+			else
+				flags &= ~LS_DIR_MISSING_OK;
+		}
+
+		if (!PG_ARGISNULL(2))
+		{
+			if (PG_GETARG_BOOL(2))
+				flags &= ~LS_DIR_SKIP_DOT_DIRS;
+			else
+				flags |= LS_DIR_SKIP_DOT_DIRS;
+		}
+	}
 
 	/* check to see if caller supports us returning a tuplestore */
 	if (rsinfo == NULL || !IsA(rsinfo, ReturnSetInfo))
@@ -558,8 +536,20 @@ pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, bool missing_ok)
 	/* The tupdesc and tuplestore must be created in ecxt_per_query_memory */
 	oldcontext = MemoryContextSwitchTo(rsinfo->econtext->ecxt_per_query_memory);
 
-	if (get_call_result_type(fcinfo, NULL, &tupdesc) != TYPEFUNC_COMPOSITE)
-		elog(ERROR, "return type must be a row type");
+	tuptype = get_call_result_type(fcinfo, NULL, &tupdesc);
+	if (flags & LS_DIR_METADATA)
+	{
+		if (tuptype != TYPEFUNC_COMPOSITE)
+			elog(ERROR, "return type must be a row type");
+	}
+	else
+	{
+		/* pg_ls_dir returns a simple scalar */
+		if (tuptype != TYPEFUNC_SCALAR)
+			elog(ERROR, "return type must be a scalar type");
+		tupdesc = CreateTemplateTupleDesc(1);
+		TupleDescInitEntry(tupdesc, (AttrNumber) 1, "column", TEXTOID, -1, 0);
+	}
 
 	randomAccess = (rsinfo->allowedModes & SFRM_Materialize_Random) != 0;
 	tupstore = tuplestore_begin_heap(randomAccess, false, work_mem);
@@ -578,20 +568,27 @@ pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, bool missing_ok)
 	if (!dirdesc)
 	{
 		/* Return empty tuplestore if appropriate */
-		if (missing_ok && errno == ENOENT)
+		if (flags & LS_DIR_MISSING_OK && errno == ENOENT)
 			return (Datum) 0;
 		/* Otherwise, we can let ReadDir() throw the error */
 	}
 
 	while ((de = ReadDir(dirdesc, dir)) != NULL)
 	{
-		Datum		values[3];
-		bool		nulls[3];
+		Datum		values[4];
+		bool		nulls[4];
 		char		path[MAXPGPATH * 2];
 		struct stat attrib;
 
-		/* Skip hidden files */
-		if (de->d_name[0] == '.')
+		/* Skip dot dirs? */
+		if (flags & LS_DIR_SKIP_DOT_DIRS &&
+			(strcmp(de->d_name, ".") == 0 ||
+			 strcmp(de->d_name, "..") == 0))
+			continue;
+
+		/* Skip hidden files? */
+		if (flags & LS_DIR_SKIP_HIDDEN &&
+			de->d_name[0] == '.')
 			continue;
 
 		/* Get the file info */
@@ -606,13 +603,34 @@ pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, bool missing_ok)
 					 errmsg("could not stat file \"%s\": %m", path)));
 		}
 
-		/* Ignore anything but regular files */
-		if (!S_ISREG(attrib.st_mode))
-			continue;
+		/* Skip dirs or special files? */
+		if (S_ISDIR(attrib.st_mode))
+		{
+			if (flags & LS_DIR_SKIP_DIRS)
+				continue;
+		}
+		else if (!S_ISREG(attrib.st_mode))
+		{
+			if (flags & LS_DIR_SKIP_SPECIAL)
+				continue;
+		}
 
 		values[0] = CStringGetTextDatum(de->d_name);
-		values[1] = Int64GetDatum((int64) attrib.st_size);
-		values[2] = TimestampTzGetDatum(time_t_to_timestamptz(attrib.st_mtime));
+		if (flags & LS_DIR_METADATA)
+		{
+			values[1] = Int64GetDatum((int64) attrib.st_size);
+			values[2] = TimestampTzGetDatum(time_t_to_timestamptz(attrib.st_mtime));
+			if (flags & LS_DIR_ISDIR)
+			{
+				values[3] = BoolGetDatum(S_ISDIR(attrib.st_mode));
+#ifdef WIN32
+				/* Links should have isdir=false */
+				if (pgwin32_is_junction(path))
+					values[3] = BoolGetDatum(false);
+#endif
+			}
+		}
+
 		memset(nulls, 0, sizeof(nulls));
 
 		tuplestore_putvalues(tupstore, tupdesc, values, nulls);
@@ -626,14 +644,14 @@ pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, bool missing_ok)
 Datum
 pg_ls_logdir(PG_FUNCTION_ARGS)
 {
-	return pg_ls_dir_files(fcinfo, Log_directory, false);
+	return pg_ls_dir_files(fcinfo, Log_directory, LS_DIR_HISTORIC);
 }
 
 /* Function to return the list of files in the WAL directory */
 Datum
 pg_ls_waldir(PG_FUNCTION_ARGS)
 {
-	return pg_ls_dir_files(fcinfo, XLOGDIR, false);
+	return pg_ls_dir_files(fcinfo, XLOGDIR, LS_DIR_HISTORIC);
 }
 
 /*
@@ -651,7 +669,8 @@ pg_ls_tmpdir(FunctionCallInfo fcinfo, Oid tblspc)
 						tblspc)));
 
 	TempTablespacePath(path, tblspc);
-	return pg_ls_dir_files(fcinfo, path, true);
+	return pg_ls_dir_files(fcinfo, path,
+			LS_DIR_HISTORIC | LS_DIR_MISSING_OK);
 }
 
 /*
@@ -680,5 +699,33 @@ pg_ls_tmpdir_1arg(PG_FUNCTION_ARGS)
 Datum
 pg_ls_archive_statusdir(PG_FUNCTION_ARGS)
 {
-	return pg_ls_dir_files(fcinfo, XLOGDIR "/archive_status", true);
+	return pg_ls_dir_files(fcinfo, XLOGDIR "/archive_status",
+			LS_DIR_HISTORIC | LS_DIR_MISSING_OK);
+}
+
+/*
+ * Function to return the list of files and metadata in an arbitrary directory.
+ */
+Datum
+pg_ls_dir_metadata(PG_FUNCTION_ARGS)
+{
+	char	*dirname = convert_and_check_filename(PG_GETARG_TEXT_PP(0));
+
+	return pg_ls_dir_files(fcinfo, dirname,
+			LS_DIR_METADATA | LS_DIR_SKIP_SPECIAL | LS_DIR_ISDIR);
+}
+
+/*
+ * Function to return the list of files and metadata in an arbitrary directory.
+ * note: this wrapper is necessary to pass the sanity check in opr_sanity,
+ * which checks that all built-in functions that share the implementing C
+ * function take the same number of arguments.
+ */
+Datum
+pg_ls_dir_metadata_1arg(PG_FUNCTION_ARGS)
+{
+	char	*dirname = convert_and_check_filename(PG_GETARG_TEXT_PP(0));
+
+	return pg_ls_dir_files(fcinfo, dirname,
+			LS_DIR_METADATA | LS_DIR_SKIP_SPECIAL | LS_DIR_ISDIR);
 }
diff --git a/src/include/catalog/pg_proc.dat b/src/include/catalog/pg_proc.dat
index 4bce3ad8de..9f46cba5ed 100644
--- a/src/include/catalog/pg_proc.dat
+++ b/src/include/catalog/pg_proc.dat
@@ -10904,6 +10904,18 @@
   proallargtypes => '{oid,text,int8,timestamptz}', proargmodes => '{i,o,o,o}',
   proargnames => '{tablespace,name,size,modification}',
   prosrc => 'pg_ls_tmpdir_1arg' },
+{ oid => '5032', descr => 'list directory with metadata',
+  proname => 'pg_ls_dir_metadata', procost => '10', prorows => '20', proretset => 't',
+  provolatile => 'v', prorettype => 'record', proargtypes => 'text bool bool',
+  proallargtypes => '{text,bool,bool,text,int8,timestamptz,bool}', proargmodes => '{i,i,i,o,o,o,o}',
+  proargnames => '{dirname,missing_ok,include_dot_dirs,name,size,modification,isdir}',
+  prosrc => 'pg_ls_dir_metadata' },
+{ oid => '5033', descr => 'list directory with metadata',
+  proname => 'pg_ls_dir_metadata', procost => '10', prorows => '20', proretset => 't',
+  provolatile => 'v', prorettype => 'record', proargtypes => 'text',
+  proallargtypes => '{text,text,int8,timestamptz,bool}', proargmodes => '{i,o,o,o,o}',
+  proargnames => '{dirname,name,size,modification,isdir}',
+  prosrc => 'pg_ls_dir_metadata_1arg' },
 
 # hash partitioning constraint function
 { oid => '5028', descr => 'hash partition CHECK constraint',
diff --git a/src/test/regress/expected/misc_functions.out b/src/test/regress/expected/misc_functions.out
index 2e87c548eb..7930909f02 100644
--- a/src/test/regress/expected/misc_functions.out
+++ b/src/test/regress/expected/misc_functions.out
@@ -219,6 +219,30 @@ select pg_ls_dir('does not exist', true, false); -- ok with missingok=true
 
 select pg_ls_dir('does not exist'); -- fails with missingok=false
 ERROR:  could not open directory "does not exist": No such file or directory
+-- This tests the missing_ok parameter, which causes pg_ls_tmpdir to succeed even if the tmpdir doesn't exist yet
+-- The name='' condition is never true, so the function runs to completion but returns zero rows.
+select * from pg_ls_tmpdir() where name='Does not exist';
+ name | size | modification 
+------+------+--------------
+(0 rows)
+
+select name, isdir from pg_ls_dir_metadata('.') where name='.';
+ name | isdir 
+------+-------
+ .    | t
+(1 row)
+
+select name, isdir from pg_ls_dir_metadata('.', false, false) where name='.'; -- include_dot_dirs=false
+ name | isdir 
+------+-------
+(0 rows)
+
+-- Check that expected columns are present
+select * from pg_ls_dir_metadata('.') limit 0;
+ name | size | modification | isdir 
+------+------+--------------+-------
+(0 rows)
+
 --
 -- Test adding a support function to a subject function
 --
diff --git a/src/test/regress/input/tablespace.source b/src/test/regress/input/tablespace.source
index a5f61a35dc..0b9cfe615e 100644
--- a/src/test/regress/input/tablespace.source
+++ b/src/test/regress/input/tablespace.source
@@ -11,6 +11,11 @@ DROP TABLESPACE regress_tblspacewith;
 -- create a tablespace we can use
 CREATE TABLESPACE regress_tblspace LOCATION '@testtablespace@';
 
+-- This tests the missing_ok parameter, which causes pg_ls_tmpdir to succeed even if the tmpdir doesn't exist yet
+-- The name='' condition is never true, so the function runs to completion but returns zero rows.
+-- The query is written to ERROR if the tablespace doesn't exist, rather than silently failing to call pg_ls_tmpdir()
+SELECT c.* FROM (SELECT oid FROM pg_tablespace b WHERE b.spcname='regress_tblspace' UNION SELECT 0 ORDER BY 1 DESC LIMIT 1) AS b , pg_ls_tmpdir(oid) AS c WHERE c.name='Does not exist';
+
 -- try setting and resetting some properties for the new tablespace
 ALTER TABLESPACE regress_tblspace SET (random_page_cost = 1.0, seq_page_cost = 1.1);
 ALTER TABLESPACE regress_tblspace SET (some_nonexistent_parameter = true);  -- fail
diff --git a/src/test/regress/output/tablespace.source b/src/test/regress/output/tablespace.source
index 162b591b31..a42714bf40 100644
--- a/src/test/regress/output/tablespace.source
+++ b/src/test/regress/output/tablespace.source
@@ -13,6 +13,14 @@ SELECT spcoptions FROM pg_tablespace WHERE spcname = 'regress_tblspacewith';
 DROP TABLESPACE regress_tblspacewith;
 -- create a tablespace we can use
 CREATE TABLESPACE regress_tblspace LOCATION '@testtablespace@';
+-- This tests the missing_ok parameter, which causes pg_ls_tmpdir to succeed even if the tmpdir doesn't exist yet
+-- The name='' condition is never true, so the function runs to completion but returns zero rows.
+-- The query is written to ERROR if the tablespace doesn't exist, rather than silently failing to call pg_ls_tmpdir()
+SELECT c.* FROM (SELECT oid FROM pg_tablespace b WHERE b.spcname='regress_tblspace' UNION SELECT 0 ORDER BY 1 DESC LIMIT 1) AS b , pg_ls_tmpdir(oid) AS c WHERE c.name='Does not exist';
+ name | size | modification 
+------+------+--------------
+(0 rows)
+
 -- try setting and resetting some properties for the new tablespace
 ALTER TABLESPACE regress_tblspace SET (random_page_cost = 1.0, seq_page_cost = 1.1);
 ALTER TABLESPACE regress_tblspace SET (some_nonexistent_parameter = true);  -- fail
diff --git a/src/test/regress/sql/misc_functions.sql b/src/test/regress/sql/misc_functions.sql
index f6857ad177..372345720d 100644
--- a/src/test/regress/sql/misc_functions.sql
+++ b/src/test/regress/sql/misc_functions.sql
@@ -65,6 +65,17 @@ select * from (select pg_ls_dir('.', false, false) as name) as ls where ls.name=
 select pg_ls_dir('does not exist', true, false); -- ok with missingok=true
 select pg_ls_dir('does not exist'); -- fails with missingok=false
 
+-- This tests the missing_ok parameter, which causes pg_ls_tmpdir to succeed even if the tmpdir doesn't exist yet
+-- The name='' condition is never true, so the function runs to completion but returns zero rows.
+select * from pg_ls_tmpdir() where name='Does not exist';
+
+select name, isdir from pg_ls_dir_metadata('.') where name='.';
+
+select name, isdir from pg_ls_dir_metadata('.', false, false) where name='.'; -- include_dot_dirs=false
+
+-- Check that expected columns are present
+select * from pg_ls_dir_metadata('.') limit 0;
+
 --
 -- Test adding a support function to a subject function
 --
-- 
2.17.0


--4LFBTxd4L5NLO6ly
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
 filename="v17-0005-pg_ls_tmpdir-to-show-directories-and-isdir-argum.patch"



^ permalink  raw  reply  [nested|flat] 68+ messages in thread

* [PATCH v16 04/10] Add pg_ls_dir_metadata to list a dir with file metadata..
@ 2020-03-10 03:40  Justin Pryzby <[email protected]>
  0 siblings, 0 replies; 68+ messages in thread

From: Justin Pryzby @ 2020-03-10 03:40 UTC (permalink / raw)

Generalize pg_ls_dir_files and retire pg_ls_dir

Need catversion bumped?
---
 doc/src/sgml/func.sgml                       |  19 ++
 src/backend/catalog/system_views.sql         |   1 +
 src/backend/utils/adt/genfile.c              | 229 +++++++++++--------
 src/include/catalog/pg_proc.dat              |  12 +
 src/test/regress/expected/misc_functions.out |  24 ++
 src/test/regress/input/tablespace.source     |   5 +
 src/test/regress/output/tablespace.source    |   8 +
 src/test/regress/sql/misc_functions.sql      |  11 +
 8 files changed, 218 insertions(+), 91 deletions(-)

diff --git a/doc/src/sgml/func.sgml b/doc/src/sgml/func.sgml
index 96b08d0500..9d617f95b9 100644
--- a/doc/src/sgml/func.sgml
+++ b/doc/src/sgml/func.sgml
@@ -25300,6 +25300,17 @@ postgres=# SELECT * FROM pg_walfile_name_offset(pg_stop_backup());
         List the contents of a directory.  Restricted to superusers by default, but other users can be granted EXECUTE to run the function.
        </entry>
       </row>
+      <row>
+       <entry>
+        <literal><function>pg_ls_dir_metadata(<parameter>dirname</parameter> <type>text</type> [, <parameter>missing_ok</parameter> <type>boolean</type>, <parameter>include_dot_dirs</parameter> <type>boolean</type>])</function></literal>
+       </entry>
+       <entry><type>setof text</type></entry>
+       <entry>
+        For each file in a directory, list the file and its metadata.
+        Restricted to superusers by default, but other users can be granted
+        EXECUTE to run the function.
+       </entry>
+      </row>
       <row>
        <entry>
         <literal><function>pg_ls_logdir()</function></literal>
@@ -25400,6 +25411,14 @@ postgres=# SELECT * FROM pg_walfile_name_offset(pg_stop_backup());
     empty directory from an non-existent directory.
    </para>
 
+   <indexterm>
+    <primary>pg_ls_dir_metadata</primary>
+   </indexterm>
+   <para>
+    <function>pg_ls_dir_metadata</function> lists the files in the specified
+    directory along with the file's metadata.
+   </para>
+
    <indexterm>
     <primary>pg_ls_logdir</primary>
    </indexterm>
diff --git a/src/backend/catalog/system_views.sql b/src/backend/catalog/system_views.sql
index 2bd5f5ea14..1c77430f0c 100644
--- a/src/backend/catalog/system_views.sql
+++ b/src/backend/catalog/system_views.sql
@@ -1467,6 +1467,7 @@ REVOKE EXECUTE ON FUNCTION pg_stat_file(text,boolean) FROM public;
 
 REVOKE EXECUTE ON FUNCTION pg_ls_dir(text) FROM public;
 REVOKE EXECUTE ON FUNCTION pg_ls_dir(text,boolean,boolean) FROM public;
+REVOKE EXECUTE ON FUNCTION pg_ls_dir_metadata(text,boolean,boolean) FROM public;
 
 --
 -- We also set up some things as accessible to standard roles.
diff --git a/src/backend/utils/adt/genfile.c b/src/backend/utils/adt/genfile.c
index 219ac160f8..4824a55480 100644
--- a/src/backend/utils/adt/genfile.c
+++ b/src/backend/utils/adt/genfile.c
@@ -36,6 +36,21 @@
 #include "utils/syscache.h"
 #include "utils/timestamp.h"
 
+static Datum pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, int flags);
+
+#define	LS_DIR_ISDIR				(1<<0) /* Show column: isdir */
+#define	LS_DIR_METADATA				(1<<1) /* Show columns: mtime, size */
+#define	LS_DIR_MISSING_OK			(1<<2) /* Ignore ENOENT if the toplevel dir is missing */
+#define	LS_DIR_SKIP_DOT_DIRS		(1<<3) /* Do not show . or .. */
+#define	LS_DIR_SKIP_HIDDEN			(1<<4) /* Do not show anything begining with . */
+#define	LS_DIR_SKIP_DIRS			(1<<5) /* Do not show directories */
+#define	LS_DIR_SKIP_SPECIAL			(1<<6) /* Do not show special file types */
+
+/*
+ * Shortcut for the historic behavior of the pg_ls_* functions (not including
+ * pg_ls_dir, which skips different files and doesn't show metadata.
+ */
+#define LS_DIR_HISTORIC				(LS_DIR_SKIP_DIRS|LS_DIR_SKIP_HIDDEN|LS_DIR_SKIP_SPECIAL|LS_DIR_METADATA)
 
 /*
  * Convert a "text" filename argument to C string, and check it's allowable.
@@ -413,6 +428,11 @@ pg_stat_file(PG_FUNCTION_ARGS)
 	values[4] = TimestampTzGetDatum(time_t_to_timestamptz(fst.st_ctime));
 #endif
 	values[5] = BoolGetDatum(S_ISDIR(fst.st_mode));
+#ifdef WIN32
+	/* Links should have isdir=false */
+	if (pgwin32_is_junction(filename))
+		values[5] = BoolGetDatum(false);
+#endif
 
 	tuple = heap_form_tuple(tupdesc, values, isnull);
 
@@ -440,79 +460,9 @@ pg_stat_file_1arg(PG_FUNCTION_ARGS)
 Datum
 pg_ls_dir(PG_FUNCTION_ARGS)
 {
-	ReturnSetInfo *rsinfo = (ReturnSetInfo *) fcinfo->resultinfo;
-	char	   *location;
-	bool		missing_ok = false;
-	bool		include_dot_dirs = false;
-	bool		randomAccess;
-	TupleDesc	tupdesc;
-	Tuplestorestate *tupstore;
-	DIR		   *dirdesc;
-	struct dirent *de;
-	MemoryContext oldcontext;
-
-	location = convert_and_check_filename(PG_GETARG_TEXT_PP(0));
-
-	/* check the optional arguments */
-	if (PG_NARGS() == 3)
-	{
-		if (!PG_ARGISNULL(1))
-			missing_ok = PG_GETARG_BOOL(1);
-		if (!PG_ARGISNULL(2))
-			include_dot_dirs = PG_GETARG_BOOL(2);
-	}
-
-	/* check to see if caller supports us returning a tuplestore */
-	if (rsinfo == NULL || !IsA(rsinfo, ReturnSetInfo))
-		ereport(ERROR,
-				(errcode(ERRCODE_FEATURE_NOT_SUPPORTED),
-				 errmsg("set-valued function called in context that cannot accept a set")));
-	if (!(rsinfo->allowedModes & SFRM_Materialize))
-		ereport(ERROR,
-				(errcode(ERRCODE_SYNTAX_ERROR),
-				 errmsg("materialize mode required, but it is not allowed in this context")));
-
-	/* The tupdesc and tuplestore must be created in ecxt_per_query_memory */
-	oldcontext = MemoryContextSwitchTo(rsinfo->econtext->ecxt_per_query_memory);
-
-	tupdesc = CreateTemplateTupleDesc(1);
-	TupleDescInitEntry(tupdesc, (AttrNumber) 1, "pg_ls_dir", TEXTOID, -1, 0);
-
-	randomAccess = (rsinfo->allowedModes & SFRM_Materialize_Random) != 0;
-	tupstore = tuplestore_begin_heap(randomAccess, false, work_mem);
-	rsinfo->returnMode = SFRM_Materialize;
-	rsinfo->setResult = tupstore;
-	rsinfo->setDesc = tupdesc;
-
-	MemoryContextSwitchTo(oldcontext);
-
-	dirdesc = AllocateDir(location);
-	if (!dirdesc)
-	{
-		/* Return empty tuplestore if appropriate */
-		if (missing_ok && errno == ENOENT)
-			return (Datum) 0;
-		/* Otherwise, we can let ReadDir() throw the error */
-	}
-
-	while ((de = ReadDir(dirdesc, location)) != NULL)
-	{
-		Datum		values[1];
-		bool		nulls[1];
-
-		if (!include_dot_dirs &&
-			(strcmp(de->d_name, ".") == 0 ||
-			 strcmp(de->d_name, "..") == 0))
-			continue;
-
-		values[0] = CStringGetTextDatum(de->d_name);
-		nulls[0] = false;
-
-		tuplestore_putvalues(tupstore, tupdesc, values, nulls);
-	}
-
-	FreeDir(dirdesc);
-	return (Datum) 0;
+	text	*filename_t = PG_GETARG_TEXT_PP(0);
+	char	*filename = convert_and_check_filename(filename_t);
+	return pg_ls_dir_files(fcinfo, filename, LS_DIR_SKIP_DOT_DIRS);
 }
 
 /*
@@ -525,7 +475,9 @@ pg_ls_dir(PG_FUNCTION_ARGS)
 Datum
 pg_ls_dir_1arg(PG_FUNCTION_ARGS)
 {
-	return pg_ls_dir(fcinfo);
+	text	*filename_t = PG_GETARG_TEXT_PP(0);
+	char	*filename = convert_and_check_filename(filename_t);
+	return pg_ls_dir_files(fcinfo, filename, LS_DIR_SKIP_DOT_DIRS);
 }
 
 /*
@@ -535,7 +487,7 @@ pg_ls_dir_1arg(PG_FUNCTION_ARGS)
  * Other unreadable-directory cases throw an error.
  */
 static Datum
-pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, bool missing_ok)
+pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, int flags)
 {
 	ReturnSetInfo *rsinfo = (ReturnSetInfo *) fcinfo->resultinfo;
 	bool		randomAccess;
@@ -544,6 +496,32 @@ pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, bool missing_ok)
 	DIR		   *dirdesc;
 	struct dirent *de;
 	MemoryContext oldcontext;
+	TypeFuncClass	tuptype ;
+
+	/* isdir depends on metadata */
+	Assert(!(flags&LS_DIR_ISDIR) || (flags&LS_DIR_METADATA));
+	/* Unreasonable to show isdir and skip dirs */
+	Assert(!(flags&LS_DIR_ISDIR) || !(flags&LS_DIR_SKIP_DIRS));
+
+	/* check the optional arguments */
+	if (PG_NARGS() == 3)
+	{
+		if (!PG_ARGISNULL(1))
+		{
+			if (PG_GETARG_BOOL(1))
+				flags |= LS_DIR_MISSING_OK;
+			else
+				flags &= ~LS_DIR_MISSING_OK;
+		}
+
+		if (!PG_ARGISNULL(2))
+		{
+			if (PG_GETARG_BOOL(2))
+				flags &= ~LS_DIR_SKIP_DOT_DIRS;
+			else
+				flags |= LS_DIR_SKIP_DOT_DIRS;
+		}
+	}
 
 	/* check to see if caller supports us returning a tuplestore */
 	if (rsinfo == NULL || !IsA(rsinfo, ReturnSetInfo))
@@ -558,8 +536,20 @@ pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, bool missing_ok)
 	/* The tupdesc and tuplestore must be created in ecxt_per_query_memory */
 	oldcontext = MemoryContextSwitchTo(rsinfo->econtext->ecxt_per_query_memory);
 
-	if (get_call_result_type(fcinfo, NULL, &tupdesc) != TYPEFUNC_COMPOSITE)
-		elog(ERROR, "return type must be a row type");
+	tuptype = get_call_result_type(fcinfo, NULL, &tupdesc);
+	if (flags & LS_DIR_METADATA)
+	{
+		if (tuptype != TYPEFUNC_COMPOSITE)
+			elog(ERROR, "return type must be a row type");
+	}
+	else
+	{
+		/* pg_ls_dir returns a simple scalar */
+		if (tuptype != TYPEFUNC_SCALAR)
+			elog(ERROR, "return type must be a scalar type");
+		tupdesc = CreateTemplateTupleDesc(1);
+		TupleDescInitEntry(tupdesc, (AttrNumber) 1, "column", TEXTOID, -1, 0);
+	}
 
 	randomAccess = (rsinfo->allowedModes & SFRM_Materialize_Random) != 0;
 	tupstore = tuplestore_begin_heap(randomAccess, false, work_mem);
@@ -578,20 +568,27 @@ pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, bool missing_ok)
 	if (!dirdesc)
 	{
 		/* Return empty tuplestore if appropriate */
-		if (missing_ok && errno == ENOENT)
+		if (flags & LS_DIR_MISSING_OK && errno == ENOENT)
 			return (Datum) 0;
 		/* Otherwise, we can let ReadDir() throw the error */
 	}
 
 	while ((de = ReadDir(dirdesc, dir)) != NULL)
 	{
-		Datum		values[3];
-		bool		nulls[3];
+		Datum		values[4];
+		bool		nulls[4];
 		char		path[MAXPGPATH * 2];
 		struct stat attrib;
 
-		/* Skip hidden files */
-		if (de->d_name[0] == '.')
+		/* Skip dot dirs? */
+		if (flags & LS_DIR_SKIP_DOT_DIRS &&
+			(strcmp(de->d_name, ".") == 0 ||
+			 strcmp(de->d_name, "..") == 0))
+			continue;
+
+		/* Skip hidden files? */
+		if (flags & LS_DIR_SKIP_HIDDEN &&
+			de->d_name[0] == '.')
 			continue;
 
 		/* Get the file info */
@@ -606,13 +603,34 @@ pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, bool missing_ok)
 					 errmsg("could not stat file \"%s\": %m", path)));
 		}
 
-		/* Ignore anything but regular files */
-		if (!S_ISREG(attrib.st_mode))
-			continue;
+		/* Skip dirs or special files? */
+		if (S_ISDIR(attrib.st_mode))
+		{
+			if (flags & LS_DIR_SKIP_DIRS)
+				continue;
+		}
+		else if (!S_ISREG(attrib.st_mode))
+		{
+			if (flags & LS_DIR_SKIP_SPECIAL)
+				continue;
+		}
 
 		values[0] = CStringGetTextDatum(de->d_name);
-		values[1] = Int64GetDatum((int64) attrib.st_size);
-		values[2] = TimestampTzGetDatum(time_t_to_timestamptz(attrib.st_mtime));
+		if (flags & LS_DIR_METADATA)
+		{
+			values[1] = Int64GetDatum((int64) attrib.st_size);
+			values[2] = TimestampTzGetDatum(time_t_to_timestamptz(attrib.st_mtime));
+			if (flags & LS_DIR_ISDIR)
+			{
+				values[3] = BoolGetDatum(S_ISDIR(attrib.st_mode));
+#ifdef WIN32
+				/* Links should have isdir=false */
+				if (pgwin32_is_junction(path))
+					values[3] = BoolGetDatum(false);
+#endif
+			}
+		}
+
 		memset(nulls, 0, sizeof(nulls));
 
 		tuplestore_putvalues(tupstore, tupdesc, values, nulls);
@@ -626,14 +644,14 @@ pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, bool missing_ok)
 Datum
 pg_ls_logdir(PG_FUNCTION_ARGS)
 {
-	return pg_ls_dir_files(fcinfo, Log_directory, false);
+	return pg_ls_dir_files(fcinfo, Log_directory, LS_DIR_HISTORIC);
 }
 
 /* Function to return the list of files in the WAL directory */
 Datum
 pg_ls_waldir(PG_FUNCTION_ARGS)
 {
-	return pg_ls_dir_files(fcinfo, XLOGDIR, false);
+	return pg_ls_dir_files(fcinfo, XLOGDIR, LS_DIR_HISTORIC);
 }
 
 /*
@@ -651,7 +669,8 @@ pg_ls_tmpdir(FunctionCallInfo fcinfo, Oid tblspc)
 						tblspc)));
 
 	TempTablespacePath(path, tblspc);
-	return pg_ls_dir_files(fcinfo, path, true);
+	return pg_ls_dir_files(fcinfo, path,
+			LS_DIR_HISTORIC | LS_DIR_MISSING_OK);
 }
 
 /*
@@ -680,5 +699,33 @@ pg_ls_tmpdir_1arg(PG_FUNCTION_ARGS)
 Datum
 pg_ls_archive_statusdir(PG_FUNCTION_ARGS)
 {
-	return pg_ls_dir_files(fcinfo, XLOGDIR "/archive_status", true);
+	return pg_ls_dir_files(fcinfo, XLOGDIR "/archive_status",
+			LS_DIR_HISTORIC | LS_DIR_MISSING_OK);
+}
+
+/*
+ * Function to return the list of files and metadata in an arbitrary directory.
+ */
+Datum
+pg_ls_dir_metadata(PG_FUNCTION_ARGS)
+{
+	char	*dirname = convert_and_check_filename(PG_GETARG_TEXT_PP(0));
+
+	return pg_ls_dir_files(fcinfo, dirname,
+			LS_DIR_METADATA | LS_DIR_SKIP_SPECIAL | LS_DIR_ISDIR);
+}
+
+/*
+ * Function to return the list of files and metadata in an arbitrary directory.
+ * note: this wrapper is necessary to pass the sanity check in opr_sanity,
+ * which checks that all built-in functions that share the implementing C
+ * function take the same number of arguments.
+ */
+Datum
+pg_ls_dir_metadata_1arg(PG_FUNCTION_ARGS)
+{
+	char	*dirname = convert_and_check_filename(PG_GETARG_TEXT_PP(0));
+
+	return pg_ls_dir_files(fcinfo, dirname,
+			LS_DIR_METADATA | LS_DIR_SKIP_SPECIAL | LS_DIR_ISDIR);
 }
diff --git a/src/include/catalog/pg_proc.dat b/src/include/catalog/pg_proc.dat
index 4bce3ad8de..9f46cba5ed 100644
--- a/src/include/catalog/pg_proc.dat
+++ b/src/include/catalog/pg_proc.dat
@@ -10904,6 +10904,18 @@
   proallargtypes => '{oid,text,int8,timestamptz}', proargmodes => '{i,o,o,o}',
   proargnames => '{tablespace,name,size,modification}',
   prosrc => 'pg_ls_tmpdir_1arg' },
+{ oid => '5032', descr => 'list directory with metadata',
+  proname => 'pg_ls_dir_metadata', procost => '10', prorows => '20', proretset => 't',
+  provolatile => 'v', prorettype => 'record', proargtypes => 'text bool bool',
+  proallargtypes => '{text,bool,bool,text,int8,timestamptz,bool}', proargmodes => '{i,i,i,o,o,o,o}',
+  proargnames => '{dirname,missing_ok,include_dot_dirs,name,size,modification,isdir}',
+  prosrc => 'pg_ls_dir_metadata' },
+{ oid => '5033', descr => 'list directory with metadata',
+  proname => 'pg_ls_dir_metadata', procost => '10', prorows => '20', proretset => 't',
+  provolatile => 'v', prorettype => 'record', proargtypes => 'text',
+  proallargtypes => '{text,text,int8,timestamptz,bool}', proargmodes => '{i,o,o,o,o}',
+  proargnames => '{dirname,name,size,modification,isdir}',
+  prosrc => 'pg_ls_dir_metadata_1arg' },
 
 # hash partitioning constraint function
 { oid => '5028', descr => 'hash partition CHECK constraint',
diff --git a/src/test/regress/expected/misc_functions.out b/src/test/regress/expected/misc_functions.out
index 2e87c548eb..7930909f02 100644
--- a/src/test/regress/expected/misc_functions.out
+++ b/src/test/regress/expected/misc_functions.out
@@ -219,6 +219,30 @@ select pg_ls_dir('does not exist', true, false); -- ok with missingok=true
 
 select pg_ls_dir('does not exist'); -- fails with missingok=false
 ERROR:  could not open directory "does not exist": No such file or directory
+-- This tests the missing_ok parameter, which causes pg_ls_tmpdir to succeed even if the tmpdir doesn't exist yet
+-- The name='' condition is never true, so the function runs to completion but returns zero rows.
+select * from pg_ls_tmpdir() where name='Does not exist';
+ name | size | modification 
+------+------+--------------
+(0 rows)
+
+select name, isdir from pg_ls_dir_metadata('.') where name='.';
+ name | isdir 
+------+-------
+ .    | t
+(1 row)
+
+select name, isdir from pg_ls_dir_metadata('.', false, false) where name='.'; -- include_dot_dirs=false
+ name | isdir 
+------+-------
+(0 rows)
+
+-- Check that expected columns are present
+select * from pg_ls_dir_metadata('.') limit 0;
+ name | size | modification | isdir 
+------+------+--------------+-------
+(0 rows)
+
 --
 -- Test adding a support function to a subject function
 --
diff --git a/src/test/regress/input/tablespace.source b/src/test/regress/input/tablespace.source
index a5f61a35dc..0b9cfe615e 100644
--- a/src/test/regress/input/tablespace.source
+++ b/src/test/regress/input/tablespace.source
@@ -11,6 +11,11 @@ DROP TABLESPACE regress_tblspacewith;
 -- create a tablespace we can use
 CREATE TABLESPACE regress_tblspace LOCATION '@testtablespace@';
 
+-- This tests the missing_ok parameter, which causes pg_ls_tmpdir to succeed even if the tmpdir doesn't exist yet
+-- The name='' condition is never true, so the function runs to completion but returns zero rows.
+-- The query is written to ERROR if the tablespace doesn't exist, rather than silently failing to call pg_ls_tmpdir()
+SELECT c.* FROM (SELECT oid FROM pg_tablespace b WHERE b.spcname='regress_tblspace' UNION SELECT 0 ORDER BY 1 DESC LIMIT 1) AS b , pg_ls_tmpdir(oid) AS c WHERE c.name='Does not exist';
+
 -- try setting and resetting some properties for the new tablespace
 ALTER TABLESPACE regress_tblspace SET (random_page_cost = 1.0, seq_page_cost = 1.1);
 ALTER TABLESPACE regress_tblspace SET (some_nonexistent_parameter = true);  -- fail
diff --git a/src/test/regress/output/tablespace.source b/src/test/regress/output/tablespace.source
index 162b591b31..a42714bf40 100644
--- a/src/test/regress/output/tablespace.source
+++ b/src/test/regress/output/tablespace.source
@@ -13,6 +13,14 @@ SELECT spcoptions FROM pg_tablespace WHERE spcname = 'regress_tblspacewith';
 DROP TABLESPACE regress_tblspacewith;
 -- create a tablespace we can use
 CREATE TABLESPACE regress_tblspace LOCATION '@testtablespace@';
+-- This tests the missing_ok parameter, which causes pg_ls_tmpdir to succeed even if the tmpdir doesn't exist yet
+-- The name='' condition is never true, so the function runs to completion but returns zero rows.
+-- The query is written to ERROR if the tablespace doesn't exist, rather than silently failing to call pg_ls_tmpdir()
+SELECT c.* FROM (SELECT oid FROM pg_tablespace b WHERE b.spcname='regress_tblspace' UNION SELECT 0 ORDER BY 1 DESC LIMIT 1) AS b , pg_ls_tmpdir(oid) AS c WHERE c.name='Does not exist';
+ name | size | modification 
+------+------+--------------
+(0 rows)
+
 -- try setting and resetting some properties for the new tablespace
 ALTER TABLESPACE regress_tblspace SET (random_page_cost = 1.0, seq_page_cost = 1.1);
 ALTER TABLESPACE regress_tblspace SET (some_nonexistent_parameter = true);  -- fail
diff --git a/src/test/regress/sql/misc_functions.sql b/src/test/regress/sql/misc_functions.sql
index f6857ad177..372345720d 100644
--- a/src/test/regress/sql/misc_functions.sql
+++ b/src/test/regress/sql/misc_functions.sql
@@ -65,6 +65,17 @@ select * from (select pg_ls_dir('.', false, false) as name) as ls where ls.name=
 select pg_ls_dir('does not exist', true, false); -- ok with missingok=true
 select pg_ls_dir('does not exist'); -- fails with missingok=false
 
+-- This tests the missing_ok parameter, which causes pg_ls_tmpdir to succeed even if the tmpdir doesn't exist yet
+-- The name='' condition is never true, so the function runs to completion but returns zero rows.
+select * from pg_ls_tmpdir() where name='Does not exist';
+
+select name, isdir from pg_ls_dir_metadata('.') where name='.';
+
+select name, isdir from pg_ls_dir_metadata('.', false, false) where name='.'; -- include_dot_dirs=false
+
+-- Check that expected columns are present
+select * from pg_ls_dir_metadata('.') limit 0;
+
 --
 -- Test adding a support function to a subject function
 --
-- 
2.17.0


--2FkSFaIQeDFoAt0B
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
 filename="v16-0005-pg_ls_tmpdir-to-show-directories-and-isdir-argum.patch"



^ permalink  raw  reply  [nested|flat] 68+ messages in thread

* [PATCH v17 04/10] Add pg_ls_dir_metadata to list a dir with file metadata..
@ 2020-03-10 03:40  Justin Pryzby <[email protected]>
  0 siblings, 0 replies; 68+ messages in thread

From: Justin Pryzby @ 2020-03-10 03:40 UTC (permalink / raw)

Generalize pg_ls_dir_files and retire pg_ls_dir

Need catversion bumped?
---
 doc/src/sgml/func.sgml                       |  21 ++
 src/backend/catalog/system_views.sql         |   1 +
 src/backend/utils/adt/genfile.c              | 229 +++++++++++--------
 src/include/catalog/pg_proc.dat              |  12 +
 src/test/regress/expected/misc_functions.out |  24 ++
 src/test/regress/input/tablespace.source     |   5 +
 src/test/regress/output/tablespace.source    |   8 +
 src/test/regress/sql/misc_functions.sql      |  11 +
 8 files changed, 220 insertions(+), 91 deletions(-)

diff --git a/doc/src/sgml/func.sgml b/doc/src/sgml/func.sgml
index d9b3598977..fc1b4ac98c 100644
--- a/doc/src/sgml/func.sgml
+++ b/doc/src/sgml/func.sgml
@@ -25714,6 +25714,27 @@ SELECT pg_size_pretty(sum(pg_relation_size(relid))) AS total_size
        </para></entry>
       </row>
 
+      <row>
+       <entry role="func_table_entry"><para role="func_signature">
+        <indexterm>
+         <primary>pg_ls_dir_metadata</primary>
+        </indexterm>
+        <function>pg_ls_dir_metadata</function> ( <parameter>dirname</parameter> <type>text</type>
+        <optional>, <parameter>missing_ok</parameter> <type>boolean</type>,
+        <parameter>include_dot_dirs</parameter> <type>boolean</type>] </optional> )
+        <returnvalue>setof record</returnvalue>
+        ( <parameter>name</parameter> <type>text</type>,
+        <parameter>size</parameter> <type>bigint</type>,
+        <parameter>modification</parameter> <type>timestamp with time zone</type> )
+       </para>
+       <para>
+        For each file in the specified directory, list the file and its
+        metadata.
+        Restricted to superusers by default, but other users can be granted
+        EXECUTE to run the function.
+       </para></entry>
+      </row>
+
       <row>
        <entry role="func_table_entry"><para role="func_signature">
         <indexterm>
diff --git a/src/backend/catalog/system_views.sql b/src/backend/catalog/system_views.sql
index 2bd5f5ea14..1c77430f0c 100644
--- a/src/backend/catalog/system_views.sql
+++ b/src/backend/catalog/system_views.sql
@@ -1467,6 +1467,7 @@ REVOKE EXECUTE ON FUNCTION pg_stat_file(text,boolean) FROM public;
 
 REVOKE EXECUTE ON FUNCTION pg_ls_dir(text) FROM public;
 REVOKE EXECUTE ON FUNCTION pg_ls_dir(text,boolean,boolean) FROM public;
+REVOKE EXECUTE ON FUNCTION pg_ls_dir_metadata(text,boolean,boolean) FROM public;
 
 --
 -- We also set up some things as accessible to standard roles.
diff --git a/src/backend/utils/adt/genfile.c b/src/backend/utils/adt/genfile.c
index 219ac160f8..4824a55480 100644
--- a/src/backend/utils/adt/genfile.c
+++ b/src/backend/utils/adt/genfile.c
@@ -36,6 +36,21 @@
 #include "utils/syscache.h"
 #include "utils/timestamp.h"
 
+static Datum pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, int flags);
+
+#define	LS_DIR_ISDIR				(1<<0) /* Show column: isdir */
+#define	LS_DIR_METADATA				(1<<1) /* Show columns: mtime, size */
+#define	LS_DIR_MISSING_OK			(1<<2) /* Ignore ENOENT if the toplevel dir is missing */
+#define	LS_DIR_SKIP_DOT_DIRS		(1<<3) /* Do not show . or .. */
+#define	LS_DIR_SKIP_HIDDEN			(1<<4) /* Do not show anything begining with . */
+#define	LS_DIR_SKIP_DIRS			(1<<5) /* Do not show directories */
+#define	LS_DIR_SKIP_SPECIAL			(1<<6) /* Do not show special file types */
+
+/*
+ * Shortcut for the historic behavior of the pg_ls_* functions (not including
+ * pg_ls_dir, which skips different files and doesn't show metadata.
+ */
+#define LS_DIR_HISTORIC				(LS_DIR_SKIP_DIRS|LS_DIR_SKIP_HIDDEN|LS_DIR_SKIP_SPECIAL|LS_DIR_METADATA)
 
 /*
  * Convert a "text" filename argument to C string, and check it's allowable.
@@ -413,6 +428,11 @@ pg_stat_file(PG_FUNCTION_ARGS)
 	values[4] = TimestampTzGetDatum(time_t_to_timestamptz(fst.st_ctime));
 #endif
 	values[5] = BoolGetDatum(S_ISDIR(fst.st_mode));
+#ifdef WIN32
+	/* Links should have isdir=false */
+	if (pgwin32_is_junction(filename))
+		values[5] = BoolGetDatum(false);
+#endif
 
 	tuple = heap_form_tuple(tupdesc, values, isnull);
 
@@ -440,79 +460,9 @@ pg_stat_file_1arg(PG_FUNCTION_ARGS)
 Datum
 pg_ls_dir(PG_FUNCTION_ARGS)
 {
-	ReturnSetInfo *rsinfo = (ReturnSetInfo *) fcinfo->resultinfo;
-	char	   *location;
-	bool		missing_ok = false;
-	bool		include_dot_dirs = false;
-	bool		randomAccess;
-	TupleDesc	tupdesc;
-	Tuplestorestate *tupstore;
-	DIR		   *dirdesc;
-	struct dirent *de;
-	MemoryContext oldcontext;
-
-	location = convert_and_check_filename(PG_GETARG_TEXT_PP(0));
-
-	/* check the optional arguments */
-	if (PG_NARGS() == 3)
-	{
-		if (!PG_ARGISNULL(1))
-			missing_ok = PG_GETARG_BOOL(1);
-		if (!PG_ARGISNULL(2))
-			include_dot_dirs = PG_GETARG_BOOL(2);
-	}
-
-	/* check to see if caller supports us returning a tuplestore */
-	if (rsinfo == NULL || !IsA(rsinfo, ReturnSetInfo))
-		ereport(ERROR,
-				(errcode(ERRCODE_FEATURE_NOT_SUPPORTED),
-				 errmsg("set-valued function called in context that cannot accept a set")));
-	if (!(rsinfo->allowedModes & SFRM_Materialize))
-		ereport(ERROR,
-				(errcode(ERRCODE_SYNTAX_ERROR),
-				 errmsg("materialize mode required, but it is not allowed in this context")));
-
-	/* The tupdesc and tuplestore must be created in ecxt_per_query_memory */
-	oldcontext = MemoryContextSwitchTo(rsinfo->econtext->ecxt_per_query_memory);
-
-	tupdesc = CreateTemplateTupleDesc(1);
-	TupleDescInitEntry(tupdesc, (AttrNumber) 1, "pg_ls_dir", TEXTOID, -1, 0);
-
-	randomAccess = (rsinfo->allowedModes & SFRM_Materialize_Random) != 0;
-	tupstore = tuplestore_begin_heap(randomAccess, false, work_mem);
-	rsinfo->returnMode = SFRM_Materialize;
-	rsinfo->setResult = tupstore;
-	rsinfo->setDesc = tupdesc;
-
-	MemoryContextSwitchTo(oldcontext);
-
-	dirdesc = AllocateDir(location);
-	if (!dirdesc)
-	{
-		/* Return empty tuplestore if appropriate */
-		if (missing_ok && errno == ENOENT)
-			return (Datum) 0;
-		/* Otherwise, we can let ReadDir() throw the error */
-	}
-
-	while ((de = ReadDir(dirdesc, location)) != NULL)
-	{
-		Datum		values[1];
-		bool		nulls[1];
-
-		if (!include_dot_dirs &&
-			(strcmp(de->d_name, ".") == 0 ||
-			 strcmp(de->d_name, "..") == 0))
-			continue;
-
-		values[0] = CStringGetTextDatum(de->d_name);
-		nulls[0] = false;
-
-		tuplestore_putvalues(tupstore, tupdesc, values, nulls);
-	}
-
-	FreeDir(dirdesc);
-	return (Datum) 0;
+	text	*filename_t = PG_GETARG_TEXT_PP(0);
+	char	*filename = convert_and_check_filename(filename_t);
+	return pg_ls_dir_files(fcinfo, filename, LS_DIR_SKIP_DOT_DIRS);
 }
 
 /*
@@ -525,7 +475,9 @@ pg_ls_dir(PG_FUNCTION_ARGS)
 Datum
 pg_ls_dir_1arg(PG_FUNCTION_ARGS)
 {
-	return pg_ls_dir(fcinfo);
+	text	*filename_t = PG_GETARG_TEXT_PP(0);
+	char	*filename = convert_and_check_filename(filename_t);
+	return pg_ls_dir_files(fcinfo, filename, LS_DIR_SKIP_DOT_DIRS);
 }
 
 /*
@@ -535,7 +487,7 @@ pg_ls_dir_1arg(PG_FUNCTION_ARGS)
  * Other unreadable-directory cases throw an error.
  */
 static Datum
-pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, bool missing_ok)
+pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, int flags)
 {
 	ReturnSetInfo *rsinfo = (ReturnSetInfo *) fcinfo->resultinfo;
 	bool		randomAccess;
@@ -544,6 +496,32 @@ pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, bool missing_ok)
 	DIR		   *dirdesc;
 	struct dirent *de;
 	MemoryContext oldcontext;
+	TypeFuncClass	tuptype ;
+
+	/* isdir depends on metadata */
+	Assert(!(flags&LS_DIR_ISDIR) || (flags&LS_DIR_METADATA));
+	/* Unreasonable to show isdir and skip dirs */
+	Assert(!(flags&LS_DIR_ISDIR) || !(flags&LS_DIR_SKIP_DIRS));
+
+	/* check the optional arguments */
+	if (PG_NARGS() == 3)
+	{
+		if (!PG_ARGISNULL(1))
+		{
+			if (PG_GETARG_BOOL(1))
+				flags |= LS_DIR_MISSING_OK;
+			else
+				flags &= ~LS_DIR_MISSING_OK;
+		}
+
+		if (!PG_ARGISNULL(2))
+		{
+			if (PG_GETARG_BOOL(2))
+				flags &= ~LS_DIR_SKIP_DOT_DIRS;
+			else
+				flags |= LS_DIR_SKIP_DOT_DIRS;
+		}
+	}
 
 	/* check to see if caller supports us returning a tuplestore */
 	if (rsinfo == NULL || !IsA(rsinfo, ReturnSetInfo))
@@ -558,8 +536,20 @@ pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, bool missing_ok)
 	/* The tupdesc and tuplestore must be created in ecxt_per_query_memory */
 	oldcontext = MemoryContextSwitchTo(rsinfo->econtext->ecxt_per_query_memory);
 
-	if (get_call_result_type(fcinfo, NULL, &tupdesc) != TYPEFUNC_COMPOSITE)
-		elog(ERROR, "return type must be a row type");
+	tuptype = get_call_result_type(fcinfo, NULL, &tupdesc);
+	if (flags & LS_DIR_METADATA)
+	{
+		if (tuptype != TYPEFUNC_COMPOSITE)
+			elog(ERROR, "return type must be a row type");
+	}
+	else
+	{
+		/* pg_ls_dir returns a simple scalar */
+		if (tuptype != TYPEFUNC_SCALAR)
+			elog(ERROR, "return type must be a scalar type");
+		tupdesc = CreateTemplateTupleDesc(1);
+		TupleDescInitEntry(tupdesc, (AttrNumber) 1, "column", TEXTOID, -1, 0);
+	}
 
 	randomAccess = (rsinfo->allowedModes & SFRM_Materialize_Random) != 0;
 	tupstore = tuplestore_begin_heap(randomAccess, false, work_mem);
@@ -578,20 +568,27 @@ pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, bool missing_ok)
 	if (!dirdesc)
 	{
 		/* Return empty tuplestore if appropriate */
-		if (missing_ok && errno == ENOENT)
+		if (flags & LS_DIR_MISSING_OK && errno == ENOENT)
 			return (Datum) 0;
 		/* Otherwise, we can let ReadDir() throw the error */
 	}
 
 	while ((de = ReadDir(dirdesc, dir)) != NULL)
 	{
-		Datum		values[3];
-		bool		nulls[3];
+		Datum		values[4];
+		bool		nulls[4];
 		char		path[MAXPGPATH * 2];
 		struct stat attrib;
 
-		/* Skip hidden files */
-		if (de->d_name[0] == '.')
+		/* Skip dot dirs? */
+		if (flags & LS_DIR_SKIP_DOT_DIRS &&
+			(strcmp(de->d_name, ".") == 0 ||
+			 strcmp(de->d_name, "..") == 0))
+			continue;
+
+		/* Skip hidden files? */
+		if (flags & LS_DIR_SKIP_HIDDEN &&
+			de->d_name[0] == '.')
 			continue;
 
 		/* Get the file info */
@@ -606,13 +603,34 @@ pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, bool missing_ok)
 					 errmsg("could not stat file \"%s\": %m", path)));
 		}
 
-		/* Ignore anything but regular files */
-		if (!S_ISREG(attrib.st_mode))
-			continue;
+		/* Skip dirs or special files? */
+		if (S_ISDIR(attrib.st_mode))
+		{
+			if (flags & LS_DIR_SKIP_DIRS)
+				continue;
+		}
+		else if (!S_ISREG(attrib.st_mode))
+		{
+			if (flags & LS_DIR_SKIP_SPECIAL)
+				continue;
+		}
 
 		values[0] = CStringGetTextDatum(de->d_name);
-		values[1] = Int64GetDatum((int64) attrib.st_size);
-		values[2] = TimestampTzGetDatum(time_t_to_timestamptz(attrib.st_mtime));
+		if (flags & LS_DIR_METADATA)
+		{
+			values[1] = Int64GetDatum((int64) attrib.st_size);
+			values[2] = TimestampTzGetDatum(time_t_to_timestamptz(attrib.st_mtime));
+			if (flags & LS_DIR_ISDIR)
+			{
+				values[3] = BoolGetDatum(S_ISDIR(attrib.st_mode));
+#ifdef WIN32
+				/* Links should have isdir=false */
+				if (pgwin32_is_junction(path))
+					values[3] = BoolGetDatum(false);
+#endif
+			}
+		}
+
 		memset(nulls, 0, sizeof(nulls));
 
 		tuplestore_putvalues(tupstore, tupdesc, values, nulls);
@@ -626,14 +644,14 @@ pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, bool missing_ok)
 Datum
 pg_ls_logdir(PG_FUNCTION_ARGS)
 {
-	return pg_ls_dir_files(fcinfo, Log_directory, false);
+	return pg_ls_dir_files(fcinfo, Log_directory, LS_DIR_HISTORIC);
 }
 
 /* Function to return the list of files in the WAL directory */
 Datum
 pg_ls_waldir(PG_FUNCTION_ARGS)
 {
-	return pg_ls_dir_files(fcinfo, XLOGDIR, false);
+	return pg_ls_dir_files(fcinfo, XLOGDIR, LS_DIR_HISTORIC);
 }
 
 /*
@@ -651,7 +669,8 @@ pg_ls_tmpdir(FunctionCallInfo fcinfo, Oid tblspc)
 						tblspc)));
 
 	TempTablespacePath(path, tblspc);
-	return pg_ls_dir_files(fcinfo, path, true);
+	return pg_ls_dir_files(fcinfo, path,
+			LS_DIR_HISTORIC | LS_DIR_MISSING_OK);
 }
 
 /*
@@ -680,5 +699,33 @@ pg_ls_tmpdir_1arg(PG_FUNCTION_ARGS)
 Datum
 pg_ls_archive_statusdir(PG_FUNCTION_ARGS)
 {
-	return pg_ls_dir_files(fcinfo, XLOGDIR "/archive_status", true);
+	return pg_ls_dir_files(fcinfo, XLOGDIR "/archive_status",
+			LS_DIR_HISTORIC | LS_DIR_MISSING_OK);
+}
+
+/*
+ * Function to return the list of files and metadata in an arbitrary directory.
+ */
+Datum
+pg_ls_dir_metadata(PG_FUNCTION_ARGS)
+{
+	char	*dirname = convert_and_check_filename(PG_GETARG_TEXT_PP(0));
+
+	return pg_ls_dir_files(fcinfo, dirname,
+			LS_DIR_METADATA | LS_DIR_SKIP_SPECIAL | LS_DIR_ISDIR);
+}
+
+/*
+ * Function to return the list of files and metadata in an arbitrary directory.
+ * note: this wrapper is necessary to pass the sanity check in opr_sanity,
+ * which checks that all built-in functions that share the implementing C
+ * function take the same number of arguments.
+ */
+Datum
+pg_ls_dir_metadata_1arg(PG_FUNCTION_ARGS)
+{
+	char	*dirname = convert_and_check_filename(PG_GETARG_TEXT_PP(0));
+
+	return pg_ls_dir_files(fcinfo, dirname,
+			LS_DIR_METADATA | LS_DIR_SKIP_SPECIAL | LS_DIR_ISDIR);
 }
diff --git a/src/include/catalog/pg_proc.dat b/src/include/catalog/pg_proc.dat
index 4bce3ad8de..9f46cba5ed 100644
--- a/src/include/catalog/pg_proc.dat
+++ b/src/include/catalog/pg_proc.dat
@@ -10904,6 +10904,18 @@
   proallargtypes => '{oid,text,int8,timestamptz}', proargmodes => '{i,o,o,o}',
   proargnames => '{tablespace,name,size,modification}',
   prosrc => 'pg_ls_tmpdir_1arg' },
+{ oid => '5032', descr => 'list directory with metadata',
+  proname => 'pg_ls_dir_metadata', procost => '10', prorows => '20', proretset => 't',
+  provolatile => 'v', prorettype => 'record', proargtypes => 'text bool bool',
+  proallargtypes => '{text,bool,bool,text,int8,timestamptz,bool}', proargmodes => '{i,i,i,o,o,o,o}',
+  proargnames => '{dirname,missing_ok,include_dot_dirs,name,size,modification,isdir}',
+  prosrc => 'pg_ls_dir_metadata' },
+{ oid => '5033', descr => 'list directory with metadata',
+  proname => 'pg_ls_dir_metadata', procost => '10', prorows => '20', proretset => 't',
+  provolatile => 'v', prorettype => 'record', proargtypes => 'text',
+  proallargtypes => '{text,text,int8,timestamptz,bool}', proargmodes => '{i,o,o,o,o}',
+  proargnames => '{dirname,name,size,modification,isdir}',
+  prosrc => 'pg_ls_dir_metadata_1arg' },
 
 # hash partitioning constraint function
 { oid => '5028', descr => 'hash partition CHECK constraint',
diff --git a/src/test/regress/expected/misc_functions.out b/src/test/regress/expected/misc_functions.out
index 2e87c548eb..7930909f02 100644
--- a/src/test/regress/expected/misc_functions.out
+++ b/src/test/regress/expected/misc_functions.out
@@ -219,6 +219,30 @@ select pg_ls_dir('does not exist', true, false); -- ok with missingok=true
 
 select pg_ls_dir('does not exist'); -- fails with missingok=false
 ERROR:  could not open directory "does not exist": No such file or directory
+-- This tests the missing_ok parameter, which causes pg_ls_tmpdir to succeed even if the tmpdir doesn't exist yet
+-- The name='' condition is never true, so the function runs to completion but returns zero rows.
+select * from pg_ls_tmpdir() where name='Does not exist';
+ name | size | modification 
+------+------+--------------
+(0 rows)
+
+select name, isdir from pg_ls_dir_metadata('.') where name='.';
+ name | isdir 
+------+-------
+ .    | t
+(1 row)
+
+select name, isdir from pg_ls_dir_metadata('.', false, false) where name='.'; -- include_dot_dirs=false
+ name | isdir 
+------+-------
+(0 rows)
+
+-- Check that expected columns are present
+select * from pg_ls_dir_metadata('.') limit 0;
+ name | size | modification | isdir 
+------+------+--------------+-------
+(0 rows)
+
 --
 -- Test adding a support function to a subject function
 --
diff --git a/src/test/regress/input/tablespace.source b/src/test/regress/input/tablespace.source
index a5f61a35dc..0b9cfe615e 100644
--- a/src/test/regress/input/tablespace.source
+++ b/src/test/regress/input/tablespace.source
@@ -11,6 +11,11 @@ DROP TABLESPACE regress_tblspacewith;
 -- create a tablespace we can use
 CREATE TABLESPACE regress_tblspace LOCATION '@testtablespace@';
 
+-- This tests the missing_ok parameter, which causes pg_ls_tmpdir to succeed even if the tmpdir doesn't exist yet
+-- The name='' condition is never true, so the function runs to completion but returns zero rows.
+-- The query is written to ERROR if the tablespace doesn't exist, rather than silently failing to call pg_ls_tmpdir()
+SELECT c.* FROM (SELECT oid FROM pg_tablespace b WHERE b.spcname='regress_tblspace' UNION SELECT 0 ORDER BY 1 DESC LIMIT 1) AS b , pg_ls_tmpdir(oid) AS c WHERE c.name='Does not exist';
+
 -- try setting and resetting some properties for the new tablespace
 ALTER TABLESPACE regress_tblspace SET (random_page_cost = 1.0, seq_page_cost = 1.1);
 ALTER TABLESPACE regress_tblspace SET (some_nonexistent_parameter = true);  -- fail
diff --git a/src/test/regress/output/tablespace.source b/src/test/regress/output/tablespace.source
index 162b591b31..a42714bf40 100644
--- a/src/test/regress/output/tablespace.source
+++ b/src/test/regress/output/tablespace.source
@@ -13,6 +13,14 @@ SELECT spcoptions FROM pg_tablespace WHERE spcname = 'regress_tblspacewith';
 DROP TABLESPACE regress_tblspacewith;
 -- create a tablespace we can use
 CREATE TABLESPACE regress_tblspace LOCATION '@testtablespace@';
+-- This tests the missing_ok parameter, which causes pg_ls_tmpdir to succeed even if the tmpdir doesn't exist yet
+-- The name='' condition is never true, so the function runs to completion but returns zero rows.
+-- The query is written to ERROR if the tablespace doesn't exist, rather than silently failing to call pg_ls_tmpdir()
+SELECT c.* FROM (SELECT oid FROM pg_tablespace b WHERE b.spcname='regress_tblspace' UNION SELECT 0 ORDER BY 1 DESC LIMIT 1) AS b , pg_ls_tmpdir(oid) AS c WHERE c.name='Does not exist';
+ name | size | modification 
+------+------+--------------
+(0 rows)
+
 -- try setting and resetting some properties for the new tablespace
 ALTER TABLESPACE regress_tblspace SET (random_page_cost = 1.0, seq_page_cost = 1.1);
 ALTER TABLESPACE regress_tblspace SET (some_nonexistent_parameter = true);  -- fail
diff --git a/src/test/regress/sql/misc_functions.sql b/src/test/regress/sql/misc_functions.sql
index f6857ad177..372345720d 100644
--- a/src/test/regress/sql/misc_functions.sql
+++ b/src/test/regress/sql/misc_functions.sql
@@ -65,6 +65,17 @@ select * from (select pg_ls_dir('.', false, false) as name) as ls where ls.name=
 select pg_ls_dir('does not exist', true, false); -- ok with missingok=true
 select pg_ls_dir('does not exist'); -- fails with missingok=false
 
+-- This tests the missing_ok parameter, which causes pg_ls_tmpdir to succeed even if the tmpdir doesn't exist yet
+-- The name='' condition is never true, so the function runs to completion but returns zero rows.
+select * from pg_ls_tmpdir() where name='Does not exist';
+
+select name, isdir from pg_ls_dir_metadata('.') where name='.';
+
+select name, isdir from pg_ls_dir_metadata('.', false, false) where name='.'; -- include_dot_dirs=false
+
+-- Check that expected columns are present
+select * from pg_ls_dir_metadata('.') limit 0;
+
 --
 -- Test adding a support function to a subject function
 --
-- 
2.17.0


--4LFBTxd4L5NLO6ly
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
 filename="v17-0005-pg_ls_tmpdir-to-show-directories-and-isdir-argum.patch"



^ permalink  raw  reply  [nested|flat] 68+ messages in thread

* [PATCH v16 04/10] Add pg_ls_dir_metadata to list a dir with file metadata..
@ 2020-03-10 03:40  Justin Pryzby <[email protected]>
  0 siblings, 0 replies; 68+ messages in thread

From: Justin Pryzby @ 2020-03-10 03:40 UTC (permalink / raw)

Generalize pg_ls_dir_files and retire pg_ls_dir

Need catversion bumped?
---
 doc/src/sgml/func.sgml                       |  19 ++
 src/backend/catalog/system_views.sql         |   1 +
 src/backend/utils/adt/genfile.c              | 229 +++++++++++--------
 src/include/catalog/pg_proc.dat              |  12 +
 src/test/regress/expected/misc_functions.out |  24 ++
 src/test/regress/input/tablespace.source     |   5 +
 src/test/regress/output/tablespace.source    |   8 +
 src/test/regress/sql/misc_functions.sql      |  11 +
 8 files changed, 218 insertions(+), 91 deletions(-)

diff --git a/doc/src/sgml/func.sgml b/doc/src/sgml/func.sgml
index 96b08d0500..9d617f95b9 100644
--- a/doc/src/sgml/func.sgml
+++ b/doc/src/sgml/func.sgml
@@ -25300,6 +25300,17 @@ postgres=# SELECT * FROM pg_walfile_name_offset(pg_stop_backup());
         List the contents of a directory.  Restricted to superusers by default, but other users can be granted EXECUTE to run the function.
        </entry>
       </row>
+      <row>
+       <entry>
+        <literal><function>pg_ls_dir_metadata(<parameter>dirname</parameter> <type>text</type> [, <parameter>missing_ok</parameter> <type>boolean</type>, <parameter>include_dot_dirs</parameter> <type>boolean</type>])</function></literal>
+       </entry>
+       <entry><type>setof text</type></entry>
+       <entry>
+        For each file in a directory, list the file and its metadata.
+        Restricted to superusers by default, but other users can be granted
+        EXECUTE to run the function.
+       </entry>
+      </row>
       <row>
        <entry>
         <literal><function>pg_ls_logdir()</function></literal>
@@ -25400,6 +25411,14 @@ postgres=# SELECT * FROM pg_walfile_name_offset(pg_stop_backup());
     empty directory from an non-existent directory.
    </para>
 
+   <indexterm>
+    <primary>pg_ls_dir_metadata</primary>
+   </indexterm>
+   <para>
+    <function>pg_ls_dir_metadata</function> lists the files in the specified
+    directory along with the file's metadata.
+   </para>
+
    <indexterm>
     <primary>pg_ls_logdir</primary>
    </indexterm>
diff --git a/src/backend/catalog/system_views.sql b/src/backend/catalog/system_views.sql
index 2bd5f5ea14..1c77430f0c 100644
--- a/src/backend/catalog/system_views.sql
+++ b/src/backend/catalog/system_views.sql
@@ -1467,6 +1467,7 @@ REVOKE EXECUTE ON FUNCTION pg_stat_file(text,boolean) FROM public;
 
 REVOKE EXECUTE ON FUNCTION pg_ls_dir(text) FROM public;
 REVOKE EXECUTE ON FUNCTION pg_ls_dir(text,boolean,boolean) FROM public;
+REVOKE EXECUTE ON FUNCTION pg_ls_dir_metadata(text,boolean,boolean) FROM public;
 
 --
 -- We also set up some things as accessible to standard roles.
diff --git a/src/backend/utils/adt/genfile.c b/src/backend/utils/adt/genfile.c
index 219ac160f8..4824a55480 100644
--- a/src/backend/utils/adt/genfile.c
+++ b/src/backend/utils/adt/genfile.c
@@ -36,6 +36,21 @@
 #include "utils/syscache.h"
 #include "utils/timestamp.h"
 
+static Datum pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, int flags);
+
+#define	LS_DIR_ISDIR				(1<<0) /* Show column: isdir */
+#define	LS_DIR_METADATA				(1<<1) /* Show columns: mtime, size */
+#define	LS_DIR_MISSING_OK			(1<<2) /* Ignore ENOENT if the toplevel dir is missing */
+#define	LS_DIR_SKIP_DOT_DIRS		(1<<3) /* Do not show . or .. */
+#define	LS_DIR_SKIP_HIDDEN			(1<<4) /* Do not show anything begining with . */
+#define	LS_DIR_SKIP_DIRS			(1<<5) /* Do not show directories */
+#define	LS_DIR_SKIP_SPECIAL			(1<<6) /* Do not show special file types */
+
+/*
+ * Shortcut for the historic behavior of the pg_ls_* functions (not including
+ * pg_ls_dir, which skips different files and doesn't show metadata.
+ */
+#define LS_DIR_HISTORIC				(LS_DIR_SKIP_DIRS|LS_DIR_SKIP_HIDDEN|LS_DIR_SKIP_SPECIAL|LS_DIR_METADATA)
 
 /*
  * Convert a "text" filename argument to C string, and check it's allowable.
@@ -413,6 +428,11 @@ pg_stat_file(PG_FUNCTION_ARGS)
 	values[4] = TimestampTzGetDatum(time_t_to_timestamptz(fst.st_ctime));
 #endif
 	values[5] = BoolGetDatum(S_ISDIR(fst.st_mode));
+#ifdef WIN32
+	/* Links should have isdir=false */
+	if (pgwin32_is_junction(filename))
+		values[5] = BoolGetDatum(false);
+#endif
 
 	tuple = heap_form_tuple(tupdesc, values, isnull);
 
@@ -440,79 +460,9 @@ pg_stat_file_1arg(PG_FUNCTION_ARGS)
 Datum
 pg_ls_dir(PG_FUNCTION_ARGS)
 {
-	ReturnSetInfo *rsinfo = (ReturnSetInfo *) fcinfo->resultinfo;
-	char	   *location;
-	bool		missing_ok = false;
-	bool		include_dot_dirs = false;
-	bool		randomAccess;
-	TupleDesc	tupdesc;
-	Tuplestorestate *tupstore;
-	DIR		   *dirdesc;
-	struct dirent *de;
-	MemoryContext oldcontext;
-
-	location = convert_and_check_filename(PG_GETARG_TEXT_PP(0));
-
-	/* check the optional arguments */
-	if (PG_NARGS() == 3)
-	{
-		if (!PG_ARGISNULL(1))
-			missing_ok = PG_GETARG_BOOL(1);
-		if (!PG_ARGISNULL(2))
-			include_dot_dirs = PG_GETARG_BOOL(2);
-	}
-
-	/* check to see if caller supports us returning a tuplestore */
-	if (rsinfo == NULL || !IsA(rsinfo, ReturnSetInfo))
-		ereport(ERROR,
-				(errcode(ERRCODE_FEATURE_NOT_SUPPORTED),
-				 errmsg("set-valued function called in context that cannot accept a set")));
-	if (!(rsinfo->allowedModes & SFRM_Materialize))
-		ereport(ERROR,
-				(errcode(ERRCODE_SYNTAX_ERROR),
-				 errmsg("materialize mode required, but it is not allowed in this context")));
-
-	/* The tupdesc and tuplestore must be created in ecxt_per_query_memory */
-	oldcontext = MemoryContextSwitchTo(rsinfo->econtext->ecxt_per_query_memory);
-
-	tupdesc = CreateTemplateTupleDesc(1);
-	TupleDescInitEntry(tupdesc, (AttrNumber) 1, "pg_ls_dir", TEXTOID, -1, 0);
-
-	randomAccess = (rsinfo->allowedModes & SFRM_Materialize_Random) != 0;
-	tupstore = tuplestore_begin_heap(randomAccess, false, work_mem);
-	rsinfo->returnMode = SFRM_Materialize;
-	rsinfo->setResult = tupstore;
-	rsinfo->setDesc = tupdesc;
-
-	MemoryContextSwitchTo(oldcontext);
-
-	dirdesc = AllocateDir(location);
-	if (!dirdesc)
-	{
-		/* Return empty tuplestore if appropriate */
-		if (missing_ok && errno == ENOENT)
-			return (Datum) 0;
-		/* Otherwise, we can let ReadDir() throw the error */
-	}
-
-	while ((de = ReadDir(dirdesc, location)) != NULL)
-	{
-		Datum		values[1];
-		bool		nulls[1];
-
-		if (!include_dot_dirs &&
-			(strcmp(de->d_name, ".") == 0 ||
-			 strcmp(de->d_name, "..") == 0))
-			continue;
-
-		values[0] = CStringGetTextDatum(de->d_name);
-		nulls[0] = false;
-
-		tuplestore_putvalues(tupstore, tupdesc, values, nulls);
-	}
-
-	FreeDir(dirdesc);
-	return (Datum) 0;
+	text	*filename_t = PG_GETARG_TEXT_PP(0);
+	char	*filename = convert_and_check_filename(filename_t);
+	return pg_ls_dir_files(fcinfo, filename, LS_DIR_SKIP_DOT_DIRS);
 }
 
 /*
@@ -525,7 +475,9 @@ pg_ls_dir(PG_FUNCTION_ARGS)
 Datum
 pg_ls_dir_1arg(PG_FUNCTION_ARGS)
 {
-	return pg_ls_dir(fcinfo);
+	text	*filename_t = PG_GETARG_TEXT_PP(0);
+	char	*filename = convert_and_check_filename(filename_t);
+	return pg_ls_dir_files(fcinfo, filename, LS_DIR_SKIP_DOT_DIRS);
 }
 
 /*
@@ -535,7 +487,7 @@ pg_ls_dir_1arg(PG_FUNCTION_ARGS)
  * Other unreadable-directory cases throw an error.
  */
 static Datum
-pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, bool missing_ok)
+pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, int flags)
 {
 	ReturnSetInfo *rsinfo = (ReturnSetInfo *) fcinfo->resultinfo;
 	bool		randomAccess;
@@ -544,6 +496,32 @@ pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, bool missing_ok)
 	DIR		   *dirdesc;
 	struct dirent *de;
 	MemoryContext oldcontext;
+	TypeFuncClass	tuptype ;
+
+	/* isdir depends on metadata */
+	Assert(!(flags&LS_DIR_ISDIR) || (flags&LS_DIR_METADATA));
+	/* Unreasonable to show isdir and skip dirs */
+	Assert(!(flags&LS_DIR_ISDIR) || !(flags&LS_DIR_SKIP_DIRS));
+
+	/* check the optional arguments */
+	if (PG_NARGS() == 3)
+	{
+		if (!PG_ARGISNULL(1))
+		{
+			if (PG_GETARG_BOOL(1))
+				flags |= LS_DIR_MISSING_OK;
+			else
+				flags &= ~LS_DIR_MISSING_OK;
+		}
+
+		if (!PG_ARGISNULL(2))
+		{
+			if (PG_GETARG_BOOL(2))
+				flags &= ~LS_DIR_SKIP_DOT_DIRS;
+			else
+				flags |= LS_DIR_SKIP_DOT_DIRS;
+		}
+	}
 
 	/* check to see if caller supports us returning a tuplestore */
 	if (rsinfo == NULL || !IsA(rsinfo, ReturnSetInfo))
@@ -558,8 +536,20 @@ pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, bool missing_ok)
 	/* The tupdesc and tuplestore must be created in ecxt_per_query_memory */
 	oldcontext = MemoryContextSwitchTo(rsinfo->econtext->ecxt_per_query_memory);
 
-	if (get_call_result_type(fcinfo, NULL, &tupdesc) != TYPEFUNC_COMPOSITE)
-		elog(ERROR, "return type must be a row type");
+	tuptype = get_call_result_type(fcinfo, NULL, &tupdesc);
+	if (flags & LS_DIR_METADATA)
+	{
+		if (tuptype != TYPEFUNC_COMPOSITE)
+			elog(ERROR, "return type must be a row type");
+	}
+	else
+	{
+		/* pg_ls_dir returns a simple scalar */
+		if (tuptype != TYPEFUNC_SCALAR)
+			elog(ERROR, "return type must be a scalar type");
+		tupdesc = CreateTemplateTupleDesc(1);
+		TupleDescInitEntry(tupdesc, (AttrNumber) 1, "column", TEXTOID, -1, 0);
+	}
 
 	randomAccess = (rsinfo->allowedModes & SFRM_Materialize_Random) != 0;
 	tupstore = tuplestore_begin_heap(randomAccess, false, work_mem);
@@ -578,20 +568,27 @@ pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, bool missing_ok)
 	if (!dirdesc)
 	{
 		/* Return empty tuplestore if appropriate */
-		if (missing_ok && errno == ENOENT)
+		if (flags & LS_DIR_MISSING_OK && errno == ENOENT)
 			return (Datum) 0;
 		/* Otherwise, we can let ReadDir() throw the error */
 	}
 
 	while ((de = ReadDir(dirdesc, dir)) != NULL)
 	{
-		Datum		values[3];
-		bool		nulls[3];
+		Datum		values[4];
+		bool		nulls[4];
 		char		path[MAXPGPATH * 2];
 		struct stat attrib;
 
-		/* Skip hidden files */
-		if (de->d_name[0] == '.')
+		/* Skip dot dirs? */
+		if (flags & LS_DIR_SKIP_DOT_DIRS &&
+			(strcmp(de->d_name, ".") == 0 ||
+			 strcmp(de->d_name, "..") == 0))
+			continue;
+
+		/* Skip hidden files? */
+		if (flags & LS_DIR_SKIP_HIDDEN &&
+			de->d_name[0] == '.')
 			continue;
 
 		/* Get the file info */
@@ -606,13 +603,34 @@ pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, bool missing_ok)
 					 errmsg("could not stat file \"%s\": %m", path)));
 		}
 
-		/* Ignore anything but regular files */
-		if (!S_ISREG(attrib.st_mode))
-			continue;
+		/* Skip dirs or special files? */
+		if (S_ISDIR(attrib.st_mode))
+		{
+			if (flags & LS_DIR_SKIP_DIRS)
+				continue;
+		}
+		else if (!S_ISREG(attrib.st_mode))
+		{
+			if (flags & LS_DIR_SKIP_SPECIAL)
+				continue;
+		}
 
 		values[0] = CStringGetTextDatum(de->d_name);
-		values[1] = Int64GetDatum((int64) attrib.st_size);
-		values[2] = TimestampTzGetDatum(time_t_to_timestamptz(attrib.st_mtime));
+		if (flags & LS_DIR_METADATA)
+		{
+			values[1] = Int64GetDatum((int64) attrib.st_size);
+			values[2] = TimestampTzGetDatum(time_t_to_timestamptz(attrib.st_mtime));
+			if (flags & LS_DIR_ISDIR)
+			{
+				values[3] = BoolGetDatum(S_ISDIR(attrib.st_mode));
+#ifdef WIN32
+				/* Links should have isdir=false */
+				if (pgwin32_is_junction(path))
+					values[3] = BoolGetDatum(false);
+#endif
+			}
+		}
+
 		memset(nulls, 0, sizeof(nulls));
 
 		tuplestore_putvalues(tupstore, tupdesc, values, nulls);
@@ -626,14 +644,14 @@ pg_ls_dir_files(FunctionCallInfo fcinfo, const char *dir, bool missing_ok)
 Datum
 pg_ls_logdir(PG_FUNCTION_ARGS)
 {
-	return pg_ls_dir_files(fcinfo, Log_directory, false);
+	return pg_ls_dir_files(fcinfo, Log_directory, LS_DIR_HISTORIC);
 }
 
 /* Function to return the list of files in the WAL directory */
 Datum
 pg_ls_waldir(PG_FUNCTION_ARGS)
 {
-	return pg_ls_dir_files(fcinfo, XLOGDIR, false);
+	return pg_ls_dir_files(fcinfo, XLOGDIR, LS_DIR_HISTORIC);
 }
 
 /*
@@ -651,7 +669,8 @@ pg_ls_tmpdir(FunctionCallInfo fcinfo, Oid tblspc)
 						tblspc)));
 
 	TempTablespacePath(path, tblspc);
-	return pg_ls_dir_files(fcinfo, path, true);
+	return pg_ls_dir_files(fcinfo, path,
+			LS_DIR_HISTORIC | LS_DIR_MISSING_OK);
 }
 
 /*
@@ -680,5 +699,33 @@ pg_ls_tmpdir_1arg(PG_FUNCTION_ARGS)
 Datum
 pg_ls_archive_statusdir(PG_FUNCTION_ARGS)
 {
-	return pg_ls_dir_files(fcinfo, XLOGDIR "/archive_status", true);
+	return pg_ls_dir_files(fcinfo, XLOGDIR "/archive_status",
+			LS_DIR_HISTORIC | LS_DIR_MISSING_OK);
+}
+
+/*
+ * Function to return the list of files and metadata in an arbitrary directory.
+ */
+Datum
+pg_ls_dir_metadata(PG_FUNCTION_ARGS)
+{
+	char	*dirname = convert_and_check_filename(PG_GETARG_TEXT_PP(0));
+
+	return pg_ls_dir_files(fcinfo, dirname,
+			LS_DIR_METADATA | LS_DIR_SKIP_SPECIAL | LS_DIR_ISDIR);
+}
+
+/*
+ * Function to return the list of files and metadata in an arbitrary directory.
+ * note: this wrapper is necessary to pass the sanity check in opr_sanity,
+ * which checks that all built-in functions that share the implementing C
+ * function take the same number of arguments.
+ */
+Datum
+pg_ls_dir_metadata_1arg(PG_FUNCTION_ARGS)
+{
+	char	*dirname = convert_and_check_filename(PG_GETARG_TEXT_PP(0));
+
+	return pg_ls_dir_files(fcinfo, dirname,
+			LS_DIR_METADATA | LS_DIR_SKIP_SPECIAL | LS_DIR_ISDIR);
 }
diff --git a/src/include/catalog/pg_proc.dat b/src/include/catalog/pg_proc.dat
index 4bce3ad8de..9f46cba5ed 100644
--- a/src/include/catalog/pg_proc.dat
+++ b/src/include/catalog/pg_proc.dat
@@ -10904,6 +10904,18 @@
   proallargtypes => '{oid,text,int8,timestamptz}', proargmodes => '{i,o,o,o}',
   proargnames => '{tablespace,name,size,modification}',
   prosrc => 'pg_ls_tmpdir_1arg' },
+{ oid => '5032', descr => 'list directory with metadata',
+  proname => 'pg_ls_dir_metadata', procost => '10', prorows => '20', proretset => 't',
+  provolatile => 'v', prorettype => 'record', proargtypes => 'text bool bool',
+  proallargtypes => '{text,bool,bool,text,int8,timestamptz,bool}', proargmodes => '{i,i,i,o,o,o,o}',
+  proargnames => '{dirname,missing_ok,include_dot_dirs,name,size,modification,isdir}',
+  prosrc => 'pg_ls_dir_metadata' },
+{ oid => '5033', descr => 'list directory with metadata',
+  proname => 'pg_ls_dir_metadata', procost => '10', prorows => '20', proretset => 't',
+  provolatile => 'v', prorettype => 'record', proargtypes => 'text',
+  proallargtypes => '{text,text,int8,timestamptz,bool}', proargmodes => '{i,o,o,o,o}',
+  proargnames => '{dirname,name,size,modification,isdir}',
+  prosrc => 'pg_ls_dir_metadata_1arg' },
 
 # hash partitioning constraint function
 { oid => '5028', descr => 'hash partition CHECK constraint',
diff --git a/src/test/regress/expected/misc_functions.out b/src/test/regress/expected/misc_functions.out
index 2e87c548eb..7930909f02 100644
--- a/src/test/regress/expected/misc_functions.out
+++ b/src/test/regress/expected/misc_functions.out
@@ -219,6 +219,30 @@ select pg_ls_dir('does not exist', true, false); -- ok with missingok=true
 
 select pg_ls_dir('does not exist'); -- fails with missingok=false
 ERROR:  could not open directory "does not exist": No such file or directory
+-- This tests the missing_ok parameter, which causes pg_ls_tmpdir to succeed even if the tmpdir doesn't exist yet
+-- The name='' condition is never true, so the function runs to completion but returns zero rows.
+select * from pg_ls_tmpdir() where name='Does not exist';
+ name | size | modification 
+------+------+--------------
+(0 rows)
+
+select name, isdir from pg_ls_dir_metadata('.') where name='.';
+ name | isdir 
+------+-------
+ .    | t
+(1 row)
+
+select name, isdir from pg_ls_dir_metadata('.', false, false) where name='.'; -- include_dot_dirs=false
+ name | isdir 
+------+-------
+(0 rows)
+
+-- Check that expected columns are present
+select * from pg_ls_dir_metadata('.') limit 0;
+ name | size | modification | isdir 
+------+------+--------------+-------
+(0 rows)
+
 --
 -- Test adding a support function to a subject function
 --
diff --git a/src/test/regress/input/tablespace.source b/src/test/regress/input/tablespace.source
index a5f61a35dc..0b9cfe615e 100644
--- a/src/test/regress/input/tablespace.source
+++ b/src/test/regress/input/tablespace.source
@@ -11,6 +11,11 @@ DROP TABLESPACE regress_tblspacewith;
 -- create a tablespace we can use
 CREATE TABLESPACE regress_tblspace LOCATION '@testtablespace@';
 
+-- This tests the missing_ok parameter, which causes pg_ls_tmpdir to succeed even if the tmpdir doesn't exist yet
+-- The name='' condition is never true, so the function runs to completion but returns zero rows.
+-- The query is written to ERROR if the tablespace doesn't exist, rather than silently failing to call pg_ls_tmpdir()
+SELECT c.* FROM (SELECT oid FROM pg_tablespace b WHERE b.spcname='regress_tblspace' UNION SELECT 0 ORDER BY 1 DESC LIMIT 1) AS b , pg_ls_tmpdir(oid) AS c WHERE c.name='Does not exist';
+
 -- try setting and resetting some properties for the new tablespace
 ALTER TABLESPACE regress_tblspace SET (random_page_cost = 1.0, seq_page_cost = 1.1);
 ALTER TABLESPACE regress_tblspace SET (some_nonexistent_parameter = true);  -- fail
diff --git a/src/test/regress/output/tablespace.source b/src/test/regress/output/tablespace.source
index 162b591b31..a42714bf40 100644
--- a/src/test/regress/output/tablespace.source
+++ b/src/test/regress/output/tablespace.source
@@ -13,6 +13,14 @@ SELECT spcoptions FROM pg_tablespace WHERE spcname = 'regress_tblspacewith';
 DROP TABLESPACE regress_tblspacewith;
 -- create a tablespace we can use
 CREATE TABLESPACE regress_tblspace LOCATION '@testtablespace@';
+-- This tests the missing_ok parameter, which causes pg_ls_tmpdir to succeed even if the tmpdir doesn't exist yet
+-- The name='' condition is never true, so the function runs to completion but returns zero rows.
+-- The query is written to ERROR if the tablespace doesn't exist, rather than silently failing to call pg_ls_tmpdir()
+SELECT c.* FROM (SELECT oid FROM pg_tablespace b WHERE b.spcname='regress_tblspace' UNION SELECT 0 ORDER BY 1 DESC LIMIT 1) AS b , pg_ls_tmpdir(oid) AS c WHERE c.name='Does not exist';
+ name | size | modification 
+------+------+--------------
+(0 rows)
+
 -- try setting and resetting some properties for the new tablespace
 ALTER TABLESPACE regress_tblspace SET (random_page_cost = 1.0, seq_page_cost = 1.1);
 ALTER TABLESPACE regress_tblspace SET (some_nonexistent_parameter = true);  -- fail
diff --git a/src/test/regress/sql/misc_functions.sql b/src/test/regress/sql/misc_functions.sql
index f6857ad177..372345720d 100644
--- a/src/test/regress/sql/misc_functions.sql
+++ b/src/test/regress/sql/misc_functions.sql
@@ -65,6 +65,17 @@ select * from (select pg_ls_dir('.', false, false) as name) as ls where ls.name=
 select pg_ls_dir('does not exist', true, false); -- ok with missingok=true
 select pg_ls_dir('does not exist'); -- fails with missingok=false
 
+-- This tests the missing_ok parameter, which causes pg_ls_tmpdir to succeed even if the tmpdir doesn't exist yet
+-- The name='' condition is never true, so the function runs to completion but returns zero rows.
+select * from pg_ls_tmpdir() where name='Does not exist';
+
+select name, isdir from pg_ls_dir_metadata('.') where name='.';
+
+select name, isdir from pg_ls_dir_metadata('.', false, false) where name='.'; -- include_dot_dirs=false
+
+-- Check that expected columns are present
+select * from pg_ls_dir_metadata('.') limit 0;
+
 --
 -- Test adding a support function to a subject function
 --
-- 
2.17.0


--2FkSFaIQeDFoAt0B
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
 filename="v16-0005-pg_ls_tmpdir-to-show-directories-and-isdir-argum.patch"



^ permalink  raw  reply  [nested|flat] 68+ messages in thread