agora inbox for pgsql-hackers@postgresql.org  
help / color / mirror / Atom feed
[PATCH v9 04/11] Add pg_ls_dir_metadata to list a dir with file metadata..
22+ messages / 2 participants
[nested] [flat]

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

* [PATCH v15 04/10] Add pg_ls_dir_metadata to list a dir with file metadata..
@ 2020-03-10 03:40 Justin Pryzby <pryzbyj@telsasoft.com>
  0 siblings, 0 replies; 22+ messages in thread

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

Generalize pg_ls_dir_files and retire pg_ls_dir

XXX: Change to use lstat() to allow pg_ls_dir_recurse to avoid infinite recursion.
That means:
 - links to dirs are NOT SHOWN (was: shown with isdir=false;)
 - NO: timestamps shown are those of the link;
 - changed pg_stat_file for consistency;

Need catversion bumped?
---
 doc/src/sgml/func.sgml                       |  17 ++
 src/backend/catalog/system_views.sql         |   1 +
 src/backend/utils/adt/genfile.c              | 227 +++++++++++--------
 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, 214 insertions(+), 91 deletions(-)

diff --git a/doc/src/sgml/func.sgml b/doc/src/sgml/func.sgml
index 630f03d7dd..ed8e1977ba 100644
--- a/doc/src/sgml/func.sgml
+++ b/doc/src/sgml/func.sgml
@@ -21372,6 +21372,15 @@ 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>
@@ -21472,6 +21481,14 @@ 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 83d00c6cde..4511f9d658 100644
--- a/src/backend/catalog/system_views.sql
+++ b/src/backend/catalog/system_views.sql
@@ -1436,6 +1436,7 @@ 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..eaa7c548e0 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 +536,18 @@ 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);
@@ -578,20 +566,27 @@ 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 +601,34 @@ 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 +642,14 @@ 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 +667,8 @@ 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 +697,33 @@ 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 a6a708cca9..e8846e1522 100644
--- a/src/include/catalog/pg_proc.dat
+++ b/src/include/catalog/pg_proc.dat
@@ -10801,6 +10801,18 @@
   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


--8w3uRX/HFJGApMzv
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
 filename="v15-0005-pg_ls_tmpdir-to-show-directories-and-isdir-argum.patch"



^ permalink  raw  reply  [nested|flat] 22+ messages in thread

* [PATCH v37 03/11] Add pg_ls_dir_metadata to list a dir with file metadata..
@ 2020-03-10 03:40 Justin Pryzby <pryzbyj@telsasoft.com>
  0 siblings, 0 replies; 22+ 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              | 205 ++++++++++++-------
 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, 217 insertions(+), 70 deletions(-)

diff --git a/doc/src/sgml/func.sgml b/doc/src/sgml/func.sgml
index d958c3e74ac..a80d794a43a 100644
--- a/doc/src/sgml/func.sgml
+++ b/doc/src/sgml/func.sgml
@@ -25900,6 +25900,27 @@ LOG:  Grand total: 1651920 bytes in 201 blocks; 622360 free (88 chunks); 1029560
        </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 30a048f6b09..b77b83017cb 100644
--- a/src/backend/catalog/system_functions.sql
+++ b/src/backend/catalog/system_functions.sql
@@ -704,6 +704,7 @@ 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 ab6f67f8747..c041c7630c8 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 beginning 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.
@@ -516,6 +531,11 @@ 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);
 
@@ -543,54 +563,9 @@ 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;
-	DIR		   *dirdesc;
-	struct dirent *de;
-
-	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);
-	}
-
-	InitMaterializedSRF(fcinfo, MAT_SRF_USE_EXPECTED_DESC);
-
-	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(rsinfo->setResult, rsinfo->setDesc,
-							 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);
 }
 
 /*
@@ -603,23 +578,55 @@ 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;
 	DIR		   *dirdesc;
 	struct dirent *de;
 
-	InitMaterializedSRF(fcinfo, 0);
+	/* 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)
+	{
+		/* missing_ok */
+		if (!PG_ARGISNULL(1))
+		{
+			if (PG_GETARG_BOOL(1))
+				flags |= LS_DIR_MISSING_OK;
+			else
+				flags &= ~LS_DIR_MISSING_OK;
+		}
+
+		/* include_dot_dirs */
+		if (!PG_ARGISNULL(2))
+		{
+			if (PG_GETARG_BOOL(2))
+				flags &= ~LS_DIR_SKIP_DOT_DIRS;
+			else
+				flags |= LS_DIR_SKIP_DOT_DIRS;
+		}
+	}
+
+	if (flags & LS_DIR_METADATA)
+		InitMaterializedSRF(fcinfo, 0);
+	else
+		InitMaterializedSRF(fcinfo, MAT_SRF_USE_EXPECTED_DESC);
 
 	/*
 	 * Now walk the directory.  Note that we must do this within a single SRF
@@ -630,20 +637,27 @@ 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 */
@@ -658,13 +672,35 @@ 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(rsinfo->setResult, rsinfo->setDesc, values, nulls);
@@ -678,14 +714,14 @@ 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);
 }
 
 /*
@@ -703,7 +739,8 @@ 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);
 }
 
 /*
@@ -732,7 +769,35 @@ 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);
 }
 
 /*
@@ -741,7 +806,7 @@ 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);
 }
 
 /*
@@ -750,7 +815,7 @@ 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);
 }
 
 /*
@@ -775,5 +840,5 @@ 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 20f5aa56eab..89892647808 100644
--- a/src/include/catalog/pg_proc.dat
+++ b/src/include/catalog/pg_proc.dat
@@ -11758,6 +11758,18 @@
   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 77d285ecc85..a656f9ad07e 100644
--- a/src/test/regress/expected/misc_functions.out
+++ b/src/test/regress/expected/misc_functions.out
@@ -539,6 +539,30 @@ 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 c52cf1cfcf9..8159c9f18f1 100644
--- a/src/test/regress/expected/tablespace.out
+++ b/src/test/regress/expected/tablespace.out
@@ -33,6 +33,14 @@ SELECT regexp_replace(pg_tablespace_location(oid), '(pg_tblspc)/(\d+)', '\1/NNN'
  pg_tblspc/NNN
 (1 row)
 
+-- 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 d299f3d8949..f808555382d 100644
--- a/src/test/regress/sql/misc_functions.sql
+++ b/src/test/regress/sql/misc_functions.sql
@@ -181,6 +181,17 @@ 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 21db433f2a8..cf683c3bf3a 100644
--- a/src/test/regress/sql/tablespace.sql
+++ b/src/test/regress/sql/tablespace.sql
@@ -27,6 +27,11 @@ CREATE TABLESPACE regress_tblspace LOCATION '';
 SELECT regexp_replace(pg_tablespace_location(oid), '(pg_tblspc)/(\d+)', '\1/NNN')
   FROM pg_tablespace  WHERE spcname = 'regress_tblspace';
 
+-- 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
-- 
2.25.1


--Pk/CTwBz1VvfPIDp
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
 filename="v37-0004-pg_ls_tmpdir-to-show-directories-and-isdir-argum.patch"



^ permalink  raw  reply  [nested|flat] 22+ messages in thread

* [PATCH v35 3/7] Add pg_ls_dir_metadata to list a dir with file metadata..
@ 2020-03-10 03:40 Justin Pryzby <pryzbyj@telsasoft.com>
  0 siblings, 0 replies; 22+ 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              | 205 ++++++++++++-------
 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, 217 insertions(+), 70 deletions(-)

diff --git a/doc/src/sgml/func.sgml b/doc/src/sgml/func.sgml
index d01aeec9f88..d2a455a3e27 100644
--- a/doc/src/sgml/func.sgml
+++ b/doc/src/sgml/func.sgml
@@ -25987,6 +25987,27 @@ 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 81bac6f5812..b4d3609cce7 100644
--- a/src/backend/catalog/system_functions.sql
+++ b/src/backend/catalog/system_functions.sql
@@ -700,6 +700,7 @@ 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 88f279d1b31..75b7bf99849 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 beginning 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.
@@ -446,6 +461,11 @@ 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);
 
@@ -473,54 +493,9 @@ 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;
-	DIR		   *dirdesc;
-	struct dirent *de;
-
-	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);
-	}
-
-	SetSingleFuncCall(fcinfo, SRF_SINGLE_USE_EXPECTED);
-
-	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(rsinfo->setResult, rsinfo->setDesc,
-							 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);
 }
 
 /*
@@ -533,23 +508,55 @@ 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;
 	DIR		   *dirdesc;
 	struct dirent *de;
 
-	SetSingleFuncCall(fcinfo, 0);
+	/* 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)
+	{
+		/* missing_ok */
+		if (!PG_ARGISNULL(1))
+		{
+			if (PG_GETARG_BOOL(1))
+				flags |= LS_DIR_MISSING_OK;
+			else
+				flags &= ~LS_DIR_MISSING_OK;
+		}
+
+		/* include_dot_dirs */
+		if (!PG_ARGISNULL(2))
+		{
+			if (PG_GETARG_BOOL(2))
+				flags &= ~LS_DIR_SKIP_DOT_DIRS;
+			else
+				flags |= LS_DIR_SKIP_DOT_DIRS;
+		}
+	}
+
+	if (flags & LS_DIR_METADATA)
+		SetSingleFuncCall(fcinfo, 0);
+	else
+		SetSingleFuncCall(fcinfo, SRF_SINGLE_USE_EXPECTED);
 
 	/*
 	 * Now walk the directory.  Note that we must do this within a single SRF
@@ -560,20 +567,27 @@ 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 */
@@ -588,13 +602,35 @@ 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(rsinfo->setResult, rsinfo->setDesc, values, nulls);
@@ -608,14 +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);
 }
 
 /*
@@ -633,7 +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);
 }
 
 /*
@@ -662,7 +699,35 @@ 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);
 }
 
 /*
@@ -671,7 +736,7 @@ 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);
 }
 
 /*
@@ -680,7 +745,7 @@ 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);
 }
 
 /*
@@ -705,5 +770,5 @@ 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 25304430f44..7307f44f371 100644
--- a/src/include/catalog/pg_proc.dat
+++ b/src/include/catalog/pg_proc.dat
@@ -11769,6 +11769,18 @@
   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 45544469af5..e54e38f54ad 100644
--- a/src/test/regress/expected/misc_functions.out
+++ b/src/test/regress/expected/misc_functions.out
@@ -475,6 +475,30 @@ 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 c52cf1cfcf9..8159c9f18f1 100644
--- a/src/test/regress/expected/tablespace.out
+++ b/src/test/regress/expected/tablespace.out
@@ -33,6 +33,14 @@ SELECT regexp_replace(pg_tablespace_location(oid), '(pg_tblspc)/(\d+)', '\1/NNN'
  pg_tblspc/NNN
 (1 row)
 
+-- 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 59d6e517503..fcb456f434d 100644
--- a/src/test/regress/sql/misc_functions.sql
+++ b/src/test/regress/sql/misc_functions.sql
@@ -155,6 +155,17 @@ 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 21db433f2a8..cf683c3bf3a 100644
--- a/src/test/regress/sql/tablespace.sql
+++ b/src/test/regress/sql/tablespace.sql
@@ -27,6 +27,11 @@ CREATE TABLESPACE regress_tblspace LOCATION '';
 SELECT regexp_replace(pg_tablespace_location(oid), '(pg_tblspc)/(\d+)', '\1/NNN')
   FROM pg_tablespace  WHERE spcname = 'regress_tblspace';
 
+-- 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


--olLTNZSltDMg5Vbm
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
 filename="v35-0004-pg_ls_tmpdir-to-show-directories-and-isdir-argum.patch"



^ permalink  raw  reply  [nested|flat] 22+ 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 <pryzbyj@telsasoft.com>
  0 siblings, 0 replies; 22+ 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] 22+ 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 <pryzbyj@telsasoft.com>
  0 siblings, 0 replies; 22+ 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] 22+ 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 <pryzbyj@telsasoft.com>
  0 siblings, 0 replies; 22+ 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] 22+ 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 <pryzbyj@telsasoft.com>
  0 siblings, 0 replies; 22+ 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] 22+ messages in thread

* [PATCH v20 04/10] Add pg_ls_dir_metadata to list a dir with file metadata..
@ 2020-03-10 03:40 Justin Pryzby <pryzbyj@telsasoft.com>
  0 siblings, 0 replies; 22+ 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 959f6a1c2f..eead88419f 100644
--- a/doc/src/sgml/func.sgml
+++ b/doc/src/sgml/func.sgml
@@ -25741,6 +25741,27 @@ 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 b6d35c2d11..360b6bda26 100644
--- a/src/backend/catalog/system_views.sql
+++ b/src/backend/catalog/system_views.sql
@@ -1478,6 +1478,7 @@ 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 46d0977c2e..3324ffe16a 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 95604e988a..186ced4a35 100644
--- a/src/include/catalog/pg_proc.dat
+++ b/src/include/catalog/pg_proc.dat
@@ -10919,6 +10919,18 @@
   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


--Z1Z8UV8BNhgCynIS
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
 filename="v20-0005-pg_ls_tmpdir-to-show-directories-and-isdir-argum.patch"



^ permalink  raw  reply  [nested|flat] 22+ messages in thread

* [PATCH v36 3/7] Add pg_ls_dir_metadata to list a dir with file metadata..
@ 2020-03-10 03:40 Justin Pryzby <pryzbyj@telsasoft.com>
  0 siblings, 0 replies; 22+ 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              | 205 ++++++++++++-------
 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, 217 insertions(+), 70 deletions(-)

diff --git a/doc/src/sgml/func.sgml b/doc/src/sgml/func.sgml
index 933118203f4..ba30faa8441 100644
--- a/doc/src/sgml/func.sgml
+++ b/doc/src/sgml/func.sgml
@@ -25940,6 +25940,27 @@ SELECT collation for ('foo' COLLATE "de_DE");
        </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 73da687d5dd..3fa6f9af0a8 100644
--- a/src/backend/catalog/system_functions.sql
+++ b/src/backend/catalog/system_functions.sql
@@ -700,6 +700,7 @@ 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 2bf52192567..c7322ed405a 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 beginning 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.
@@ -446,6 +461,11 @@ 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);
 
@@ -473,54 +493,9 @@ 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;
-	DIR		   *dirdesc;
-	struct dirent *de;
-
-	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);
-	}
-
-	SetSingleFuncCall(fcinfo, SRF_SINGLE_USE_EXPECTED);
-
-	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(rsinfo->setResult, rsinfo->setDesc,
-							 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);
 }
 
 /*
@@ -533,23 +508,55 @@ 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;
 	DIR		   *dirdesc;
 	struct dirent *de;
 
-	SetSingleFuncCall(fcinfo, 0);
+	/* 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)
+	{
+		/* missing_ok */
+		if (!PG_ARGISNULL(1))
+		{
+			if (PG_GETARG_BOOL(1))
+				flags |= LS_DIR_MISSING_OK;
+			else
+				flags &= ~LS_DIR_MISSING_OK;
+		}
+
+		/* include_dot_dirs */
+		if (!PG_ARGISNULL(2))
+		{
+			if (PG_GETARG_BOOL(2))
+				flags &= ~LS_DIR_SKIP_DOT_DIRS;
+			else
+				flags |= LS_DIR_SKIP_DOT_DIRS;
+		}
+	}
+
+	if (flags & LS_DIR_METADATA)
+		SetSingleFuncCall(fcinfo, 0);
+	else
+		SetSingleFuncCall(fcinfo, SRF_SINGLE_USE_EXPECTED);
 
 	/*
 	 * Now walk the directory.  Note that we must do this within a single SRF
@@ -560,20 +567,27 @@ 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 */
@@ -588,13 +602,35 @@ 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(rsinfo->setResult, rsinfo->setDesc, values, nulls);
@@ -608,14 +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);
 }
 
 /*
@@ -633,7 +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);
 }
 
 /*
@@ -662,7 +699,35 @@ 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);
 }
 
 /*
@@ -671,7 +736,7 @@ 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);
 }
 
 /*
@@ -680,7 +745,7 @@ 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);
 }
 
 /*
@@ -705,5 +770,5 @@ 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 87aa571a331..0e5c977e69f 100644
--- a/src/include/catalog/pg_proc.dat
+++ b/src/include/catalog/pg_proc.dat
@@ -11815,6 +11815,18 @@
   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 45544469af5..e54e38f54ad 100644
--- a/src/test/regress/expected/misc_functions.out
+++ b/src/test/regress/expected/misc_functions.out
@@ -475,6 +475,30 @@ 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 c52cf1cfcf9..8159c9f18f1 100644
--- a/src/test/regress/expected/tablespace.out
+++ b/src/test/regress/expected/tablespace.out
@@ -33,6 +33,14 @@ SELECT regexp_replace(pg_tablespace_location(oid), '(pg_tblspc)/(\d+)', '\1/NNN'
  pg_tblspc/NNN
 (1 row)
 
+-- 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 59d6e517503..fcb456f434d 100644
--- a/src/test/regress/sql/misc_functions.sql
+++ b/src/test/regress/sql/misc_functions.sql
@@ -155,6 +155,17 @@ 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 21db433f2a8..cf683c3bf3a 100644
--- a/src/test/regress/sql/tablespace.sql
+++ b/src/test/regress/sql/tablespace.sql
@@ -27,6 +27,11 @@ CREATE TABLESPACE regress_tblspace LOCATION '';
 SELECT regexp_replace(pg_tablespace_location(oid), '(pg_tblspc)/(\d+)', '\1/NNN')
   FROM pg_tablespace  WHERE spcname = 'regress_tblspace';
 
+-- 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


--4ybNbZnZ8tziJ7D6
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
 filename="v36-0004-pg_ls_tmpdir-to-show-directories-and-isdir-argum.patch"



^ permalink  raw  reply  [nested|flat] 22+ 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 <pryzbyj@telsasoft.com>
  0 siblings, 0 replies; 22+ 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] 22+ 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 <pryzbyj@telsasoft.com>
  0 siblings, 0 replies; 22+ 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] 22+ messages in thread

* [PATCH v15 04/10] Add pg_ls_dir_metadata to list a dir with file metadata..
@ 2020-03-10 03:40 Justin Pryzby <pryzbyj@telsasoft.com>
  0 siblings, 0 replies; 22+ messages in thread

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

Generalize pg_ls_dir_files and retire pg_ls_dir

XXX: Change to use lstat() to allow pg_ls_dir_recurse to avoid infinite recursion.
That means:
 - links to dirs are NOT SHOWN (was: shown with isdir=false;)
 - NO: timestamps shown are those of the link;
 - changed pg_stat_file for consistency;

Need catversion bumped?
---
 doc/src/sgml/func.sgml                       |  17 ++
 src/backend/catalog/system_views.sql         |   1 +
 src/backend/utils/adt/genfile.c              | 227 +++++++++++--------
 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, 214 insertions(+), 91 deletions(-)

diff --git a/doc/src/sgml/func.sgml b/doc/src/sgml/func.sgml
index 630f03d7dd..ed8e1977ba 100644
--- a/doc/src/sgml/func.sgml
+++ b/doc/src/sgml/func.sgml
@@ -21372,6 +21372,15 @@ 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>
@@ -21472,6 +21481,14 @@ 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 83d00c6cde..4511f9d658 100644
--- a/src/backend/catalog/system_views.sql
+++ b/src/backend/catalog/system_views.sql
@@ -1436,6 +1436,7 @@ 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..eaa7c548e0 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 +536,18 @@ 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);
@@ -578,20 +566,27 @@ 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 +601,34 @@ 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 +642,14 @@ 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 +667,8 @@ 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 +697,33 @@ 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 a6a708cca9..e8846e1522 100644
--- a/src/include/catalog/pg_proc.dat
+++ b/src/include/catalog/pg_proc.dat
@@ -10801,6 +10801,18 @@
   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


--8w3uRX/HFJGApMzv
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
 filename="v15-0005-pg_ls_tmpdir-to-show-directories-and-isdir-argum.patch"



^ permalink  raw  reply  [nested|flat] 22+ 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 <pryzbyj@telsasoft.com>
  0 siblings, 0 replies; 22+ 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] 22+ 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 <pryzbyj@telsasoft.com>
  0 siblings, 0 replies; 22+ 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] 22+ messages in thread

* [PATCH v21 04/10] Add pg_ls_dir_metadata to list a dir with file metadata..
@ 2020-03-10 03:40 Justin Pryzby <pryzbyj@telsasoft.com>
  0 siblings, 0 replies; 22+ 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 959f6a1c2f..eead88419f 100644
--- a/doc/src/sgml/func.sgml
+++ b/doc/src/sgml/func.sgml
@@ -25741,6 +25741,27 @@ 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 8625cbeab6..b9b2a6aa20 100644
--- a/src/backend/catalog/system_views.sql
+++ b/src/backend/catalog/system_views.sql
@@ -1478,6 +1478,7 @@ 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..24d858d1bc 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 95604e988a..83927b510a 100644
--- a/src/include/catalog/pg_proc.dat
+++ b/src/include/catalog/pg_proc.dat
@@ -10919,6 +10919,18 @@
   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


--Tcb1KvpfnM4LxW2s
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
 filename="v21-0005-pg_ls_tmpdir-to-show-directories-and-isdir-argum.patch"



^ permalink  raw  reply  [nested|flat] 22+ 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 <pryzbyj@telsasoft.com>
  0 siblings, 0 replies; 22+ 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] 22+ messages in thread

* [PATCH v34 03/15] Add pg_ls_dir_metadata to list a dir with file metadata..
@ 2020-03-10 03:40 Justin Pryzby <pryzbyj@telsasoft.com>
  0 siblings, 0 replies; 22+ 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              | 203 ++++++++++++-------
 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, 215 insertions(+), 70 deletions(-)

diff --git a/doc/src/sgml/func.sgml b/doc/src/sgml/func.sgml
index 0be4743e3ed..c2ff924d021 100644
--- a/doc/src/sgml/func.sgml
+++ b/doc/src/sgml/func.sgml
@@ -25976,6 +25976,27 @@ 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 81bac6f5812..b4d3609cce7 100644
--- a/src/backend/catalog/system_functions.sql
+++ b/src/backend/catalog/system_functions.sql
@@ -700,6 +700,7 @@ 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 1ed01620a1b..1878bac7eb3 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 beginning 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.
@@ -446,6 +461,11 @@ 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);
 
@@ -473,54 +493,9 @@ 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;
-	DIR		   *dirdesc;
-	struct dirent *de;
-
-	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);
-	}
-
-	SetSingleFuncCall(fcinfo, SRF_SINGLE_USE_EXPECTED);
-
-	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(rsinfo->setResult, rsinfo->setDesc,
-							 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);
 }
 
 /*
@@ -533,23 +508,53 @@ 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;
 	DIR		   *dirdesc;
 	struct dirent *de;
 
-	SetSingleFuncCall(fcinfo, 0);
+	/* 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;
+		}
+	}
+
+	if (flags & LS_DIR_METADATA)
+		SetSingleFuncCall(fcinfo, 0);
+	else
+		SetSingleFuncCall(fcinfo, SRF_SINGLE_USE_EXPECTED);
 
 	/*
 	 * Now walk the directory.  Note that we must do this within a single SRF
@@ -560,20 +565,27 @@ 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 */
@@ -588,13 +600,35 @@ 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(rsinfo->setResult, rsinfo->setDesc, values, nulls);
@@ -608,14 +642,14 @@ 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);
 }
 
 /*
@@ -633,7 +667,8 @@ 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);
 }
 
 /*
@@ -662,7 +697,35 @@ 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);
 }
 
 /*
@@ -671,7 +734,7 @@ 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);
 }
 
 /*
@@ -680,7 +743,7 @@ 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);
 }
 
 /*
@@ -705,5 +768,5 @@ 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 d8e8715ed1c..43ee0cdc454 100644
--- a/src/include/catalog/pg_proc.dat
+++ b/src/include/catalog/pg_proc.dat
@@ -11677,6 +11677,18 @@
   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 b08e7c4a6d0..d18f401248a 100644
--- a/src/test/regress/expected/misc_functions.out
+++ b/src/test/regress/expected/misc_functions.out
@@ -452,6 +452,30 @@ 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 ebb0352ac68..05794e1ae45 100644
--- a/src/test/regress/sql/misc_functions.sql
+++ b/src/test/regress/sql/misc_functions.sql
@@ -147,6 +147,17 @@ 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


--smOfPzt+Qjm5bNGJ
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
 filename="v34-0004-pg_ls_tmpdir-to-show-directories-and-isdir-argum.patch"



^ permalink  raw  reply  [nested|flat] 22+ messages in thread

* [PATCH v2] Cleanup explicit PREPARE query strings
@ 2025-12-24 14:31 Julien Rouhaud <julien.rouhaud@free.fr>
  0 siblings, 0 replies; 22+ messages in thread

From: Julien Rouhaud @ 2025-12-24 14:31 UTC (permalink / raw)

When a multi statements query string contains one PREPARE statement (or
multiple), the whole query string was saved in the cached plan.  This is
wasteful as that string can be artitrarily big, but it can also confusing as
some other parts like pg_prepared_statements will output the saved query string
as-is.

This commit changes this behavior and only stores the part of the query string
that correspond to any given PREPARED statement, similarly to how it's already
done in pg_stat_statements.
---
 contrib/auto_explain/t/001_auto_explain.pl    |  8 +--
 contrib/pg_stat_statements/Makefile           |  2 +-
 .../pg_stat_statements/expected/prepare.out   | 53 +++++++++++++++++++
 contrib/pg_stat_statements/meson.build        |  1 +
 contrib/pg_stat_statements/sql/prepare.sql    | 15 ++++++
 src/backend/commands/prepare.c                | 45 ++++++++++++++--
 src/test/regress/expected/prepare.out         | 44 +++++++++------
 src/test/regress/sql/prepare.sql              |  2 +-
 8 files changed, 143 insertions(+), 27 deletions(-)
 create mode 100644 contrib/pg_stat_statements/expected/prepare.out
 create mode 100644 contrib/pg_stat_statements/sql/prepare.sql

diff --git a/contrib/auto_explain/t/001_auto_explain.pl b/contrib/auto_explain/t/001_auto_explain.pl
index 5f673bd14c1..f2d8625f8bb 100644
--- a/contrib/auto_explain/t/001_auto_explain.pl
+++ b/contrib/auto_explain/t/001_auto_explain.pl
@@ -60,7 +60,7 @@ $log_contents = query_log($node,
 
 like(
 	$log_contents,
-	qr/Query Text: PREPARE get_proc\(name\) AS SELECT \* FROM pg_proc WHERE proname = \$1;/,
+	qr/Query Text: PREPARE get_proc\(name\) AS SELECT \* FROM pg_proc WHERE proname = \$1/,
 	"prepared query text logged, text mode");
 
 like(
@@ -82,7 +82,7 @@ $log_contents = query_log(
 
 like(
 	$log_contents,
-	qr/Query Text: PREPARE get_type\(name\) AS SELECT \* FROM pg_type WHERE typname = \$1;/,
+	qr/Query Text: PREPARE get_type\(name\) AS SELECT \* FROM pg_type WHERE typname = \$1/,
 	"prepared query text logged, text mode");
 
 like(
@@ -98,7 +98,7 @@ $log_contents = query_log(
 
 like(
 	$log_contents,
-	qr/Query Text: PREPARE get_type\(name\) AS SELECT \* FROM pg_type WHERE typname = \$1;/,
+	qr/Query Text: PREPARE get_type\(name\) AS SELECT \* FROM pg_type WHERE typname = \$1/,
 	"prepared query text logged, text mode");
 
 unlike(
@@ -164,7 +164,7 @@ $log_contents = query_log(
 
 like(
 	$log_contents,
-	qr/"Query Text": "PREPARE get_class\(name\) AS SELECT \* FROM pg_class WHERE relname = \$1;"/,
+	qr/"Query Text": "PREPARE get_class\(name\) AS SELECT \* FROM pg_class WHERE relname = \$1"/,
 	"prepared query text logged, json mode");
 
 like(
diff --git a/contrib/pg_stat_statements/Makefile b/contrib/pg_stat_statements/Makefile
index fe0478ac552..9a6bda90070 100644
--- a/contrib/pg_stat_statements/Makefile
+++ b/contrib/pg_stat_statements/Makefile
@@ -21,7 +21,7 @@ LDFLAGS_SL += $(filter -lm, $(LIBS))
 REGRESS_OPTS = --temp-config $(top_srcdir)/contrib/pg_stat_statements/pg_stat_statements.conf
 REGRESS = select dml cursors utility level_tracking planning \
 	user_activity wal entry_timestamp privileges extended \
-	parallel plancache cleanup oldextversions squashing
+	parallel plancache cleanup oldextversions squashing prepare
 # Disabled because these tests require "shared_preload_libraries=pg_stat_statements",
 # which typical installcheck users do not have (e.g. buildfarm clients).
 NO_INSTALLCHECK = 1
diff --git a/contrib/pg_stat_statements/expected/prepare.out b/contrib/pg_stat_statements/expected/prepare.out
new file mode 100644
index 00000000000..010e289c1b0
--- /dev/null
+++ b/contrib/pg_stat_statements/expected/prepare.out
@@ -0,0 +1,53 @@
+-- Tests for PREPARE
+SELECT pg_stat_statements_reset() IS NOT NULL AS t;
+ t 
+---
+ t
+(1 row)
+
+-- Test that prepared statements in a multi-query string behaves as expected
+SELECT 1\;PREPARE p1 AS SELECT 1\; PREPARE p2(int) AS SELECT 2 * $1\; SELECT 1, 1;
+ ?column? 
+----------
+        1
+(1 row)
+
+ ?column? | ?column? 
+----------+----------
+        1 |        1
+(1 row)
+
+SELECT calls, rows, query FROM pg_stat_statements ORDER BY query COLLATE "C";
+ calls | rows |                       query                        
+-------+------+----------------------------------------------------
+     1 |    1 | SELECT $1
+     1 |    1 | SELECT $1, $2
+     1 |    1 | SELECT pg_stat_statements_reset() IS NOT NULL AS t
+(3 rows)
+
+SELECT pg_stat_statements_reset() IS NOT NULL AS t;
+ t 
+---
+ t
+(1 row)
+
+EXECUTE p1;
+ ?column? 
+----------
+        1
+(1 row)
+
+EXECUTE p2(0);
+ ?column? 
+----------
+        0
+(1 row)
+
+SELECT calls, rows, query FROM pg_stat_statements ORDER BY query COLLATE "C";
+ calls | rows |                       query                        
+-------+------+----------------------------------------------------
+     1 |    1 | PREPARE p1 AS SELECT 1
+     1 |    1 | PREPARE p2(int) AS SELECT 2 * $1
+     1 |    1 | SELECT pg_stat_statements_reset() IS NOT NULL AS t
+(3 rows)
+
diff --git a/contrib/pg_stat_statements/meson.build b/contrib/pg_stat_statements/meson.build
index 079a8eb5afc..65258651bd2 100644
--- a/contrib/pg_stat_statements/meson.build
+++ b/contrib/pg_stat_statements/meson.build
@@ -59,6 +59,7 @@ tests += {
       'cleanup',
       'oldextversions',
       'squashing',
+      'prepare',
     ],
     'regress_args': ['--temp-config', files('pg_stat_statements.conf')],
     # Disabled because these tests require
diff --git a/contrib/pg_stat_statements/sql/prepare.sql b/contrib/pg_stat_statements/sql/prepare.sql
new file mode 100644
index 00000000000..a6bc5de4430
--- /dev/null
+++ b/contrib/pg_stat_statements/sql/prepare.sql
@@ -0,0 +1,15 @@
+-- Tests for PREPARE
+
+SELECT pg_stat_statements_reset() IS NOT NULL AS t;
+
+-- Test that prepared statements in a multi-query string behaves as expected
+SELECT 1\;PREPARE p1 AS SELECT 1\; PREPARE p2(int) AS SELECT 2 * $1\; SELECT 1, 1;
+
+SELECT calls, rows, query FROM pg_stat_statements ORDER BY query COLLATE "C";
+
+SELECT pg_stat_statements_reset() IS NOT NULL AS t;
+
+EXECUTE p1;
+EXECUTE p2(0);
+
+SELECT calls, rows, query FROM pg_stat_statements ORDER BY query COLLATE "C";
diff --git a/src/backend/commands/prepare.c b/src/backend/commands/prepare.c
index 5b86a727587..b8fe3dd5302 100644
--- a/src/backend/commands/prepare.c
+++ b/src/backend/commands/prepare.c
@@ -27,6 +27,7 @@
 #include "commands/prepare.h"
 #include "funcapi.h"
 #include "nodes/nodeFuncs.h"
+#include "nodes/queryjumble.h"
 #include "parser/parse_coerce.h"
 #include "parser/parse_collate.h"
 #include "parser/parse_expr.h"
@@ -64,6 +65,7 @@ PrepareQuery(ParseState *pstate, PrepareStmt *stmt,
 	Oid		   *argtypes = NULL;
 	int			nargs;
 	List	   *query_list;
+	const char *new_query;
 
 	/*
 	 * Disallow empty-string statement name (conflicts with protocol-level
@@ -80,14 +82,49 @@ PrepareQuery(ParseState *pstate, PrepareStmt *stmt,
 	 */
 	rawstmt = makeNode(RawStmt);
 	rawstmt->stmt = stmt->query;
-	rawstmt->stmt_location = stmt_location;
-	rawstmt->stmt_len = stmt_len;
+
+	/*
+	 * Extract the query text if possible.
+	 *
+	 * If we have a statement location, we can extract the relevant part of the
+	 * possibly multi-statement query string.  If not just use what we were
+	 * given.
+	 */
+	if (stmt_location < 0)
+	{
+		rawstmt->stmt_location = stmt_location;
+		rawstmt->stmt_len = stmt_len;
+		new_query = pstate->p_sourcetext;
+	}
+	else
+	{
+		const char *cleaned;
+		char	   *tmp;
+
+		rawstmt->stmt_len = stmt_len;
+		cleaned = CleanQuerytext(pstate->p_sourcetext, &stmt_location,
+								 &rawstmt->stmt_len);
+
+		if (rawstmt->stmt_len == 0)
+			rawstmt->stmt_len = strlen(cleaned);
+
+		/*
+		 * CleanQuerytext() removes any leading whitespace and returns a
+		 * pointer to the first actual character, so the cleaned query string
+		 * is guaranteed to start at offset 0.
+		 */
+		rawstmt->stmt_location = 0;
+		tmp = palloc(rawstmt->stmt_len + 1);
+		strlcpy(tmp, cleaned, rawstmt->stmt_len + 1);
+
+		new_query = tmp;
+	}
 
 	/*
 	 * Create the CachedPlanSource before we do parse analysis, since it needs
 	 * to see the unmodified raw parse tree.
 	 */
-	plansource = CreateCachedPlan(rawstmt, pstate->p_sourcetext,
+	plansource = CreateCachedPlan(rawstmt, new_query,
 								  CreateCommandTag(stmt->query));
 
 	/* Transform list of TypeNames to array of type OIDs */
@@ -116,7 +153,7 @@ PrepareQuery(ParseState *pstate, PrepareStmt *stmt,
 	 * information about unknown parameters to be deduced from context.
 	 * Rewrite the query. The result could be 0, 1, or many queries.
 	 */
-	query_list = pg_analyze_and_rewrite_varparams(rawstmt, pstate->p_sourcetext,
+	query_list = pg_analyze_and_rewrite_varparams(rawstmt, new_query,
 												  &argtypes, &nargs, NULL);
 
 	/* Finish filling in the CachedPlanSource */
diff --git a/src/test/regress/expected/prepare.out b/src/test/regress/expected/prepare.out
index 5815e17b39c..c645a4e5d0e 100644
--- a/src/test/regress/expected/prepare.out
+++ b/src/test/regress/expected/prepare.out
@@ -6,7 +6,17 @@ SELECT name, statement, parameter_types, result_types FROM pg_prepared_statement
 ------+-----------+-----------------+--------------
 (0 rows)
 
-PREPARE q1 AS SELECT 1 AS a;
+SELECT 'bingo'\;  PREPARE q1 AS SELECT 1 AS a \; SELECT 42;
+ ?column? 
+----------
+ bingo
+(1 row)
+
+ ?column? 
+----------
+       42
+(1 row)
+
 EXECUTE q1;
  a 
 ---
@@ -14,9 +24,9 @@ EXECUTE q1;
 (1 row)
 
 SELECT name, statement, parameter_types, result_types FROM pg_prepared_statements;
- name |          statement           | parameter_types | result_types 
-------+------------------------------+-----------------+--------------
- q1   | PREPARE q1 AS SELECT 1 AS a; | {}              | {integer}
+ name |          statement          | parameter_types | result_types 
+------+-----------------------------+-----------------+--------------
+ q1   | PREPARE q1 AS SELECT 1 AS a | {}              | {integer}
 (1 row)
 
 -- should fail
@@ -33,18 +43,18 @@ EXECUTE q1;
 
 PREPARE q2 AS SELECT 2 AS b;
 SELECT name, statement, parameter_types, result_types FROM pg_prepared_statements;
- name |          statement           | parameter_types | result_types 
-------+------------------------------+-----------------+--------------
- q1   | PREPARE q1 AS SELECT 2;      | {}              | {integer}
- q2   | PREPARE q2 AS SELECT 2 AS b; | {}              | {integer}
+ name |          statement          | parameter_types | result_types 
+------+-----------------------------+-----------------+--------------
+ q1   | PREPARE q1 AS SELECT 2      | {}              | {integer}
+ q2   | PREPARE q2 AS SELECT 2 AS b | {}              | {integer}
 (2 rows)
 
 -- sql92 syntax
 DEALLOCATE PREPARE q1;
 SELECT name, statement, parameter_types, result_types FROM pg_prepared_statements;
- name |          statement           | parameter_types | result_types 
-------+------------------------------+-----------------+--------------
- q2   | PREPARE q2 AS SELECT 2 AS b; | {}              | {integer}
+ name |          statement          | parameter_types | result_types 
+------+-----------------------------+-----------------+--------------
+ q2   | PREPARE q2 AS SELECT 2 AS b | {}              | {integer}
 (1 row)
 
 DEALLOCATE PREPARE q2;
@@ -168,20 +178,20 @@ SELECT name, statement, parameter_types, result_types FROM pg_prepared_statement
 ------+------------------------------------------------------------------+----------------------------------------------------+--------------------------------------------------------------------------------------------------------------------------
  q2   | PREPARE q2(text) AS                                             +| {text}                                             | {name,boolean,boolean}
       |         SELECT datname, datistemplate, datallowconn             +|                                                    | 
-      |         FROM pg_database WHERE datname = $1;                     |                                                    | 
+      |         FROM pg_database WHERE datname = $1                      |                                                    | 
  q3   | PREPARE q3(text, int, float, boolean, smallint) AS              +| {text,integer,"double precision",boolean,smallint} | {integer,integer,integer,integer,integer,integer,integer,integer,integer,integer,integer,integer,integer,name,name,name}
       |         SELECT * FROM tenk1 WHERE string4 = $1 AND (four = $2 OR+|                                                    | 
       |         ten = $3::bigint OR true = $4 OR odd = $5::int)         +|                                                    | 
-      |         ORDER BY unique1;                                        |                                                    | 
+      |         ORDER BY unique1                                         |                                                    | 
  q5   | PREPARE q5(int, text) AS                                        +| {integer,text}                                     | {integer,integer,integer,integer,integer,integer,integer,integer,integer,integer,integer,integer,integer,name,name,name}
       |         SELECT * FROM tenk1 WHERE unique1 = $1 OR stringu1 = $2 +|                                                    | 
-      |         ORDER BY unique1;                                        |                                                    | 
+      |         ORDER BY unique1                                         |                                                    | 
  q6   | PREPARE q6 AS                                                   +| {integer,name}                                     | {integer,integer,integer,integer,integer,integer,integer,integer,integer,integer,integer,integer,integer,name,name,name}
-      |     SELECT * FROM tenk1 WHERE unique1 = $1 AND stringu1 = $2;    |                                                    | 
+      |     SELECT * FROM tenk1 WHERE unique1 = $1 AND stringu1 = $2     |                                                    | 
  q7   | PREPARE q7(unknown) AS                                          +| {path}                                             | {text,path}
-      |     SELECT * FROM road WHERE thepath = $1;                       |                                                    | 
+      |     SELECT * FROM road WHERE thepath = $1                        |                                                    | 
  q8   | PREPARE q8 AS                                                   +| {integer,name}                                     | 
-      |     UPDATE tenk1 SET stringu1 = $2 WHERE unique1 = $1;           |                                                    | 
+      |     UPDATE tenk1 SET stringu1 = $2 WHERE unique1 = $1            |                                                    | 
 (6 rows)
 
 -- test DEALLOCATE ALL;
diff --git a/src/test/regress/sql/prepare.sql b/src/test/regress/sql/prepare.sql
index c6098dc95ce..0e7fe44725e 100644
--- a/src/test/regress/sql/prepare.sql
+++ b/src/test/regress/sql/prepare.sql
@@ -4,7 +4,7 @@
 
 SELECT name, statement, parameter_types, result_types FROM pg_prepared_statements;
 
-PREPARE q1 AS SELECT 1 AS a;
+SELECT 'bingo'\;  PREPARE q1 AS SELECT 1 AS a \; SELECT 42;
 EXECUTE q1;
 
 SELECT name, statement, parameter_types, result_types FROM pg_prepared_statements;
-- 
2.52.0


--soawQPnav6xcFtRL--





^ permalink  raw  reply  [nested|flat] 22+ messages in thread


end of thread, other threads:[~2025-12-24 14:31 UTC | newest]

Thread overview: 22+ messages (download: mbox mbox.gz follow: Atom feed)
-- links below jump to the message on this page --
2020-03-08 22:15 [PATCH v9 04/11] Add pg_ls_dir_metadata to list a dir with file metadata.. Justin Pryzby <pryzbyj@telsasoft.com>
2020-03-08 22:15 [PATCH v9 04/11] Add pg_ls_dir_metadata to list a dir with file metadata.. Justin Pryzby <pryzbyj@telsasoft.com>
2020-03-10 03:40 [PATCH v33 03/11] Add pg_ls_dir_metadata to list a dir with file metadata.. Justin Pryzby <pryzbyj@telsasoft.com>
2020-03-10 03:40 [PATCH v30 03/11] Add pg_ls_dir_metadata to list a dir with file metadata.. Justin Pryzby <pryzbyj@telsasoft.com>
2020-03-10 03:40 [PATCH v15 04/10] Add pg_ls_dir_metadata to list a dir with file metadata.. Justin Pryzby <pryzbyj@telsasoft.com>
2020-03-10 03:40 [PATCH v37 03/11] Add pg_ls_dir_metadata to list a dir with file metadata.. Justin Pryzby <pryzbyj@telsasoft.com>
2020-03-10 03:40 [PATCH v35 3/7] Add pg_ls_dir_metadata to list a dir with file metadata.. Justin Pryzby <pryzbyj@telsasoft.com>
2020-03-10 03:40 [PATCH v30 03/11] Add pg_ls_dir_metadata to list a dir with file metadata.. Justin Pryzby <pryzbyj@telsasoft.com>
2020-03-10 03:40 [PATCH v28 03/11] Add pg_ls_dir_metadata to list a dir with file metadata.. Justin Pryzby <pryzbyj@telsasoft.com>
2020-03-10 03:40 [PATCH v19 04/10] Add pg_ls_dir_metadata to list a dir with file metadata.. Justin Pryzby <pryzbyj@telsasoft.com>
2020-03-10 03:40 [PATCH v31 03/11] Add pg_ls_dir_metadata to list a dir with file metadata.. Justin Pryzby <pryzbyj@telsasoft.com>
2020-03-10 03:40 [PATCH v20 04/10] Add pg_ls_dir_metadata to list a dir with file metadata.. Justin Pryzby <pryzbyj@telsasoft.com>
2020-03-10 03:40 [PATCH v36 3/7] Add pg_ls_dir_metadata to list a dir with file metadata.. Justin Pryzby <pryzbyj@telsasoft.com>
2020-03-10 03:40 [PATCH v32 03/11] Add pg_ls_dir_metadata to list a dir with file metadata.. Justin Pryzby <pryzbyj@telsasoft.com>
2020-03-10 03:40 [PATCH v24 04/11] Add pg_ls_dir_metadata to list a dir with file metadata.. Justin Pryzby <pryzbyj@telsasoft.com>
2020-03-10 03:40 [PATCH v15 04/10] Add pg_ls_dir_metadata to list a dir with file metadata.. Justin Pryzby <pryzbyj@telsasoft.com>
2020-03-10 03:40 [PATCH v31 03/11] Add pg_ls_dir_metadata to list a dir with file metadata.. Justin Pryzby <pryzbyj@telsasoft.com>
2020-03-10 03:40 [PATCH v22 04/10] Add pg_ls_dir_metadata to list a dir with file metadata.. Justin Pryzby <pryzbyj@telsasoft.com>
2020-03-10 03:40 [PATCH v21 04/10] Add pg_ls_dir_metadata to list a dir with file metadata.. Justin Pryzby <pryzbyj@telsasoft.com>
2020-03-10 03:40 [PATCH v25 04/11] Add pg_ls_dir_metadata to list a dir with file metadata.. Justin Pryzby <pryzbyj@telsasoft.com>
2020-03-10 03:40 [PATCH v34 03/15] Add pg_ls_dir_metadata to list a dir with file metadata.. Justin Pryzby <pryzbyj@telsasoft.com>
2025-12-24 14:31 [PATCH v2] Cleanup explicit PREPARE query strings Julien Rouhaud <julien.rouhaud@free.fr>

This inbox is served by agora; see mirroring instructions
for how to clone and mirror all data and code used for this inbox