public inbox for [email protected]
help / color / mirror / Atom feed[PATCH 2/4] Change-tmplinit-argument
6+ messages / 5 participants
[nested] [flat]
* [PATCH 2/4] Change-tmplinit-argument
@ 2019-01-17 12:05 Arthur Zakirov <[email protected]>
0 siblings, 0 replies; 6+ messages in thread
From: Arthur Zakirov @ 2019-01-17 12:05 UTC (permalink / raw)
---
contrib/dict_int/dict_int.c | 4 +-
contrib/dict_xsyn/dict_xsyn.c | 4 +-
contrib/unaccent/unaccent.c | 4 +-
src/backend/commands/tsearchcmds.c | 10 ++++-
src/backend/snowball/dict_snowball.c | 4 +-
src/backend/tsearch/dict_ispell.c | 4 +-
src/backend/tsearch/dict_simple.c | 4 +-
src/backend/tsearch/dict_synonym.c | 4 +-
src/backend/tsearch/dict_thesaurus.c | 4 +-
src/backend/utils/cache/ts_cache.c | 13 +++++-
src/include/tsearch/ts_cache.h | 4 ++
src/include/tsearch/ts_public.h | 67 ++++++++++++++++++++++++++--
12 files changed, 105 insertions(+), 21 deletions(-)
diff --git a/contrib/dict_int/dict_int.c b/contrib/dict_int/dict_int.c
index 628b9769c3..ddde55eee4 100644
--- a/contrib/dict_int/dict_int.c
+++ b/contrib/dict_int/dict_int.c
@@ -30,7 +30,7 @@ PG_FUNCTION_INFO_V1(dintdict_lexize);
Datum
dintdict_init(PG_FUNCTION_ARGS)
{
- List *dictoptions = (List *) PG_GETARG_POINTER(0);
+ DictInitData *init_data = (DictInitData *) PG_GETARG_POINTER(0);
DictInt *d;
ListCell *l;
@@ -38,7 +38,7 @@ dintdict_init(PG_FUNCTION_ARGS)
d->maxlen = 6;
d->rejectlong = false;
- foreach(l, dictoptions)
+ foreach(l, init_data->dict_options)
{
DefElem *defel = (DefElem *) lfirst(l);
diff --git a/contrib/dict_xsyn/dict_xsyn.c b/contrib/dict_xsyn/dict_xsyn.c
index 509e14aee0..15b1a0033a 100644
--- a/contrib/dict_xsyn/dict_xsyn.c
+++ b/contrib/dict_xsyn/dict_xsyn.c
@@ -140,7 +140,7 @@ read_dictionary(DictSyn *d, const char *filename)
Datum
dxsyn_init(PG_FUNCTION_ARGS)
{
- List *dictoptions = (List *) PG_GETARG_POINTER(0);
+ DictInitData *init_data = (DictInitData *) PG_GETARG_POINTER(0);
DictSyn *d;
ListCell *l;
char *filename = NULL;
@@ -153,7 +153,7 @@ dxsyn_init(PG_FUNCTION_ARGS)
d->matchsynonyms = false;
d->keepsynonyms = true;
- foreach(l, dictoptions)
+ foreach(l, init_data->dict_options)
{
DefElem *defel = (DefElem *) lfirst(l);
diff --git a/contrib/unaccent/unaccent.c b/contrib/unaccent/unaccent.c
index fc5176e338..f3663cefd0 100644
--- a/contrib/unaccent/unaccent.c
+++ b/contrib/unaccent/unaccent.c
@@ -270,12 +270,12 @@ PG_FUNCTION_INFO_V1(unaccent_init);
Datum
unaccent_init(PG_FUNCTION_ARGS)
{
- List *dictoptions = (List *) PG_GETARG_POINTER(0);
+ DictInitData *init_data = (DictInitData *) PG_GETARG_POINTER(0);
TrieChar *rootTrie = NULL;
bool fileloaded = false;
ListCell *l;
- foreach(l, dictoptions)
+ foreach(l, init_data->dict_options)
{
DefElem *defel = (DefElem *) lfirst(l);
diff --git a/src/backend/commands/tsearchcmds.c b/src/backend/commands/tsearchcmds.c
index 8e5eec22b5..30c5eb72a2 100644
--- a/src/backend/commands/tsearchcmds.c
+++ b/src/backend/commands/tsearchcmds.c
@@ -389,17 +389,25 @@ verify_dictoptions(Oid tmplId, List *dictoptions)
}
else
{
+ DictInitData init_data;
+
/*
* Copy the options just in case init method thinks it can scribble on
* them ...
*/
dictoptions = copyObject(dictoptions);
+ init_data.dict_options = dictoptions;
+ init_data.dict.id = InvalidOid;
+ init_data.dict.xmin = InvalidTransactionId;
+ init_data.dict.xmax = InvalidTransactionId;
+ ItemPointerSetInvalid(&init_data.dict.tid);
+
/*
* Call the init method and see if it complains. We don't worry about
* it leaking memory, since our command will soon be over anyway.
*/
- (void) OidFunctionCall1(initmethod, PointerGetDatum(dictoptions));
+ (void) OidFunctionCall1(initmethod, PointerGetDatum(&init_data));
}
ReleaseSysCache(tup);
diff --git a/src/backend/snowball/dict_snowball.c b/src/backend/snowball/dict_snowball.c
index 5166738310..f30f29865c 100644
--- a/src/backend/snowball/dict_snowball.c
+++ b/src/backend/snowball/dict_snowball.c
@@ -201,14 +201,14 @@ locate_stem_module(DictSnowball *d, const char *lang)
Datum
dsnowball_init(PG_FUNCTION_ARGS)
{
- List *dictoptions = (List *) PG_GETARG_POINTER(0);
+ DictInitData *init_data = (DictInitData *) PG_GETARG_POINTER(0);
DictSnowball *d;
bool stoploaded = false;
ListCell *l;
d = (DictSnowball *) palloc0(sizeof(DictSnowball));
- foreach(l, dictoptions)
+ foreach(l, init_data->dict_options)
{
DefElem *defel = (DefElem *) lfirst(l);
diff --git a/src/backend/tsearch/dict_ispell.c b/src/backend/tsearch/dict_ispell.c
index 8b05a477f1..fc9a96abca 100644
--- a/src/backend/tsearch/dict_ispell.c
+++ b/src/backend/tsearch/dict_ispell.c
@@ -29,7 +29,7 @@ typedef struct
Datum
dispell_init(PG_FUNCTION_ARGS)
{
- List *dictoptions = (List *) PG_GETARG_POINTER(0);
+ DictInitData *init_data = (DictInitData *) PG_GETARG_POINTER(0);
DictISpell *d;
bool affloaded = false,
dictloaded = false,
@@ -40,7 +40,7 @@ dispell_init(PG_FUNCTION_ARGS)
NIStartBuild(&(d->obj));
- foreach(l, dictoptions)
+ foreach(l, init_data->dict_options)
{
DefElem *defel = (DefElem *) lfirst(l);
diff --git a/src/backend/tsearch/dict_simple.c b/src/backend/tsearch/dict_simple.c
index 2f62ef00c8..c92744641b 100644
--- a/src/backend/tsearch/dict_simple.c
+++ b/src/backend/tsearch/dict_simple.c
@@ -29,7 +29,7 @@ typedef struct
Datum
dsimple_init(PG_FUNCTION_ARGS)
{
- List *dictoptions = (List *) PG_GETARG_POINTER(0);
+ DictInitData *init_data = (DictInitData *) PG_GETARG_POINTER(0);
DictSimple *d = (DictSimple *) palloc0(sizeof(DictSimple));
bool stoploaded = false,
acceptloaded = false;
@@ -37,7 +37,7 @@ dsimple_init(PG_FUNCTION_ARGS)
d->accept = true; /* default */
- foreach(l, dictoptions)
+ foreach(l, init_data->dict_options)
{
DefElem *defel = (DefElem *) lfirst(l);
diff --git a/src/backend/tsearch/dict_synonym.c b/src/backend/tsearch/dict_synonym.c
index b6226df940..d3f5f0da3f 100644
--- a/src/backend/tsearch/dict_synonym.c
+++ b/src/backend/tsearch/dict_synonym.c
@@ -91,7 +91,7 @@ compareSyn(const void *a, const void *b)
Datum
dsynonym_init(PG_FUNCTION_ARGS)
{
- List *dictoptions = (List *) PG_GETARG_POINTER(0);
+ DictInitData *init_data = (DictInitData *) PG_GETARG_POINTER(0);
DictSyn *d;
ListCell *l;
char *filename = NULL;
@@ -104,7 +104,7 @@ dsynonym_init(PG_FUNCTION_ARGS)
char *line = NULL;
uint16 flags = 0;
- foreach(l, dictoptions)
+ foreach(l, init_data->dict_options)
{
DefElem *defel = (DefElem *) lfirst(l);
diff --git a/src/backend/tsearch/dict_thesaurus.c b/src/backend/tsearch/dict_thesaurus.c
index 75f8deef6a..8962e252e0 100644
--- a/src/backend/tsearch/dict_thesaurus.c
+++ b/src/backend/tsearch/dict_thesaurus.c
@@ -604,7 +604,7 @@ compileTheSubstitute(DictThesaurus *d)
Datum
thesaurus_init(PG_FUNCTION_ARGS)
{
- List *dictoptions = (List *) PG_GETARG_POINTER(0);
+ DictInitData *init_data = (DictInitData *) PG_GETARG_POINTER(0);
DictThesaurus *d;
char *subdictname = NULL;
bool fileloaded = false;
@@ -612,7 +612,7 @@ thesaurus_init(PG_FUNCTION_ARGS)
d = (DictThesaurus *) palloc0(sizeof(DictThesaurus));
- foreach(l, dictoptions)
+ foreach(l, init_data->dict_options)
{
DefElem *defel = (DefElem *) lfirst(l);
diff --git a/src/backend/utils/cache/ts_cache.c b/src/backend/utils/cache/ts_cache.c
index 0545efc75b..8bc8d82c76 100644
--- a/src/backend/utils/cache/ts_cache.c
+++ b/src/backend/utils/cache/ts_cache.c
@@ -39,6 +39,7 @@
#include "catalog/pg_ts_template.h"
#include "commands/defrem.h"
#include "tsearch/ts_cache.h"
+#include "tsearch/ts_public.h"
#include "utils/builtins.h"
#include "utils/catcache.h"
#include "utils/fmgroids.h"
@@ -311,11 +312,15 @@ lookup_ts_dictionary_cache(Oid dictId)
MemSet(entry, 0, sizeof(TSDictionaryCacheEntry));
entry->dictId = dictId;
entry->dictCtx = saveCtx;
+ entry->dict_xmin = HeapTupleHeaderGetRawXmin(tpdict->t_data);
+ entry->dict_xmax = HeapTupleHeaderGetRawXmax(tpdict->t_data);
+ entry->dict_tid = tpdict->t_self;
entry->lexizeOid = template->tmpllexize;
if (OidIsValid(template->tmplinit))
{
+ DictInitData init_data;
List *dictoptions;
Datum opt;
bool isnull;
@@ -335,9 +340,15 @@ lookup_ts_dictionary_cache(Oid dictId)
else
dictoptions = deserialize_deflist(opt);
+ init_data.dict_options = dictoptions;
+ init_data.dict.id = dictId;
+ init_data.dict.xmin = entry->dict_xmin;
+ init_data.dict.xmax = entry->dict_xmax;
+ init_data.dict.tid = entry->dict_tid;
+
entry->dictData =
DatumGetPointer(OidFunctionCall1(template->tmplinit,
- PointerGetDatum(dictoptions)));
+ PointerGetDatum(&init_data)));
MemoryContextSwitchTo(oldcontext);
}
diff --git a/src/include/tsearch/ts_cache.h b/src/include/tsearch/ts_cache.h
index 77e325d101..2298e0a275 100644
--- a/src/include/tsearch/ts_cache.h
+++ b/src/include/tsearch/ts_cache.h
@@ -54,6 +54,10 @@ typedef struct TSDictionaryCacheEntry
Oid dictId;
bool isvalid;
+ TransactionId dict_xmin; /* XMIN of the dictionary's tuple */
+ TransactionId dict_xmax; /* XMAX of the dictionary's tuple */
+ ItemPointerData dict_tid; /* TID of the dictionary's tuple */
+
/* most frequent fmgr call */
Oid lexizeOid;
FmgrInfo lexize;
diff --git a/src/include/tsearch/ts_public.h b/src/include/tsearch/ts_public.h
index b325fa122c..db028ed6ad 100644
--- a/src/include/tsearch/ts_public.h
+++ b/src/include/tsearch/ts_public.h
@@ -13,6 +13,8 @@
#ifndef _PG_TS_PUBLIC_H_
#define _PG_TS_PUBLIC_H_
+#include "nodes/pg_list.h"
+#include "storage/itemptr.h"
#include "tsearch/ts_type.h"
/*
@@ -81,10 +83,69 @@ extern void readstoplist(const char *fname, StopList *s,
extern bool searchstoplist(StopList *s, char *key);
/*
- * Interface with dictionaries
+ * API for text search dictionaries.
+ *
+ * API functions to manage a text search dictionary are defined by a text search
+ * template. Currently an existing template cannot be altered in order to use
+ * different functions. API consists of the following functions:
+ *
+ * init function
+ * -------------
+ * - optional function which initializes internal structures of the dictionary
+ * - accepts DictInitData structure as an argument and must return a custom
+ * palloc'd structure which stores content of the processed dictionary and
+ * is used by lexize function
+ *
+ * lexize function
+ * ---------------
+ * - normalizes a single word (token) using specific dictionary
+ * - returns a palloc'd array of TSLexeme, with a terminating NULL entry
+ * - accepts the following arguments:
+ *
+ * - dictData - pointer to a structure returned by init function or NULL if
+ * init function wasn't defined by the template
+ * - token - string to normalize (not null-terminated)
+ * - length - length of the token
+ * - dictState - pointer to a DictSubState structure storing current
+ * state of a set of tokens processing and allows to normalize phrases
+ */
+
+/*
+ * A preprocessed dictionary can be stored in shared memory using DSM - this is
+ * decided in the init function. A DSM segment is released after altering or
+ * dropping the dictionary. The segment may still leak, when a backend uses the
+ * dictionary right before dropping - in that case the backend will hold the DSM
+ * untill it disconnects or calls lookup_ts_dictionary_cache().
+ *
+ * DictEntryData represents DSM segment with a preprocessed dictionary. We need
+ * to ensure the content of the DSM segment is still valid, which is what xmin,
+ * xmax and tid are for.
+ */
+typedef struct
+{
+ Oid id; /* OID of the dictionary */
+ TransactionId xmin; /* XMIN of the dictionary's tuple */
+ TransactionId xmax; /* XMAX of the dictionary's tuple */
+ ItemPointerData tid; /* TID of the dictionary's tuple */
+} DictEntryData;
+
+/*
+ * API structure for a dictionary initialization. It is passed as an argument
+ * to a template's init function.
*/
+typedef struct
+{
+ /* List of options for a template's init method */
+ List *dict_options;
+
+ /* Data used to allocate, search and release the DSM segment */
+ DictEntryData dict;
+} DictInitData;
-/* return struct for any lexize function */
+/*
+ * Return struct for any lexize function. They are combined into an array, the
+ * last entry is the terminating entry.
+ */
typedef struct
{
/*----------
@@ -108,7 +169,7 @@ typedef struct
uint16 flags; /* See flag bits below */
- char *lexeme; /* C string */
+ char *lexeme; /* C string (NULL for terminating entry) */
} TSLexeme;
/* Flag bits that can appear in TSLexeme.flags */
--
2.20.1
--------------9E285E8AF9A980CD1515771F
Content-Type: text/x-patch;
name="0003-Retrieve-shared-location-for-dict-v18.patch"
Content-Transfer-Encoding: 7bit
Content-Disposition: attachment;
filename="0003-Retrieve-shared-location-for-dict-v18.patch"
^ permalink raw reply [nested|flat] 6+ messages in thread
* [PATCH v1 2/7] Row pattern recognition patch (parse/analysis).
@ 2023-06-25 11:48 Tatsuo Ishii <[email protected]>
0 siblings, 0 replies; 6+ messages in thread
From: Tatsuo Ishii @ 2023-06-25 11:48 UTC (permalink / raw)
---
src/backend/parser/parse_agg.c | 7 ++
src/backend/parser/parse_clause.c | 196 +++++++++++++++++++++++++++++-
src/backend/parser/parse_expr.c | 4 +
src/backend/parser/parse_func.c | 3 +
4 files changed, 209 insertions(+), 1 deletion(-)
diff --git a/src/backend/parser/parse_agg.c b/src/backend/parser/parse_agg.c
index 85cd47b7ae..aa7a1cee80 100644
--- a/src/backend/parser/parse_agg.c
+++ b/src/backend/parser/parse_agg.c
@@ -564,6 +564,10 @@ check_agglevels_and_constraints(ParseState *pstate, Node *expr)
errkind = true;
break;
+ case EXPR_KIND_RPR_DEFINE:
+ errkind = true;
+ break;
+
/*
* There is intentionally no default: case here, so that the
* compiler will warn if we add a new ParseExprKind without
@@ -953,6 +957,9 @@ transformWindowFuncCall(ParseState *pstate, WindowFunc *wfunc,
case EXPR_KIND_CYCLE_MARK:
errkind = true;
break;
+ case EXPR_KIND_RPR_DEFINE:
+ errkind = true;
+ break;
/*
* There is intentionally no default: case here, so that the
diff --git a/src/backend/parser/parse_clause.c b/src/backend/parser/parse_clause.c
index f61f794755..5bb11add3c 100644
--- a/src/backend/parser/parse_clause.c
+++ b/src/backend/parser/parse_clause.c
@@ -100,7 +100,10 @@ static WindowClause *findWindowClause(List *wclist, const char *name);
static Node *transformFrameOffset(ParseState *pstate, int frameOptions,
Oid rangeopfamily, Oid rangeopcintype, Oid *inRangeFunc,
Node *clause);
-
+static void transformRPR(ParseState *pstate, WindowClause *wc, WindowDef *windef);
+static List *transformDefineClause(ParseState *pstate, WindowClause *wc, WindowDef *windef);
+static List *transformPatternClause(ParseState *pstate, WindowClause *wc, WindowDef *windef);
+static List *transformMeasureClause(ParseState *pstate, WindowClause *wc, WindowDef *windef);
/*
* transformFromClause -
@@ -2949,6 +2952,10 @@ transformWindowDefinitions(ParseState *pstate,
rangeopfamily, rangeopcintype,
&wc->endInRangeFunc,
windef->endOffset);
+
+ /* Process Row Pattern Recognition related clauses */
+ transformRPR(pstate, wc, windef);
+
wc->runCondition = NIL;
wc->winref = winref;
@@ -3814,3 +3821,190 @@ transformFrameOffset(ParseState *pstate, int frameOptions,
return node;
}
+
+/*
+ * transformRPR
+ * Process Row Pattern Recognition related clauses
+ */
+static void
+transformRPR(ParseState *pstate, WindowClause *wc, WindowDef *windef)
+{
+ /* Check Frame option. Frame must start at current row */
+
+ /*
+ * Window definition exists?
+ */
+ if (windef == NULL)
+ return;
+
+ /*
+ * Row Pattern Common Syntax clause exists?
+ */
+ if (windef->rpCommonSyntax == NULL)
+ return;
+
+ if ((wc->frameOptions & FRAMEOPTION_START_CURRENT_ROW) == 0)
+ ereport(ERROR,
+ (errcode(ERRCODE_SYNTAX_ERROR),
+ errmsg("FRAME must start at current row when row patttern recognition is used")));
+
+ /* Transform AFTER MACH SKIP TO clause */
+ wc->rpSkipTo = windef->rpCommonSyntax->rpSkipTo;
+ if (wc->rpSkipTo != ST_NEXT_ROW)
+ ereport(ERROR,
+ (errcode(ERRCODE_SYNTAX_ERROR),
+ errmsg("only AFTER MATCH SKIP TO NEXT_ROW is supported")));
+
+ /* Transform SEEK or INITIAL clause */
+ wc->initial = windef->rpCommonSyntax->initial;
+
+ /* Transform DEFINE clause into list of TargetEntry's */
+ wc->defineClause = transformDefineClause(pstate, wc, windef);
+
+ /* Check PATTERN clause and copy to patternClause */
+ transformPatternClause(pstate, wc, windef);
+
+ /* Transform MEASURE clause */
+ transformMeasureClause(pstate, wc, windef);
+}
+
+/*
+ * transformDefineClause Process DEFINE clause and transform ResTarget into
+ * list of TargetEntry.
+ *
+ * XXX we only support column reference in row pattern definition search
+ * condition, e.g. "price". <row pattern definition variable name>.<column
+ * reference> is not supported, e.g. "A.price".
+ */
+static List *
+transformDefineClause(ParseState *pstate, WindowClause *wc, WindowDef *windef)
+{
+ ListCell *lc;
+ ResTarget *restarget, *r;
+ List *restargets;
+
+
+ /*
+ * If Row Definition Common Syntax exists, DEFINE clause must exist.
+ * (the raw parser should have already checked it.)
+ */
+ Assert(windef->rpCommonSyntax->rpDefs != NULL);
+
+ /*
+ * Check for duplicate row pattern definition variables. The standard
+ * requires that no two row pattern definition variable names shall be
+ * equivalent.
+ */
+ restargets = NIL;
+ foreach(lc, windef->rpCommonSyntax->rpDefs)
+ {
+ char *name;
+ ListCell *l;
+
+ restarget = (ResTarget *)lfirst(lc);
+ name = restarget->name;
+
+ /*
+ * Make sure that row pattern definition search condition is a boolean
+ * expression.
+ */
+ transformWhereClause(pstate, restarget->val,
+ EXPR_KIND_RPR_DEFINE, "DEFINE");
+
+ foreach(l, restargets)
+ {
+ char *n;
+
+ r = (ResTarget *) lfirst(l);
+ n = r->name;
+
+ if (!strcmp(n, name))
+ ereport(ERROR,
+ (errcode(ERRCODE_SYNTAX_ERROR),
+ errmsg("row pattern definition variable name \"%s\" appears more than once in DEFINE clause",
+ name),
+ parser_errposition(pstate, exprLocation((Node *)r))));
+ }
+ restargets = lappend(restargets, restarget);
+ }
+ list_free(restargets);
+
+ return transformTargetList(pstate, windef->rpCommonSyntax->rpDefs,
+ EXPR_KIND_RPR_DEFINE);
+}
+
+/*
+ * transformPatternClause
+ * Process PATTERN clause and return PATTERN clause in the raw parse tree
+ */
+static List *
+transformPatternClause(ParseState *pstate, WindowClause *wc, WindowDef *windef)
+{
+ List *patterns;
+ ListCell *lc, *l;
+
+ /*
+ * Row Pattern Common Syntax clause exists?
+ */
+ if (windef->rpCommonSyntax == NULL)
+ return NULL;
+
+ /*
+ * Primary row pattern variable names in PATTERN clause must appear in
+ * DEFINE clause as row pattern definition variable names.
+ */
+ wc->patternVariable = NIL;
+ wc->patternRegexp = NIL;
+ foreach(lc, windef->rpCommonSyntax->rpPatterns)
+ {
+ A_Expr *a;
+ char *name;
+ char *regexp;
+ bool found = false;
+
+ if (!IsA(lfirst(lc), A_Expr))
+ ereport(ERROR,
+ errmsg("node type is not A_Expr"));
+
+ a = (A_Expr *)lfirst(lc);
+ name = strVal(a->lexpr);
+
+ foreach(l, windef->rpCommonSyntax->rpDefs)
+ {
+ ResTarget *restarget = (ResTarget *)lfirst(l);
+
+ if (!strcmp(restarget->name, name))
+ {
+ found = true;
+ break;
+ }
+ }
+
+ if (!found)
+ ereport(ERROR,
+ (errcode(ERRCODE_SYNTAX_ERROR),
+ errmsg("primary row pattern variable name \"%s\" does not appear in DEFINE clause",
+ name),
+ parser_errposition(pstate, exprLocation((Node *)a))));
+ wc->patternVariable = lappend(wc->patternVariable, makeString(pstrdup(name)));
+ regexp = strVal(lfirst(list_head(a->name)));
+ wc->patternRegexp = lappend(wc->patternRegexp, makeString(pstrdup(regexp)));
+ }
+ return patterns;
+}
+
+/*
+ * transformMeasureClause
+ * Process MEASURE clause
+ */
+static List *
+transformMeasureClause(ParseState *pstate, WindowClause *wc, WindowDef *windef)
+{
+ if (windef->rowPatternMeasures == NIL)
+ return NIL;
+
+ ereport(ERROR,
+ (errcode(ERRCODE_SYNTAX_ERROR),
+ errmsg("%s","MEASURE clause is not supported yet"),
+ parser_errposition(pstate, exprLocation((Node *)windef->rowPatternMeasures))));
+}
diff --git a/src/backend/parser/parse_expr.c b/src/backend/parser/parse_expr.c
index 346fd272b6..20231d9ec0 100644
--- a/src/backend/parser/parse_expr.c
+++ b/src/backend/parser/parse_expr.c
@@ -541,6 +541,7 @@ transformColumnRef(ParseState *pstate, ColumnRef *cref)
case EXPR_KIND_COPY_WHERE:
case EXPR_KIND_GENERATED_COLUMN:
case EXPR_KIND_CYCLE_MARK:
+ case EXPR_KIND_RPR_DEFINE:
/* okay */
break;
@@ -1754,6 +1755,7 @@ transformSubLink(ParseState *pstate, SubLink *sublink)
case EXPR_KIND_VALUES:
case EXPR_KIND_VALUES_SINGLE:
case EXPR_KIND_CYCLE_MARK:
+ case EXPR_KIND_RPR_DEFINE:
/* okay */
break;
case EXPR_KIND_CHECK_CONSTRAINT:
@@ -3133,6 +3135,8 @@ ParseExprKindName(ParseExprKind exprKind)
return "GENERATED AS";
case EXPR_KIND_CYCLE_MARK:
return "CYCLE";
+ case EXPR_KIND_RPR_DEFINE:
+ return "DEFINE";
/*
* There is intentionally no default: case here, so that the
diff --git a/src/backend/parser/parse_func.c b/src/backend/parser/parse_func.c
index b3f0b6a137..2ff3699538 100644
--- a/src/backend/parser/parse_func.c
+++ b/src/backend/parser/parse_func.c
@@ -2656,6 +2656,9 @@ check_srf_call_placement(ParseState *pstate, Node *last_srf, int location)
case EXPR_KIND_CYCLE_MARK:
errkind = true;
break;
+ case EXPR_KIND_RPR_DEFINE:
+ errkind = true;
+ break;
/*
* There is intentionally no default: case here, so that the
--
2.25.1
----Next_Part(Sun_Jun_25_21_05_09_2023_126)--
Content-Type: Text/X-Patch; charset=us-ascii
Content-Transfer-Encoding: 7bit
Content-Disposition: inline;
filename="v1-0003-Row-pattern-recognition-patch-planner.patch"
^ permalink raw reply [nested|flat] 6+ messages in thread
* Re: [18] Policy on IMMUTABLE functions and Unicode updates
@ 2024-07-24 19:12 Jeff Davis <[email protected]>
0 siblings, 1 reply; 6+ messages in thread
From: Jeff Davis @ 2024-07-24 19:12 UTC (permalink / raw)
To: Robert Haas <[email protected]>; +Cc: Peter Eisentraut <[email protected]>; Laurenz Albe <[email protected]>; pgsql-hackers; Daniel Verite <[email protected]>; Noah Misch <[email protected]>; Tom Lane <[email protected]>; Jeremy Schneider <[email protected]>
On Wed, 2024-07-24 at 14:47 -0400, Robert Haas wrote:
> On Wed, Jul 24, 2024 at 1:45 PM Jeff Davis <[email protected]> wrote:
> > There's a qualitative difference between a collation update which
> > can
> > break your PKs and FKs, and a ctype update which definitely will
> > not.
>
> I don't think that's true. All you need is a unique index on
> UPPER(somecol).
Primary keys are on plain column references, not expressions; and don't
support WHERE clauses, so I don't see how a ctype update would affect a
PK.
In any case, you are correct that Unicode updates could put some
constraints at risk, including unique indexes, CHECK, and partition
constraints. But someone has to actually use one of the affected
functions somewhere, and that's the main distinction that I'm trying to
draw.
The reason why collation is qualitatively a much bigger problem is
because there's no obvious indication that you are doing anything
related to collation at all. A very plain "CREATE TABLE x(t text
PRIMARY KEY)" is at risk.
Regards,
Jeff Davis
^ permalink raw reply [nested|flat] 6+ messages in thread
* Re: [18] Policy on IMMUTABLE functions and Unicode updates
@ 2024-07-24 19:19 Robert Haas <[email protected]>
parent: Jeff Davis <[email protected]>
0 siblings, 1 reply; 6+ messages in thread
From: Robert Haas @ 2024-07-24 19:19 UTC (permalink / raw)
To: Jeff Davis <[email protected]>; +Cc: Peter Eisentraut <[email protected]>; Laurenz Albe <[email protected]>; pgsql-hackers; Daniel Verite <[email protected]>; Noah Misch <[email protected]>; Tom Lane <[email protected]>; Jeremy Schneider <[email protected]>
On Wed, Jul 24, 2024 at 3:12 PM Jeff Davis <[email protected]> wrote:
> In any case, you are correct that Unicode updates could put some
> constraints at risk, including unique indexes, CHECK, and partition
> constraints. But someone has to actually use one of the affected
> functions somewhere, and that's the main distinction that I'm trying to
> draw.
>
> The reason why collation is qualitatively a much bigger problem is
> because there's no obvious indication that you are doing anything
> related to collation at all. A very plain "CREATE TABLE x(t text
> PRIMARY KEY)" is at risk.
Well, I don't know. I agree that collation is a much bigger problem,
but not for that reason. I think a user who is familiar with the
problems in this area will see the danger either way, and one who
isn't, won't. For me, the only real difference is that a unique index
on a text column is a lot more common than one that involves UPPER.
--
Robert Haas
EDB: http://www.enterprisedb.com
^ permalink raw reply [nested|flat] 6+ messages in thread
* Re: [18] Policy on IMMUTABLE functions and Unicode updates
@ 2024-07-24 19:43 Jeremy Schneider <[email protected]>
parent: Robert Haas <[email protected]>
0 siblings, 1 reply; 6+ messages in thread
From: Jeremy Schneider @ 2024-07-24 19:43 UTC (permalink / raw)
To: Robert Haas <[email protected]>; +Cc: Daniel Verite <[email protected]>; Jeff Davis <[email protected]>; Laurenz Albe <[email protected]>; Noah Misch <[email protected]>; Peter Eisentraut <[email protected]>; Tom Lane <[email protected]>; pgsql-hackers
On Wed, Jul 24, 2024 at 12:47 PM Robert Haas <[email protected]> wrote:
On Wed, Jul 24, 2024 at 1:45 PM Jeff Davis <[email protected]> wrote:
> There's a qualitative difference between a collation update which can
> break your PKs and FKs, and a ctype update which definitely will not.
I don't think that's true. All you need is a unique index on UPPER(somecol).
I doubt it’s common to have unique on upper()
But non-unique indexes for case insensitive searches will be more common.
Historically this is the most common way people did case insensitive on
oracle.
Changing ctype would mean these queries can return wrong results
The impact would be similar to the critical problem TripAdvisor hit in 2014
with their read replicas, in the Postgres email thread I linked above
-Jeremy
^ permalink raw reply [nested|flat] 6+ messages in thread
* Re: [18] Policy on IMMUTABLE functions and Unicode updates
@ 2024-07-24 19:47 Robert Haas <[email protected]>
parent: Jeremy Schneider <[email protected]>
0 siblings, 0 replies; 6+ messages in thread
From: Robert Haas @ 2024-07-24 19:47 UTC (permalink / raw)
To: Jeremy Schneider <[email protected]>; +Cc: Daniel Verite <[email protected]>; Jeff Davis <[email protected]>; Laurenz Albe <[email protected]>; Noah Misch <[email protected]>; Peter Eisentraut <[email protected]>; Tom Lane <[email protected]>; pgsql-hackers
On Wed, Jul 24, 2024 at 3:43 PM Jeremy Schneider
<[email protected]> wrote:
> But non-unique indexes for case insensitive searches will be more common. Historically this is the most common way people did case insensitive on oracle.
>
> Changing ctype would mean these queries can return wrong results
Yeah. I mentioned earlier that I very recently saw a customer query
with UPPER() in the join condition. If someone is doing foo JOIN bar
ON upper(foo.x) = upper(bar.x), it is not unlikely that one or both of
those expressions are indexed. Not guaranteed, of course, but very
plausible.
--
Robert Haas
EDB: http://www.enterprisedb.com
^ permalink raw reply [nested|flat] 6+ messages in thread
end of thread, other threads:[~2024-07-24 19:47 UTC | newest]
Thread overview: 6+ messages (download: mbox mbox.gz follow: Atom feed)
-- links below jump to the message on this page --
2019-01-17 12:05 [PATCH 2/4] Change-tmplinit-argument Arthur Zakirov <[email protected]>
2023-06-25 11:48 [PATCH v1 2/7] Row pattern recognition patch (parse/analysis). Tatsuo Ishii <[email protected]>
2024-07-24 19:12 Re: [18] Policy on IMMUTABLE functions and Unicode updates Jeff Davis <[email protected]>
2024-07-24 19:19 ` Re: [18] Policy on IMMUTABLE functions and Unicode updates Robert Haas <[email protected]>
2024-07-24 19:43 ` Re: [18] Policy on IMMUTABLE functions and Unicode updates Jeremy Schneider <[email protected]>
2024-07-24 19:47 ` Re: [18] Policy on IMMUTABLE functions and Unicode updates Robert Haas <[email protected]>
This inbox is served by agora; see mirroring instructions
for how to clone and mirror all data and code used for this inbox