From 47eddc78c263590e7a7ce7f002ec11bddba43a54 Mon Sep 17 00:00:00 2001 From: hari krishna Date: Tue, 10 Jan 2023 15:35:38 +0530 Subject: [PATCH v13] Implement input functions for extended statistics types In order restore a dumped extended statistics (stxdndistinct, stxddependencies, stxdmcv) we need to provide input functions to parse pg_distinct/pg_dependency/pg_mcv_list strings. Today we get the ERROR "cannot accept a value of type pg_ndistinct/pg_dependencies/pg_mcv_list" when we try to do an insert of any type. Approach : - Using yacc grammar file (statistics_gram.y) to parse the input string to its internal format for the types pg_distinct and pg_dependencies - We are just calling byteain() for serialized input text of type pg_mcv_list. - Add regress testcases for intput functions pg_dependencies_in and pg_ndistinct_in Usecase: - Helps for reproducing complex customer issues locally - dump and restore using pg_upgrade and pg_restore Co-authored-by: David Kimura dkimura@vmware.com --- src/backend/statistics/Makefile | 5 +- src/backend/statistics/dependencies.c | 28 +- src/backend/statistics/mcv.c | 14 +- src/backend/statistics/mvdistinct.c | 25 +- src/backend/statistics/statistics_gram.y | 219 ++++++++++++++ src/backend/statistics/statistics_scanner.l | 118 ++++++++ .../statistics/extended_stats_internal.h | 13 + src/include/statistics/statistics.h | 4 + src/test/regress/expected/stats_ext.out | 277 ++++++++++++++++++ src/test/regress/sql/stats_ext.sql | 98 +++++++ 10 files changed, 771 insertions(+), 30 deletions(-) create mode 100644 src/backend/statistics/statistics_gram.y create mode 100644 src/backend/statistics/statistics_scanner.l diff --git a/src/backend/statistics/Makefile b/src/backend/statistics/Makefile index 89cf8c2797..cd4bb37734 100644 --- a/src/backend/statistics/Makefile +++ b/src/backend/statistics/Makefile @@ -16,6 +16,9 @@ OBJS = \ dependencies.o \ extended_stats.o \ mcv.o \ - mvdistinct.o + mvdistinct.o \ + statistics_gram.o + +statistics_gram.o: statistics_scanner.c include $(top_srcdir)/src/backend/common.mk diff --git a/src/backend/statistics/dependencies.c b/src/backend/statistics/dependencies.c index e6e2835345..2bcc86bfd2 100644 --- a/src/backend/statistics/dependencies.c +++ b/src/backend/statistics/dependencies.c @@ -88,6 +88,7 @@ static Selectivity clauselist_apply_dependencies(PlannerInfo *root, List *clause int ndependencies, AttrNumber *list_attnums, Bitmapset **estimatedclauses); +extern void statistics_scanner_init(const char *query_string); static void generate_dependencies_recurse(DependencyGenerator state, int index, @@ -648,21 +649,26 @@ statext_dependencies_load(Oid mvoid, bool inh) /* * pg_dependencies_in - input routine for type pg_dependencies. * - * pg_dependencies is real enough to be a table column, but it has no operations - * of its own, and disallows input too + * converts the dependencies from the external format in "string" to its + * internal format. */ Datum pg_dependencies_in(PG_FUNCTION_ARGS) { - /* - * pg_node_list stores the data in binary form and parsing text input is - * not needed, so disallow this. - */ - ereport(ERROR, - (errcode(ERRCODE_FEATURE_NOT_SUPPORTED), - errmsg("cannot accept a value of type %s", "pg_dependencies"))); - - PG_RETURN_VOID(); /* keep compiler quiet */ + char *str = PG_GETARG_CSTRING(0); + MVDependencies *mvdependencies; + int parse_rc; + + statistics_scanner_init(str); + parse_rc = statistic_yyparse(); + if (parse_rc != 0) + ereport(ERROR, + (errcode(ERRCODE_INVALID_TEXT_REPRESENTATION), + errmsg("failed to parse a value of type %s", "pg_dependencies"))); + statistic_scanner_finish(); + mvdependencies = mvdependencies_parse_result; + + PG_RETURN_MVNDistinct_P(statext_dependencies_serialize(mvdependencies)); } /* diff --git a/src/backend/statistics/mcv.c b/src/backend/statistics/mcv.c index 2d2a87d3a6..452a760683 100644 --- a/src/backend/statistics/mcv.c +++ b/src/backend/statistics/mcv.c @@ -1469,21 +1469,13 @@ pg_stats_ext_mcvlist_items(PG_FUNCTION_ARGS) /* * pg_mcv_list_in - input routine for type pg_mcv_list. * - * pg_mcv_list is real enough to be a table column, but it has no operations - * of its own, and disallows input too + * converts serialized text MCV lists into a byte values by simply + * calling byeain(). */ Datum pg_mcv_list_in(PG_FUNCTION_ARGS) { - /* - * pg_mcv_list stores the data in binary form and parsing text input is - * not needed, so disallow this. - */ - ereport(ERROR, - (errcode(ERRCODE_FEATURE_NOT_SUPPORTED), - errmsg("cannot accept a value of type %s", "pg_mcv_list"))); - - PG_RETURN_VOID(); /* keep compiler quiet */ + PG_RETURN_MCVList_P(byteain(fcinfo)); } diff --git a/src/backend/statistics/mvdistinct.c b/src/backend/statistics/mvdistinct.c index 13301a3157..0548fda85a 100644 --- a/src/backend/statistics/mvdistinct.c +++ b/src/backend/statistics/mvdistinct.c @@ -42,6 +42,8 @@ static double estimate_ndistinct(double totalrows, int numrows, int d, int f1); static int n_choose_k(int n, int k); static int num_combinations(int n); +extern void statistics_scanner_init(const char *query_string); + /* size of the struct header fields (magic, type, nitems) */ #define SizeOfHeader (3 * sizeof(uint32)) @@ -333,17 +335,26 @@ statext_ndistinct_deserialize(bytea *data) * pg_ndistinct_in * input routine for type pg_ndistinct * - * pg_ndistinct is real enough to be a table column, but it has no - * operations of its own, and disallows input (just like pg_node_tree). + * converts the distinct from the external format in "string" to its internal + * format. */ Datum pg_ndistinct_in(PG_FUNCTION_ARGS) { - ereport(ERROR, - (errcode(ERRCODE_FEATURE_NOT_SUPPORTED), - errmsg("cannot accept a value of type %s", "pg_ndistinct"))); - - PG_RETURN_VOID(); /* keep compiler quiet */ + char *str = PG_GETARG_CSTRING(0); + MVNDistinct *mvndistinct; + int parse_rc; + + statistics_scanner_init(str); + parse_rc = statistic_yyparse(); + if (parse_rc != 0) + ereport(ERROR, + (errcode(ERRCODE_INVALID_TEXT_REPRESENTATION), + errmsg("failed to parse a value of type %s", "pg_ndistinct"))); + statistic_scanner_finish(); + mvndistinct = mvndistinct_parse_result; + + PG_RETURN_MVNDistinct_P(statext_ndistinct_serialize(mvndistinct)); } /* diff --git a/src/backend/statistics/statistics_gram.y b/src/backend/statistics/statistics_gram.y new file mode 100644 index 0000000000..022fba6ca9 --- /dev/null +++ b/src/backend/statistics/statistics_gram.y @@ -0,0 +1,219 @@ +% +{ +#include "postgres.h" + +#include "statistics/extended_stats_internal.h" +#include "statistics/statistics.h" + + MVNDistinct *mvndistinct_parse_result; + MVDependencies *mvdependencies_parse_result; + +/* + * Bison doesn't allocate anything that needs to live across parser calls, + * so we can easily have it use palloc instead of malloc. This prevents + * memory leaks if we error out during parsing. Note this only works with + * bison >= 2.0. However, in bison 1.875 the default is to use alloca() + * if possible, so there's not really much problem anyhow, at least if + * you're building with gcc. + */ +#define YYMALLOC palloc +#define YYFREE pfree + + int attrCount = 0; + + % +} + +%expect 0 +% name - prefix = "statistic_yy" + + +% union +{ + uint32 uintval; + double doubleval; + + MVNDistinct *ndistinct; + MVNDistinctItem *ndistinct_item; + + MVDependencies *dependencies; + MVDependency *dependency; + + Bitmapset *bitmap; + List *list; +} + +/* Non-keyword tokens */ + %token < uintval > UCONST +% token < doubleval > DOUBLE +% token ARROW + +% type < ndistinct > ndistinct +% type < ndistinct_item > ndistinct_item +% type < list > ndistinct_item_list + +% type < dependencies > dependencies +% type < list > dependency_item_list +% type < dependency > dependency_item +% type < bitmap > dependency_attrs + +% type < bitmap > attrs + +%% + + extended_statistic: + ndistinct +{ +} | + + dependencies +{ +} + + ; + +/* + * "ndistinct" rule helps to parse the input string recursively and stores the output into MVNDistinct structure. + * Exmple: + * intput : '{"1, 2": 1,"2, 3": 2, "3, 1", 2}' + * output : returns MVNDistinct object +*/ +ndistinct: +'{' ndistinct_item_list '}' +{ + $$ = palloc0(MAXALIGN(offsetof(MVNDistinct, items)) + + list_length($2) * sizeof(MVNDistinctItem)); + mvndistinct_parse_result = $$; + $$->magic = STATS_NDISTINCT_MAGIC; + $$->type = STATS_NDISTINCT_TYPE_BASIC; + $$->nitems = list_length($2); + + ListCell *cell; + MVNDistinctItem *pointer = $$->items; + + foreach(cell, $2) + { + memcpy(pointer, lfirst(cell), sizeof(MVNDistinctItem)); + pointer += 1; + } +} + +; + +ndistinct_item_list: +ndistinct_item_list ',' ndistinct_item +{ + $$ = lappend($1, $3); +} + +|ndistinct_item +{ + $$ = lappend(NIL, $1); +} + +; + +ndistinct_item: +'"' attrs '"' ':' UCONST +{ + $$ = (MVNDistinctItem *) palloc0(sizeof(MVNDistinctItem)); + $$->attrs = $2; + $$->ndistinct = $5; +} + +; + +attrs: +attrs ',' UCONST +{ + $$ = bms_add_member($1, $3); + attrCount += 1; + +} + +|UCONST ',' UCONST +{ + $$ = bms_make_singleton($1); + $$ = bms_add_member($$, $3); + + attrCount += 2; +} + +; + +/* + * "dependencies" rule helps to parse the input string recursively and stores the output into MVDependencies structure. + * example: + * intput : '{"1 => 2": 1.000000, "2 => 3": 2.000000}' + * output : returns MVDependencies object +*/ +dependencies: +'{' dependency_item_list '}' +{ + $$ = palloc0(MAXALIGN(offsetof(MVDependencies, deps)) + list_length($2) * sizeof(MVDependency *)); + mvdependencies_parse_result = $$; + + $$->magic = STATS_DEPS_MAGIC; + $$->type = STATS_DEPS_TYPE_BASIC; + $$->ndeps = list_length($2); + + for (int i = 0; i < $$->ndeps; i++) + { + $$->deps[i] = list_nth($2, i); + } + attrCount = 0; +} + +; + +dependency_item_list: +dependency_item_list ',' dependency_item +{ + $$ = lappend($1, $3); +} + +|dependency_item +{ + $$ = lappend(NIL, $1); +} + +; + +dependency_item: +'"' dependency_attrs '"' ':' DOUBLE +{ + $$ = (MVDependency *) palloc0(sizeof(MVDependency)); + $$->degree = $5; + $$->nattributes = attrCount; + + AttrNumber *ptr = build_attnums_array($2, &$$->nattributes); + + for (int i = 0; i < $$->nattributes; i++) + { + $$->attributes[i] = *(ptr + i); + } + attrCount = 0; +} + +; + +dependency_attrs: + +UCONST ARROW UCONST +{ + $$ = bms_make_singleton($1); + $$ = bms_add_member($$, $3); + attrCount += 2; +} + +|attrs ARROW UCONST +{ + $$ = bms_add_member($1, $3); + attrCount += 1; +} + +; + +%% + +#include "statistics_scanner.c" diff --git a/src/backend/statistics/statistics_scanner.l b/src/backend/statistics/statistics_scanner.l new file mode 100644 index 0000000000..849a970153 --- /dev/null +++ b/src/backend/statistics/statistics_scanner.l @@ -0,0 +1,118 @@ +% +{ +#include "postgres.h" + +#include "utils/builtins.h" +#include "parser/scansup.h" + +/* Handle to the buffer that the lexer uses internally */ + static YY_BUFFER_STATE scanbufhandle; + + % +} + +%option noyywrap +% option prefix = "statistic_yy" + +digit[0 - 9] + +double[ 0 - 9.] + +%% + +[ \t \ n]; + +"{" +{ + return '{'; +} + +"}" +{ + return '}'; +} + +"," +{ + return ','; +} + +"\"" +{ + return '"'; +} + +":" +{ + return ':'; +} + +"=>" +{ + return ARROW; +} + +{ + digit +} + + +{ + yylval.uintval = strtoul(yytext, NULL, 10); + return UCONST; +} + +{ + double +} + + +{ + yylval.doubleval = strtod(yytext, NULL); + return DOUBLE; +} + + %% + + void +yyerror(const char *message) +{ + if (*yytext == YY_END_OF_BUFFER_CHAR) + { + ereport(ERROR, (errcode(ERRCODE_SYNTAX_ERROR), + errmsg("invalid input syntax for extended stats type"), + errdetail("%s at end of input", message))); + } + else + { + ereport(ERROR, + (errcode(ERRCODE_SYNTAX_ERROR), + errmsg("invalid input syntax for extended stats type"), + errdetail("%s at or near \"%s\"", message, yytext))); + } +} + +void +statistics_scanner_init(const char *str) +{ + Size slen = strlen(str); + char *scanbuf; + + /* + * Might be left over after ereport() + */ + if (YY_CURRENT_BUFFER) + yy_delete_buffer(YY_CURRENT_BUFFER); + + /* + * Make a scan buffer with special termination needed by flex. + */ + scanbuf = (char *) palloc(slen + 2); + memcpy(scanbuf, str, slen); + scanbuf[slen] = scanbuf[slen + 1] = YY_END_OF_BUFFER_CHAR; + scanbufhandle = yy_scan_buffer(scanbuf, slen + 2); +} + +void +statistic_scanner_finish(void) +{ + yy_delete_buffer(scanbufhandle); + scanbufhandle = NULL; +} diff --git a/src/include/statistics/extended_stats_internal.h b/src/include/statistics/extended_stats_internal.h index 7b55eb8ffa..1d40436119 100644 --- a/src/include/statistics/extended_stats_internal.h +++ b/src/include/statistics/extended_stats_internal.h @@ -127,4 +127,17 @@ extern Selectivity mcv_clause_selectivity_or(PlannerInfo *root, Selectivity *overlap_basesel, Selectivity *totalsel); +/* + * Internal functions for parsing the statistics grammar, in statiatics_gram.y and + * statistics_scanner.l + */ +extern int statistic_yyparse(void); +extern int statistic_yylex(void); +extern void statistic_yyerror(const char *str) pg_attribute_noreturn(); +extern void statistic_scanner_init(const char *query_string); +extern void statistic_scanner_finish(void); + +extern MVNDistinct *mvndistinct_parse_result; +extern MVDependencies *mvdependencies_parse_result; + #endif /* EXTENDED_STATS_INTERNAL_H */ diff --git a/src/include/statistics/statistics.h b/src/include/statistics/statistics.h index 17e3e7f881..b57b991419 100644 --- a/src/include/statistics/statistics.h +++ b/src/include/statistics/statistics.h @@ -22,6 +22,10 @@ #define STATS_NDISTINCT_MAGIC 0xA352BFA4 /* struct identifier */ #define STATS_NDISTINCT_TYPE_BASIC 1 /* struct version */ +#define PG_RETURN_MVNDistinct_P(X) return PointerGetDatum(X) +#define PG_RETURN_MVDependencies_P(X) return PointerGetDatum(X) +#define PG_RETURN_MCVList_P(X) return PointerGetDatum(X) + /* MVNDistinctItem represents a single combination of columns */ typedef struct MVNDistinctItem { diff --git a/src/test/regress/expected/stats_ext.out b/src/test/regress/expected/stats_ext.out index 03880874c1..67f5bfd212 100644 --- a/src/test/regress/expected/stats_ext.out +++ b/src/test/regress/expected/stats_ext.out @@ -3290,3 +3290,280 @@ NOTICE: drop cascades to 2 other objects DETAIL: drop cascades to table tststats.priv_test_tbl drop cascades to view tststats.priv_test_view DROP USER regress_stats_user1; + -- Test pg_ndistinct_in +drop table if exists tbl_distinct; +NOTICE: table "tbl_distinct" does not exist, skipping +create table tbl_distinct(i int, ii pg_ndistinct); +NOTICE: Table doesn't have 'DISTRIBUTED BY' clause -- Using column named 'i' as the Greenplum Database data distribution key for this table. +HINT: The 'DISTRIBUTED BY' clause determines the distribution of data. Make sure column(s) chosen are the optimal data distribution key to minimize skew. +insert into tbl_distinct values (1, '{"1, 2": 1}'); +insert into tbl_distinct values (2, '{"1, 2": 2, "1, 3": 3, "2, 3": 2, "1, 2, 3": 3}'); +insert into tbl_distinct values (3, '{"123, 234": 11}'); +select * from tbl_distinct; + i | ii +---+------------------------------------------------- + 2 | {"1, 2": 2, "1, 3": 3, "2, 3": 2, "1, 2, 3": 3} + 3 | {"123, 234": 11} + 1 | {"1, 2": 1} +(3 rows) + +-- leading space +insert into tbl_distinct values (1, ' {"1, 2": 1}'); +-- trailing space +insert into tbl_distinct values (1, '{"1, 2": 1} '); +-- unmatched quote +insert into tbl_distinct values (1, '{"1", 2": 1} '); +ERROR: invalid input syntax for extended stats type +LINE 1: insert into tbl_distinct values (1, '{"1", 2": 1} '); + ^ +DETAIL: syntax error at or near """ +-- space in attribute list +insert into tbl_distinct values (1, '{"1 3, 2": 1} '); +ERROR: invalid input syntax for extended stats type +LINE 1: insert into tbl_distinct values (1, '{"1 3, 2": 1} '); + ^ +DETAIL: syntax error at or near "3" +-- colon in attribute list +insert into tbl_distinct values (1, '{"1: 2": 1}'); +ERROR: invalid input syntax for extended stats type +LINE 1: insert into tbl_distinct values (1, '{"1: 2": 1}'); + ^ +DETAIL: syntax error at or near ":" +insert into tbl_distinct values (1, '{"1, 2:" 1}'); +ERROR: invalid input syntax for extended stats type +LINE 1: insert into tbl_distinct values (1, '{"1, 2:" 1}'); + ^ +DETAIL: syntax error at or near ":" +insert into tbl_distinct values (1, '{":1 2": 1}'); +ERROR: invalid input syntax for extended stats type +LINE 1: insert into tbl_distinct values (1, '{":1 2": 1}'); + ^ +DETAIL: syntax error at or near ":" +-- zero/single item attribute list +insert into tbl_distinct values (1, '{"1": 1}'); +ERROR: invalid input syntax for extended stats type +LINE 1: insert into tbl_distinct values (1, '{"1": 1}'); + ^ +DETAIL: syntax error at or near """ +insert into tbl_distinct values (1, '{: 1}'); +ERROR: invalid input syntax for extended stats type +LINE 1: insert into tbl_distinct values (1, '{: 1}'); + ^ +DETAIL: syntax error at or near ":" +insert into tbl_distinct values (1, '{"": 1}'); +ERROR: invalid input syntax for extended stats type +LINE 1: insert into tbl_distinct values (1, '{"": 1}'); + ^ +DETAIL: syntax error at or near """ +insert into tbl_distinct values (1, '{" ": 1}'); +ERROR: invalid input syntax for extended stats type +LINE 1: insert into tbl_distinct values (1, '{" ": 1}'); + ^ +DETAIL: syntax error at or near """ +insert into tbl_distinct values (1, '{}'); +ERROR: invalid input syntax for extended stats type +LINE 1: insert into tbl_distinct values (1, '{}'); + ^ +DETAIL: syntax error at or near "}" +insert into tbl_distinct values (1, '{:}'); +ERROR: invalid input syntax for extended stats type +LINE 1: insert into tbl_distinct values (1, '{:}'); + ^ +DETAIL: syntax error at or near ":" +-- illegal character +insert into tbl_distinct values (1, '{"1,| 2": 1}'); +-- multiple consecutive characters +insert into tbl_distinct values (1, '{"1,, 2": 1}'); +ERROR: invalid input syntax for extended stats type +LINE 1: insert into tbl_distinct values (1, '{"1,, 2": 1}'); + ^ +DETAIL: syntax error at or near "," +insert into tbl_distinct values (1, '{"1": 1}'); +ERROR: invalid input syntax for extended stats type +LINE 1: insert into tbl_distinct values (1, '{"1": 1}'); + ^ +DETAIL: syntax error at or near """ +-- Need to add check on catalog table insert that atribute numbers are legal +-- (e.g. there shouldn't be attribute number 100 for a table with only 2 +-- columns also it should match) +select * from tbl_distinct; + i | ii +---+------------------------------------------------- + 1 | {"1, 2": 1} + 1 | {"1, 2": 1} + 1 | {"1, 2": 1} + 1 | {"1, 2": 1} + 2 | {"1, 2": 2, "1, 3": 3, "2, 3": 2, "1, 2, 3": 3} + 3 | {"123, 234": 11} +(6 rows) + + -- Test pg_dependencies_in +drop table if exists tbl_dependencies; +NOTICE: table "tbl_dependencies" does not exist, skipping +create table tbl_dependencies(i int, ii pg_dependencies); +NOTICE: Table doesn't have 'DISTRIBUTED BY' clause -- Using column named 'i' as the Greenplum Database data distribution key for this table. +HINT: The 'DISTRIBUTED BY' clause determines the distribution of data. Make sure column(s) chosen are the optimal data distribution key to minimize skew. +insert into tbl_dependencies values (1, '{"1 => 2": 1.000000}'); +insert into tbl_dependencies values (2, '{"1 => 2": 2.000000, "1 => 3": 3.000000, "2 => 3": 2.000000, "1, 2 => 3": 3.000000}'); +select * from tbl_dependencies; + i | ii +---+------------------------------------------------------------------------------------- + 2 | {"1 => 2": 2.000000, "1 => 3": 3.000000, "2 => 3": 2.000000, "1, 2 => 3": 3.000000} + 1 | {"1 => 2": 1.000000} +(2 rows) + +-- leading space +insert into tbl_dependencies values (1, ' {"1 => 2": 1.000000}'); +-- trailing space +insert into tbl_dependencies values (1, '{"1 => 2": 1.000000} '); +-- unmatched quote +insert into tbl_dependencies values (1, '{"1" => 2": 1.000000} '); +ERROR: invalid input syntax for extended stats type +LINE 1: insert into tbl_dependencies values (1, '{"1" => 2": 1.00000... + ^ +DETAIL: syntax error at or near """ +-- Wrong format +insert into tbl_dependencies values (1, '{"1, 2": 1.000000}'); +ERROR: invalid input syntax for extended stats type +LINE 1: insert into tbl_dependencies values (1, '{"1, 2": 1.000000}'... + ^ +DETAIL: syntax error at or near "1.000000" +insert into tbl_dependencies values (1, '{"1 => 2": 1}'); +ERROR: invalid input syntax for extended stats type +LINE 1: insert into tbl_dependencies values (1, '{"1 => 2": 1}'); + ^ +DETAIL: syntax error at or near "1" +insert into tbl_dependencies values (1, '{"1 => 2": 1.000000, " 2 => 1": 2}'); +ERROR: invalid input syntax for extended stats type +LINE 1: insert into tbl_dependencies values (1, '{"1 => 2": 1.000000... + ^ +DETAIL: syntax error at or near "2" +-- space in attribute list +insert into tbl_dependencies values (1, '{"1 3 => 2": 1.000000} '); +ERROR: invalid input syntax for extended stats type +LINE 1: insert into tbl_dependencies values (1, '{"1 3 => 2": 1.0000... + ^ +DETAIL: syntax error at or near "3" +-- colon in attribute list +insert into tbl_dependencies values (1, '{"1: 2": 1.000000}'); +ERROR: invalid input syntax for extended stats type +LINE 1: insert into tbl_dependencies values (1, '{"1: 2": 1.000000}'... + ^ +DETAIL: syntax error at or near ":" +insert into tbl_dependencies values (1, '{"1 => 2:" 1.000000}'); +ERROR: invalid input syntax for extended stats type +LINE 1: insert into tbl_dependencies values (1, '{"1 => 2:" 1.000000... + ^ +DETAIL: syntax error at or near ":" +insert into tbl_dependencies values (1, '{":1 2": 1.000000}'); +ERROR: invalid input syntax for extended stats type +LINE 1: insert into tbl_dependencies values (1, '{":1 2": 1.000000}'... + ^ +DETAIL: syntax error at or near ":" +insert into tbl_dependencies values (1, '{"1, 2" 1.000000}'); +ERROR: invalid input syntax for extended stats type +LINE 1: insert into tbl_dependencies values (1, '{"1, 2" 1.000000}')... + ^ +DETAIL: syntax error at or near "1.000000" +-- zero/single item attribute list +insert into tbl_dependencies values (1, '{"1": 1.000000}'); +ERROR: invalid input syntax for extended stats type +LINE 1: insert into tbl_dependencies values (1, '{"1": 1.000000}'); + ^ +DETAIL: syntax error at or near """ +insert into tbl_dependencies values (1, '{: 1.000000}'); +ERROR: invalid input syntax for extended stats type +LINE 1: insert into tbl_dependencies values (1, '{: 1.000000}'); + ^ +DETAIL: syntax error at or near ":" +insert into tbl_dependencies values (1, '{"": 1.000000}'); +ERROR: invalid input syntax for extended stats type +LINE 1: insert into tbl_dependencies values (1, '{"": 1.000000}'); + ^ +DETAIL: syntax error at or near """ +insert into tbl_dependencies values (1, '{" ": 1.000000}'); +ERROR: invalid input syntax for extended stats type +LINE 1: insert into tbl_dependencies values (1, '{" ": 1.000000}'); + ^ +DETAIL: syntax error at or near """ +insert into tbl_dependencies values (1, '{}'); +ERROR: invalid input syntax for extended stats type +LINE 1: insert into tbl_dependencies values (1, '{}'); + ^ +DETAIL: syntax error at or near "}" +insert into tbl_dependencies values (1, '{:}'); +ERROR: invalid input syntax for extended stats type +LINE 1: insert into tbl_dependencies values (1, '{:}'); + ^ +DETAIL: syntax error at or near ":" +-- multiple consecutive characters +insert into tbl_dependencies values (1, '{"1 =>=> 2": 1.000000}'); +ERROR: invalid input syntax for extended stats type +LINE 1: insert into tbl_dependencies values (1, '{"1 =>=> 2": 1.0000... + ^ +DETAIL: syntax error at or near "=>" +insert into tbl_dependencies values (1, '{"1": 1.000000}'); +ERROR: invalid input syntax for extended stats type +LINE 1: insert into tbl_dependencies values (1, '{"1": 1.000000}'); + ^ +DETAIL: syntax error at or near """ +select * from tbl_dependencies; + i | ii +---+------------------------------------------------------------------------------------- + 2 | {"1 => 2": 2.000000, "1 => 3": 3.000000, "2 => 3": 2.000000, "1, 2 => 3": 3.000000} + 1 | {"1 => 2": 1.000000} + 1 | {"1 => 2": 1.000000} + 1 | {"1 => 2": 1.000000} +(4 rows) + +-- Test a table with columns of type pg_ndistinct and pg_dependencies +drop table if exists tbl_dist_dep; +NOTICE: table "tbl_dist_dep" does not exist, skipping +create table tbl_dist_dep(i pg_ndistinct, ii pg_dependencies); +NOTICE: Table doesn't have 'DISTRIBUTED BY' clause -- Using column named 'i' as the Greenplum Database data distribution key for this table. +HINT: The 'DISTRIBUTED BY' clause determines the distribution of data. Make sure column(s) chosen are the optimal data distribution key to minimize skew. +insert into tbl_dist_dep values ('{"1, 2" : 1}', '{"1 => 2": 1.000000}'); +insert into tbl_dist_dep values ('{"1, 2" : 3, "1, 3" : 3, "2, 3" : 2, "1, 2, 3" : 3}', '{"1 => 2": 2.000000, "1 => 3": 3.000000, "2 => 3": 2.000000, "1, 2 => 3": 3.000000}'); +-- unmatched quote +insert into tbl_dist_dep values ('{1, 2 : 1}', '{"1 => 2": 1.000000}'); +ERROR: invalid input syntax for extended stats type +LINE 1: insert into tbl_dist_dep values ('{1, 2 : 1}', '{"1 => 2": 1... + ^ +DETAIL: syntax error at or near "1" +insert into tbl_dist_dep values ('{"1, 2" : 1}', '{1 => 2: 1.000000}'); +ERROR: invalid input syntax for extended stats type +LINE 1: insert into tbl_dist_dep values ('{"1, 2" : 1}', '{1 => 2: 1... + ^ +DETAIL: syntax error at or near "1" +insert into tbl_dist_dep values ('{1, 2 : 1}', '{1 => 2: 1.000000}'); +ERROR: invalid input syntax for extended stats type +LINE 1: insert into tbl_dist_dep values ('{1, 2 : 1}', '{1 => 2: 1.0... + ^ +DETAIL: syntax error at or near "1" +-- Invalid type +insert into tbl_dist_dep values ('{"1, 2" : 1}', '{"1 => 2": 1}'); +ERROR: invalid input syntax for extended stats type +LINE 1: insert into tbl_dist_dep values ('{"1, 2" : 1}', '{"1 => 2":... + ^ +DETAIL: syntax error at or near "1" +insert into tbl_dist_dep values ('{"1, 2" : 1.000000}', '{"1 => 2": 1.000000}'); +ERROR: invalid input syntax for extended stats type +LINE 1: insert into tbl_dist_dep values ('{"1, 2" : 1.000000}', '{"1... + ^ +DETAIL: syntax error at or near "1.000000" +insert into tbl_dist_dep values ('{"1, 2" : 1}', '{"1.000000 => 2.000000": 1.000000}'); +ERROR: invalid input syntax for extended stats type +LINE 1: insert into tbl_dist_dep values ('{"1, 2" : 1}', '{"1.000000... + ^ +DETAIL: syntax error at or near "1.000000" +insert into tbl_dist_dep values ('{"1, 2.000000" : 1}', '{"1 => 2": 1.000000}'); +ERROR: invalid input syntax for extended stats type +LINE 1: insert into tbl_dist_dep values ('{"1, 2.000000" : 1}', '{"1... + ^ +DETAIL: syntax error at or near "2.000000" +select * from tbl_dist_dep; + i | ii +-------------------------------------------------+------------------------------------------------------------------------------------- + {"1, 2": 3, "1, 3": 3, "2, 3": 2, "1, 2, 3": 3} | {"1 => 2": 2.000000, "1 => 3": 3.000000, "2 => 3": 2.000000, "1, 2 => 3": 3.000000} + {"1, 2": 1} | {"1 => 2": 1.000000} +(2 rows) diff --git a/src/test/regress/sql/stats_ext.sql b/src/test/regress/sql/stats_ext.sql index d0d42cd013..5acf0f5465 100644 --- a/src/test/regress/sql/stats_ext.sql +++ b/src/test/regress/sql/stats_ext.sql @@ -1663,3 +1663,101 @@ DROP FUNCTION op_leak(int, int); RESET SESSION AUTHORIZATION; DROP SCHEMA tststats CASCADE; DROP USER regress_stats_user1; + + -- Test pg_ndistinct_in +drop table if exists tbl_distinct; +create table tbl_distinct(i int, ii pg_ndistinct); +insert into tbl_distinct values (1, '{"1, 2": 1}'); +insert into tbl_distinct values (2, '{"1, 2": 2, "1, 3": 3, "2, 3": 2, "1, 2, 3": 3}'); +insert into tbl_distinct values (3, '{"123, 234": 11}'); +select * from tbl_distinct; + +-- leading space +insert into tbl_distinct values (1, ' {"1, 2": 1}'); +-- trailing space +insert into tbl_distinct values (1, '{"1, 2": 1} '); +-- unmatched quote +insert into tbl_distinct values (1, '{"1", 2": 1} '); +-- space in attribute list +insert into tbl_distinct values (1, '{"1 3, 2": 1} '); +-- colon in attribute list +insert into tbl_distinct values (1, '{"1: 2": 1}'); +insert into tbl_distinct values (1, '{"1, 2:" 1}'); +insert into tbl_distinct values (1, '{":1 2": 1}'); +-- zero/single item attribute list +insert into tbl_distinct values (1, '{"1": 1}'); +insert into tbl_distinct values (1, '{: 1}'); +insert into tbl_distinct values (1, '{"": 1}'); +insert into tbl_distinct values (1, '{" ": 1}'); +insert into tbl_distinct values (1, '{}'); +insert into tbl_distinct values (1, '{:}'); +-- illegal character +insert into tbl_distinct values (1, '{"1,| 2": 1}'); + +-- multiple consecutive characters +insert into tbl_distinct values (1, '{"1,, 2": 1}'); +insert into tbl_distinct values (1, '{"1": 1}'); + +-- Need to add check on catalog table insert that atribute numbers are legal +-- (e.g. there shouldn't be attribute number 100 for a table with only 2 +-- columns also it should match) + +select * from tbl_distinct; + + +-- Test pg_dependencies_in +drop table if exists tbl_dependencies; +create table tbl_dependencies(i int, ii pg_dependencies); +insert into tbl_dependencies values (1, '{"1 => 2": 1.000000}'); +insert into tbl_dependencies values (2, '{"1 => 2": 2.000000, "1 => 3": 3.000000, "2 => 3": 2.000000, "1, 2 => 3": 3.000000}'); + +select * from tbl_dependencies; + +-- leading space +insert into tbl_dependencies values (1, ' {"1 => 2": 1.000000}'); +-- trailing space +insert into tbl_dependencies values (1, '{"1 => 2": 1.000000} '); +-- unmatched quote +insert into tbl_dependencies values (1, '{"1" => 2": 1.000000} '); +-- Wrong format +insert into tbl_dependencies values (1, '{"1, 2": 1.000000}'); +insert into tbl_dependencies values (1, '{"1 => 2": 1}'); +insert into tbl_dependencies values (1, '{"1 => 2": 1.000000, " 2 => 1": 2}'); +-- space in attribute list +insert into tbl_dependencies values (1, '{"1 3 => 2": 1.000000} '); +-- colon in attribute list +insert into tbl_dependencies values (1, '{"1: 2": 1.000000}'); +insert into tbl_dependencies values (1, '{"1 => 2:" 1.000000}'); +insert into tbl_dependencies values (1, '{":1 2": 1.000000}'); +insert into tbl_dependencies values (1, '{"1, 2" 1.000000}'); +-- zero/single item attribute list +insert into tbl_dependencies values (1, '{"1": 1.000000}'); +insert into tbl_dependencies values (1, '{: 1.000000}'); +insert into tbl_dependencies values (1, '{"": 1.000000}'); +insert into tbl_dependencies values (1, '{" ": 1.000000}'); +insert into tbl_dependencies values (1, '{}'); +insert into tbl_dependencies values (1, '{:}'); + +-- multiple consecutive characters +insert into tbl_dependencies values (1, '{"1 =>=> 2": 1.000000}'); +insert into tbl_dependencies values (1, '{"1": 1.000000}'); + +select * from tbl_dependencies; + + +-- Test a table with columns of type pg_ndistinct and pg_dependencies +drop table if exists tbl_dist_dep; +create table tbl_dist_dep(i pg_ndistinct, ii pg_dependencies); +insert into tbl_dist_dep values ('{"1, 2" : 1}', '{"1 => 2": 1.000000}'); +insert into tbl_dist_dep values ('{"1, 2" : 3, "1, 3" : 3, "2, 3" : 2, "1, 2, 3" : 3}', '{"1 => 2": 2.000000, "1 => 3": 3.000000, "2 => 3": 2.000000, "1, 2 => 3": 3.000000}'); +-- unmatched quote +insert into tbl_dist_dep values ('{1, 2 : 1}', '{"1 => 2": 1.000000}'); +insert into tbl_dist_dep values ('{"1, 2" : 1}', '{1 => 2: 1.000000}'); +insert into tbl_dist_dep values ('{1, 2 : 1}', '{1 => 2: 1.000000}'); +-- Invalid type +insert into tbl_dist_dep values ('{"1, 2" : 1}', '{"1 => 2": 1}'); +insert into tbl_dist_dep values ('{"1, 2" : 1.000000}', '{"1 => 2": 1.000000}'); +insert into tbl_dist_dep values ('{"1, 2" : 1}', '{"1.000000 => 2.000000": 1.000000}'); +insert into tbl_dist_dep values ('{"1, 2.000000" : 1}', '{"1 => 2": 1.000000}'); + +select * from tbl_dist_dep; -- 2.30.1 (Apple Git-130)