From 9593de5aedd8934ae6b12895e015cbe31ff5a946 Mon Sep 17 00:00:00 2001 From: Anthonin Bonnefoy Date: Thu, 3 Oct 2024 08:52:02 +0200 Subject: Track location to extract relevant part in nested statement Previously, Query generated through transform would have unset stmt_location. Extensions relying on the statement location to extract the relevant part of the statement would fallback to use the whole statement instead, thus showing the same string in the top and nested level which was a source of confusion. This patch fixes the issue by keeping track of the statement locations and propagate it to Query during transform, allowing pgss to only show the relevant part of the query for nested query. --- .../expected/level_tracking.out | 118 +++++++++--------- .../pg_stat_statements/expected/planning.out | 10 +- .../pg_stat_statements/expected/select.out | 2 +- .../pg_stat_statements/expected/utility.out | 2 +- contrib/pg_stat_statements/sql/planning.sql | 4 +- src/backend/optimizer/util/clauses.c | 2 +- src/backend/parser/analyze.c | 96 +++++++++----- src/backend/parser/gram.y | 71 ++++++++++- src/backend/parser/parse_merge.c | 2 + src/include/nodes/parsenodes.h | 10 ++ src/include/parser/analyze.h | 3 +- src/include/parser/parse_node.h | 2 + 12 files changed, 217 insertions(+), 105 deletions(-) diff --git a/contrib/pg_stat_statements/expected/level_tracking.out b/contrib/pg_stat_statements/expected/level_tracking.out index 8c5c1b5b55d..183a8cfc195 100644 --- a/contrib/pg_stat_statements/expected/level_tracking.out +++ b/contrib/pg_stat_statements/expected/level_tracking.out @@ -221,19 +221,19 @@ SELECT toplevel, calls, query FROM pg_stat_statements t | 1 | explain (costs off) TABLE stats_track_tab t | 1 | explain (costs off) UPDATE stats_track_tab SET x=$1 WHERE x=$2 t | 1 | explain (costs off) VALUES($1) - f | 1 | explain (costs off) (SELECT $1, $2); - f | 1 | explain (costs off) (TABLE test_table); - f | 1 | explain (costs off) (VALUES($1, $2)); - f | 1 | explain (costs off) DELETE FROM stats_track_tab; - f | 1 | explain (costs off) INSERT INTO stats_track_tab VALUES (($1)); - f | 1 | explain (costs off) MERGE INTO stats_track_tab USING (SELECT id FROM generate_series($1, $2) id) ON x = id+ + f | 1 | DELETE FROM stats_track_tab + f | 1 | INSERT INTO stats_track_tab VALUES (($1)) + f | 1 | MERGE INTO stats_track_tab USING (SELECT id FROM generate_series($1, $2) id) ON x = id + | | WHEN MATCHED THEN UPDATE SET x = id + - | | WHEN NOT MATCHED THEN INSERT (x) VALUES (id); - f | 1 | explain (costs off) SELECT $1 UNION SELECT $2; - f | 1 | explain (costs off) SELECT $1; - f | 1 | explain (costs off) TABLE stats_track_tab; - f | 1 | explain (costs off) UPDATE stats_track_tab SET x=$1 WHERE x=$2; - f | 1 | explain (costs off) VALUES($1); + | | WHEN NOT MATCHED THEN INSERT (x) VALUES (id) + f | 1 | SELECT $1 + f | 1 | SELECT $1 UNION SELECT $2 + f | 1 | SELECT $1, $2 + f | 1 | TABLE stats_track_tab + f | 1 | TABLE test_table + f | 1 | UPDATE stats_track_tab SET x=$1 WHERE x=$2 + f | 1 | VALUES($1) + f | 1 | VALUES($1, $2) (23 rows) -- Explain - top-level tracking. @@ -480,8 +480,8 @@ explain (costs off) SELECT 1, 2 UNION SELECT 3, 4\; explain (costs off) (SELECT SELECT toplevel, calls, query FROM pg_stat_statements ORDER BY toplevel desc, query COLLATE "C"; - toplevel | calls | query -----------+-------+---------------------------------------------------------------------------------------------------------------------------------- + toplevel | calls | query +----------+-------+------------------------------------------------------------------------------------------------------------ t | 1 | SELECT pg_stat_statements_reset() IS NOT NULL AS t t | 1 | explain (costs off) (SELECT $1, $2, $3) t | 1 | explain (costs off) (SELECT $1, $2, $3) UNION SELECT $4, $5, $6 @@ -492,8 +492,8 @@ SELECT toplevel, calls, query FROM pg_stat_statements t | 1 | explain (costs off) DELETE FROM stats_track_tab WHERE x=$1 t | 1 | explain (costs off) INSERT INTO stats_track_tab VALUES ($1), ($2) t | 1 | explain (costs off) INSERT INTO stats_track_tab VALUES (($1)) - t | 1 | explain (costs off) MERGE INTO stats_track_tab USING (SELECT id FROM generate_series($1, $2) id) ON x = id + - | | WHEN MATCHED THEN UPDATE SET x = id + + t | 1 | explain (costs off) MERGE INTO stats_track_tab USING (SELECT id FROM generate_series($1, $2) id) ON x = id+ + | | WHEN MATCHED THEN UPDATE SET x = id + | | WHEN NOT MATCHED THEN INSERT (x) VALUES (id) t | 1 | explain (costs off) SELECT $1 t | 1 | explain (costs off) SELECT $1, $2 @@ -503,28 +503,26 @@ SELECT toplevel, calls, query FROM pg_stat_statements t | 1 | explain (costs off) UPDATE stats_track_tab SET x=$1 t | 1 | explain (costs off) UPDATE stats_track_tab SET x=$1 WHERE x=$2 t | 1 | explain (costs off) VALUES($1) - f | 1 | explain (costs off) (SELECT $1, $2, $3); explain (costs off) (SELECT 1, 2, 3, 4); - f | 1 | explain (costs off) (SELECT 1, 2, 3); explain (costs off) (SELECT $1, $2, $3, $4); - f | 1 | explain (costs off) DELETE FROM stats_track_tab; explain (costs off) DELETE FROM stats_track_tab WHERE x=$1; - f | 1 | explain (costs off) DELETE FROM stats_track_tab; explain (costs off) DELETE FROM stats_track_tab WHERE x=1; - f | 1 | explain (costs off) INSERT INTO stats_track_tab VALUES (($1)); explain (costs off) INSERT INTO stats_track_tab VALUES (1), (2); - f | 1 | explain (costs off) INSERT INTO stats_track_tab VALUES ((1)); explain (costs off) INSERT INTO stats_track_tab VALUES ($1), ($2); - f | 1 | explain (costs off) MERGE INTO stats_track_tab USING (SELECT id FROM generate_series($1, $2) id) ON x = id + - | | WHEN MATCHED THEN UPDATE SET x = id + - | | WHEN NOT MATCHED THEN INSERT (x) VALUES (id); explain (costs off) SELECT 1, 2, 3, 4, 5; - f | 1 | explain (costs off) MERGE INTO stats_track_tab USING (SELECT id FROM generate_series(1, 10) id) ON x = id + - | | WHEN MATCHED THEN UPDATE SET x = id + - | | WHEN NOT MATCHED THEN INSERT (x) VALUES (id); explain (costs off) SELECT $1, $2, $3, $4, $5; - f | 1 | explain (costs off) SELECT $1, $2 UNION SELECT $3, $4; explain (costs off) (SELECT 1, 2, 3) UNION SELECT 3, 4, 5; - f | 1 | explain (costs off) SELECT $1; explain (costs off) SELECT 1, 2; - f | 1 | explain (costs off) SELECT 1, 2 UNION SELECT 3, 4; explain (costs off) (SELECT $1, $2, $3) UNION SELECT $4, $5, $6; - f | 1 | explain (costs off) SELECT 1; explain (costs off) SELECT $1, $2; - f | 1 | explain (costs off) TABLE stats_track_tab; explain (costs off) (TABLE test_table); - f | 1 | explain (costs off) TABLE stats_track_tab; explain (costs off) (TABLE test_table); - f | 1 | explain (costs off) UPDATE stats_track_tab SET x=$1 WHERE x=$2; explain (costs off) UPDATE stats_track_tab SET x=1; - f | 1 | explain (costs off) UPDATE stats_track_tab SET x=1 WHERE x=1; explain (costs off) UPDATE stats_track_tab SET x=$1; - f | 1 | explain (costs off) VALUES($1); explain (costs off) (VALUES(1, 2)); - f | 1 | explain (costs off) VALUES(1); explain (costs off) (VALUES($1, $2)); + f | 1 | (SELECT $1, $2, $3) UNION SELECT $4, $5, $6 + f | 1 | DELETE FROM stats_track_tab + f | 1 | DELETE FROM stats_track_tab WHERE x=$1 + f | 1 | INSERT INTO stats_track_tab VALUES ($1), ($2) + f | 1 | INSERT INTO stats_track_tab VALUES (($1)) + f | 1 | MERGE INTO stats_track_tab USING (SELECT id FROM generate_series($1, $2) id) ON x = id + + | | WHEN MATCHED THEN UPDATE SET x = id + + | | WHEN NOT MATCHED THEN INSERT (x) VALUES (id) + f | 1 | SELECT $1 + f | 1 | SELECT $1, $2 + f | 1 | SELECT $1, $2 UNION SELECT $3, $4 + f | 1 | SELECT $1, $2, $3 + f | 1 | SELECT $1, $2, $3, $4 + f | 1 | SELECT $1, $2, $3, $4, $5 + f | 1 | TABLE stats_track_tab + f | 1 | TABLE test_table + f | 1 | UPDATE stats_track_tab SET x=$1 + f | 1 | UPDATE stats_track_tab SET x=$1 WHERE x=$2 + f | 1 | VALUES($1) + f | 1 | VALUES($1, $2) (37 rows) -- Explain - top-level tracking with multi statement. @@ -762,15 +760,15 @@ SELECT toplevel, calls, query FROM pg_stat_statements t | 1 | explain (costs off) WITH a AS (select $1) SELECT $2 t | 1 | explain (costs off) WITH a AS (select $1) SELECT $2 UNION SELECT $3 t | 1 | explain (costs off) WITH a AS (select $1) UPDATE stats_track_tab SET x=$2 WHERE x=$3 - f | 1 | explain (costs off) (WITH a AS (select $1) (SELECT $2, $3)); - f | 1 | explain (costs off) WITH a AS (select $1) DELETE FROM stats_track_tab; - f | 1 | explain (costs off) WITH a AS (select $1) INSERT INTO stats_track_tab VALUES (($2)); - f | 1 | explain (costs off) WITH a AS (select $1) MERGE INTO stats_track_tab USING (SELECT id FROM generate_series($2, $3) id) ON x = id+ + f | 1 | WITH a AS (select $1) (SELECT $2, $3) + f | 1 | WITH a AS (select $1) DELETE FROM stats_track_tab + f | 1 | WITH a AS (select $1) INSERT INTO stats_track_tab VALUES (($2)) + f | 1 | WITH a AS (select $1) MERGE INTO stats_track_tab USING (SELECT id FROM generate_series($2, $3) id) ON x = id + | | WHEN MATCHED THEN UPDATE SET x = id + - | | WHEN NOT MATCHED THEN INSERT (x) VALUES (id); - f | 1 | explain (costs off) WITH a AS (select $1) SELECT $2 UNION SELECT $3; - f | 1 | explain (costs off) WITH a AS (select $1) SELECT $2; - f | 1 | explain (costs off) WITH a AS (select $1) UPDATE stats_track_tab SET x=$2 WHERE x=$3; + | | WHEN NOT MATCHED THEN INSERT (x) VALUES (id) + f | 1 | WITH a AS (select $1) SELECT $2 + f | 1 | WITH a AS (select $1) SELECT $2 UNION SELECT $3 + f | 1 | WITH a AS (select $1) UPDATE stats_track_tab SET x=$2 WHERE x=$3 (15 rows) -- Explain with CTE - top-level tracking @@ -882,7 +880,7 @@ SELECT toplevel, calls, query FROM pg_stat_statements t | 1 | EXPLAIN (ANALYZE, COSTS OFF, SUMMARY OFF, TIMING OFF) DECLARE foocur CURSOR FOR SELECT * FROM stats_track_tab t | 1 | EXPLAIN (ANALYZE, COSTS OFF, SUMMARY OFF, TIMING OFF) SELECT $1 t | 1 | SELECT pg_stat_statements_reset() IS NOT NULL AS t - f | 1 | EXPLAIN (ANALYZE, COSTS OFF, SUMMARY OFF, TIMING OFF) SELECT $1; + f | 1 | SELECT $1 (4 rows) -- Explain analyze, top tracking. @@ -1000,7 +998,7 @@ SELECT toplevel, calls, query FROM pg_stat_statements t | 1 | CREATE TEMPORARY TABLE pgss_ctas_1 AS SELECT $1 t | 1 | CREATE TEMPORARY TABLE pgss_ctas_2 AS EXECUTE test_prepare_pgss t | 1 | SELECT pg_stat_statements_reset() IS NOT NULL AS t - f | 1 | PREPARE test_prepare_pgss AS select generate_series($1, $2) + f | 1 | select generate_series($1, $2) (4 rows) -- Create Table As, top-level tracking. @@ -1151,25 +1149,25 @@ COPY (DELETE FROM stats_track_tab WHERE x=2 RETURNING x) to stdout; 2 SELECT toplevel, calls, query FROM pg_stat_statements ORDER BY toplevel desc, query COLLATE "C"; - toplevel | calls | query -----------+-------+-------------------------------------------------------------------------- + toplevel | calls | query +----------+-------+------------------------------------------------------------------------- t | 1 | COPY (DELETE FROM stats_track_tab WHERE x=2 RETURNING x) to stdout t | 1 | COPY (INSERT INTO stats_track_tab (x) VALUES (1) RETURNING x) to stdout - t | 1 | COPY (MERGE INTO stats_track_tab USING (SELECT 1 id) ON x = id + - | | WHEN MATCHED THEN UPDATE SET x = id + + t | 1 | COPY (MERGE INTO stats_track_tab USING (SELECT 1 id) ON x = id + + | | WHEN MATCHED THEN UPDATE SET x = id + | | WHEN NOT MATCHED THEN INSERT (x) VALUES (id) RETURNING x) to stdout t | 1 | COPY (SELECT 1 UNION SELECT 2) to stdout t | 1 | COPY (SELECT 1) to stdout t | 1 | COPY (UPDATE stats_track_tab SET x=2 WHERE x=1 RETURNING x) to stdout t | 1 | SELECT pg_stat_statements_reset() IS NOT NULL AS t - f | 1 | COPY (DELETE FROM stats_track_tab WHERE x=$1 RETURNING x) to stdout - f | 1 | COPY (INSERT INTO stats_track_tab (x) VALUES ($1) RETURNING x) to stdout - f | 1 | COPY (MERGE INTO stats_track_tab USING (SELECT $1 id) ON x = id + - | | WHEN MATCHED THEN UPDATE SET x = id + - | | WHEN NOT MATCHED THEN INSERT (x) VALUES (id) RETURNING x) to stdout - f | 1 | COPY (SELECT $1 UNION SELECT $2) to stdout - f | 1 | COPY (SELECT $1) to stdout - f | 1 | COPY (UPDATE stats_track_tab SET x=$1 WHERE x=$2 RETURNING x) to stdout + f | 1 | DELETE FROM stats_track_tab WHERE x=$1 RETURNING x + f | 1 | INSERT INTO stats_track_tab (x) VALUES ($1) RETURNING x + f | 1 | MERGE INTO stats_track_tab USING (SELECT $1 id) ON x = id + + | | WHEN MATCHED THEN UPDATE SET x = id + + | | WHEN NOT MATCHED THEN INSERT (x) VALUES (id) RETURNING x + f | 1 | SELECT $1 + f | 1 | SELECT $1 UNION SELECT $2 + f | 1 | UPDATE stats_track_tab SET x=$1 WHERE x=$2 RETURNING x (13 rows) -- COPY - top-level tracking. diff --git a/contrib/pg_stat_statements/expected/planning.out b/contrib/pg_stat_statements/expected/planning.out index 9effd11fdc8..3ee1928cbe9 100644 --- a/contrib/pg_stat_statements/expected/planning.out +++ b/contrib/pg_stat_statements/expected/planning.out @@ -58,7 +58,7 @@ SELECT 42; (1 row) SELECT plans, calls, rows, query FROM pg_stat_statements - WHERE query NOT LIKE 'PREPARE%' ORDER BY query COLLATE "C"; + WHERE query NOT LIKE 'SELECT COUNT%' ORDER BY query COLLATE "C"; plans | calls | rows | query -------+-------+------+---------------------------------------------------------- 0 | 1 | 0 | ALTER TABLE stats_plan_test ADD COLUMN x int @@ -72,10 +72,10 @@ SELECT plans, calls, rows, query FROM pg_stat_statements -- for the prepared statement we expect at least one replan, but cache -- invalidations could force more SELECT plans >= 2 AND plans <= calls AS plans_ok, calls, rows, query FROM pg_stat_statements - WHERE query LIKE 'PREPARE%' ORDER BY query COLLATE "C"; - plans_ok | calls | rows | query -----------+-------+------+------------------------------------------------------- - t | 4 | 4 | PREPARE prep1 AS SELECT COUNT(*) FROM stats_plan_test + WHERE query LIKE 'SELECT COUNT%' ORDER BY query COLLATE "C"; + plans_ok | calls | rows | query +----------+-------+------+-------------------------------------- + t | 4 | 4 | SELECT COUNT(*) FROM stats_plan_test (1 row) -- Cleanup diff --git a/contrib/pg_stat_statements/expected/select.out b/contrib/pg_stat_statements/expected/select.out index dd6c756f67d..217a2c0b2bc 100644 --- a/contrib/pg_stat_statements/expected/select.out +++ b/contrib/pg_stat_statements/expected/select.out @@ -127,7 +127,6 @@ DEALLOCATE pgss_test; SELECT calls, rows, query FROM pg_stat_statements ORDER BY query COLLATE "C"; calls | rows | query -------+------+------------------------------------------------------------------------------ - 1 | 1 | PREPARE pgss_test (int) AS SELECT $1, $2 LIMIT $3 4 | 4 | SELECT $1 + | | -- multiline + | | AS "text" @@ -137,6 +136,7 @@ SELECT calls, rows, query FROM pg_stat_statements ORDER BY query COLLATE "C"; 2 | 2 | SELECT $1 AS "int" 1 | 2 | SELECT $1 AS i UNION SELECT $2 ORDER BY i 1 | 1 | SELECT $1 || $2 + 1 | 1 | SELECT $1, $2 LIMIT $3 0 | 0 | SELECT calls, rows, query FROM pg_stat_statements ORDER BY query COLLATE "C" 1 | 1 | SELECT pg_stat_statements_reset() IS NOT NULL AS t 1 | 2 | WITH t(f) AS ( + diff --git a/contrib/pg_stat_statements/expected/utility.out b/contrib/pg_stat_statements/expected/utility.out index 060d4416dd7..aa4f0f7e628 100644 --- a/contrib/pg_stat_statements/expected/utility.out +++ b/contrib/pg_stat_statements/expected/utility.out @@ -540,7 +540,7 @@ SELECT calls, rows, query FROM pg_stat_statements ORDER BY query COLLATE "C"; -------+------+---------------------------------------------------- 2 | 0 | DEALLOCATE $1 2 | 0 | DEALLOCATE ALL - 2 | 2 | PREPARE stat_select AS SELECT $1 AS a + 2 | 2 | SELECT $1 AS a 1 | 1 | SELECT $1 as a 1 | 1 | SELECT pg_stat_statements_reset() IS NOT NULL AS t (5 rows) diff --git a/contrib/pg_stat_statements/sql/planning.sql b/contrib/pg_stat_statements/sql/planning.sql index 46f5d9b951c..9cfe206b3b0 100644 --- a/contrib/pg_stat_statements/sql/planning.sql +++ b/contrib/pg_stat_statements/sql/planning.sql @@ -20,11 +20,11 @@ SELECT 42; SELECT 42; SELECT 42; SELECT plans, calls, rows, query FROM pg_stat_statements - WHERE query NOT LIKE 'PREPARE%' ORDER BY query COLLATE "C"; + WHERE query NOT LIKE 'SELECT COUNT%' ORDER BY query COLLATE "C"; -- for the prepared statement we expect at least one replan, but cache -- invalidations could force more SELECT plans >= 2 AND plans <= calls AS plans_ok, calls, rows, query FROM pg_stat_statements - WHERE query LIKE 'PREPARE%' ORDER BY query COLLATE "C"; + WHERE query LIKE 'SELECT COUNT%' ORDER BY query COLLATE "C"; -- Cleanup DROP TABLE stats_plan_test; diff --git a/src/backend/optimizer/util/clauses.c b/src/backend/optimizer/util/clauses.c index b4e085e9d4b..8fc4da11483 100644 --- a/src/backend/optimizer/util/clauses.c +++ b/src/backend/optimizer/util/clauses.c @@ -4674,7 +4674,7 @@ inline_function(Oid funcid, Oid result_type, Oid result_collid, pstate->p_sourcetext = src; sql_fn_parser_setup(pstate, pinfo); - querytree = transformTopLevelStmt(pstate, linitial(raw_parsetree_list)); + querytree = transformOptionalSelectInto(pstate, ((RawStmt *) linitial(raw_parsetree_list))->stmt); free_parsestate(pstate); } diff --git a/src/backend/parser/analyze.c b/src/backend/parser/analyze.c index 2d3d8fcf769..0c7e63b783f 100644 --- a/src/backend/parser/analyze.c +++ b/src/backend/parser/analyze.c @@ -58,7 +58,6 @@ /* Hook for plugins to get control at end of parse analysis */ post_parse_analyze_hook_type post_parse_analyze_hook = NULL; -static Query *transformOptionalSelectInto(ParseState *pstate, Node *parseTree); static Query *transformDeleteStmt(ParseState *pstate, DeleteStmt *stmt); static Query *transformInsertStmt(ParseState *pstate, InsertStmt *stmt); static OnConflictExpr *transformOnConflictClause(ParseState *pstate, @@ -113,13 +112,15 @@ parse_analyze_fixedparams(RawStmt *parseTree, const char *sourceText, Assert(sourceText != NULL); /* required as of 8.4 */ pstate->p_sourcetext = sourceText; + pstate->p_stmt_len = parseTree->stmt_len; + pstate->p_stmt_location = parseTree->stmt_location; if (numParams > 0) setup_parse_fixed_parameters(pstate, paramTypes, numParams); pstate->p_queryEnv = queryEnv; - query = transformTopLevelStmt(pstate, parseTree); + query = transformOptionalSelectInto(pstate, parseTree->stmt); if (IsQueryIdEnabled()) jstate = JumbleQuery(query); @@ -153,12 +154,14 @@ parse_analyze_varparams(RawStmt *parseTree, const char *sourceText, Assert(sourceText != NULL); /* required as of 8.4 */ pstate->p_sourcetext = sourceText; + pstate->p_stmt_len = parseTree->stmt_len; + pstate->p_stmt_location = parseTree->stmt_location; setup_parse_variable_parameters(pstate, paramTypes, numParams); pstate->p_queryEnv = queryEnv; - query = transformTopLevelStmt(pstate, parseTree); + query = transformOptionalSelectInto(pstate, parseTree->stmt); /* make sure all is well with parameter types */ check_variable_parameters(pstate, query); @@ -195,10 +198,12 @@ parse_analyze_withcb(RawStmt *parseTree, const char *sourceText, Assert(sourceText != NULL); /* required as of 8.4 */ pstate->p_sourcetext = sourceText; + pstate->p_stmt_len = parseTree->stmt_len; + pstate->p_stmt_location = parseTree->stmt_location; pstate->p_queryEnv = queryEnv; (*parserSetup) (pstate, parserSetupArg); - query = transformTopLevelStmt(pstate, parseTree); + query = transformOptionalSelectInto(pstate, parseTree->stmt); if (IsQueryIdEnabled()) jstate = JumbleQuery(query); @@ -238,27 +243,6 @@ parse_sub_analyze(Node *parseTree, ParseState *parentParseState, return query; } -/* - * transformTopLevelStmt - - * transform a Parse tree into a Query tree. - * - * This function is just responsible for transferring statement location data - * from the RawStmt into the finished Query. - */ -Query * -transformTopLevelStmt(ParseState *pstate, RawStmt *parseTree) -{ - Query *result; - - /* We're at top level, so allow SELECT INTO */ - result = transformOptionalSelectInto(pstate, parseTree->stmt); - - result->stmt_location = parseTree->stmt_location; - result->stmt_len = parseTree->stmt_len; - - return result; -} - /* * transformOptionalSelectInto - * If SELECT has INTO, convert it to CREATE TABLE AS. @@ -269,7 +253,7 @@ transformTopLevelStmt(ParseState *pstate, RawStmt *parseTree) * of the parse tree, and so we only try it before entering the recursive * transformStmt() processing. */ -static Query * +Query * transformOptionalSelectInto(ParseState *pstate, Node *parseTree) { if (IsA(parseTree, SelectStmt)) @@ -417,7 +401,9 @@ transformStmt(ParseState *pstate, Node *parseTree) */ result = makeNode(Query); result->commandType = CMD_UTILITY; - result->utilityStmt = (Node *) parseTree; + result->utilityStmt = parseTree; + result->stmt_location = pstate->p_stmt_location; + result->stmt_len = pstate->p_stmt_len; break; } @@ -506,6 +492,37 @@ analyze_requires_snapshot(RawStmt *parseTree) return stmt_requires_parse_analysis(parseTree); } +/* + * setQueryStmtLen + * Set stmt_len in Query. + * + * Some statements, like PreparableStmt, can be located within parentheses. + * For example "(SELECT 1)" or "COPY (UPDATE ...) to x;". For those, we can't + * use the whole string from the statement's location or the SQL string will + * yield "SELECT 1)". The parser will set stmt_len, reflecting the size of the + * statement within the parentheses. Thus, when stmt_len is available, we use it + * for the Query's stmt_len. + * + * For other cases, the parser can't provide the length of individual statements. + * However, we have the statement's location plus the length (p_stmt_len) and + * location (p_stmt_location) of the top level RawStmt, stored in pstate. Thus, + * the statement's length is the RawStmt's length minus how much we've advanced + * in the RawStmt's string. + */ +void +setQueryStmtLen(ParseState *pstate, Query *qry, int stmt_len) +{ + if (stmt_len > 0) + /* Statement's length is known, use it */ + qry->stmt_len = stmt_len; + else + /* + * Compute the statement's length from statement's location and RawStmt's + * length and location + */ + qry->stmt_len = pstate->p_stmt_len - (qry->stmt_location - pstate->p_stmt_location); +} + /* * transformDeleteStmt - * transforms a Delete Statement @@ -518,6 +535,8 @@ transformDeleteStmt(ParseState *pstate, DeleteStmt *stmt) Node *qual; qry->commandType = CMD_DELETE; + qry->stmt_location = stmt->stmt_location; + setQueryStmtLen(pstate, qry, stmt->stmt_len); /* process the WITH clause independently of all else */ if (stmt->withClause) @@ -606,6 +625,8 @@ transformInsertStmt(ParseState *pstate, InsertStmt *stmt) Assert(pstate->p_ctenamespace == NIL); qry->commandType = CMD_INSERT; + qry->stmt_location = stmt->stmt_location; + setQueryStmtLen(pstate, qry, stmt->stmt_len); pstate->p_is_insert = true; /* process the WITH clause independently of all else */ @@ -1331,7 +1352,6 @@ count_rowexpr_columns(ParseState *pstate, Node *expr) return -1; } - /* * transformSelectStmt - * transforms a Select Statement @@ -1347,6 +1367,8 @@ transformSelectStmt(ParseState *pstate, SelectStmt *stmt) ListCell *l; qry->commandType = CMD_SELECT; + qry->stmt_location = stmt->stmt_location; + setQueryStmtLen(pstate, qry, stmt->stmt_len); /* process the WITH clause independently of all else */ if (stmt->withClause) @@ -1499,6 +1521,8 @@ transformValuesClause(ParseState *pstate, SelectStmt *stmt) int i; qry->commandType = CMD_SELECT; + qry->stmt_location = stmt->stmt_location; + setQueryStmtLen(pstate, qry, stmt->stmt_len); /* Most SELECT stuff doesn't apply in a VALUES clause */ Assert(stmt->distinctClause == NIL); @@ -1730,6 +1754,8 @@ transformSetOperationStmt(ParseState *pstate, SelectStmt *stmt) int tllen; qry->commandType = CMD_SELECT; + qry->stmt_location = stmt->stmt_location; + setQueryStmtLen(pstate, qry, stmt->stmt_len); /* * Find leftmost leaf SelectStmt. We currently only need to do this in @@ -2397,6 +2423,8 @@ transformReturnStmt(ParseState *pstate, ReturnStmt *stmt) Query *qry = makeNode(Query); qry->commandType = CMD_SELECT; + qry->stmt_location = pstate->p_stmt_location; + qry->stmt_len = pstate->p_stmt_len; qry->isReturn = true; qry->targetList = list_make1(makeTargetEntry((Expr *) transformExpr(pstate, stmt->returnval, EXPR_KIND_SELECT_TARGET), @@ -2430,6 +2458,8 @@ transformUpdateStmt(ParseState *pstate, UpdateStmt *stmt) Node *qual; qry->commandType = CMD_UPDATE; + qry->stmt_location = stmt->stmt_location; + setQueryStmtLen(pstate, qry, stmt->stmt_len); pstate->p_is_insert = false; /* process the WITH clause independently of all else */ @@ -2677,6 +2707,8 @@ transformPLAssignStmt(ParseState *pstate, PLAssignStmt *stmt) * consider WITH or INTO, and we build a targetlist our own way. */ qry->commandType = CMD_SELECT; + qry->stmt_location = pstate->p_stmt_location; + qry->stmt_len = pstate->p_stmt_len; pstate->p_is_insert = false; /* make FOR UPDATE/FOR SHARE info available to addRangeTableEntry */ @@ -2948,6 +2980,8 @@ transformDeclareCursorStmt(ParseState *pstate, DeclareCursorStmt *stmt) /* represent the command as a utility Query */ result = makeNode(Query); result->commandType = CMD_UTILITY; + result->stmt_location = pstate->p_stmt_location; + result->stmt_len = pstate->p_stmt_len; result->utilityStmt = (Node *) stmt; return result; @@ -3003,6 +3037,8 @@ transformExplainStmt(ParseState *pstate, ExplainStmt *stmt) /* represent the command as a utility Query */ result = makeNode(Query); result->commandType = CMD_UTILITY; + result->stmt_location = pstate->p_stmt_location; + result->stmt_len = pstate->p_stmt_len; result->utilityStmt = (Node *) stmt; return result; @@ -3083,6 +3119,8 @@ transformCreateTableAsStmt(ParseState *pstate, CreateTableAsStmt *stmt) /* represent the command as a utility Query */ result = makeNode(Query); result->commandType = CMD_UTILITY; + result->stmt_location = pstate->p_stmt_location; + result->stmt_len = pstate->p_stmt_len; result->utilityStmt = (Node *) stmt; return result; @@ -3207,6 +3245,8 @@ transformCallStmt(ParseState *pstate, CallStmt *stmt) /* represent the command as a utility Query */ result = makeNode(Query); result->commandType = CMD_UTILITY; + result->stmt_location = pstate->p_stmt_location; + result->stmt_len = pstate->p_stmt_len; result->utilityStmt = (Node *) stmt; return result; diff --git a/src/backend/parser/gram.y b/src/backend/parser/gram.y index 4aa8646af7b..d00f4b47c4a 100644 --- a/src/backend/parser/gram.y +++ b/src/backend/parser/gram.y @@ -167,6 +167,7 @@ static void base_yyerror(YYLTYPE *yylloc, core_yyscan_t yyscanner, const char *msg); static RawStmt *makeRawStmt(Node *stmt, int stmt_location); static void updateRawStmtEnd(RawStmt *rs, int end_location); +static void updatePreparableStmtEnd(Node *n, int end_location); static Node *makeColumnRef(char *colname, List *indirection, int location, core_yyscan_t yyscanner); static Node *makeTypeCast(Node *arg, TypeName *typename, int location); @@ -190,7 +191,7 @@ static void insertSelectOptions(SelectStmt *stmt, SelectLimit *limitClause, WithClause *withClause, core_yyscan_t yyscanner); -static Node *makeSetOp(SetOperation op, bool all, Node *larg, Node *rarg); +static Node *makeSetOp(SetOperation op, bool all, Node *larg, Node *rarg, int location); static Node *doNegate(Node *n, int location); static void doNegateFloat(Float *v); static Node *makeAndExpr(Node *lexpr, Node *rexpr, int location); @@ -3403,6 +3404,7 @@ CopyStmt: COPY opt_binary qualified_name opt_column_list { CopyStmt *n = makeNode(CopyStmt); + updatePreparableStmtEnd($3, @4); n->relation = NULL; n->query = $3; n->attlist = NIL; @@ -12170,6 +12172,9 @@ InsertStmt: $5->onConflictClause = $6; $5->returningList = $7; $5->withClause = $1; + if (@$ < 0) /* see comments for YYLLOC_DEFAULT */ + @$ = @2; + $5->stmt_location = @$; $$ = (Node *) $5; } ; @@ -12323,6 +12328,9 @@ DeleteStmt: opt_with_clause DELETE_P FROM relation_expr_opt_alias n->whereClause = $6; n->returningList = $7; n->withClause = $1; + if (@$ < 0) /* see comments for YYLLOC_DEFAULT */ + @$ = @2; + n->stmt_location = @$; $$ = (Node *) n; } ; @@ -12397,6 +12405,9 @@ UpdateStmt: opt_with_clause UPDATE relation_expr_opt_alias n->whereClause = $7; n->returningList = $8; n->withClause = $1; + if (@$ < 0) /* see comments for YYLLOC_DEFAULT */ + @$ = @2; + n->stmt_location = @$; $$ = (Node *) n; } ; @@ -12474,6 +12485,9 @@ MergeStmt: m->joinCondition = $8; m->mergeWhenClauses = $9; m->returningList = $10; + if (@$ < 0) /* see comments for YYLLOC_DEFAULT */ + @$ = @2; + m->stmt_location = @$; $$ = (Node *) m; } @@ -12714,7 +12728,12 @@ SelectStmt: select_no_parens %prec UMINUS ; select_with_parens: - '(' select_no_parens ')' { $$ = $2; } + '(' select_no_parens ')' + { + SelectStmt *n = (SelectStmt *) $2; + n->stmt_len = @3 - @2; + $$ = $2; + } | '(' select_with_parens ')' { $$ = $2; } ; @@ -12836,6 +12855,7 @@ simple_select: n->groupDistinct = ($7)->distinct; n->havingClause = $8; n->windowClause = $9; + n->stmt_location = @1; $$ = (Node *) n; } | SELECT distinct_clause target_list @@ -12853,6 +12873,7 @@ simple_select: n->groupDistinct = ($7)->distinct; n->havingClause = $8; n->windowClause = $9; + n->stmt_location = @1; $$ = (Node *) n; } | values_clause { $$ = $1; } @@ -12873,19 +12894,20 @@ simple_select: n->targetList = list_make1(rt); n->fromClause = list_make1($2); + n->stmt_location = @1; $$ = (Node *) n; } | select_clause UNION set_quantifier select_clause { - $$ = makeSetOp(SETOP_UNION, $3 == SET_QUANTIFIER_ALL, $1, $4); + $$ = makeSetOp(SETOP_UNION, $3 == SET_QUANTIFIER_ALL, $1, $4, @1); } | select_clause INTERSECT set_quantifier select_clause { - $$ = makeSetOp(SETOP_INTERSECT, $3 == SET_QUANTIFIER_ALL, $1, $4); + $$ = makeSetOp(SETOP_INTERSECT, $3 == SET_QUANTIFIER_ALL, $1, $4, @1); } | select_clause EXCEPT set_quantifier select_clause { - $$ = makeSetOp(SETOP_EXCEPT, $3 == SET_QUANTIFIER_ALL, $1, $4); + $$ = makeSetOp(SETOP_EXCEPT, $3 == SET_QUANTIFIER_ALL, $1, $4, @1); } ; @@ -13443,6 +13465,7 @@ values_clause: { SelectStmt *n = makeNode(SelectStmt); + n->stmt_location = @1; n->valuesLists = list_make1($3); $$ = (Node *) n; } @@ -18585,6 +18608,39 @@ updateRawStmtEnd(RawStmt *rs, int end_location) rs->stmt_len = end_location - rs->stmt_location; } +/* Adjust a PreparableStmt to reflect that it doesn't run to the end of the string */ +static void +updatePreparableStmtEnd(Node *n, int end_location) +{ + if (IsA(n, SelectStmt)) + { + SelectStmt *stmt = (SelectStmt *)n; + stmt->stmt_len = end_location - stmt->stmt_location; + } + else if (IsA(n, InsertStmt)) + { + InsertStmt *stmt = (InsertStmt *)n; + stmt->stmt_len = end_location - stmt->stmt_location; + } + else if (IsA(n, UpdateStmt)) + { + UpdateStmt *stmt = (UpdateStmt *)n; + stmt->stmt_len = end_location - stmt->stmt_location; + } + else if (IsA(n, DeleteStmt)) + { + DeleteStmt *stmt = (DeleteStmt *)n; + stmt->stmt_len = end_location - stmt->stmt_location; + } + else if (IsA(n, MergeStmt)) + { + MergeStmt *stmt = (MergeStmt *)n; + stmt->stmt_len = end_location - stmt->stmt_location; + } + else + elog(ERROR, "unexpected node type %d", (int) n->type); +} + static Node * makeColumnRef(char *colname, List *indirection, int location, core_yyscan_t yyscanner) @@ -18963,11 +19019,13 @@ insertSelectOptions(SelectStmt *stmt, errmsg("multiple WITH clauses not allowed"), parser_errposition(exprLocation((Node *) withClause)))); stmt->withClause = withClause; + /* Update SelectStmt's location to the start of the with clause */ + stmt->stmt_location = withClause->location; } } static Node * -makeSetOp(SetOperation op, bool all, Node *larg, Node *rarg) +makeSetOp(SetOperation op, bool all, Node *larg, Node *rarg, int location) { SelectStmt *n = makeNode(SelectStmt); @@ -18975,6 +19033,7 @@ makeSetOp(SetOperation op, bool all, Node *larg, Node *rarg) n->all = all; n->larg = (SelectStmt *) larg; n->rarg = (SelectStmt *) rarg; + n->stmt_location = location; return (Node *) n; } diff --git a/src/backend/parser/parse_merge.c b/src/backend/parser/parse_merge.c index 87df79027d7..fdfa61868cf 100644 --- a/src/backend/parser/parse_merge.c +++ b/src/backend/parser/parse_merge.c @@ -118,6 +118,8 @@ transformMergeStmt(ParseState *pstate, MergeStmt *stmt) Assert(pstate->p_ctenamespace == NIL); qry->commandType = CMD_MERGE; + qry->stmt_location = stmt->stmt_location; + setQueryStmtLen(pstate, qry, stmt->stmt_len); qry->hasRecursive = false; /* process the WITH clause independently of all else */ diff --git a/src/include/nodes/parsenodes.h b/src/include/nodes/parsenodes.h index c92cef3d16d..b40b661ec8a 100644 --- a/src/include/nodes/parsenodes.h +++ b/src/include/nodes/parsenodes.h @@ -2046,6 +2046,8 @@ typedef struct InsertStmt List *returningList; /* list of expressions to return */ WithClause *withClause; /* WITH clause */ OverridingKind override; /* OVERRIDING clause */ + ParseLoc stmt_location; /* start location, or -1 if unknown */ + ParseLoc stmt_len; /* length in bytes; 0 means "rest of string" */ } InsertStmt; /* ---------------------- @@ -2060,6 +2062,8 @@ typedef struct DeleteStmt Node *whereClause; /* qualifications */ List *returningList; /* list of expressions to return */ WithClause *withClause; /* WITH clause */ + ParseLoc stmt_location; /* start location, or -1 if unknown */ + ParseLoc stmt_len; /* length in bytes; 0 means "rest of string" */ } DeleteStmt; /* ---------------------- @@ -2075,6 +2079,8 @@ typedef struct UpdateStmt List *fromClause; /* optional from clause for more tables */ List *returningList; /* list of expressions to return */ WithClause *withClause; /* WITH clause */ + ParseLoc stmt_location; /* start location, or -1 if unknown */ + ParseLoc stmt_len; /* length in bytes; 0 means "rest of string" */ } UpdateStmt; /* ---------------------- @@ -2090,6 +2096,8 @@ typedef struct MergeStmt List *mergeWhenClauses; /* list of MergeWhenClause(es) */ List *returningList; /* list of expressions to return */ WithClause *withClause; /* WITH clause */ + ParseLoc stmt_location; /* start location, or -1 if unknown */ + ParseLoc stmt_len; /* length in bytes; 0 means "rest of string" */ } MergeStmt; /* ---------------------- @@ -2159,6 +2167,8 @@ typedef struct SelectStmt bool all; /* ALL specified? */ struct SelectStmt *larg; /* left child */ struct SelectStmt *rarg; /* right child */ + ParseLoc stmt_location; /* start location, or -1 if unknown */ + ParseLoc stmt_len; /* length in bytes; 0 means "rest of string" */ /* Eventually add fields for CORRESPONDING spec here */ } SelectStmt; diff --git a/src/include/parser/analyze.h b/src/include/parser/analyze.h index 28b66fccb43..8ba4e050aff 100644 --- a/src/include/parser/analyze.h +++ b/src/include/parser/analyze.h @@ -46,8 +46,9 @@ extern List *transformUpdateTargetList(ParseState *pstate, List *origTlist); extern List *transformReturningList(ParseState *pstate, List *returningList, ParseExprKind exprKind); -extern Query *transformTopLevelStmt(ParseState *pstate, RawStmt *parseTree); +extern Query *transformOptionalSelectInto(ParseState *pstate, Node *parseTree); extern Query *transformStmt(ParseState *pstate, Node *parseTree); +extern void setQueryStmtLen(ParseState *pstate, Query *qry, int stmt_len); extern bool stmt_requires_parse_analysis(RawStmt *parseTree); extern bool analyze_requires_snapshot(RawStmt *parseTree); diff --git a/src/include/parser/parse_node.h b/src/include/parser/parse_node.h index 543df568147..ba572b3aea2 100644 --- a/src/include/parser/parse_node.h +++ b/src/include/parser/parse_node.h @@ -193,6 +193,8 @@ struct ParseState { ParseState *parentParseState; /* stack link */ const char *p_sourcetext; /* source text, or NULL if not available */ + ParseLoc p_stmt_location; /* start location, or -1 if unknown */ + ParseLoc p_stmt_len; /* length in bytes; 0 means "rest of string" */ List *p_rtable; /* range table so far */ List *p_rteperminfos; /* list of RTEPermissionInfo nodes for each * RTE_RELATION entry in rtable */ -- 2.39.5 (Apple Git-154)