v6-0007-refactor-ReadArrayStr.patch

From 827f6d75a97a065374a368f5ba772f14c53b72b1 Mon Sep 17 00:00:00 2001
From: pgaddict <jian.universality@gmail.com>
Date: Mon, 11 Sep 2023 12:31:39 +0800
Subject: [PATCH v6 7/7] refactor ReadArrayStr.

array ending error can happen in any of cases(quoted_element, unquoted_element, with backslash, non-backslash).
conslidate it, using a goto ending_errror.
add more tests.
expect_delim state is not enought for case like '{{},}'. You need tracking
current token and previous token to solve cases where a delimiter followed by
a closing brace.
---
 src/backend/utils/adt/arrayfuncs.c   | 55 ++++++++++------------------
 src/test/regress/expected/arrays.out | 28 ++++++++++++++
 src/test/regress/sql/arrays.sql      |  7 ++++
 3 files changed, 54 insertions(+), 36 deletions(-)

diff --git a/src/backend/utils/adt/arrayfuncs.c b/src/backend/utils/adt/arrayfuncs.c
index 79f4e1b4..b8f5d59c 100644
--- a/src/backend/utils/adt/arrayfuncs.c
+++ b/src/backend/utils/adt/arrayfuncs.c
@@ -625,7 +625,7 @@ ReadArrayStr(char **srcptr,
 	do
 	{
 		ArrayToken	tok;
-
+		ArrayToken	prev_tok;
 		tok = ReadArrayToken(srcptr, elembuf, typdelim, origStr, escontext);
 
 		switch (tok)
@@ -655,7 +655,8 @@ ReadArrayStr(char **srcptr,
 				}
 				break;
 			case ATOK_LEVEL_END:
-				if (nest_level == 0)
+				/* cannot precede with a delim. nest_level should larger than 0 */
+				if (prev_tok == ATOK_DELIM || nest_level == 0)
 					ereturn(escontext, false,
 							(errcode(ERRCODE_INVALID_TEXT_REPRESENTATION),
 							 errmsg("malformed array literal: \"%s\"", origStr),
@@ -739,6 +740,7 @@ ReadArrayStr(char **srcptr,
 			case ATOK_ERROR:
 				return false;
 		}
+		prev_tok = tok;
 	} while (nest_level > 0);
 
 	pfree(elembuf);
@@ -788,12 +790,7 @@ ReadArrayToken(char **srcptr, char *elembuf, char typdelim,
 		switch (*p)
 		{
 			case '\0':
-				errsave(escontext,
-						(errcode(ERRCODE_INVALID_TEXT_REPRESENTATION),
-						 errmsg("malformed array literal: \"%s\"",
-								origStr)));
-				*srcptr = p;
-				return ATOK_ERROR;
+				goto ending_error;
 			case '{':
 				*srcptr = p + 1;
 				return ATOK_LEVEL_START;
@@ -825,24 +822,12 @@ quoted_element:
 		switch (*p)
 		{
 			case '\0':
-				errsave(escontext,
-						(errcode(ERRCODE_INVALID_TEXT_REPRESENTATION),
-						 errmsg("malformed array literal: \"%s\"",
-								origStr)));
-				*srcptr = p;
-				return ATOK_ERROR;
+				goto ending_error;
 			case '\\':
 				/* Skip backslash, copy next character as-is. */
 				p++;
 				if (*p == '\0')
-				{
-					errsave(escontext,
-							(errcode(ERRCODE_INVALID_TEXT_REPRESENTATION),
-							 errmsg("malformed array literal: \"%s\"",
-									origStr)));
-					*srcptr = p;
-					return ATOK_ERROR;
-				}
+					goto ending_error;
 				*dst++ = *p++;
 				break;
 			case '"':
@@ -863,25 +848,14 @@ unquoted_element:
 		{
 			case '\0':
 			case '{':
-				errsave(escontext,
-						(errcode(ERRCODE_INVALID_TEXT_REPRESENTATION),
-						 errmsg("malformed array literal: \"%s\"",
-								origStr)));
-				*srcptr = p;
-				return ATOK_ERROR;
+				goto ending_error;
 			case '\\':
 				/* Skip backslash, copy next character as-is. */
 				p++;
 				if (*p == '\0')
-				{
-					errsave(escontext,
-							(errcode(ERRCODE_INVALID_TEXT_REPRESENTATION),
-							 errmsg("malformed array literal: \"%s\"",
-									origStr)));
-					*srcptr = p;
-					return ATOK_ERROR;
-				}
+					goto ending_error;
 				*dst++ = *p++;
+				dstendptr = dst;
 				break;
 			case '"':
 				errsave(escontext,
@@ -907,6 +881,15 @@ unquoted_element:
 				p++;
 		}
 	}
+
+ending_error:
+	errsave(escontext,
+			(errcode(ERRCODE_INVALID_TEXT_REPRESENTATION),
+				errmsg("malformed array literal: \"%s\"",
+					origStr),
+				errdetail("Unexpected end of input.")));
+	*srcptr = p;
+	return ATOK_ERROR;
 }
 
 /*
diff --git a/src/test/regress/expected/arrays.out b/src/test/regress/expected/arrays.out
index bea0d00b..9bb9924f 100644
--- a/src/test/regress/expected/arrays.out
+++ b/src/test/regress/expected/arrays.out
@@ -1544,6 +1544,34 @@ ERROR:  cannot determine type of empty array
 LINE 1: select array[];
                ^
 HINT:  Explicitly cast to the desired type, for example ARRAY[]::integer[].
+select '{{1,},{1},}'::text[];
+ERROR:  malformed array literal: "{{1,},{1},}"
+LINE 1: select '{{1,},{1},}'::text[];
+               ^
+DETAIL:  Unexpected "}" character.
+select '{{1,},{1}}'::text[];
+ERROR:  malformed array literal: "{{1,},{1}}"
+LINE 1: select '{{1,},{1}}'::text[];
+               ^
+DETAIL:  Unexpected "}" character.
+select '{{1,}}'::text[];
+ERROR:  malformed array literal: "{{1,}}"
+LINE 1: select '{{1,}}'::text[];
+               ^
+DETAIL:  Unexpected "}" character.
+select '{1,}'::text[];
+ERROR:  malformed array literal: "{1,}"
+LINE 1: select '{1,}'::text[];
+               ^
+DETAIL:  Unexpected "}" character.
+select '[21474836488:21474836489]={1,2}'::int[];
+ERROR:  array bound is out of range
+LINE 1: select '[21474836488:21474836489]={1,2}'::int[];
+               ^
+select '[-2147483649:-2147483648]={1,2}'::int[];
+ERROR:  array bound is out of range
+LINE 1: select '[-2147483649:-2147483648]={1,2}'::int[];
+               ^
 -- none of the above should be accepted
 -- all of the following should be accepted
 select '{}'::text[];
diff --git a/src/test/regress/sql/arrays.sql b/src/test/regress/sql/arrays.sql
index 5eacb351..6bf18b7d 100644
--- a/src/test/regress/sql/arrays.sql
+++ b/src/test/regress/sql/arrays.sql
@@ -468,6 +468,13 @@ select '[2147483646:2147483647]={1,2}'::int[];
 select '[1:-1]={}'::int[];
 select '[1:0]={1}'::int[];
 select array[];
+select '{{1,},{1},}'::text[];
+select '{{1,},{1}}'::text[];
+select '{{1,}}'::text[];
+select '{1,}'::text[];
+select '[21474836488:21474836489]={1,2}'::int[];
+select '[-2147483649:-2147483648]={1,2}'::int[];
+
 -- none of the above should be accepted
 
 -- all of the following should be accepted
-- 
2.34.1