Fix quoted-substring handling in format parsing for to_char/to_number/etc.

This code evidently intended to treat backslash as an escape character
within double-quoted substrings, but it was sufficiently confused that
cases like ..."foo\\"... did not work right: the second backslash
managed to quote the double-quote after it, despite being quoted itself.
Rewrite to get that right, while preserving the existing behavior
outside double-quoted substrings, which is that backslash isn't special
except in the combination \".

Comparing to Oracle, it seems that their version of to_char() for
timestamps allows literal alphanumerics only within double quotes, while
non-alphanumerics are allowed outside quotes; backslashes aren't special
anywhere; there is no way at all to emit a literal double quote.
(Bizarrely, their to_char() for numbers is different; it doesn't allow
literal text at all AFAICT.)  The fact that they don't treat backslash
as special justifies our existing behavior for backslash outside double
quotes.  I considered making backslash inside double quotes act the same
way (ie, special only if before "), which in a green field would be a
more consistent behavior.  But that would likely break more existing SQL
code than what this patch does.

Add some test cases illustrating this behavior.  (Only the last new
case actually changes behavior in this commit.)

Little of this behavior was documented, either, so fix that.

Discussion: https://postgr.es/m/3626.1510949486@sss.pgh.pa.us

1 files changed, 26 insertions, 44 deletions

diff --git a/src/backend/utils/adt/formatting.c b/src/backend/utils/adt/formatting.c
index 5afc293a5a0..cb0dbf748e5 100644
--- a/src/backend/utils/adt/formatting.c
+++ b/src/backend/utils/adt/formatting.c
@@ -1227,11 +1227,7 @@ static void
 parse_format(FormatNode *node, const char *str, const KeyWord *kw,
 			 const KeySuffix *suf, const int *index, int ver, NUMDesc *Num)
 {
-	const KeySuffix *s;
 	FormatNode *n;
-	int			node_set = 0,
-				suffix,
-				last = 0;
 
 #ifdef DEBUG_TO_FROM_CHAR
 	elog(DEBUG_elog_output, "to_char/number(): run parser");
@@ -1241,12 +1237,14 @@ parse_format(FormatNode *node, const char *str, const KeyWord *kw,
 
 	while (*str)
 	{
-		suffix = 0;
+		int			suffix = 0;
+		const KeySuffix *s;
 
 		/*
 		 * Prefix
 		 */
-		if (ver == DCH_TYPE && (s = suff_search(str, suf, SUFFTYPE_PREFIX)) != NULL)
+		if (ver == DCH_TYPE &&
+			(s = suff_search(str, suf, SUFFTYPE_PREFIX)) != NULL)
 		{
 			suffix |= s->id;
 			if (s->len)
@@ -1259,8 +1257,7 @@ parse_format(FormatNode *node, const char *str, const KeyWord *kw,
 		if (*str && (n->key = index_seq_search(str, kw, index)) != NULL)
 		{
 			n->type = NODE_TYPE_ACTION;
-			n->suffix = 0;
-			node_set = 1;
+			n->suffix = suffix;
 			if (n->key->len)
 				str += n->key->len;
 
@@ -1273,71 +1270,56 @@ parse_format(FormatNode *node, const char *str, const KeyWord *kw,
 			/*
 			 * Postfix
 			 */
-			if (ver == DCH_TYPE && *str && (s = suff_search(str, suf, SUFFTYPE_POSTFIX)) != NULL)
+			if (ver == DCH_TYPE && *str &&
+				(s = suff_search(str, suf, SUFFTYPE_POSTFIX)) != NULL)
 			{
-				suffix |= s->id;
+				n->suffix |= s->id;
 				if (s->len)
 					str += s->len;
 			}
+
+			n++;
 		}
 		else if (*str)
 		{
 			/*
-			 * Special characters '\' and '"'
+			 * Process double-quoted literal string, if any
 			 */
-			if (*str == '"' && last != '\\')
+			if (*str == '"')
 			{
-				int			x = 0;
-
 				while (*(++str))
 				{
-					if (*str == '"' && x != '\\')
+					if (*str == '"')
 					{
 						str++;
 						break;
 					}
-					else if (*str == '\\' && x != '\\')
-					{
-						x = '\\';
-						continue;
-					}
+					/* backslash quotes the next character, if any */
+					if (*str == '\\' && *(str + 1))
+						str++;
 					n->type = NODE_TYPE_CHAR;
 					n->character = *str;
 					n->key = NULL;
 					n->suffix = 0;
-					++n;
-					x = *str;
+					n++;
 				}
-				node_set = 0;
-				suffix = 0;
-				last = 0;
 			}
-			else if (*str && *str == '\\' && last != '\\' && *(str + 1) == '"')
-			{
-				last = *str;
-				str++;
-			}
-			else if (*str)
+			else
 			{
+				/*
+				 * Outside double-quoted strings, backslash is only special if
+				 * it immediately precedes a double quote.
+				 */
+				if (*str == '\\' && *(str + 1) == '"')
+					str++;
 				n->type = NODE_TYPE_CHAR;
 				n->character = *str;
 				n->key = NULL;
-				node_set = 1;
-				last = 0;
+				n->suffix = 0;
+				n++;
 				str++;
 			}
 		}
-
-		/* end */
-		if (node_set)
-		{
-			if (n->type == NODE_TYPE_ACTION)
-				n->suffix = suffix;
-			++n;
-
-			n->suffix = 0;
-			node_set = 0;
-		}
 	}
 
 	n->type = NODE_TYPE_END;