Avoid character classification in regex escape parsing.

For regex escape sequences, just test directly for the relevant ASCII
characters rather than using locale-sensitive character
classification.

This fixes an assertion failure when a locale considers a non-ASCII
character, such as "൧", to be a digit.

Reported-by: Richard Guo
Discussion: https://postgr.es/m/CAMbWs49Q6UoKGeT8pBkMtJGJd+16CBFZaaWUk9Du+2ERE5g_YA@mail.gmail.com
Backpatch-through: 11

diff --git a/src/backend/regex/regc_lex.c b/src/backend/regex/regc_lex.c
index d573032db67a610b3bd95191ad14ed79f3a8bec5..4e6fa5c61325df7ab183b00000164dac8974b3b0 100644 (file)
--- a/src/backend/regex/regc_lex.c
+++ b/src/backend/regex/regc_lex.c
@@ -750,7 +750,11 @@ lexescape(struct vars *v)
 
        assert(!ATEOS());
        c = *v->now++;
-       if (!iscalnum(c))
+
+       /* if it's not alphanumeric ASCII, treat it as a plain character */
+       if (!('a' <= c && c <= 'z') &&
+               !('A' <= c && c <= 'Z') &&
+               !('0' <= c && c <= '9'))
                RETV(PLAIN, c);
 
        NOTE(REG_UNONPOSIX);
@@ -892,8 +896,11 @@ lexescape(struct vars *v)
                        RETV(PLAIN, c);
                        break;
                default:
-                       assert(iscalpha(c));
-                       FAILW(REG_EESCAPE); /* unknown alphabetic escape */
+                       /*
+                        * Throw an error for unrecognized ASCII alpha escape sequences,
+                        * which reserves them for future use if needed.
+                        */
+                       FAILW(REG_EESCAPE);
                        break;
        }
        assert(NOTREACHED);