Commit
e39f99046 moved some code up closer to the start of
DecodeInterval(), without noticing that it had been implicitly
relying on previous checks to reject the case of empty input.
Given empty input, we'd now dereference a pointer that hadn't been
set, possibly leading to a core dump. (But if we fail to provoke
a SIGSEGV, nothing bad happens, and the expected syntax error is
thrown a bit later.)
Per bug #17788 from Alexander Lakhin. Back-patch to v15 where
the fault was introduced.
Discussion: https://postgr.es/m/17788-
dabac9f98f7eafd5@postgresql.org
* to dump in postgres style, not SQL style.)
*----------
*/
- if (IntervalStyle == INTSTYLE_SQL_STANDARD && *field[0] == '-')
+ if (IntervalStyle == INTSTYLE_SQL_STANDARD && nf > 0 && *field[0] == '-')
{
force_negative = true;
/* Check for additional explicit signs */
-23:45:12.34 | -1 23:45:12.34 | -1-2 -1 -23:45:12.34 | -0-10 +1 +23:45:12.34
(1 row)
+-- edge case for sign-matching rules
+SELECT interval ''; -- error
+ERROR: invalid input syntax for type interval: ""
+LINE 1: SELECT interval '';
+ ^
-- test outputting iso8601 intervals
SET IntervalStyle to iso_8601;
select interval '0' AS "zero",
interval '-1 year 2 months 1 day 23 hours 45 min 12.34 sec',
interval '-1 year 2 months 1 day 23 hours 45 min +12.34 sec';
+-- edge case for sign-matching rules
+SELECT interval ''; -- error
+
-- test outputting iso8601 intervals
SET IntervalStyle to iso_8601;
select interval '0' AS "zero",