Hi Paul,
Paul Eggert <[email protected]> writes:
> On 2025-06-08 17:39, Collin Funk wrote:
>> Posting on bug-gnulib since this test failure seems to be caused by
>> parse-datetime despite being caught by tests/date/date-debug.sh in
>> Coreutils.
>
> I installed the attached patch to Gnulib, along with a minor testcase
> patch in Coreutils. Please give it a try.
This fails with the following diff:
+ diff -u exp2 out2
--- exp2 2025-06-10 04:15:06.472911681 +0000
+++ out2 2025-06-10 04:15:06.518214174 +0000
@@ -4,10 +4,10 @@
date: using specified time as starting value: '02:30:00'
date: error: invalid date/time value:
date: user provided time: '(Y-M-D) 2006-04-02 02:30:00'
-date: normalized time: '(Y-M-D) 2006-04-02 XX:XX:XX'
-date: --
+date: time could not be normalized
date: possible reasons:
date: nonexistent due to daylight-saving time;
date: invalid day/month combination;
+date: numeric values overflow;
date: missing timezone
date: invalid date 'TZ="America/Edmonton" 2006-04-02 02:30:00'
The cause can be seen in 'gdb'. On GNU/Linux:
debug_mktime_not_ok (tm0=0x7fffffffcbd0, tm1=0x7fffffffcb90,
pc=0x7fffffffcde0, time_zone_seen=false) at
/home/collin/.local/src/coreutils/parse-datetime.y:1698
1698 if (mktime_failed)
(gdb) print mktime_failed
$1 = false
(gdb) print *tm1
$2 = {
tm_sec = 0,
tm_min = 30,
tm_hour = 3,
tm_mday = 2,
tm_mon = 3,
tm_year = 106,
tm_wday = 0,
tm_yday = 91,
tm_isdst = 1,
tm_gmtoff = -21600,
tm_zone = 0x4184da "MDT"
}
On NetBSD 10.0:
debug_mktime_not_ok (time_zone_seen=false, pc=0xffffffcc6940,
tm1=0xffffffcc6728, tm0=0xffffffcc6760) at parse-datetime.y:1698
1698 if (mktime_failed)
(gdb) print *tm1
$1 = {tm_sec = 0, tm_min = 30, tm_hour = 2, tm_mday = 2, tm_mon = 3,
tm_year = 106, tm_wday = -1, tm_yday = 0, tm_isdst = -1, tm_gmtoff =
281474674135736,
tm_zone = 0x15000 <error: Cannot access memory at address 0x15000>}
(gdb) print mktime_failed
$2 = true
Collin
