Re: "at " doesn't work as documented
On Sat, 16 Feb 2019 01:48:11 +0100, Christian Weisgerber wrote: > I think tod() already has the required logic, but it doesn't handle > the case where the time is only past by minutes. Even better. OK millert@ - todd
Re: "at " doesn't work as documented
That looks like the correct fix.
ok deraadt
> > The manual is correct. Below is one way to fix this, though the
> > logic probably belongs in tod() itself.
>
> I think tod() already has the required logic, but it doesn't handle
> the case where the time is only past by minutes.
>
> Index: parsetime.c
> ===
> RCS file: /cvs/src/usr.bin/at/parsetime.c,v
> retrieving revision 1.26
> diff -u -p -r1.26 parsetime.c
> --- parsetime.c 11 Nov 2015 17:42:51 - 1.26
> +++ parsetime.c 16 Feb 2019 00:44:27 -
> @@ -432,7 +432,8 @@ tod(struct tm *tm)
>* a relative offset, it's okay to bump things
>*/
> if ((sc_tokid == EOF || sc_tokid == PLUS || sc_tokid == NEXT) &&
> - tm->tm_hour > hour) {
> + (tm->tm_hour > hour ||
> + (tm->tm_hour == hour && tm->tm_min > minute))) {
> tm->tm_mday++;
> tm->tm_wday++;
> }
> --
> Christian "naddy" Weisgerber na...@mips.inka.de
>
Re: "at " doesn't work as documented
Todd C. Miller:
> The manual is correct. Below is one way to fix this, though the
> logic probably belongs in tod() itself.
I think tod() already has the required logic, but it doesn't handle
the case where the time is only past by minutes.
Index: parsetime.c
===
RCS file: /cvs/src/usr.bin/at/parsetime.c,v
retrieving revision 1.26
diff -u -p -r1.26 parsetime.c
--- parsetime.c 11 Nov 2015 17:42:51 - 1.26
+++ parsetime.c 16 Feb 2019 00:44:27 -
@@ -432,7 +432,8 @@ tod(struct tm *tm)
* a relative offset, it's okay to bump things
*/
if ((sc_tokid == EOF || sc_tokid == PLUS || sc_tokid == NEXT) &&
- tm->tm_hour > hour) {
+ (tm->tm_hour > hour ||
+ (tm->tm_hour == hour && tm->tm_min > minute))) {
tm->tm_mday++;
tm->tm_wday++;
}
--
Christian "naddy" Weisgerber na...@mips.inka.de
Re: "at " doesn't work as documented
On Fri, 15 Feb 2019 22:36:08 +0100, Christian Weisgerber wrote:
> The at(1) man page says:
> at allows some moderately complex timespec specifications. It accepts
> times of the form HHMM or HH:MM to run a job at a specific time of day.
> (If that time is already past, the next day is assumed.)
The manual is correct. Below is one way to fix this, though the
logic probably belongs in tod() itself.
- todd
Index: usr.bin/at/parsetime.c
===
RCS file: /cvs/src/usr.bin/at/parsetime.c,v
retrieving revision 1.26
diff -u -p -u -r1.26 parsetime.c
--- usr.bin/at/parsetime.c 11 Nov 2015 17:42:51 - 1.26
+++ usr.bin/at/parsetime.c 15 Feb 2019 22:16:36 -
@@ -652,8 +652,22 @@ parsetime(int argc, char **argv)
break;
case NUMBER:
- if (tod(&runtime) != 0 || month(&runtime) != 0)
+ if (tod(&runtime) != 0)
return (-1);
+
+ if (sc_tokid == EOF) {
+ /* If time in form HH:MM is past, assume tomorrow. */
+ if (runtime.tm_mday == nowtime.tm_mday &&
+ runtime.tm_wday == nowtime.tm_wday &&
+ (runtime.tm_hour < nowtime.tm_hour ||
+ (runtime.tm_hour == nowtime.tm_hour &&
+ runtime.tm_min < nowtime.tm_min))) {
+ runtime.tm_mday++;
+ runtime.tm_wday++;
+ }
+ } else if (month(&runtime) != 0) {
+ return (-1);
+ }
break;
/*
"at " doesn't work as documented
OpenBSD 6.4-current (GENERIC.MP) #2: Thu Feb 14 21:19:17 CET 2019 na...@ariolc.mips.inka.de:/sys/arch/amd64/compile/GENERIC.MP The at(1) man page says: at allows some moderately complex timespec specifications. It accepts times of the form HHMM or HH:MM to run a job at a specific time of day. (If that time is already past, the next day is assumed.) However: $ date Fri Feb 15 22:32:41 CET 2019 $ at 22:00 at: cannot schedule jobs in the past Rather than assuming the next day, at(1) refuses the job. I don't know if that is an implementation or documentation bug. -- Christian "naddy" Weisgerber na...@mips.inka.de
