commands following asynchronous list are executed twice
While reading commands from stdin, if job controls are enabled,
commands following an asynchronous list in `{ cmd; } &` fashion are
executed twice. See below extract:
$ bash -m <<\EOD
{ sleep 1; } &
wait
echo $BASHPID
EOD
10615
10615
When job controls are disabled, or an external utility is called in
the same line this behavior disappears. E.g.:
$ bash -m <<\EOD
{ sleep 1; } & x
wait
echo $BASHPID
EOD
./bash: line 1: x: command not found
10618
This is reproducible on 4.4.20(1)-release, 5.0.11(1)-release and 5.0.16(1)-maint
--
Oğuz
verbosity of DEBUG trap following edit-and-execute-command
After the editor invoked by edit-and-execute-command exits, the about-to-be-executed command (as edited by the editor) is echoed, but seemingly as if [set -v] was set, causing a command executed before the edited command as the result of a trap DEBUG to be echoed, as well. This is annoying for example when using the DEBUG trap as a way to change colors as demonstrated in https://nigeltao.github.io/blog/2018/colorful-text.html#shell-prompts It would be nicer if the set -v was undone before the trap's handler was executed so that the printf wasn't shown to the user in the example below. Repro recipe follows. The initial echo demonstrates that the trap is silent in the absence of edit-and-execute-command. The second echo command has C-x C-e hit after the "echo He" is typed, and the line containing printf (third from the bottom) is what I'm hoping to avoid. ~ $ docker run -it bash:5.0.16 bash-5.0# EDITOR=ed bash-5.0# trap 'printf "\e[0m"' DEBUG bash-5.0# echo Hello Hello bash-5.0# echo He "/tmp/bash-fc.INbEkF", 1 lines, 8 chars : 1s/He/Hello/ : w "/tmp/bash-fc.INbEkF", 1 lines, 11 chars : q echo Hello printf "\e[0m" Hello bash-5.0# Note this also repros in 5.0.11(1) as well as 3.2.57(1). Cheers, -a
Re: commands following asynchronous list are executed twice
On 4/2/20 8:09 AM, Oğuz wrote:
> While reading commands from stdin, if job controls are enabled,
> commands following an asynchronous list in `{ cmd; } &` fashion are
> executed twice.
Thanks for the report. This will be fixed in the next devel branch push.
--
``The lyf so short, the craft so long to lerne.'' - Chaucer
``Ars longa, vita brevis'' - Hippocrates
Chet Ramey, UTech, CWRUc...@case.eduhttp://tiswww.cwru.edu/~chet/
Re: Obscure issue with process substitution since bash-5.0.016 upgrade
On 4/1/20 10:15 PM, Joan Bruguera Micó wrote:
> I'm having a problem with a script that used to work, which I reduced
> to the following test case:
>
> #!/bin/bash
>
> function docat() {
> for x in 1 2 3 ; do true; done
> cat "$@"
> }
>
> for x in $(seq 25); do
> docat <(echo a) <(echo a) <(echo a) <(echo a) <(echo a) <(echo a)
> done
>
> Expected behaviour: A lot of lines with an "a" are printed
Thanks for the report.
Yes, this is a problem, as you discovered, with bash-5.0 patch 16. There is
a better way to solve the problem that patch attempts to solve, and the
enclosed patch to bash-5.0.16 does it.
Chet
--
``The lyf so short, the craft so long to lerne.'' - Chaucer
``Ars longa, vita brevis'' - Hippocrates
Chet Ramey, UTech, CWRUc...@case.eduhttp://tiswww.cwru.edu/~chet/
*** ../bash-5.0-patched/subst.c 2019-08-29 11:16:49.0 -0400
--- subst.c 2020-04-02 16:24:19.0 -0400
***
*** 5337,5341
}
! char *
copy_fifo_list (sizep)
int *sizep;
--- 5337,5341
}
! void *
copy_fifo_list (sizep)
int *sizep;
***
*** 5343,5347
if (sizep)
*sizep = 0;
! return (char *)NULL;
}
--- 5343,5347
if (sizep)
*sizep = 0;
! return (void *)NULL;
}
***
*** 5409,5414
if (fifo_list[i].file)
{
! fifo_list[j].file = fifo_list[i].file;
! fifo_list[j].proc = fifo_list[i].proc;
j++;
}
--- 5409,5419
if (fifo_list[i].file)
{
! if (i != j)
! {
! fifo_list[j].file = fifo_list[i].file;
! fifo_list[j].proc = fifo_list[i].proc;
! fifo_list[i].file = (char *)NULL;
! fifo_list[i].proc = 0;
! }
j++;
}
***
*** 5426,5433
void
close_new_fifos (list, lsize)
! char *list;
int lsize;
{
int i;
if (list == 0)
--- 5431,5439
void
close_new_fifos (list, lsize)
! void *list;
int lsize;
{
int i;
+ char *plist;
if (list == 0)
***
*** 5437,5442
}
! for (i = 0; i < lsize; i++)
! if (list[i] == 0 && i < fifo_list_size && fifo_list[i].proc != -1)
unlink_fifo (i);
--- 5443,5448
}
! for (plist = (char *)list, i = 0; i < lsize; i++)
! if (plist[i] == 0 && i < fifo_list_size && fifo_list[i].proc != -1)
unlink_fifo (i);
***
*** 5560,5568
}
! char *
copy_fifo_list (sizep)
int *sizep;
{
! char *ret;
if (nfds == 0 || totfds == 0)
--- 5566,5574
}
! void *
copy_fifo_list (sizep)
int *sizep;
{
! void *ret;
if (nfds == 0 || totfds == 0)
***
*** 5570,5579
if (sizep)
*sizep = 0;
! return (char *)NULL;
}
if (sizep)
*sizep = totfds;
! ret = (char *)xmalloc (totfds * sizeof (pid_t));
return (memcpy (ret, dev_fd_list, totfds * sizeof (pid_t)));
}
--- 5576,5585
if (sizep)
*sizep = 0;
! return (void *)NULL;
}
if (sizep)
*sizep = totfds;
! ret = xmalloc (totfds * sizeof (pid_t));
return (memcpy (ret, dev_fd_list, totfds * sizeof (pid_t)));
}
***
*** 5648,5655
void
close_new_fifos (list, lsize)
! char *list;
int lsize;
{
int i;
if (list == 0)
--- 5654,5662
void
close_new_fifos (list, lsize)
! void *list;
int lsize;
{
int i;
+ pid_t *plist;
if (list == 0)
***
*** 5659,5664
}
! for (i = 0; i < lsize; i++)
! if (list[i] == 0 && i < totfds && dev_fd_list[i])
unlink_fifo (i);
--- 5666,5671
}
! for (plist = (pid_t *)list, i = 0; i < lsize; i++)
! if (plist[i] == 0 && i < totfds && dev_fd_list[i])
unlink_fifo (i);
*** ../bash-5.0-patched/subst.h 2018-10-21 18:46:09.0 -0400
--- subst.h 2020-04-02 16:29:28.0 -0400
***
*** 274,280
extern void unlink_fifo __P((int));
! extern char *copy_fifo_list __P((int *));
! extern void unlink_new_fifos __P((char *, int));
! extern void close_new_fifos __P((char *, int));
extern void clear_fifo_list __P((void));
--- 274,279
extern void unlink_fifo __P((int));
! extern void *copy_fifo_list __P((int *));
! extern void close_new_fifos __P((void *, int));
extern void clear_fifo_list __P((void));
*** ../bash-5.0-patched/execute_cmd.c 2020-02-06 20:16:48.0 -0500
--- execute_cmd.c 2020-04-02 17:00:10.0 -0400
***
*** 565,569
#if defined (PROCESS_SUBSTITUTION)
volatile int ofifo, nfifo, osize, saved_fifo;
! volatile char *ofifo_list;
#endif
--- 565,569
#if defined (PROCESS_SUBSTITUTION)
volatile int ofifo, nfifo, osize, saved_fifo;
! volatile vo
Re: Obscure issue with process substitution since bash-5.0.016 upgrade
Thanks for your attention and the provided reduced patch. I can
confirm that, applied over bash-5.0.16, this solves my issue, both on
the test case as well as the bigger application from which it was
extracted.
Regards,
- Joan Bruguera
On Thu, 2 Apr 2020 at 23:20, Chet Ramey wrote:
>
> On 4/1/20 10:15 PM, Joan Bruguera Micó wrote:
> > I'm having a problem with a script that used to work, which I reduced
> > to the following test case:
> >
> > #!/bin/bash
> >
> > function docat() {
> > for x in 1 2 3 ; do true; done
> > cat "$@"
> > }
> >
> > for x in $(seq 25); do
> > docat <(echo a) <(echo a) <(echo a) <(echo a) <(echo a) <(echo a)
> > done
> >
> > Expected behaviour: A lot of lines with an "a" are printed
>
> Thanks for the report.
>
> Yes, this is a problem, as you discovered, with bash-5.0 patch 16. There is
> a better way to solve the problem that patch attempts to solve, and the
> enclosed patch to bash-5.0.16 does it.
>
> Chet
>
> --
> ``The lyf so short, the craft so long to lerne.'' - Chaucer
> ``Ars longa, vita brevis'' - Hippocrates
> Chet Ramey, UTech, CWRUc...@case.eduhttp://tiswww.cwru.edu/~chet/
--
Joan Bruguera Micó - PGP key id: 88A7 A061 6B47 0CE1 EB4E D431 8744
44D1 21CE B72E
