zsh-workers
 help / color / mirror / code / Atom feed
* Test release: 5.8.1.2-test
@ 2022-04-09 20:07 dana
  2022-04-09 20:18 ` Bart Schaefer
                   ` (4 more replies)
  0 siblings, 5 replies; 18+ messages in thread
From: dana @ 2022-04-09 20:07 UTC (permalink / raw)
  To: Zsh hackers list

Hello,

I've tagged 5.8.1.2-test (test release for the upcoming zsh 5.9) and
uploaded the artefacts to:

  https://sourceforge.net/projects/zsh/files/zsh-test/5.8.1.2-test/

If you have the time, please test and report any issues found.

dana

PS: The version string format for dev/test releases has changed slightly
since 5.8 was released; see the replies to workers/45028 if you're curious


^ permalink raw reply	[flat|nested] 18+ messages in thread

* Re: Test release: 5.8.1.2-test
  2022-04-09 20:07 Test release: 5.8.1.2-test dana
@ 2022-04-09 20:18 ` Bart Schaefer
  2022-04-10 20:07 ` Peter Stephenson
                   ` (3 subsequent siblings)
  4 siblings, 0 replies; 18+ messages in thread
From: Bart Schaefer @ 2022-04-09 20:18 UTC (permalink / raw)
  To: dana; +Cc: Zsh hackers list

[-- Attachment #1: Type: text/plain, Size: 16 bytes --]

Thank you, Dana

[-- Attachment #2: Type: text/html, Size: 38 bytes --]

^ permalink raw reply	[flat|nested] 18+ messages in thread

* Re: Test release: 5.8.1.2-test
  2022-04-09 20:07 Test release: 5.8.1.2-test dana
  2022-04-09 20:18 ` Bart Schaefer
@ 2022-04-10 20:07 ` Peter Stephenson
  2022-04-12  9:16   ` Peter Stephenson
  2022-04-10 20:21 ` Luna Jernberg
                   ` (2 subsequent siblings)
  4 siblings, 1 reply; 18+ messages in thread
From: Peter Stephenson @ 2022-04-10 20:07 UTC (permalink / raw)
  To: zsh-workers

On Sat, 2022-04-09 at 15:07 -0500, dana wrote:
> I've tagged 5.8.1.2-test (test release for the upcoming zsh 5.9) and
> uploaded the artefacts to:
> 
>   https://sourceforge.net/projects/zsh/files/zsh-test/5.8.1.2-test/
> 
> If you have the time, please test and report any issues found.

Thanks very much for this, fine on my Ubuntu system.  I'll be able to
try out on RedHat and Cygwin tomorrow.

pws



^ permalink raw reply	[flat|nested] 18+ messages in thread

* Re: Test release: 5.8.1.2-test
  2022-04-09 20:07 Test release: 5.8.1.2-test dana
  2022-04-09 20:18 ` Bart Schaefer
  2022-04-10 20:07 ` Peter Stephenson
@ 2022-04-10 20:21 ` Luna Jernberg
  2022-04-11  0:32 ` Axel Beckert
  2022-04-11 10:44 ` Kamil Dudka
  4 siblings, 0 replies; 18+ messages in thread
From: Luna Jernberg @ 2022-04-10 20:21 UTC (permalink / raw)
  To: dana; +Cc: Zsh hackers list, droidbittin

Works as it should on Raspberry Pi OS (Debian) :)

On 4/9/22, dana <dana@dana.is> wrote:
> Hello,
>
> I've tagged 5.8.1.2-test (test release for the upcoming zsh 5.9) and
> uploaded the artefacts to:
>
>   https://sourceforge.net/projects/zsh/files/zsh-test/5.8.1.2-test/
>
> If you have the time, please test and report any issues found.
>
> dana
>
> PS: The version string format for dev/test releases has changed slightly
> since 5.8 was released; see the replies to workers/45028 if you're curious
>
>


^ permalink raw reply	[flat|nested] 18+ messages in thread

* Re: Test release: 5.8.1.2-test
  2022-04-09 20:07 Test release: 5.8.1.2-test dana
                   ` (2 preceding siblings ...)
  2022-04-10 20:21 ` Luna Jernberg
@ 2022-04-11  0:32 ` Axel Beckert
  2022-04-11  8:30   ` Axel Beckert
  2022-04-11 10:44 ` Kamil Dudka
  4 siblings, 1 reply; 18+ messages in thread
From: Axel Beckert @ 2022-04-11  0:32 UTC (permalink / raw)
  To: Zsh hackers list

Hi,

On Sat, Apr 09, 2022 at 03:07:05PM -0500, dana wrote:
> I've tagged 5.8.1.2-test (test release for the upcoming zsh 5.9) and
> uploaded the artefacts to:
> 
>   https://sourceforge.net/projects/zsh/files/zsh-test/5.8.1.2-test/
> 
> If you have the time, please test and report any issues found.

Builds fine on Debian Unstable locally and in a local, clean chroot,
but seems to fail to build in Debian's Gitlab CI due to one test
failure: https://salsa.debian.org/debian/zsh/-/jobs/2661258

The failure I noticed in the log:

7266 Running test: POSIX: width in %s should be computed in bytes, not in characters
7267 Test ../../Test/D07multibyte.ztst was expected to fail, but passed.
7268 Was testing: POSIX: width in %s should be computed in bytes, not in characters
7269 ../../Test/D07multibyte.ztst: test XPassed.
7270 The following may (or may not) help identifying the cause:
7271  This is considered a bugfix in zsh
7272 ../../Test/D08cmdsubst.ztst: starting.

And in the end:

8850 61 successful test scripts, 1 failure, 2 skipped

Hopefully that's the failure mentioned in the summary, because I found
no other.

I've uploaded it to Debian Experimental anyway to see if it builds
fine on the official build daemons. Will see later today.

BTW: I expected that scp tab completion fails with OpenSSH 9.0 as it
uses SFTP as backend now which needs less quoting. But to my surprise
it still seems to work.

		Kind regards, Axel
-- 
PGP: 2FF9CD59612616B5      /~\  Plain Text Ribbon Campaign, http://arc.pasp.de/
Mail: abe@deuxchevaux.org  \ /  Say No to HTML in E-Mail and Usenet
Mail+Jabber: abe@noone.org  X
https://axel.beckert.ch/   / \  I love long mails: https://email.is-not-s.ms/


^ permalink raw reply	[flat|nested] 18+ messages in thread

* Re: Test release: 5.8.1.2-test
  2022-04-11  0:32 ` Axel Beckert
@ 2022-04-11  8:30   ` Axel Beckert
  2022-04-11 19:54     ` Bart Schaefer
  0 siblings, 1 reply; 18+ messages in thread
From: Axel Beckert @ 2022-04-11  8:30 UTC (permalink / raw)
  To: zsh-workers

Hi,

On Mon, Apr 11, 2022 at 02:32:57AM +0200, Axel Beckert wrote:
> >   https://sourceforge.net/projects/zsh/files/zsh-test/5.8.1.2-test/
> > 
> > If you have the time, please test and report any issues found.
> 
> Builds fine on Debian Unstable locally and in a local, clean chroot,
> but seems to fail to build in Debian's Gitlab CI due to one test
> failure: https://salsa.debian.org/debian/zsh/-/jobs/2661258
> 
> The failure I noticed in the log:
> 
> 7266 Running test: POSIX: width in %s should be computed in bytes, not in characters
> 7267 Test ../../Test/D07multibyte.ztst was expected to fail, but passed.
> 7268 Was testing: POSIX: width in %s should be computed in bytes, not in characters
> 7269 ../../Test/D07multibyte.ztst: test XPassed.
> 7270 The following may (or may not) help identifying the cause:
> 7271  This is considered a bugfix in zsh
> 7272 ../../Test/D08cmdsubst.ztst: starting.
> 
> And in the end:
> 
> 8850 61 successful test scripts, 1 failure, 2 skipped
> 
> Hopefully that's the failure mentioned in the summary, because I found
> no other.
> 
> I've uploaded it to Debian Experimental anyway to see if it builds
> fine on the official build daemons. Will see later today.

Looks bad: Test suite failures on most architectures on the build daemons:

https://buildd.debian.org/status/package.php?p=zsh&suite=experimental

Those were the test suite didn't fail seem only 32-bit architectures
(although I'm not sure if all of them really are 32-bit architectures):

i386, hppa, hurd-i386, m68k (and "all" which is actually the
architecture-independent stuff like docs, etc. where the test suite
isn't running).

All other architectures have 1 test suite failure according to the
summary:

61 successful test scripts, 1 failure, 2 skipped

Haven't checked yet, if it always was the same issue (the one cited
above from the "git push" triggered CI), but I currently assume so.

		Kind regards, Axel
-- 
PGP: 2FF9CD59612616B5      /~\  Plain Text Ribbon Campaign, http://arc.pasp.de/
Mail: abe@deuxchevaux.org  \ /  Say No to HTML in E-Mail and Usenet
Mail+Jabber: abe@noone.org  X
https://axel.beckert.ch/   / \  I love long mails: https://email.is-not-s.ms/


^ permalink raw reply	[flat|nested] 18+ messages in thread

* Re: Test release: 5.8.1.2-test
  2022-04-09 20:07 Test release: 5.8.1.2-test dana
                   ` (3 preceding siblings ...)
  2022-04-11  0:32 ` Axel Beckert
@ 2022-04-11 10:44 ` Kamil Dudka
  4 siblings, 0 replies; 18+ messages in thread
From: Kamil Dudka @ 2022-04-11 10:44 UTC (permalink / raw)
  To: dana; +Cc: zsh-workers

On Saturday, April 9, 2022 10:07:05 PM CEST dana wrote:
> Hello,
> 
> I've tagged 5.8.1.2-test (test release for the upcoming zsh 5.9) and
> uploaded the artefacts to:
> 
>   https://sourceforge.net/projects/zsh/files/zsh-test/5.8.1.2-test/
> 
> If you have the time, please test and report any issues found.
> 
> dana
> 
> PS: The version string format for dev/test releases has changed slightly
> since 5.8 was released; see the replies to workers/45028 if you're curious

I was able to successfully build it and run the tests in Fedora and EPEL
build roots:

    https://copr.fedorainfracloud.org/coprs/kdudka/zsh-next/build/4172134/

Thank you!

Kamil




^ permalink raw reply	[flat|nested] 18+ messages in thread

* Re: Test release: 5.8.1.2-test
  2022-04-11  8:30   ` Axel Beckert
@ 2022-04-11 19:54     ` Bart Schaefer
  2022-04-12  2:39       ` Jun T
  0 siblings, 1 reply; 18+ messages in thread
From: Bart Schaefer @ 2022-04-11 19:54 UTC (permalink / raw)
  To: Zsh hackers list

On Mon, Apr 11, 2022 at 1:31 AM Axel Beckert <abe@deuxchevaux.org> wrote:
>
> Hi,
>
> On Mon, Apr 11, 2022 at 02:32:57AM +0200, Axel Beckert wrote:
> >
> > Builds fine on Debian Unstable locally and in a local, clean chroot,
> > but seems to fail to build in Debian's Gitlab CI due to one test
> > failure: https://salsa.debian.org/debian/zsh/-/jobs/2661258

Assuming this is the same failure on the other architectures you tried ...

Can you find out which UTF-8 local the %prep section of D07 is choosing?
Any recent (a year?) changes to mbrtowc() in libraries?
Could the gitlab build be overriding environment locale settings?

That test is really pretty inconsequential, and was recently moved
from the E03posix test to D07multibyte.


^ permalink raw reply	[flat|nested] 18+ messages in thread

* Re: Test release: 5.8.1.2-test
  2022-04-11 19:54     ` Bart Schaefer
@ 2022-04-12  2:39       ` Jun T
  2022-04-12  3:58         ` Bart Schaefer
  0 siblings, 1 reply; 18+ messages in thread
From: Jun T @ 2022-04-12  2:39 UTC (permalink / raw)
  To: zsh-workers


> 2022/04/12 4:54, Bart Schaefer <schaefer@brasslantern.com> wrote:
> 
>> On Mon, Apr 11, 2022 at 02:32:57AM +0200, Axel Beckert wrote:
>>> 
>>> but seems to fail to build in Debian's Gitlab CI due to one test
>>> failure: https://salsa.debian.org/debian/zsh/-/jobs/2661258
> 
> Assuming this is the same failure on the other architectures you tried ...

I've looked into a few failed logs, and they all fail at the same test
(D07, "considered as bugfix").

> Can you find out which UTF-8 local the %prep section of D07 is choosing?

The log says:
7215: Testing multibyte with locale C.UTF-8

The same locale is used when I run the test on my Debian-11.0
(and the test passes, i.e., fails as expected).


^ permalink raw reply	[flat|nested] 18+ messages in thread

* Re: Test release: 5.8.1.2-test
  2022-04-12  2:39       ` Jun T
@ 2022-04-12  3:58         ` Bart Schaefer
  2022-04-12  6:30           ` Jun T
  0 siblings, 1 reply; 18+ messages in thread
From: Bart Schaefer @ 2022-04-12  3:58 UTC (permalink / raw)
  To: Jun T; +Cc: Zsh hackers list

On Mon, Apr 11, 2022 at 7:40 PM Jun T <takimoto-j@kba.biglobe.ne.jp> wrote:
>
> The log says:
> 7215: Testing multibyte with locale C.UTF-8
>
> The same locale is used when I run the test on my Debian-11.0
> (and the test passes, i.e., fails as expected).

The two tests for printf are the only ones that spawn
$ZTST_testdir/../Src/zsh ... perhaps something in /etc/*rc is
re-asserting an LC_* parameter?


^ permalink raw reply	[flat|nested] 18+ messages in thread

* Re: Test release: 5.8.1.2-test
  2022-04-12  3:58         ` Bart Schaefer
@ 2022-04-12  6:30           ` Jun T
  2022-04-12  7:17             ` Jun T
  2022-04-12  9:05             ` Axel Beckert
  0 siblings, 2 replies; 18+ messages in thread
From: Jun T @ 2022-04-12  6:30 UTC (permalink / raw)
  To: zsh-workers

I've found that the test Xpassed if LANG was unset:

% unset LANG
% make TESTNUM=D07 check
./D07multibyte.ztst: test XPassed.

It seems we need to export LANG (in ztst.zsh) to pass it to
$ZTST_testdir/../Src/zsh.


diff --git a/Test/ztst.zsh b/Test/ztst.zsh
index 89fe69b5b..263573333 100755
--- a/Test/ztst.zsh
+++ b/Test/ztst.zsh
@@ -30,6 +30,7 @@ emulate -R zsh
 [[ -n $LC_NUMERIC ]] && LC_NUMERIC=C
 [[ -n $LC_MESSAGES ]] && LC_MESSAGES=C
 [[ -n $LANG ]] && LANG=C
+export LANG
 
 # Don't propagate variables that are set by default in the shell.
 typeset +x WORDCHARS






^ permalink raw reply	[flat|nested] 18+ messages in thread

* Re: Test release: 5.8.1.2-test
  2022-04-12  6:30           ` Jun T
@ 2022-04-12  7:17             ` Jun T
  2022-04-12  9:05             ` Axel Beckert
  1 sibling, 0 replies; 18+ messages in thread
From: Jun T @ 2022-04-12  7:17 UTC (permalink / raw)
  To: zsh-workers


> 2022/04/12 15:30, I wrote:
> 
> It seems we need to export LANG (in ztst.zsh) to pass it to
> $ZTST_testdir/../Src/zsh.

Or is it better to export it in the %prep section of D07
after determining LANG to use?


^ permalink raw reply	[flat|nested] 18+ messages in thread

* Re: Test release: 5.8.1.2-test
  2022-04-12  6:30           ` Jun T
  2022-04-12  7:17             ` Jun T
@ 2022-04-12  9:05             ` Axel Beckert
  2022-04-12 17:55               ` [PATCH] " Axel Beckert
  1 sibling, 1 reply; 18+ messages in thread
From: Axel Beckert @ 2022-04-12  9:05 UTC (permalink / raw)
  To: zsh-workers

[-- Attachment #1: Type: text/plain, Size: 994 bytes --]

Hi,

On Tue, Apr 12, 2022 at 03:30:28PM +0900, Jun T wrote:
> diff --git a/Test/ztst.zsh b/Test/ztst.zsh
> index 89fe69b5b..263573333 100755
> --- a/Test/ztst.zsh
> +++ b/Test/ztst.zsh
> @@ -30,6 +30,7 @@ emulate -R zsh
>  [[ -n $LC_NUMERIC ]] && LC_NUMERIC=C
>  [[ -n $LC_MESSAGES ]] && LC_MESSAGES=C
>  [[ -n $LANG ]] && LANG=C
> +export LANG
>  
>  # Don't propagate variables that are set by default in the shell.
>  typeset +x WORDCHARS

Yay, this patch at least fixed the Gitlab runner builds! See
https://salsa.debian.org/debian/zsh/-/pipelines/367955

I will soon make another upload to the official build infrastructure
on all architectures and will report back how it went.

		Kind regards, Axel
-- 
PGP: 2FF9CD59612616B5      /~\  Plain Text Ribbon Campaign, http://arc.pasp.de/
Mail: abe@deuxchevaux.org  \ /  Say No to HTML in E-Mail and Usenet
Mail+Jabber: abe@noone.org  X
https://axel.beckert.ch/   / \  I love long mails: https://email.is-not-s.ms/

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 833 bytes --]

^ permalink raw reply	[flat|nested] 18+ messages in thread

* Re: Test release: 5.8.1.2-test
  2022-04-10 20:07 ` Peter Stephenson
@ 2022-04-12  9:16   ` Peter Stephenson
  0 siblings, 0 replies; 18+ messages in thread
From: Peter Stephenson @ 2022-04-12  9:16 UTC (permalink / raw)
  To: zsh-workers

> On 10 April 2022 at 21:07 Peter Stephenson <p.w.stephenson@ntlworld.com> wrote:
> On Sat, 2022-04-09 at 15:07 -0500, dana wrote:
> > I've tagged 5.8.1.2-test (test release for the upcoming zsh 5.9) and
> > uploaded the artefacts to:
> > 
> >   https://sourceforge.net/projects/zsh/files/zsh-test/5.8.1.2-test/
> > 
> > If you have the time, please test and report any issues found.
> 
> Thanks very much for this, fine on my Ubuntu system.  I'll be able to
> try out on RedHat and Cygwin tomorrow.

Seems entirely clean on Ubuntu 16.04 and 18.04 and RedHat EL7.

Compiles on Cygwin but the test system is a bit sick on my laptop ---
not actually that unusual.  Unlikely to have time to look at this.

A couple of fixes will remove compilation warnings; it looks like
POSIX systems are pretty safe about signed characters these days
so Cygwin is about the only place where they still cause issues.

pws

diff --git a/Src/exec.c b/Src/exec.c
index 27d49e005..47753da48 100644
--- a/Src/exec.c
+++ b/Src/exec.c
@@ -561,7 +561,7 @@ zexecve(char *pth, char **argv, char **newenvp)
                         isbinary = 1;
                         hasletter = 0;
                         for (ptr = execvebuf; ptr < ptr2; ptr++) {
-                            if (islower(*ptr) || *ptr == '$' || *ptr == '`')
+                            if (islower(STOUC(*ptr)) || *ptr == '$' || *ptr == '`')
                                 hasletter = 1;
                             if (hasletter && *ptr == '\n') {
                                 isbinary = 0;
diff --git a/Src/prompt.c b/Src/prompt.c
index 738c7fc7a..092de63a4 100644
--- a/Src/prompt.c
+++ b/Src/prompt.c
@@ -1666,7 +1666,7 @@ match_colour(const char **teststrp, int is_fg, int colour)
 	tc = TCBGCOLOUR;
     }
     if (teststrp) {
-	if (**teststrp == '#' && isxdigit((*teststrp)[1])) {
+	if (**teststrp == '#' && isxdigit(STOUC((*teststrp)[1]))) {
 	    struct color_rgb color;
 	    char *end;
 	    zlong col = zstrtol(*teststrp+1, &end, 16);


^ permalink raw reply	[flat|nested] 18+ messages in thread

* [PATCH] Re: Test release: 5.8.1.2-test
  2022-04-12  9:05             ` Axel Beckert
@ 2022-04-12 17:55               ` Axel Beckert
  2022-04-12 18:40                 ` Axel Beckert
  0 siblings, 1 reply; 18+ messages in thread
From: Axel Beckert @ 2022-04-12 17:55 UTC (permalink / raw)
  To: zsh-workers

[-- Attachment #1: Type: text/plain, Size: 3797 bytes --]

Hi,

On Tue, Apr 12, 2022 at 11:05:56AM +0200, Axel Beckert wrote:
> On Tue, Apr 12, 2022 at 03:30:28PM +0900, Jun T wrote:
> > diff --git a/Test/ztst.zsh b/Test/ztst.zsh
> > index 89fe69b5b..263573333 100755
> > --- a/Test/ztst.zsh
> > +++ b/Test/ztst.zsh
> > @@ -30,6 +30,7 @@ emulate -R zsh
> >  [[ -n $LC_NUMERIC ]] && LC_NUMERIC=C
> >  [[ -n $LC_MESSAGES ]] && LC_MESSAGES=C
> >  [[ -n $LANG ]] && LANG=C
> > +export LANG
> >  
> >  # Don't propagate variables that are set by default in the shell.
> >  typeset +x WORDCHARS
> 
> Yay, this patch at least fixed the Gitlab runner builds! See
> https://salsa.debian.org/debian/zsh/-/pipelines/367955
> 
> I will soon make another upload to the official build infrastructure
> on all architectures and will report back how it went.

Done now, looks good so far:
https://buildd.debian.org/status/package.php?p=zsh&suite=experimental

I though noticed one odd warning on STDERR when running the test suite
against the installed zsh-static binary in the log from
https://salsa.debian.org/debian/zsh/-/jobs/2665273:

872 **************************************
873 47 successful test scripts, 0 failures, 15 skipped
874 **************************************
875 run-testsuite-static FAIL stderr: (eval):7: permission denied: /file
                                                ^^^^^^^^^^^^^^^^^^^^^^^^

(run-testsuite-static is the script at
https://salsa.debian.org/debian/zsh/-/blob/debian/debian/tests/run-testsuite-static
— This script is meant to run the test suite against /bin/zsh-static,
i.e. to do as-installed-testing. We also skip all PCRE-related tests
as this doesn't seem to work with a really static zsh binary.)

The warning looks a lot as if in a _single_ case of e.g.
"$tst_dir/file" the variable $tst_dir is empty or undefined. And given
that "$tst_dir" only seems to appear in Test/V14system.ztst, it
probably must be in there somewhere. I can also reproduce this warning
with this command:

  % zsh-static +Z -f Test/ztst.zsh Test/V14system.ztst
  Test/V14system.ztst: starting.
  (eval):7: permission denied: /file
  Test/V14system.ztst: skipped (the zsh/system and zsh/zselect modules are not available)

And I think I found the issue:

      3 %prep
      4
      5   if zmodload -s zsh/system && zmodload -s zsh/zselect; then
      6     tst_dir=V14.tmp
      7     mkdir -p -- $tst_dir
      8   else
      9     ZTST_unimplemented='the zsh/system and zsh/zselect modules are not available'
     10   fi
     11   : > $tst_dir/file # File on which to acquire flock.

So line 11 works fine of those modules are there. But if they aren't,
$tst_dir never gets set and this explains very well the odd error
message above.

The question is how to fix this properly. My suggestion would be to
move "tst_dir=V14.tmp" as well as "mkdir -p -- $tst_dir" out of the
conditional, something along these lines:

diff --git a/Test/V14system.ztst b/Test/V14system.ztst
index 100daab08..1e683b5b9 100644
--- a/Test/V14system.ztst
+++ b/Test/V14system.ztst
@@ -2,10 +2,9 @@
 
 %prep
 
-  if zmodload -s zsh/system && zmodload -s zsh/zselect; then
-    tst_dir=V14.tmp
-    mkdir -p -- $tst_dir
-  else
+  tst_dir=V14.tmp
+  mkdir -p -- $tst_dir
+  if ! ( zmodload -s zsh/system && zmodload -s zsh/zselect ); then
     ZTST_unimplemented='the zsh/system and zsh/zselect modules are not available'
   fi
   : > $tst_dir/file # File on which to acquire flock.

		Kind regards, Axel
-- 
PGP: 2FF9CD59612616B5      /~\  Plain Text Ribbon Campaign, http://arc.pasp.de/
Mail: abe@deuxchevaux.org  \ /  Say No to HTML in E-Mail and Usenet
Mail+Jabber: abe@noone.org  X
https://axel.beckert.ch/   / \  I love long mails: https://email.is-not-s.ms/

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 833 bytes --]

^ permalink raw reply	[flat|nested] 18+ messages in thread

* Re: [PATCH] Re: Test release: 5.8.1.2-test
  2022-04-12 17:55               ` [PATCH] " Axel Beckert
@ 2022-04-12 18:40                 ` Axel Beckert
  2022-04-13  1:17                   ` [PATCH] " Jun T
  0 siblings, 1 reply; 18+ messages in thread
From: Axel Beckert @ 2022-04-12 18:40 UTC (permalink / raw)
  To: zsh-workers

Hi again,

On Tue, Apr 12, 2022 at 07:55:14PM +0200, Axel Beckert wrote:
> 875 run-testsuite-static FAIL stderr: (eval):7: permission denied: /file
[…]
> The question is how to fix this properly. My suggestion would be to
> move "tst_dir=V14.tmp" as well as "mkdir -p -- $tst_dir" out of the
> conditional, something along these lines:
> 
> diff --git a/Test/V14system.ztst b/Test/V14system.ztst
> index 100daab08..1e683b5b9 100644
> --- a/Test/V14system.ztst
> +++ b/Test/V14system.ztst
> @@ -2,10 +2,9 @@
>  
>  %prep
>  
> -  if zmodload -s zsh/system && zmodload -s zsh/zselect; then
> -    tst_dir=V14.tmp
> -    mkdir -p -- $tst_dir
> -  else
> +  tst_dir=V14.tmp
> +  mkdir -p -- $tst_dir
> +  if ! ( zmodload -s zsh/system && zmodload -s zsh/zselect ); then
>      ZTST_unimplemented='the zsh/system and zsh/zselect modules are not available'
>    fi
>    : > $tst_dir/file # File on which to acquire flock.

Nope, that test now fails with with non-static zsh as follows:

../../Test/V14system.ztst: starting.
Running test: zsystem flock valid time arguments
Test ../../Test/V14system.ztst failed: bad status 127, expected 0 from:
  (
    zsystem flock -t 0 -i 0.000001 $tst_dir/file    &&
    zsystem flock -t 0.1 -i 0.000001 $tst_dir/file  &&
    zsystem flock -t 0.1 -i 0.0000001 $tst_dir/file &&
    zsystem flock -t 1 -i 0.000001 $tst_dir/file
  )
Error output:
(eval):2: command not found: zsystem
Was testing: zsystem flock valid time arguments
../../Test/V14system.ztst: test failed.

So I currently have no idea how to fix that properly.

		Kind regards, Axel
-- 
PGP: 2FF9CD59612616B5      /~\  Plain Text Ribbon Campaign, http://arc.pasp.de/
Mail: abe@deuxchevaux.org  \ /  Say No to HTML in E-Mail and Usenet
Mail+Jabber: abe@noone.org  X
https://axel.beckert.ch/   / \  I love long mails: https://email.is-not-s.ms/


^ permalink raw reply	[flat|nested] 18+ messages in thread

* Re: [PATCH] Test release: 5.8.1.2-test
  2022-04-12 18:40                 ` Axel Beckert
@ 2022-04-13  1:17                   ` Jun T
  2022-04-13  9:34                     ` Axel Beckert
  0 siblings, 1 reply; 18+ messages in thread
From: Jun T @ 2022-04-13  1:17 UTC (permalink / raw)
  To: zsh-workers


> 2022/04/13 3:40, Axel Beckert <abe@deuxchevaux.org> wrote:

>> +  if ! ( zmodload -s zsh/system && zmodload -s zsh/zselect ); then

Calling zmodload in a subshell (...) does not load the module to the main zsh.


diff --git a/Test/V14system.ztst b/Test/V14system.ztst
index 100daab08..ffdb730a4 100644
--- a/Test/V14system.ztst
+++ b/Test/V14system.ztst
@@ -5,10 +5,10 @@
   if zmodload -s zsh/system && zmodload -s zsh/zselect; then
     tst_dir=V14.tmp
     mkdir -p -- $tst_dir
+    : > $tst_dir/file # File on which to acquire flock.
   else
     ZTST_unimplemented='the zsh/system and zsh/zselect modules are not available'
   fi
-  : > $tst_dir/file # File on which to acquire flock.
 
 %test
 





^ permalink raw reply	[flat|nested] 18+ messages in thread

* Re: [PATCH] Test release: 5.8.1.2-test
  2022-04-13  1:17                   ` [PATCH] " Jun T
@ 2022-04-13  9:34                     ` Axel Beckert
  0 siblings, 0 replies; 18+ messages in thread
From: Axel Beckert @ 2022-04-13  9:34 UTC (permalink / raw)
  To: zsh-workers

Hi,

On Wed, Apr 13, 2022 at 10:17:38AM +0900, Jun T wrote:
> > 2022/04/13 3:40, Axel Beckert <abe@deuxchevaux.org> wrote:
> 
> >> +  if ! ( zmodload -s zsh/system && zmodload -s zsh/zselect ); then
> 
> Calling zmodload in a subshell (...) does not load the module to the main zsh.

Argh, thanks! I missed the forest for the trees. I was just thinking
in boolean logic there, not in subshells.

> diff --git a/Test/V14system.ztst b/Test/V14system.ztst
> index 100daab08..ffdb730a4 100644
> --- a/Test/V14system.ztst
> +++ b/Test/V14system.ztst
> @@ -5,10 +5,10 @@
>    if zmodload -s zsh/system && zmodload -s zsh/zselect; then
>      tst_dir=V14.tmp
>      mkdir -p -- $tst_dir
> +    : > $tst_dir/file # File on which to acquire flock.
>    else
>      ZTST_unimplemented='the zsh/system and zsh/zselect modules are not available'
>    fi
> -  : > $tst_dir/file # File on which to acquire flock.
>  
>  %test

Helped, thanks!

		Kind regards, Axel
-- 
PGP: 2FF9CD59612616B5      /~\  Plain Text Ribbon Campaign, http://arc.pasp.de/
Mail: abe@deuxchevaux.org  \ /  Say No to HTML in E-Mail and Usenet
Mail+Jabber: abe@noone.org  X
https://axel.beckert.ch/   / \  I love long mails: https://email.is-not-s.ms/


^ permalink raw reply	[flat|nested] 18+ messages in thread

end of thread, other threads:[~2022-04-13 10:12 UTC | newest]

Thread overview: 18+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2022-04-09 20:07 Test release: 5.8.1.2-test dana
2022-04-09 20:18 ` Bart Schaefer
2022-04-10 20:07 ` Peter Stephenson
2022-04-12  9:16   ` Peter Stephenson
2022-04-10 20:21 ` Luna Jernberg
2022-04-11  0:32 ` Axel Beckert
2022-04-11  8:30   ` Axel Beckert
2022-04-11 19:54     ` Bart Schaefer
2022-04-12  2:39       ` Jun T
2022-04-12  3:58         ` Bart Schaefer
2022-04-12  6:30           ` Jun T
2022-04-12  7:17             ` Jun T
2022-04-12  9:05             ` Axel Beckert
2022-04-12 17:55               ` [PATCH] " Axel Beckert
2022-04-12 18:40                 ` Axel Beckert
2022-04-13  1:17                   ` [PATCH] " Jun T
2022-04-13  9:34                     ` Axel Beckert
2022-04-11 10:44 ` Kamil Dudka

Code repositories for project(s) associated with this inbox:

	https://git.vuxu.org/mirror/zsh/

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).