GNU bug report logs - #49983
grep-3.6.27-20b4 tests fail on Cygwin

Previous Next

To add a comment to this bug, you must first unarchive it, by sending
a message to control AT debbugs.gnu.org, with unarchive 49983 in the body.
You can then email your comments to 49983 AT debbugs.gnu.org in the normal way.

Toggle the display of automated, internal messages from the tracker.

Message #5 received at submit <at> debbugs.gnu.org (full text, mbox):

From: Gary Johnson <garyjohn <at> spocom.com>
To: bug-grep <at> gnu.org
Subject: grep-3.6.27-20b4 tests fail on Cygwin
Date: Tue, 10 Aug 2021 08:54:26 -0700

I built grep-3.6.27-20b4 on a machine running up-to-date Cygwin
3.2.0 and Windows 10 Enterprise Version 10.0.18363 Build 18363.
Running "make check" resulted in the following summary.

    ============================================================================
    Testsuite summary for GNU grep 3.6.27-20b4
    ============================================================================
    # TOTAL: 117
    # PASS:  93
    # SKIP:  21
    # XFAIL: 2
    # FAIL:  1
    # XPASS: 0
    # ERROR: 0
    ============================================================================
    See tests/test-suite.log
    Please report to bug-grep <at> gnu.org
    ============================================================================

Here is a summary of the log.

    $ grep FAIL tests/test-suite.log
    # XFAIL: 2
    # FAIL:  1
    XFAIL: equiv-classes
    XFAIL equiv-classes (exit status: 1)
    FAIL: surrogate-pair
    FAIL surrogate-pair (exit status: 1)
    XFAIL: triple-backref
    XFAIL triple-backref (exit status: 1)

    $ wc tests/test-suite.log
      3000  11841 128417 tests/test-suite.log

The log is pretty big.  Would you like it posted here?

Regards,
Gary

Message #8 received at 49983 <at> debbugs.gnu.org (full text, mbox):

From: Paul Eggert <eggert <at> cs.ucla.edu>
To: Gary Johnson <garyjohn <at> spocom.com>
Cc: 49983 <at> debbugs.gnu.org
Subject: Re: bug#49983: grep-3.6.27-20b4 tests fail on Cygwin
Date: Tue, 10 Aug 2021 14:25:46 -0700

On 8/10/21 8:54 AM, Gary Johnson wrote:
> The log is pretty big.  Would you like it posted here?

It'd be helpful to see tests/surrogate-pair.log. If it's long perhaps 
you could compress it and attach the compressed version to your email.

Message #11 received at 49983 <at> debbugs.gnu.org (full text, mbox):

From: Gary Johnson <garyjohn <at> spocom.com>
To: Paul Eggert <eggert <at> cs.ucla.edu>
Cc: 49983 <at> debbugs.gnu.org
Subject: Re: bug#49983: grep-3.6.27-20b4 tests fail on Cygwin
Date: Tue, 10 Aug 2021 15:09:20 -0700

[Message part 1 (text/plain, inline)]

On 2021-08-10, Paul Eggert wrote:
> On 8/10/21 8:54 AM, Gary Johnson wrote:
> >The log is pretty big.  Would you like it posted here?
> 
> It'd be helpful to see tests/surrogate-pair.log. If it's long perhaps
> you could compress it and attach the compressed version to your
> email.

It's not very long.  Here it is.

Regards,
Gary

[surrogate-pair.log (text/plain, attachment)]

Message #14 received at 49983 <at> debbugs.gnu.org (full text, mbox):

From: Paul Eggert <eggert <at> cs.ucla.edu>
To: Gary Johnson <garyjohn <at> spocom.com>
Cc: 49983 <at> debbugs.gnu.org
Subject: Re: bug#49983: grep-3.6.27-20b4 tests fail on Cygwin
Date: Tue, 10 Aug 2021 15:47:59 -0700

Here is a shell transcript on my platform. Can you compare it to what 
happens on your platform? I am guessing that the first 'echo $?' outputs 
something different but would like to know what it is. Plus any other 
differences.

$ export LC_ALL=en_US.UTF-8
$ s_pair=$(printf '\360\220\220\205')
$ printf '%s\n' "$s_pair" > in
$ src/grep --file=in in >out
$ echo $?
0
$ od -t o1 in
0000000 360 220 220 205 012
0000005
$ cmp in out
$ echo $?
0

Message #17 received at 49983 <at> debbugs.gnu.org (full text, mbox):

From: Gary Johnson <garyjohn <at> spocom.com>
To: Paul Eggert <eggert <at> cs.ucla.edu>
Cc: 49983 <at> debbugs.gnu.org
Subject: Re: bug#49983: grep-3.6.27-20b4 tests fail on Cygwin
Date: Tue, 10 Aug 2021 16:21:45 -0700

On 2021-08-10, Paul Eggert wrote:
> Here is a shell transcript on my platform. Can you compare it to what
> happens on your platform? I am guessing that the first 'echo $?'
> outputs something different but would like to know what it is. Plus
> any other differences.
> 
> $ export LC_ALL=en_US.UTF-8
> $ s_pair=$(printf '\360\220\220\205')
> $ printf '%s\n' "$s_pair" > in
> $ src/grep --file=in in >out
> $ echo $?
> 0
> $ od -t o1 in
> 0000000 360 220 220 205 012
> 0000005
> $ cmp in out
> $ echo $?
> 0

This is a pretty drastic difference.  I don't think I missed
anything, though.

$ export LC_ALL=en_US.UTF-8
$ s_pair=$(printf '\360\220\220\205')
$ printf '%s\n' "$s_pair" > in
$ src/grep --file=in in >out
$ echo $?
1
$ od -t o1 in
0000000 360 220 220 205 012
0000005
$ cmp in out
cmp: EOF on out which is empty
$ echo $?
1

Regards,
Gary

Message #20 received at 49983 <at> debbugs.gnu.org (full text, mbox):

From: Paul Eggert <eggert <at> cs.ucla.edu>
To: Gary Johnson <garyjohn <at> spocom.com>
Cc: 49983 <at> debbugs.gnu.org
Subject: Re: bug#49983: grep-3.6.27-20b4 tests fail on Cygwin
Date: Tue, 10 Aug 2021 23:19:33 -0700

On 8/10/21 4:21 PM, Gary Johnson wrote:
> $ src/grep --file=in in >out
> $ echo $?
> 1
> $ od -t o1 in
> 0000000 360 220 220 205 012
> 0000005
> $ cmp in out
> cmp: EOF on out which is empty

OK, so grep is simply not finding a match.

At this point it's probably time to fire up a debugger and see why 
grep's matchers are saying "no match". That would require some debugging 
expertise and time, though; do you have those available?

Message #23 received at 49983 <at> debbugs.gnu.org (full text, mbox):

From: Gary Johnson <garyjohn <at> spocom.com>
To: Paul Eggert <eggert <at> cs.ucla.edu>
Cc: 49983 <at> debbugs.gnu.org
Subject: Re: bug#49983: grep-3.6.27-20b4 tests fail on Cygwin
Date: Wed, 11 Aug 2021 00:20:32 -0700

On 2021-08-10, Paul Eggert wrote:
> On 8/10/21 4:21 PM, Gary Johnson wrote:
> >$ src/grep --file=in in >out
> >$ echo $?
> >1
> >$ od -t o1 in
> >0000000 360 220 220 205 012
> >0000005
> >$ cmp in out
> >cmp: EOF on out which is empty
> 
> OK, so grep is simply not finding a match.
> 
> At this point it's probably time to fire up a debugger and see why
> grep's matchers are saying "no match". That would require some
> debugging expertise and time, though; do you have those available?

I'm familiar with gdb, but not an expert.  I've used it mostly for
debugging Vim on Linux and for debugging an embedded, bare-metal
RISC-V project.  I hacked grep once to add an option to control
whether underscore was part of a word or not, but that was a long
time ago.  I don't remember anything about grep's internals.  The
main thing I'm lacking is time.  I'm married and employed full time,
though I've managed to squeeze in some time for Vim enhancements on
weekends when it's too hot and/or smokey to work outside.  I can
take a stab at it and let you know if it's taking too much time.

Regards,
Gary

Message #26 received at 49983 <at> debbugs.gnu.org (full text, mbox):

From: Paul Eggert <eggert <at> cs.ucla.edu>
To: Gary Johnson <garyjohn <at> spocom.com>
Cc: 49983 <at> debbugs.gnu.org
Subject: Re: bug#49983: grep-3.6.27-20b4 tests fail on Cygwin
Date: Wed, 11 Aug 2021 00:35:27 -0700

On 8/11/21 12:20 AM, Gary Johnson wrote:
> The
> main thing I'm lacking is time.

I *quite* understand.

Unfortunately I don't use Cygwin so can't help much on Cygwin-specific 
stuff.

Message #29 received at 49983 <at> debbugs.gnu.org (full text, mbox):

From: Jim Meyering <jim <at> meyering.net>
To: Paul Eggert <eggert <at> cs.ucla.edu>
Cc: Gary Johnson <garyjohn <at> spocom.com>, 49983 <at> debbugs.gnu.org
Subject: Re: bug#49983: grep-3.6.27-20b4 tests fail on Cygwin
Date: Sat, 14 Aug 2021 12:48:07 +0200

On Wed, Aug 11, 2021 at 9:36 AM Paul Eggert <eggert <at> cs.ucla.edu> wrote:
>
> On 8/11/21 12:20 AM, Gary Johnson wrote:
> > The
> > main thing I'm lacking is time.
>
> I *quite* understand.
>
> Unfortunately I don't use Cygwin so can't help much on Cygwin-specific
> stuff.

Thanks for reporting and investigating.

While understanding/fixing this surrogate-pair test failure on Cygwin
would be nice, it is niche enough that it need not hold up the
release.

Previous Next

GNU bug tracking system
Copyright (C) 1999 Darren O. Benham, 1997,2003 nCipher Corporation Ltd, 1994-97 Ian Jackson.