GNU bug report logs - #10441
The testsuite assumes that ln -s really creates a symlink

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 10441 in the body.
You can then email your comments to 10441 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: Peter Rosin <peda <at> lysator.liu.se>
To: bug-automake <at> gnu.org
Subject: The testsuite assumes that ln -s really creates a symlink
Date: Thu, 05 Jan 2012 23:13:22 +0100

Hi!

"ln -s" on MSYS does not create a symlink, instead it makes a
regular copy.  Lots of test cases (on master) FAIL due to this.

"test -h foo" will simply not return success on MSYS, ever.

Affected tests (that I have noticed):
add-missing.tap (lots of failing test cases)
copy.test

Cheers,
Peter

Message #10 received at 10441 <at> debbugs.gnu.org (full text, mbox):

From: Stefano Lattarini <stefano.lattarini <at> gmail.com>
To: Peter Rosin <peda <at> lysator.liu.se>
Cc: 10441 <at> debbugs.gnu.org
Subject: Re: bug#10441: The testsuite assumes that ln -s really creates a
	symlink
Date: Sun, 08 Jan 2012 13:56:36 +0100

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

On 01/05/2012 11:13 PM, Peter Rosin wrote:
> Hi!
>
Hi Peter, thanks for the report.

> "ln -s" on MSYS does not create a symlink, instead it makes a
> regular copy.  Lots of test cases (on master) FAIL due to this.
> 
> "test -h foo" will simply not return success on MSYS, ever.
> 
> Affected tests (that I have noticed):
> add-missing.tap (lots of failing test cases)
> copy.test
> 
The attached patch should take care of the problem.  Tested using
this script in PATH as the `ln' program:

  #!/bin/bash
  declare -a args=()
  for i in "$@"; do
    test x"$i" = x"-s" && continue
    args=("${args[@]}" "$i")
  done
  exec /bin/ln "${args[@]}"

I will push after your ACK or in 48 hours, whichever comes first.

Regards,
  Stefano

[0001-tests-avoid-many-spurious-failures-on-MSYS-due-to-la.patch (text/x-diff, attachment)]

Message #13 received at 10441 <at> debbugs.gnu.org (full text, mbox):

From: Peter Rosin <peda <at> lysator.liu.se>
To: Stefano Lattarini <stefano.lattarini <at> gmail.com>
Cc: 10441 <at> debbugs.gnu.org
Subject: Re: bug#10441: The testsuite assumes that ln -s really creates a
	symlink
Date: Sun, 08 Jan 2012 16:49:13 +0100

Stefano Lattarini skrev 2012-01-08 13:56:
> On 01/05/2012 11:13 PM, Peter Rosin wrote:
>> Hi!
>>
> Hi Peter, thanks for the report.
> 
>> "ln -s" on MSYS does not create a symlink, instead it makes a
>> regular copy.  Lots of test cases (on master) FAIL due to this.
>>
>> "test -h foo" will simply not return success on MSYS, ever.
>>
>> Affected tests (that I have noticed):
>> add-missing.tap (lots of failing test cases)
>> copy.test
>>
> The attached patch should take care of the problem.  Tested using
> this script in PATH as the `ln' program:
> 
>   #!/bin/bash
>   declare -a args=()
>   for i in "$@"; do
>     test x"$i" = x"-s" && continue
>     args=("${args[@]}" "$i")
>   done
>   exec /bin/ln "${args[@]}"
> 
> I will push after your ACK or in 48 hours, whichever comes first.

add-missing.tap: 351 PASS, 50 skip
copy.test: PASS

Thanks!

Cheers,
Peter

Message #16 received at submit <at> debbugs.gnu.org (full text, mbox):

From: Peter Johansson <trojkan <at> gmail.com>
To: bug-automake <at> gnu.org
Subject: Re: bug#10441: The testsuite assumes that ln -s really creates a
	symlink
Date: Sun, 08 Jan 2012 11:06:42 -0500

On 1/8/12 7:56 AM, Stefano Lattarini wrote:
>
>
> The attached patch should take care of the problem.  Tested using
> this script in PATH as the `ln' program:

Hi Stefano,

I'm just curious if there is a reason not to use AC_PROG_LN_S as 
provided by Autoconf. I thought one could call that macro in Autoconf 
and place a line

LN_S=@LN_S@

in 'defs.in' and then replace all 'ln -s' with '$LN_S' in tests. In that 
way you avoid having another script to maintain. Am I missing something?

Cheers,
Peter

Message #19 received at 10441 <at> debbugs.gnu.org (full text, mbox):

From: Stefano Lattarini <stefano.lattarini <at> gmail.com>
To: Peter Johansson <trojkan <at> gmail.com>
Cc: 10441 <at> debbugs.gnu.org
Subject: Re: bug#10441: The testsuite assumes that ln -s really creates a
	symlink
Date: Sun, 08 Jan 2012 17:57:58 +0100

On 01/08/2012 05:06 PM, Peter Johansson wrote:
> On 1/8/12 7:56 AM, Stefano Lattarini wrote:
>>
>>
>> The attached patch should take care of the problem.  Tested using
>> this script in PATH as the `ln' program:
> 
> Hi Stefano,
> 
Hi Peter.

> I'm just curious if there is a reason not to use AC_PROG_LN_S as provided
> by Autoconf.
>
> I thought one could call that macro in Autoconf and place a line
> 
> LN_S=@LN_S@
> 
> in 'defs.in' and then replace all 'ln -s' with '$LN_S' in tests.
>
Ah, but the "ln -s" in the testsuite are used only in places where we want
*real* symlinks (and want to skip some tests if they are not available), so
using $LN_S would quite defy the purpose.

> In that way you avoid having another script to maintain.
>
Oh, don't worry: that script was only a temporary hack to allow me testing
the patch on GNU/Linux, since I don't have easy access to a MinGW system.
It has already been deleted and forgotten.  Sorry if I haven't been clear.

> Am I missing something?
> 
I hope the above clarifies your doubts.

Thanks,
  Stefano

Message #22 received at 10441 <at> debbugs.gnu.org (full text, mbox):

From: Stefano Lattarini <stefano.lattarini <at> gmail.com>
To: Peter Rosin <peda <at> lysator.liu.se>
Cc: 10441 <at> debbugs.gnu.org
Subject: Re: bug#10441: The testsuite assumes that ln -s really creates a
	symlink
Date: Sun, 08 Jan 2012 18:05:59 +0100

tags 10441 patch
close 10441
thanks

On 01/08/2012 04:49 PM, Peter Rosin wrote:
> 
> add-missing.tap: 351 PASS, 50 skip
> copy.test: PASS
>
>  Thanks!
>
Good!  Patch pushed, and bug report closed.

Thanks again for all the testing,
  Stefano

Previous Next

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