Re: regtests for previous stables failing

[Thomas Morley]

2017-10-27 23:13 GMT+02:00 David Kastrup <address@hidden>:
> Thomas Morley <address@hidden> writes:
>
>> But in the light of David's last mail I revived my old LilyDev3 which has:
>> $ make --version && gcc --version
>> GNU Make 3.81
>> Copyright (C) 2006  Free Software Foundation, Inc.
>> This is free software; see the source for copying conditions.
>> There is NO warranty; not even for MERCHANTABILITY or FITNESS FOR A
>> PARTICULAR PURPOSE.
>>
>> This program built for i486-pc-linux-gnu
>> gcc (Debian 4.7.2-5) 4.7.2
>> Copyright (C) 2012 Free Software Foundation, Inc.
>> This is free software; see the source for copying conditions.  There is NO
>> warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.
>>
>> All worked there.
>>
>> Obviously David was right about the make-version. Do we need to fix
>> something? At least with more current code all is working as well.
>>
>> The results of the exercise at
>> https://sourceforge.net/p/testlilyissues/issues/5217/#ae0c/7c60/8bbe
>> ff
>
> It's a bit tricky.  Maybe we should cherry-pick the necessary
> compatibility patches to the stable branches' tips?  If you want to
> bisect, you'd need to skim them in as well, I guess.
>
> --
> David Kastrup



The compability patch is likely:

commit [r2545fab01a601743bb3ecc18942cf14263ccfc6c]
Author: David Kastrup <address@hidden>
Date:   Sat May 2 00:35:35 2015 +0200

    Issue 4364: Allow ImageMagick's compare to exit with status 1

    Apparently Ubuntu 15.04 has a version of "compare" that cannot be easily
    persuaded to return anything but exit status 1 (which indicates
    dissimilar images but no actual error condition) so we allow this in
    script/build/output-distance.py in order to keep "make check" from
    failing.

    This patch is somewhat artless but does the trick.

Putting this one on top of my local copy of remotes/origin/stable/2.18
makes it work with GNU Make 4.1

Though, not for my 2.16.-branch.
I get:
~/lilypond-git/build (dev/2-16-test)$ make -j5
/home/hermann/lilypond-git/stepmake/stepmake/po-targets.make:41: ***
recipe commences before first target.  Stop.
Not sure what it makes it fail here but succeeds for 2.18.


Cheers,
  Harm