[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: 2.19.22 not building
From: |
Thomas Spuhler |
Subject: |
Re: 2.19.22 not building |
Date: |
Thu, 06 Aug 2015 13:59:49 -0700 |
User-agent: |
KMail/4.14.5 (Linux/3.19.8-desktop-3.mga5; KDE/4.14.5; x86_64; ; ) |
On Thursday, August 06, 2015 06:27:25 PM you wrote:
> Thomas Spuhler <address@hidden> writes:
> > On Monday, July 13, 2015 10:31:18 AM David Kastrup wrote:
> >> Thomas Spuhler <address@hidden> writes:
> >> > On Friday, July 10, 2015 02:38:14 AM David Kastrup wrote:
> >> >> Thomas Spuhler <address@hidden> writes:
> >> >> > On Friday, July 10, 2015 01:43:22 AM David Kastrup wrote:
> >> >> But for packaging, GUILE 2.0 is definitely the wrong choice at the
> >> >> current point of time.
> >> >
> >> > I am coming back to Guile-2. I upgraded to vers 2.0.11 and rebuild
> >> > lily 2.19.21. I still cannot
> >> > build 2.19.22. It chokes at the documentation.
> >>
> >> That's more or less chance. Neither version will pass the regtests.
> >>
> >> > But lilypond 2.19.21 (built with guile-2.0.11) compiles the testfile
> >> > lily-0e752a19.ly
> >> > looks good, no crash, nothing special
> >>
> >> Oh good grief. Distributing some build of software on the rationale
> >> that it happens to compile some small file without crashing is not
> >> solid engineering.
> >
> > Good news, version 2.19.24 is the first one that built since
> > vers. 2.19.21 Thanks for fixing it.
>
> The last seriously GUILE-related change has been made in version 2.19.23
> already (issue 4462). However, no GUILE 2 related changes have been
> made since considerably before 2.19.20.
>
> So any changes you may have witnessed are not the outcome of "fixes" on
> the LilyPond side and are more likely than not of temporary duration.
> You are still faring better here than the LilyPond developers
> themselves.
Well, all I did was downloading the next release and trying to build it,
locally and on the Mageia build server. 2.19.22 and 2.19.23 didn't build. I am
already limit the build to one cpu core as a parallel build have been failing
for years. So it shouldn't be random.
--
Best regards
Thomas Spuhler
All of my e-mails have a valid digital signature
ID 60114E63
signature.asc
Description: This is a digitally signed message part.