lilypond-devel
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: GUB dblatex breaks compile


From: Julien Rioux
Subject: Re: GUB dblatex breaks compile
Date: Thu, 16 Feb 2012 15:07:53 +0100

On Thu, Feb 16, 2012 at 2:54 PM, Graham Percival
<address@hidden> wrote:
> On Thu, Feb 16, 2012 at 02:39:58PM +0100, Julien Rioux wrote:
>> On Thu, Feb 16, 2012 at 12:14 PM, Graham Percival
>> <address@hidden> wrote:
>> > The correct bugfix may be simply to update the version of dblatex
>> > in GUB, but that is rarely as simple as it sounds.
>>
>> Either fix your dblatex installation (it should depend on the package
>> python-apt, but otherwise you can install python-apt manually)
>
> GUB is an (almost)-complete separate beast.  It uses the base OS
> to build gcc/make/etc, but thereafter it compiles (almost)
> everything using only self-compiled programs.  The intent is that
> it should be more reliable; as long as GUB's "bootstrap" (i.e. the
> toolchain) compiles, it shouldn't matter what distro you're using.
> It certainly doesn't use apt.
>

But it uses a dblatex that looks for a python apt module?

>> or
>> remove dblatex completely (in this case there was indeed a bug in
>> lilypond master, but I pushed a fix at
>> e8bb4dcdf244ccda234b63df0a78635cf066be4c and it was tested and
>> verified).
>
> The bottom line is that GUB could build release-2.15.29, but it
> cannot build the current release/unstable.  This means that we
> cannot produce any binary releases at the moment.
>
> Maybe GUB needs to be tweaked so that it never looks for programs
> on the host OS.  Maybe dblatex needs to be added to GUB (it's not
> there right now).  Either of those options will take 1-10 hours.
>

This should definitely be looked into.

> This is unfortunate.
>
> - Graham

I can think of two really quick options: either remove the
suffix-lyxml.lyxml regtest from the source tree where you are running
this from, or remove dblatex from the host system running gub.

Cheers,
Julien



reply via email to

[Prev in Thread] Current Thread [Next in Thread]