[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: bison-3.0 fix lilypond
From: |
David Kastrup |
Subject: |
Re: bison-3.0 fix lilypond |
Date: |
Sun, 01 Sep 2013 14:28:40 +0200 |
User-agent: |
Gnus/5.13 (Gnus v5.13) Emacs/24.3.50 (gnu/linux) |
address@hidden writes:
[...]
>> Can you check which (if any) of parser.cc and parser.hh get rebuilt for
>> you? It's likely that one of the two is kept in the version from the
>> tarball while the other is recreated with a different Bison, and this
>> mismatch is what is causing the problems.
>
> Looks like you got it: parser.cc is regenerated (and, of course,
> different from the original copy), while parser.hh is not:
>
> Sep 1 12:16 ./lilypond-2.17.25/lily/out/parser.cc
> Aug 25 16:42 ./lilypond-2.17.25/lily/out/parser.cc.ORIG
> Aug 25 16:42 ./lilypond-2.17.25/lily/out/parser.hh
> Aug 25 16:42 ./lilypond-2.17.25/lily/out/parser.hh.ORIG
Well, I don't think "I got it": the real question is _what_ causes this
discrepancy. We have a clear _bug_ here, and a design question.
The _bug_ is that the dependencies for both of those files seem to be
met under different conditions.
The design question is whether we should be distributing any
bison-generated file in the first place. I think that the best answer
to the design question is likely "no". But if we have broken
dependencies, we should fix them first anyway as they are still going to
cause problems for people _rebuilding_ LilyPond.
--
David Kastrup