Eric Blake <[EMAIL PROTECTED]> writes:
> OK to apply? Also, should this be modified to print an INFO message warning
> the user if git-version-gen returned UNKNOWN (normally an important warning,
> but one which can be safely ignored in a gnulib-tool testdir)?
I noticed you had pushed this, an
Eric Blake <[EMAIL PROTECTED]> wrote:
...
> Actually, that isn't right either. Rather, git-version-gen has the
> requirement
> that _IF_ the first argument does not exist as a file, or exists but does not
> have a valid version string, _THEN_ it must be invoked from within a VCS
> checkout in ord
Bruno Haible clisp.org> writes:
>
> Eric Blake wrote:
> > it is not GNUmakefile that is wrong, it is the fact that you are
> > including git-version-gen but violating the requirements of git-version-gen.
>
> I disagree. git-version-gen does not have the requirement that a file
> .tarball-versi
Eric Blake wrote:
> it is not GNUmakefile that is wrong, it is the fact that you are
> including git-version-gen but violating the requirements of git-version-gen.
I disagree. git-version-gen does not have the requirement that a file
.tarball-version does exist. git-version-gen has the requiremen
Simon Josefsson wrote:
> It seems the patch isn't enough, the infloop still happens. The reason
> seems to be that maintainer-makefile is not filtered out, and it depends
> on gnumakefile.
Ouch. So I backed out my gnulib-tool patch.
> I'm not sure these gnulib-tool hacks are the right solution t
Eric Blake wrote:
> But it is NOT gnumakefile in isolation that breaks the build. It is the
> combination of gnumakefile AND git-version-gen.
OK. But while I can imagine that 'git-version-gen' gets a unit test, I cannot
see how a 'gnumakefile' unit test could look like. Therefore 'gnumakefile' is
Eric Blake <[EMAIL PROTECTED]> writes:
> Simon Josefsson josefsson.org> writes:
>
>> > The 'gnumakefile' in particular does not provide functionality to the user
>> > of the package, only to the maintainer. It does not change the way "make"
> and
>> > "make check" work. It has no test suite. The
Simon Josefsson josefsson.org> writes:
> > The 'gnumakefile' in particular does not provide functionality to the user
> > of the package, only to the maintainer. It does not change the way "make"
and
> > "make check" work. It has no test suite. Therefore it doesn't make much
sense
> > to includ
Bruno Haible <[EMAIL PROTECTED]> writes:
> Eric Blake wrote:
>> In fact, it might be worth a patch to gnulib-tool, rather than to
>> GNUmakefile, to guarantee that --create-testdir creates .tarball-version
>> to simulate the 'make dist' process of any package which uses the
>> GNUmakefile module
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
According to Bruno Haible on 6/4/2008 4:41 AM:
| @@ -3418,8 +3418,9 @@
| # Except config-h, which breaks all modules which use HAVE_CONFIG_H.
| # Except fnmatch-posix, which conflicts with fnmatch-gnu. FIXME.
| # Except ftruncate, mount
Eric Blake wrote:
> In fact, it might be worth a patch to gnulib-tool, rather than to
> GNUmakefile, to guarantee that --create-testdir creates .tarball-version
> to simulate the 'make dist' process of any package which uses the
> GNUmakefile module.
I disagree: gnulib-tool should not do special
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
According to Simon Josefsson on 6/3/2008 9:37 PM:
|>
|> That's the correct workaround - you are indeed creating the equivalent of
|> a tarball outside the VCS, so you should have a .tarball-version present.
|
| If that is the case, the gnumakefile mod
Eric Blake <[EMAIL PROTECTED]> writes:
> According to Simon Josefsson on 6/3/2008 10:29 AM:
> |
> | The failure can be reproduced with:
> |
> | $ gnulib-tool --create-testdir --with-tests --dir=foo
>
> |
> | As a work around, my daily builder just does 'echo 0 > .tarball-version'
> | and that make
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
According to Simon Josefsson on 6/3/2008 10:29 AM:
|
| The failure can be reproduced with:
|
| $ gnulib-tool --create-testdir --with-tests --dir=foo
|
| As a work around, my daily builder just does 'echo 0 > .tarball-version'
| and that makes everyth
Jim Meyering <[EMAIL PROTECTED]> writes:
> Simon Josefsson <[EMAIL PROTECTED]> wrote:
>> 'make dist' gets into a infloop in gnulib daily builds for me since
>> there is no .tarball-version file. This should fix it. Ok to push?
>
> Hi Simon,
>
> Thanks for the report.
> In daily builds of gnulib
Simon Josefsson <[EMAIL PROTECTED]> wrote:
> 'make dist' gets into a infloop in gnulib daily builds for me since
> there is no .tarball-version file. This should fix it. Ok to push?
Hi Simon,
Thanks for the report.
In daily builds of gnulib itself? built how?
In normal practice, the .tarball-
'make dist' gets into a infloop in gnulib daily builds for me since
there is no .tarball-version file. This should fix it. Ok to push?
/Simon
2008-06-02 Simon Josefsson <[EMAIL PROTECTED]>
* top/GNUmakefile: Don't infloop if .tarball-version doesn't exist.
diff --git a/top/GNUmakefi
17 matches
Mail list logo
