Message ID | 1337170229-29684-1-git-send-email-spdawson@gmail.com |
---|---|
State | Superseded |
Headers | show |
>>>>> "spdawson" == spdawson <spdawson@gmail.com> writes:
spdawson> From: Simon Dawson <spdawson@gmail.com>
spdawson> Signed-off-by: Simon Dawson <spdawson@gmail.com>
spdawson> ---
spdawson> package/vala/vala-fix-dirty-version-suffix.patch | 15 +++++++++++++++
spdawson> 1 file changed, 15 insertions(+)
spdawson> create mode 100644 package/vala/vala-fix-dirty-version-suffix.patch
spdawson> diff --git a/package/vala/vala-fix-dirty-version-suffix.patch b/package/vala/vala-fix-dirty-version-suffix.patch
spdawson> new file mode 100644
spdawson> index 0000000..4b0044d
spdawson> --- /dev/null
spdawson> +++ b/package/vala/vala-fix-dirty-version-suffix.patch
spdawson> @@ -0,0 +1,15 @@
spdawson> +Do not append a -dirty suffix to the valac version number. The presence of
spdawson> +the suffix can caused problems for packages (e.g. midori) which check the
spdawson> +valac version number during configuration.
Why is this needed? We don't apply any patches to vala, and I don't see
any "-dirty" suffix in valac --version:
./output/host/usr/bin/valac --version
Vala 0.15.2
> Why is this needed? We don't apply any patches to vala, and I don't see > any "-dirty" suffix in valac --version: > > ./output/host/usr/bin/valac --version > Vala 0.15.2 It may be because the Buildroot checkout in which you are testing has no uncommitted changes. The valac build process calls $ git diff-index --name-only HEAD 2>/dev/null and appends the -dirty version suffix if anything is returned. In my case --- with uncommitted changes to my Buildroot Git checkout --- the -dirty suffix does get applied. Consequently, for me: $ ./output/host/usr/bin/valac --version Vala 0.15.2-dirty In essence, I think the patch is necessary in order to prevent the valac build process from falling foul of uncommitted changes in the local Buildroot checkout. Of course, if you've got uncommitted changes in the repository in which you're testing, then my theory goes out of the window... Simon.
>>>>> "Simon" == Simon Dawson <spdawson@gmail.com> writes:
Hi,
Simon> In essence, I think the patch is necessary in order to prevent the
Simon> valac build process from falling foul of uncommitted changes in the
Simon> local Buildroot checkout. Of course, if you've got uncommitted changes
Simon> in the repository in which you're testing, then my theory goes out of
Simon> the window...
Ahh, ok. That was not clear to me from your commit message. We have
something similar for autoconf
(autoconf-2.65-dont-add-dirty-to-version.patch). Care to adjust patch
description and commit message and resend?
diff --git a/package/vala/vala-fix-dirty-version-suffix.patch b/package/vala/vala-fix-dirty-version-suffix.patch new file mode 100644 index 0000000..4b0044d --- /dev/null +++ b/package/vala/vala-fix-dirty-version-suffix.patch @@ -0,0 +1,15 @@ +Do not append a -dirty suffix to the valac version number. The presence of +the suffix can caused problems for packages (e.g. midori) which check the +valac version number during configuration. + +diff -Nur a/build-aux/git-version-gen b/build-aux/git-version-gen +--- a/build-aux/git-version-gen 2010-08-15 12:49:03.000000000 +0100 ++++ b/build-aux/git-version-gen 2012-05-14 10:17:19.977204570 +0100 +@@ -135,7 +135,6 @@ + *) # Append the suffix only if there isn't one already. + case $v in + *-dirty) ;; +- *) v="$v-dirty" ;; + esac ;; + esac +