I done s/spacewalk.releng./tito./ in spacewalk.git, but it may reside in some (private) branches.
So let keep it for little bit more releases, but print warnings.
because output of:
rpm -q --qf '%{name}' --eval '%undefine scl' --specfile rubygem-robotex.spec
is:
rubygem-robotex
(i.e. two empty lines, then the output)
* Store the --test flag on the releaser and pass it to the builder
* With --test in effect, have the builder update the spec file
* When the builder does so it also updates the build_version to include git hash
Uploading new sources with fedpkg seems to make files read only after
uploaded in an attempt to maintain their checksum most likely.
tito already generates files with consistent checksums, and this is
breaking if you release to Fedora/RHEL and then try to release anywhere
else.
"git diff" can accept --binary, which will produce diff with binary content. But patch does not understood
such diff, so we are doomed. Diff with binary file will produce
Binary files /dev/null and b/app/assets/images/false.png differ
diff --git a/app/assets/images/foreman.png b/app/assets/images/foreman.png
new file mode 100644
index 0000000..8815824
And patch will accept such syntax but will not create the file. Not even empty one.
So this result in silently throwing away some changes.
The VersionTagger fails in the spec file contains a number followed
by and underscore. This patch fixes that issue.
increase_version also returned an emtpy vlaue when nothing matched,
which led to a broken spec file. It now returns the original,
unmodified value instead.
This is for supporting tagging software collections (SC), if you have
installed meta package of SC then then name of package is different.
I.e. if you are tagging rubygem-foo package and you have ruby193-build
installed then the name of package is ruby193-rubygem-foo.
We use suffix in relase number to indicate build in a branch. Builds from master have no suffix.
When we had package-1.2.3-4h built in branch and was trying to tag package-1.2.3-4 in master
the check failed as the new tag was substring of the other tag and grep matches that.
