| Commit message (Collapse) | Author | Age |
| |
|
|
|
|
|
|
|
| |
With aa8548d5 the newly introduced gbp-dch option --upstream-branch
got introduced.
Closes: #gh10
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
In some unfortunate cases you might have something like this:
$ git tag
v1.0
vyatta/1.0-0something1
... where later tag is closer to describe current HEAD.
With upstream-tag set to v%(version)s find_tag() is going to propose
something based on vyatta/1.0-0something1 which is not expected.
The problem is that find_tag() returns tags, which match the
upstream-tag format, but which are not on the upstream-branch at all.
To fix this find_tag()/git-describe should not be used to determine
the closest upstream-tag.
Closes: gh#7
|
|
|
|
|
|
|
| |
Signed-off-by: Martin Erik Werner <martinerikwerner@gmail.com>
Signed-off-by: Guido Günther <agx@sigxcpu.org>
Closes: #783013
|
|
|
|
| |
Signed-off-by: Markus Lehtonen <markus.lehtonen@linux.intel.com>
|
|
|
|
|
|
|
|
| |
Strip extra linefeeds from stdout and stderr output (if captured) when
generating the error message, so that the error message itself would not
have any unwanted linefeeds.
Signed-off-by: Markus Lehtonen <markus.lehtonen@linux.intel.com>
|
|
|
|
|
|
|
| |
The new name more precisely describes it as it can be used to format any
strings (e.g. tag names) and not only messages.
Signed-off-by: Markus Lehtonen <markus.lehtonen@linux.intel.com>
|
|
|
|
|
|
| |
In order to be able to re-use it in gbp-buildpackage-rpm.
Signed-off-by: Markus Lehtonen <markus.lehtonen@linux.intel.com>
|
|
|
|
| |
needed by gbp import-srpm
|
|
|
|
| |
since the SrcRpmFile tests require it
|
|
|
|
|
|
| |
so we can better debug failures like
http://honk.sigxcpu.org:8001/job/git-buildpackage/230/changes
|
|
|
|
|
|
| |
Some SUSE-specific RPM tags are not supported anymore.
Signed-off-by: Markus Lehtonen <markus.lehtonen@linux.intel.com>
|
|
|
|
| |
Signed-off-by: Markus Lehtonen <markus.lehtonen@linux.intel.com>
|
|
|
|
| |
Signed-off-by: Markus Lehtonen <markus.lehtonen@linux.intel.com>
|
|
|
|
| |
for Python3 compatibility
|
|
|
|
|
| |
We want to get there eventually but having this on now is
just too much noise.
|
|
|
|
| |
for Python3 compatibility
|
|
|
|
|
|
| |
to work towards Python3 support
Gbp-Dch: Ignore
|
|
|
|
| |
and use it to test the help output
|
|
|
|
| |
The values weren't correctly seperated
|
|
|
|
|
| |
Capture its output and only dump it on failure. This makes "gbp
import-orig" and test runs less verbose.
|
|
|
|
|
|
| |
We allow to substitute stderr, stdout and error_reason in run_error now.
These changes the API for derived classses slightly so fix them up as
well.
|
|
|
|
| |
We want to test no prefix, 'gbp-' and 'git-'.
|
|
|
|
|
| |
So far we were modifying the list of handlers in place which resulted in
all handlers being removed at the second call to _capture_log only.
|
| |
|
| |
|
|
|
|
| |
Remove unused imports, remove unused variables or actually use them
|
|
|
|
|
|
|
| |
Get rid of __run and use the same codepaths for both. This also makes
the same instance variables available after both calls.
The funtions still differ in their default logging behaviour though.
|
|
|
|
| |
For symmetry with capture_stderr
|
| |
|
| |
|
| |
|
| |
|
|
|
|
| |
It's much simpler to spot the cause of test failures this way.
|
|
|
|
| |
Git-Dch: Ignore
|
| |
|
|
|
|
|
|
|
|
|
|
| |
enigmail upstream uses tags named enigmail-1-8 for 1.8. Other
upstreams have used similar conventions, likely as holdovers from CVS
(e.g. gnupg 1.4.2 was tagged with V1-4-2). This patch helps packagers
work with these upstreams.
Closes: #780679
Signed-off-by: Guido Günther <agx@sigxcpu.org>
|
|
|
|
|
|
|
|
|
|
| |
Also pass it through version_to_tag so
upstream-vcs-tag = libgpg-error-%(version)s
properly expands the version replacement.
Closes: #780602
|
|
|
|
|
|
|
| |
For defining a Git treeish which to look into, instead of the current
working copy.
Signed-off-by: Markus Lehtonen <markus.lehtonen@linux.intel.com>
|
|
|
|
| |
Signed-off-by: Markus Lehtonen <markus.lehtonen@linux.intel.com>
|
|
|
|
|
|
|
|
|
|
|
|
| |
Usage of six.moves.configparser in the previous way failed at least with
some versions of python-six (e.g. 1.4.1) like
> >>> from six.moves.configparser import SafeConfigParser
> Traceback (most recent call last):
> File "<stdin>", line 1, in <module>
> ImportError: No module named configparser
Signed-off-by: Markus Lehtonen <markus.lehtonen@linux.intel.com>
|
|
|
|
| |
and pylint exception that doesn't trigger
|
| |
|
|
|
|
|
| |
Types are classes in python3 so we get 'class' instead of 'type'.
Ignore that detail in the matcher since it's not relevanthere.
|
|
|
|
| |
Signed-off-by: Markus Lehtonen <markus.lehtonen@linux.intel.com>
|
|
|
|
|
|
| |
Returns False if tristate is 'off', otherwise True ('on' or 'auto').
Signed-off-by: Markus Lehtonen <markus.lehtonen@linux.intel.com>
|
|
|
|
|
|
|
| |
Provides better error messages as is more consistent as eq_ was already
used in some test cases.
Signed-off-by: Markus Lehtonen <markus.lehtonen@linux.intel.com>
|
|
|
|
| |
Signed-off-by: Markus Lehtonen <markus.lehtonen@linux.intel.com>
|
|
|
|
| |
Signed-off-by: Markus Lehtonen <markus.lehtonen@linux.intel.com>
|
| |
|