| Commit message (Collapse) | Author | Age | Files | Lines |
|
|
|
|
|
|
|
| |
Differentiate output when $QUILT_PATCHES/series is not found from
when $QUILT_PATCHES/series is empty.
This has originally been reported as
http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=557621
|
|
|
|
|
|
|
|
|
|
|
| |
Add an option to the test case runner script to generate an output
suitable for parallel runs:
* Print the full output at the end of the test case, all at once.
* Only output the detailed commands on failed test cases.
This makes it possible and convenient to run the test suite in
parallel on SMP systems.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Let the test suite runner script create a separate working
directory for each test. This makes it possible to run several
tests in parallel, speeding up testing on SMP systems. It also
ensures that a broken test case can no longer cause another test
case to fail.
At this point, it is possible to run the test suite in parallel,
the final result (success or failure) will be correct, however
the output is hardly readable because the progress of all tests
are interlaced. This will be improved later.
As a nice side effect, this means we can drop the setup and
clean-up steps that were present in all test cases.
|
|
|
|
| |
QUILT_PATCHES/QUILT_SERIES
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
http://savannah.nongnu.org/bugs/?19477 and
http://lists.nongnu.org/archive/html/quilt-dev/2009-09/msg00010.html
reported that the current quilt does not work with file name with
spaces. we all know that it is tough to handle it with bash script and
fix all quilt commands.
this patch only fixes add, files, and remove command with normal code
path. no excessive test has been done but a few test code is also
added. hope this starts fixing quilt to handle all possible path
pattern.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
I think the reason is that apply_patch_temporarily() uses backup-files
to create the temporary files, and by default backup-files uses hard
links. For files which are modified by the patch, this isn't a problem
because patch will unlink the file before modifying it. But for files
which are not modified by the patch, the hard link is preserved.
Then quilt revert also makes a hard link to restore the file, on the
assumption that the temporary copy will be deleted right away. At
this point, files not modified by the patch will have a link count
of 3 (.pc/<patch>/<file>, d.*/<file> and <file>). When the temporary
file is deleted, the link count is still 2 (.pc/<patch>/<file> and
<file>).
A simple way to fix this is to always use cp instead of ln when
restoring the files. Performance is not as good, but at least it is
always correct. We could optimize the code to use ln when possible
and cp for the other cases, but honestly I don't think it is worth
the extra complexity. The speed of "quilt revert" is hardly
critical, given how infrequently this command is used.
|
|
|
|
|
| |
Recognize the output of more recent versions of patch, and set the
coreutils quoting style instead of matching different quoting styles.
|
| |
|
|
|
|
|
|
| |
"The test for import did not always recreate a patches-dir
inside the temporary location, so it would fail if you
modified quilt using quilt."
|
|
|
|
|
| |
- Only run test/patch-wrapper.test if the --with-patch-wrapper
config option is used.
|
|
|
|
|
|
| |
sure that they won't end up in the series file in reverse order.
This happened because each new patch was inserted before the
"next" patch, which made it the new "next" patch.
|
|
|
|
|
| |
used internally from the resulting emails; not sure how that
was missed so far.
|
|
|
|
|
|
|
|
|
|
| |
subjects won't get expanded by accident (this was broken). Add
test case.
- quilt/mail.in: ``formail -x Foo'' will extract all Foo* headers,
so use ``formail -x Foo:'' instead.
- quilt/mail.in: Recognize sequences of common subject prefixes
like Fwd:, Fw:, Re:, Aw:, Tr:, [foo] only at the beginning of
subjects.
|
|
|
|
|
| |
the signature used. (The test suite still depended on the
.signature file of the user runnig it; fixed now.)
|
|
|
|
|
|
|
|
| |
specifying the location of the package sources. Pass the source
dir to inspect as well. Fix a test suite bug for quilt setup.
Minor cosmetic changes.
- Fix a quilt mail test suite bug: specify a dummy --sender in
test.quiltrc.
|
|
|
|
|
|
|
|
|
|
| |
that one. Only mail all patches to the end of the series if the
last patch is '-', or no patches have been specified.
- Revert command: change semantics to really revert the changes
relative to the patch instead of removing the file from the
patch. That's much more useful; the other behavior is available
as ``quilt diff <file> | patch -p1'', anyway.
- Update German translation.
|
|
|
|
|
| |
ambiguities with the delete command, and more closely matches
what the command does.
|
|
|
|
|
|
|
| |
possible: previously, a ``quilt add'' usually linked the file
to he backup and then copied the backup over the file to ensure
a link count of one. Recognize this case, and create a copy in
the first place instead.
|
|
|
|
|
| |
for reverse-applying a patch. Original patch contributed by
Randy Dunlap.
|
|
|
|
|
| |
put such comments in our test cases, and it prevents legitimate uses of
the "#" character (e.g. in a string, inside quotes) in commands.
|
|
|
|
| |
new patch instead of modifying the existing patch.
|
|
|
|
|
| |
"quilt delete" with no argument and no patch is applied.
- test/delete.test: Add a test for this specific case.
|
|
|
|
|
|
| |
file, for it breaks compatibility with previous versions of
quilt, and is inconsistent.
- test/dir-a-b.test: Drop test case for "-pab" in series file.
|
| |
|
|
|
|
| |
suite, as it may or may not be available.
|
|
|
|
| |
from Ville Syrjala.
|
| |
|
|
|
|
| |
- test/dir-a-b.test: Add a regression test case for this fix.
|
|
|
|
|
|
|
| |
produce a/file and b/file filenams in headers instead of
dir.orig/file and dir/file, similar to BK, git, hg patches.
- Only touch .pc/$patch/.timestamp when a path actually is being
refreshed.
|
|
|
|
|
|
|
|
|
|
|
| |
patches. Report and original patch by Arnaud Patard.
- quilt/push.in: Create .pc/$patch directory when pushing missing
or empty patches. If we don't, some commands such as rename can
fail later.
- test/missing.test: Test rename command on missing patches.
- test/empty.test: New test case similar to missing.test, but for
empty patches. Somewhat based on a contribution by Arnaud
Patard.
|
|
|
|
|
|
| |
find_applied_patch, which was causing a bogus error message
on failed pop.
- test/three.test: Test failed pop.
|
| |
|
| |
|
| |
|
| |
|
|
|
|
| |
- test/import2.test: New test case for compressed patches.
|
|
|
|
| |
patches to end up half-reversed (introduced on 22 Jan).
|
| |
|
|
|
|
|
|
|
| |
path. Fix a bug in check_for_existing_files. When a series file
is given as an argument, symlink to that series file in the
same way as we symlink to the patches. Improve the symlink
creation algorithm. Add testcase test/setup.test.
|
|
|
|
|
| |
in subdirectories.
- Add testcase test/edit.test to catch this kind of bug.
|
|
|
|
|
|
|
|
|
|
| |
most common options and mimics GNU patch's behavior and output,
and creates the quilt metadata as if quilt push was used to
apply the patch. When options are used that this wrapper does
not recognize, GNU patch is used directly, and no quilt metadata
will get created.
- lib/backup-files.c: Don't fail if a directory specified with
-B does not exist.
|
|
|
|
|
| |
directories. We occasionally have this condition in spec files,
and it would be really great if quilt setup could cope better.
|
|
|
|
|
|
|
| |
with `+'.
- test/run: Adjust $ENV{PWD} in cd command; add FIXME comment in
export command: in commands like `export foo=$var/bar', $var is
not evaluated.
|
| |
|
| |
|
|
|
|
| |
platforms on which $(hostname -f) doesn't work.
|
|
|
|
| |
hostname -f on platforms that don't support -f.
|
| |
|
|
|
|
| |
subject headers.
|
| |
|