| Commit message (Collapse) | Author | Age |
|
|
|
|
|
| |
No functional change.
Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
|
|
|
|
|
|
|
| |
This means that tests can capture the output from t-dgit without going
wrong.
Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
|
|
|
|
|
|
|
|
|
| |
A `git worktree' has a file for .git. We still want to fsck them. We
can simply run the fsck in the directory containing .git, instead.
That works for ordinary trees with a .git directory, and for
worktrees.
Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
|
|
|
|
|
|
| |
This is really helpful when debugging.
Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
|
|
|
|
|
|
|
| |
This means that t-defdistro works without t-chain-test and
reinvocation.
Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
|
|
|
|
| |
Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
|
|
|
|
|
|
| |
We need this for the .dscs we have in tests/pkg-srcs/.
Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
|
|
|
|
|
|
|
| |
When we invoke these, this makes their output easier to find in the
middle of the test suite debug scrool.
Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
We want .save-env to be the env on entry to the original test, not
on execution of t-chain-test with a modified environment.
Without this the effects of t-defdistro might leak into the execution
of tests/setup/examplegit. This is no good because
tests/setup/examplegit relies in the -dtest-dummy which is our usual
dgit runes. (Or to put it another way, examplegit uses many suites
which are not (yet) set up properly by t-defdistro.)
Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
|
|
|
|
| |
Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
|
|
|
|
| |
Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
|
|
|
|
|
|
|
| |
Otherwise if we run this again in a fresh environment, the wrapper may
be on PATH even though DGIT_TEST_REAL_<FOO> is not set.
Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
|
|
|
|
| |
Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
|
|
|
|
| |
Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
|
|
|
|
| |
Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
|
|
|
|
| |
Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
|
|
|
|
| |
Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
|
|
|
|
|
|
| |
No overall functional change.
Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
|
|
|
|
| |
Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
|
|
|
|
|
|
|
|
|
|
|
| |
This is very like a commit generated by #849041. The wrongness is the
same. The difference is that it's not in a dgit-generated merge, but
the fixup script doesn't care about that.
The existing test case which uses this (drs-push-rejects) still sees
badness and is therefore still happy.
Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
|
|
|
|
| |
Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
|
|
|
|
|
|
|
|
|
|
| |
We must tolerate
notice: HEAD points to an unborn branch (master)
notice: No default references
which are generated by some of our existing test cases and are pretty
much harmless.
Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
|
|
|
|
|
|
| |
No functional change.
Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
|
|
|
|
|
|
| |
This will make it possible to be stricter.
Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
|
|\
| |
| |
| |
| |
| |
| |
| |
| |
| | |
Fix up semantic conflict: the new git-fsck test (in 2.14) fails on the
new broken commits (in the `defence in dgit-repos-server' branch).
We need to disable this. Ideally we would make a tighter test, but
that's too much to do in a merge.
Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
|
| |
| |
| |
| | |
Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
|
|/
|
|
|
|
|
|
|
|
|
| |
No functional change.
perl -i~ -pe 's/^echo ok\.$/t-ok/' tests/tests/*[^~]
perl -i~ -pe 's/^echo ok\.$/t-ok/' tests/setup/*[^~]
perl -i~ -pe 's/^echo done\.$/t-ok/' tests/tests/*[^~]
perl -i~ -pe 's/^echo done\.$/t-ok/' tests/setup/*[^~]
Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
|
|
|
|
|
|
|
|
|
| |
This avoids lintian complaining about our stupid test packages.
We need the stunt debuild because debuild likes to launder the
environment, removing our stunt lintian from PATH !
Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
|
|
|
|
| |
Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
|
|
|
|
|
|
| |
If it's wrong, we know what we're doing.
Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
|
|
|
|
|
|
|
|
|
| |
* Indirect t-archive-none and t-archive-query via this variable
* Rename t-archive-updated; it is for callers that have updated
aq/package.$suite.$p. There is one out-of-lib caller.
Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
|
|
|
|
|
|
|
|
|
| |
This is only used by the inarchivecopy test. And that ought not to
fail if there is no into-suite pseudomerge: that is, when the suite
branch contains the intended commit, rather than something merging the
intended commit into the suite branch.
Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
|
|
|
|
| |
Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
|
|
|
|
| |
Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
|
|
|
|
| |
Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
|
|
|
|
|
|
|
|
|
|
| |
Previously, t_ref_val="" was meant to mean "no refs specified yet" but
could also arise from "ref was specified but did not exist".
Distinguish these two cases, by using t_ref_val="" to mean only the
latter. "No refs specified yet" is represented as t_ref_val unset.
Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
|
|
|
|
|
|
| |
No callers yet.
Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
|
|
|
|
| |
Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
|
|
|
|
|
|
|
|
|
|
|
| |
The previous rune wouldn't process any new .origs. And in fact there
is no good reason to restrict this.
certainly not right now, because right now we are going to want to
process whatever is in incoming so that the orig-include-exclude test
can easily see what dgit push uploaded.
Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
|
|
|
|
|
|
|
| |
Even in non-split-brain mode.
Also change the test suite to expect this.
Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
|
|
|
|
|
|
|
| |
To work around #841143 and improve performance by amortising gnupg
migration cost.
Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
|
|
|
|
|
|
|
| |
This stops t-setup-import wrongly deciding we are in ADT non-shared
mode.
Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
|
|
|
|
|
|
|
|
|
|
| |
Name the IMPORT file after the setup name.
This is important when running under autopkgtest, where all the setups
share the tmp directory with each other and with the main test. The
setups need distinct IMPORTS.
Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
|
|
|
|
|
|
| |
No functional change.
Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
|
|
|
|
| |
Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
|
|
|
|
|
|
| |
Broken ever since it was introduced.
Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
|
|
|
|
| |
Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
|
|
|
|
|
|
|
|
|
|
|
|
| |
Specifically:
* Properly dereference the variable when looking for the stunt
dir to be there already
* Do not add "." to the path if the path starts out empty.
* Crash _with the right message_ in dpkg-parsechangelog if PERLLIB is
unset. (PERLLIB mustn't be unset because we're trying to strip
our own value from it.)
Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
|
|
|
|
|
|
|
| |
No overall functional change just yet, but it means we can wipe our
test archives.
Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
|
|
|
|
|
|
|
|
| |
Since actually several tests which use t-newtag are quilty split brain
ones which technically require it, although nothing checks this right
now.
Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
|