Package: tcl8.6-doc
Version: 8.6.6+dfsg-1
Severity: minor
lindex(3tcl) says:
In extracting the element, lindex observes
the same rules concerning braces and quotes and backslashes as the Tcl
command interpreter; however, variable substitution and command
hen there is no other build system, then upstream are doing
the same manual thing that we are expecting ourselves, our users, and
our downstreams to do.
Thanks,
Ian.
--
Ian JacksonThese opinions are my own.
If I emailed you from an address @fyvzl.net or @evade.org.uk, that is
a private address which bypasses my fierce spamfilter.
rvices that
>were started by the build process. Services started by the build
> process must be shut down after use.
LGTM.
Ian.
--
Ian JacksonThese opinions are my own.
If I emailed you from an address @fyvzl.net or @evade.org.uk, that is
a private address which bypasses my fierce spamfilter.
Sean Whitton writes ("Bug#904306: clone-reprepro test: "apt updated wrong
number of Release files (), erk""):
> On Mon 23 Jul 2018 at 10:47AM +0100, Ian Jackson wrote:
> > Is your filesystem mounted `noatime' ? This might be #851873.
>
> It is indeed.
case
just for this. I think it would be better to add the configuration
option for a cross-section of existing tests.
Probably, that would be,
* introduce a new t-buildproductsdir-config subroutine
* add it to the top of a selection of existing tests
Do you want me to do that ?
Thanks,
Ian.
--
Ian
dgit: error: failed to obtain unstable: failed with error exit status 255
Is your filesystem mounted `noatime' ? This might be #851873.
Ian.
--
Ian JacksonThese opinions are my own.
If I emailed you from an address @fyvzl.net or @evade.org.uk, that is
a private address which bypasses my fierce spamfilter.
Ian Jackson writes ("[PATCH] test suite: unset EDITOR, which interferes."):
> Closes:#904308.
The Subject line of this patch was wrong but I think the code is
right.
Sean, can you test it please ?
Ian.
Closes:#904308.
Signed-off-by: Ian Jackson
---
debian/changelog | 2 +-
tests/lib| 1 +
2 files changed, 2 insertions(+), 1 deletion(-)
diff --git a/debian/changelog b/debian/changelog
index 1d496bbc..c1ab7e75 100644
--- a/debian/changelog
+++ b/debian/changelog
@@ -1,6 +1,6 @@
dgit
Ian Jackson writes ("Re: Bug#904308: test suite: git invokes
$EDITOR/$VISUAL/sensible-editor"):
> Can you please send me the output of
>printenv | sort
> ?
Never mind. The problem is that a couple of the tests set EDITOR but
do not set VISUAL. So it will run your VIS
Sean Whitton writes ("Bug#904308: test suite: git invokes
$EDITOR/$VISUAL/sensible-editor"):
> Source: dgit
> Version: 5.10
>
> When the test suite runs the user's configured editor can be invoked.
> This can pop up new X windows and sometimes tests will fail. Sometimes
> it is git that is
Sean Whitton writes ("Bug#628515: Seeking seconds for patch to recommend
verbose logs and define DEB_BUILD_OPTIONS=terse"):
> I think that the use of 'maximally' is fine given that the previous
> sentence is now qualified with 'reasonably'.
Yes.
> Here is the revised patch; David and Andrey,
Sean Whitton writes ("Bug#628515: Seeking seconds for patch to recommend
verbose logs and define DEB_BUILD_OPTIONS=terse"):
> > +The package build should be as verbose as possible, except where the
Are you sure "as verbose as {+reasonably+} possible" would not be
better ?
Imagine a build that,
Debian Bug Tracking System writes ("Bug#841414 closed by Ian Jackson
(Bug#841414: fixed in dgit 5.10)"):
> #841414: Should occasionally run git-gc in dgit-repos
FAOD this is not only fixed in dgit in Debian sid, but also deployed
on the Debian dgit server infrastructure.
Ian.
e dgit view.
Tangentially relevant is this change in dgit 2.13:
Changed behaviour:
* quilt fixup: Permit creation of patches which delete files, by psssing
--include-removal to dpkg-source, and tolerating it when we do our
quilt fixup analysis. dpkg-source has supported this since at
at all, given that that in
that bug the effect of the faketime wrapper can apparently be produced
simply by setting env vars.)
Ian.
--
Ian JacksonThese opinions are my own.
If I emailed you from an address @fyvzl.net or @evade.org.uk, that is
a private address which bypasses my fierce
Ian Jackson writes ("Re: Bug#903441: dgit: autopkgtest failures in Ubuntu [and
1 more messages]"):
> Sorry, this is due to #903696 in autopkgtest, which I just filed. I
> should have CC'd you.
>
> 5.9+exp4 is supposed to fix it. This time, for sure.
Wow! It actually D
Mattia Rizzolo writes ("Re: Bug#903441: dgit: autopkgtest failures in Ubuntu
[and 1 more messages]"):
> Doesn't look like it's enough, ci.d.n still fails due to:
> |cp: cannot overwrite directory
> '/tmp/tmp.ugPG6m6qcG/autopkgtest-incoming/unstable/amd64/d/dgit/601985/artifacts/example/.git'
>
the test.
Ian.
--
Ian JacksonThese opinions are my own.
If I emailed you from an address @fyvzl.net or @evade.org.uk, that is
a private address which bypasses my fierce spamfilter.
Ian Jackson writes ("Re: Bug#903441: dgit: autopkgtest failures in Ubuntu [and
1 more messages]"):
> This is now implemented in 5.9+exp1, just uploaded to Debian
> experimental. Can you give it a try in Ubuntu please ?
This trips what seems to be a bug in adt-virt-lxc; see #
dgit 5.9+exp1,
which copies some non-world-readable files into
$AUTOPKGTEST_ARTIFACTS.
I will add an appropriate chmod a+r.
Ian.
--
Ian JacksonThese opinions are my own.
If I emailed you from an address @fyvzl.net or @evade.org.uk, that is
a private address which bypasses my fierce spamfilter.
Ian Jackson writes ("Re: Bug#903598: devscripts: dch -r [something]"):
> Mattia Rizzolo writes ("Re: Bug#903598: devscripts: dch -r [something]"):
> > Ton of things use '' (i.e. empty string). Including debchange's
> > testsuite :3
>
> Well, that'
nything with the supplied
> string
>
> (so that unstable is affected as well…)
Ah, yes. Indeed.
> On Wed, Jul 11, 2018 at 06:37:27PM +0100, Ian Jackson wrote:
> > Consider
> > dch -r `foo'
> >
> > dch ignores `foo', which is `TEXT', in the sense that `foo' does n
Mattia Rizzolo writes ("Re: Bug#903441: dgit: autopkgtest failures in Ubuntu"):
> On Wed, Jul 11, 2018 at 06:33:45PM +0100, Ian Jackson wrote:
> > > Do you think it would be (easily) possible to have the test dir saved to
> > > the artifacts only in case of failures
Package: devscripts
Version: 2.17.6+deb9u1
Severity: minor
Consider
dch -r `foo'
dch ignores `foo', which is `TEXT', in the sense that `foo' does not
appear in the resulting changelog entry. *But* if you don't supply
`foo' dch runs an editor. And there's no other way to stop it doing
so
Mattia Rizzolo writes ("Re: Bug#903441: dgit: autopkgtest failures in Ubuntu"):
> On Tue, Jul 10, 2018 at 10:57:03PM +0100, Ian Jackson wrote:
> > Can you run the test on bionic with this attached patch ?
>
> Alas, apparently the infra doesn't like it. It seems the t
Mattia Rizzolo writes ("Re: Bug#903441: dgit: autopkgtest failures in Ubuntu"):
> It just so happens that I never really read dch's manpage in full, so I
> don't remember all the options, but looking at it now, -r doesn't really
> take any option. If you need to force a distribution you need to
/lib
@@ -40,6 +40,10 @@ testname="${DGIT_TEST_TESTNAME-${0##*/}}"
export DGIT_TEST_TROOT=$troot
tmp=$ADTTMP
+if [ x"$AUTOPKGTEST_ARTIFACTS" != x ]; then
+ tmp=$AUTOPKGTEST_ARTIFACTS/tmp
+ mkdir -p -- "$tmp"
+fi
if [ x"$tmp" = x ]; then
export DG
Johannes Schauer writes ("Bug#841414: Should occasionally run git-gc in
dgit-repos"):
> for a couple of days I have tried to do the following:
>
> $ dgit clone botch
> canonical suite name for unstable is sid
> fetching existing git history
> remote: warning: suboptimal pack - out of memory
>
Mattia Rizzolo writes ("Re: Bug#903441: dgit: autopkgtest failures in Ubuntu"):
> On Tue, Jul 10, 2018 at 11:55:57AM +0100, Ian Jackson wrote:
> > I don't think it's possible to add something to the artifacts, from
> > within a test. I didn't provide any way to do tha
Mattia Rizzolo writes ("Re: Bug#903441: dgit: autopkgtest failures in Ubuntu"):
> On Tue, Jul 10, 2018 at 11:12:01AM +0100, Ian Jackson wrote:
> > Thanks. I don't have an install of cosmic and the test logs are not
> > quite sufficient. Can I get a copy of the temporar
rectory after
one of these test runs ?
I suspect that what is going on is that cosmic's dch is putting
`cosmic' into the changelog. If so, depending on dch putting
`unstable' in would be a bug in the test suite.
Ian.
--
Ian JacksonThese opinions are my own.
If I emailed you from an addr
es which were git rm'd, but not committed,
and then recreated.
Ian.
--
Ian JacksonThese opinions are my own.
If I emailed you from an address @fyvzl.net or @evade.org.uk, that is
a private address which bypasses my fierce spamfilter.
ehow avoid stuffing all that into the
maintainer's .git but I can't see how. (Also, all the data ends up
being uploaded twice, at least on first upload.) I think this is a
problem for the future, and so that is not what this bug is about.)
Ian.
--
Ian JacksonThese opinions are my own.
If I e
tag
corresponds to dgit/dgit/sid (except for debian/patches).
But this is complicated and fiddly. I suggest that we document this
problem, instead.
Ian.
--
Ian JacksonThese opinions are my own.
If I emailed you from an address @fyvzl.net or @evade.org.uk, that is
a private address which
too. On the gripping hand, this
function might be necessary even with a pure-git workflow when dgit is
not needed.
Ian.
--
Ian JacksonThese opinions are my own.
If I emailed you from an address @fyvzl.net or @evade.org.uk, that is
a private address which bypasses my fierce spamfilter.
Ian Jackson writes ("Re: Bug#903132: -fdiverged when `git debrebase
convert-from-gbp` after `dgit push --gbp`"):
> Ian Jackson writes ("Re: Bug#903132: -fdiverged when `git debrebase
> convert-from-gbp` after `dgit push --gbp`"):
> > Also what should it do if the
Ian Jackson writes ("Re: Bug#903132: -fdiverged when `git debrebase
convert-from-gbp` after `dgit push --gbp`"):
> I'm not entirely sure about this system of interlocking snags, where
> the behaviour varies depending on which set of them you chose to
> force.
On
Sean Whitton writes ("Bug#903132: -fdiverged when `git debrebase
convert-from-gbp` after `dgit push --gbp`"):
> Either they should not require passing it, or dgit-maint-debrebase(7)
> should explain that it will be needed. I'm not yet sure which is right.
I have thought about this and:
In most
Sean Whitton writes ("Re: Bug#812721: gbp could filter out Files-Excluded:
entries when committing to the pristine-tar branch"):
> It is useful to provide the full upstream git history to users because
> they can use tools like `git bisect` to figure out how to fix bugs. The
> packaging history
Mattia Rizzolo writes ("Re: Bug#902920: diffoscope: be triggered by all the
reverse-recommends"):
> Mh, in our case we have this entry:
> Tests: pytest
> Depends: diffoscope, python3-pytest
> Restrictions: needs-recommends
> If I go down the route of autogenerating d/tests/control I
Sean Whitton writes ("Bug#903127: git-debrebase new-upstream does not include a
Debian revision in version"):
> Package: git-debrebase
> Version: 5.2
> Severity: normal
>
> The changelog entry generated by `git debrebase new-upstream` does not
> contain a Debian revision. It should append -1 to
Control: clone -1 -2
Control: retitle -2 Want to spot source-only binaries-NEW upload to Debian
Control: reopen -2
Control: found -2 dgit/5.7
Johannes Schauer writes ("Bug#801435: closed by Ian Jackson
(Bug#801435: fixed in dgit 5.7)"):
> Quoting Ian Jackson (2018-07-05 17:04:44
Johannes Schauer writes ("Bug#801435: closed by Ian Jackson
(Bug#801435: fixed in dgit 5.7)"):
> Quoting Debian Bug Tracking System (2018-07-05 16:30:04)
> >* dgit: Check that entirely-new uploads to Debian are not
> > source-only-uploads, as those are REJECTed
use dgit.
>
> Yes, it's only compatible with gbp user.
dgit and gbp work very well together. IMO all gbp users should be
using dgit.
Ian.
--
Ian JacksonThese opinions are my own.
If I emailed you from an address @fyvzl.net or @evade.org.uk, that is
a private address which bypasses my fierce spamfilter.
Shengjing Zhu writes ("Re: Bug#812721: gbp could filter out Files-Excluded:
entries when committing to the pristine-tar branch"):
> Here's my thoughts about the workflow with Files-Excluded.
>
> 1. Upstream git history is imported *as is* in upstream branch. If we
> want to maintain a upstream
.UTF-8"
> LC_MEASUREMENT="en_GB.UTF-8"
> LC_IDENTIFICATION="en_GB.UTF-8"
> LC_ALL=
>
> ... which is surely the same as yours?
No, my usual locale is C.UTF-8. I can't stand non-C LC_COLLATE.
Anyway, I think I have fixed it.
However, there is somethi
-dgit build
t-dgit push --deliberately-fresh-repo
--
Ian JacksonThese opinions are my own.
If I emailed you from an address @fyvzl.net or @evade.org.uk, that is
a private address which bypasses my fierce spamfilter.
ST FAILED
>
> I can provide the full log by private mail.
Yes, please. Also, next time, please don't snip the end of the test
log. That random-looking junk at the end has the line numbers in it
:-).
Ian.
--
Ian JacksonThese opinions are my own.
If I emailed you from an a
Control: tags -1 confirmed
Control: usertags -1 rsn
Sean Whitton writes ("Bug#903006: Indeterminacy issues with import-nonnative
and import-tarbomb tests"):
> On my machine, outside of autopkgtest only, these tests are failing
> because of something to do with file ordering:
I have reproduced
dgit rpush ought to work fairly well for the stuations where debrsign
might be used, and does not have this ssh trust reversal problem.
(It still operates rather as a signing oracle but the signing oracle
is much more restricted.)
Ian.
--
Ian JacksonThese opinions are my own.
If I emailed
Control: close -1
This wnpp bug is nonsense, because this module is already there in the
package libanyevent-perl.
--
Ian JacksonThese opinions are my own.
If I emailed you from an address @fyvzl.net or @evade.org.uk, that is
a private address which bypasses my fierce spamfilter.
Package: wnpp
Severity: wishlist
Owner: Ian Jackson
* Package name: libanyevent-socket-perl
Version : 7.14
Upstream Author : Marc A. Lehmann
* URL : https://metacpan.org/pod/AnyEvent::Socket
* License : Artistic / GPL1+
Programming Lang: Perl
Description
Package: wnpp
Severity: wishlist
Owner: Ian Jackson
* Package name: libanyevent-websocket-client-perl
Version : 0.48
Upstream Author : Marc A. Lehmann
* URL : https://metacpan.org/pod/AnyEvent::WebSocket::Client
* License : Artistic / GPL1+
Programming Lang
utting something like that in your debian/tests/control
is that autopkgtest will always skip the test, but the Depends get
copied to Testsuite-Triggers and used by ci.d.n.
HTH.
Regards,
Ian.
--
Ian JacksonThese opinions are my own.
If I emailed you from an address @fyvzl.net or @evade.org.uk, that is
a private address which bypasses my fierce spamfilter.
ipts" to explicitly
> permit wrapper scripts §9.9 written by the maintainer?
I think shell scripts are programs.
Ian.
--
Ian JacksonThese opinions are my own.
If I emailed you from an address @fyvzl.net or @evade.org.uk, that is
a private address which bypasses my fierce spamfilter.
Package: dgit
Version: 5.6
Usertags: rsn
dgit clone does this:
set -o pipefail
git ls-tree -r --name-only -z HEAD | \
xargs -0r touch -h -r . --
The explanation is this:
commit 4c09c958f20263601483e0150caf4787a064aa5e
Author: Ian Jackson
Date: Sun Oct 30 16:34:11
to follow the instructions, is better than
(nearly) none.
Thanks,
Ian.
--
Ian JacksonThese opinions are my own.
If I emailed you from an address @fyvzl.net or @evade.org.uk, that is
a private address which bypasses my fierce spamfilter.
tderr.
The test is entitled to write non-UTF-8 to these logfiles.
autopkgtest should not check, launder or convert the character set.
Neither should whatever means are used to publish the files.
* A new restriction needs-log-dir can be declared by tests which
unconditionally use AUTOP
Ian Jackson writes ("Re: Bug#902316: gnupg failing completely in dgit
autopkgtests [and 1 more messages]"):
> I attempted to work around this problem by explicitly starting and
> stopping the gpg-agent. This has dramatically reduced the failure
> probability.
...
> I am
Sean Whitton writes ("Bug#902612: Packages should not touch users' home
directories"):
> Do you think we could resolve this by adding "... except when that user
> was created by the package or a closely related package." ?
Maybe the concept of an "unrelated&q
-i agent | less -j20 +/'^ *11904'
FYI, attached is the diff, ie the output of
git-diff dgit/dgit/sid..dgit/dgit/experimental tests
in dgit.
I am going to try even more ridiculous workarounds.
workaround.diff
Description: workaround patch
--
Ian JacksonThese opinions are my own.
If I
Sean Whitton writes ("Bug#902559: some tests can fail if gnupg happens not to
be installed"):
> On Wed, Jun 27 2018, Ian Jackson wrote:
> > Here is a better patch. This has passed more tests. If all is well I
> > will upload it shortly.
>
> Ftr cherry-picking the
t most people won't ignore a
"should" unless they feel they understand why it's there.
Ian.
--
Ian JacksonThese opinions are my own.
If I emailed you from an address @fyvzl.net or @evade.org.uk, that is
a private address which bypasses my fierce spamfilter.
e
there are not going to be big arguments about this.
Ian.
--
Ian JacksonThese opinions are my own.
If I emailed you from an address @fyvzl.net or @evade.org.uk, that is
a private address which bypasses my fierce spamfilter.
Ian Jackson writes ("Re: Bug#902559: some tests can fail if gnupg happens not
to be installed"):
> Here is a better patch. This has passed more tests. If all is well I
> will upload it shortly.
>From 03d9c0e7c24a96cd792c051026afd46a7fd02bfc Mon Sep 17 00:00:00 2001
From: Ian
Control: tags -1 + patch
Here is a better patch. This has passed more tests. If all is well I
will upload it shortly.
Sean, you can probably get autopkgtest to run the new test suite
against the old source code, if you want to test your backport. It
may be easier to wait for 5.4 though.
Ian.
Control: tags -1 - patch
This patch does not work properly.
rom cb8e3355cf9816b86a16108fd22405624aa5a8f4 Mon Sep 17 00:00:00 2001
From: Ian Jackson
Date: Wed, 27 Jun 2018 21:41:38 +0100
Subject: [PATCH] test suite: gdr-viagit, gdr-newupstream: do not auto-import
t-setup-gnupg. This fixes them in chroots without gnupg.
---
debian/changelog |
nupg even though (i) it doesn't need
to (ii) there is no gnupg in, or implied, by the test dependencies.
I will upload a fix RSN.
Ian.
--
Ian JacksonThese opinions are my own.
If I emailed you from an address @fyvzl.net or @evade.org.uk, that is
a private address which bypasses my fierce spamfilter.
So I think the
most likely explanation is a bug in the test suite. Of course that
means that these tests were not run.
I will see if I can repro it in adt with a stretch chroot.
Ian.
--
Ian JacksonThese opinions are my own.
If I emailed you from an address @fyvzl.net or @evade.org.uk, tha
Ian Jackson writes ("Re: Bug#902534: dgit: needs a better integration story
with patches-unapplied maintainer trees"):
> I think maybe the fact that I am writing this all here suggests we
> need better wording in the --overwrite section of dgit(1).
How about this ?
--ove
Sean Whitton writes ("Bug#869146: dgit: should never add .orig to the .changes
more than once"):
> On Tue, Jun 26 2018, Ian Jackson wrote:
> > Of course if you had followed dgit's advice you would not have
> > experienced this but :-)
>
> I must have been sure
onto the end of your queue, without rebasing.
Also, I'm not sure what you mean by "rebase pain". Maybe you expect
more pain than git-debrebase would inflict ?
> Perl is definitely not my strong suit, but I hope at least this
> discussion will end up providing a clearer view
Ian Jackson writes ("Re: Bug#902534: dgit: needs a better integration story
with patches-unapplied maintainer trees"):
> Control: forcemerge -1 842614
...
I just had a thought, though:
It might be worth seeing what git-debrebase convert-to-gbp does on
some of the the dgit dsc i
Control: forcemerge -1 842614
Felipe Sateler writes ("Bug#902534: dgit: needs a better integration story with
patches-unapplied maintainer trees"):
> Currently, dgit-maint-gbp(7) says:
> > dgit pull can't yet incorporate NMUs into patches-unapplied gbp branches.
> > You can just apply the NMU
Control: severity -1 wishlist
Since the VM for git.dgit.d.o was enalrged, there have been no
operational problems. This bug report is a request for a performance
optimisation which does not, currently, seem to be at all urgent.
Ian.
--
Ian JacksonThese opinions are my own.
If I emailed
bit more prominent.
Looking at the top of git-debrebase(5), the 2nd half of INTRODUCTION
is rather discursive. Maybe add a section heading just before,
DISCUSSION or something, and put the gitk thing there ?
Ian.
--
Ian JacksonThese opinions are my own.
If I emailed you from an address @fyvzl.n
[ message from private email ruthlessly trimmed for #901927 ]
Sean Whitton writes ("Re: Manpage suggestions"):
> On Wed, Jun 20 2018, Ian Jackson wrote:
> > #901927 [w| | ] [dgit, git-debrebase] Add gitk tricks to documentation
>
> I'm not sure where it should go
.
Ian.
--
Ian JacksonThese opinions are my own.
If I emailed you from an address @fyvzl.net or @evade.org.uk, that is
a private address which bypasses my fierce spamfilter.
And add cross-references to dgit(1) and the actual error.
Somewhat apropos of #842614.
Signed-off-by: Ian Jackson
---
debian/changelog | 3 ++
dgit | 3 +-
dgit.1 | 5
dgit.7 | 84
4 files changed
lanation of
this error message to the quilt-fixup section of dgit(1), and a
cross-reference to the error message.
Ian.
--
Ian JacksonThese opinions are my own.
If I emailed you from an address @fyvzl.net or @evade.org.uk, that is
a private address which bypasses my fierce spamfilter.
-
maint-*(7).
Ian.
--
Ian JacksonThese opinions are my own.
If I emailed you from an address @fyvzl.net or @evade.org.uk, that is
a private address which bypasses my fierce spamfilter.
Ian Jackson writes ("dgit: should never add .orig to the .changes more than
once"):
> Anyway, I will fix the bug.
Having investigated, I have discovered that this might also happen if
you didn't use dgit build, and whatever build process you used
included some .origs. dgit would
xperienced this but :-). Anyway, I will fix the bug.
Thanks,
Ian.
--
Ian JacksonThese opinions are my own.
If I emailed you from an address @fyvzl.net or @evade.org.uk, that is
a private address which bypasses my fierce spamfilter.
Control: block -1 by 851679
--
Ian JacksonThese opinions are my own.
If I emailed you from an address @fyvzl.net or @evade.org.uk, that is
a private address which bypasses my fierce spamfilter.
Control: fixed -1 3.3
Control: close -1
Ian Jackson writes ("Bug#869675: extracting upstream source."):
> I hope this is of some use. I appreciate that it's a bit late to be
> replying to this now. Anyway, the above is not 100% convenient but
> ISTM that it is good enough gi
Sean Whitton writes ("Bug#886969: Problems with export-dsc and push-source on
arbitrary refs"):
> On Mon, Jun 25 2018, Ian Jackson wrote:
> > export-dsc has no push and that makes the whole split brain mode seem
> > rather odd.
>
> Hmm, why would it be odd s
(dropping the gnupg bug again)
Paul Gevers writes ("Re: Bug#902316: gnupg failing completely in dgit
autopkgtests"):
> Adding Martin and Antonio into the loop, as I don't know the details of
> autopkgtest so well. Question is near the bottom.
Thanks for your attention, everyone.
> PS: for you
Sean Whitton writes ("Re: Bug#886969: Problems with export-dsc and push-source
on arbitrary refs"):
> On Fri, Jun 22 2018, Ian Jackson wrote:
> > So I think for push-source:
> >
> > 886966 push-source should be usable on arbitrary refs 886625
> > push-source
emd239-1
+systemd-sysv 239-1
Has something else changed ?
Ian.
[1] My complete adt-run has not yet finished but it's obviously not
behaving like the one on ci.debian.net.
--
Ian JacksonThese opinions are my own.
If I emailed you from an address @fyvzl.net or @evade.org.uk, that is
a
Paul Gevers writes ("Re: Bug#901897 closed by Jonathan Nieder
(Bug#901897: fixed in git 1:2.18.0~rc2-2)"):
> On 22-06-18 18:05, Ian Jackson wrote:
> It currently happens quite often, but often the test suite from testing
> runs happily with the package from unstable. If th
g about here ? If you are
willing to wait a couple of weeks, I think sid/testing's dgit will
have settled down.
Ian.
[1] git://git.chiark.greenend.org.uk/~ianmdlvl/dgit.git#wip (rebasing)
--
Ian JacksonThese opinions are my own.
If I emailed you from an address @fyvzl.net or @evade.org.u
a lot of workaround. It would be
better to treat britney as free software that we can modify to DTRT.
But I'm not offering to do anh of the work so I will let you get on
with your workaroud and wish you luck!
Regards,
Ian.
--
Ian JacksonThese opinions are my own.
If I emailed you from an addre
Ian Jackson writes ("Re: Bug#901897 closed by Jonathan Nieder
(Bug#901897: fixed in git 1:2.18.0~rc2-2)"):
> I think it _is_ optional. At least, it was in adt-run. I think
> autopkgtest has much of the same logic. It's a question of putting
> together the right rune to sp
Paul Gevers writes ("Re: Bug#901897 closed by Jonathan Nieder
(Bug#901897: fixed in git 1:2.18.0~rc2-2)"):
> On 22-06-18 15:02, Ian Jackson wrote:
> > I don't know exactly what autopkgtest command line rune you are using
> > but certainly it was designed to be able to
Paul Gevers writes ("Re: Bug#901897 closed by Jonathan Nieder
(Bug#901897: fixed in git 1:2.18.0~rc2-2)"):
> On 22-06-18 12:45, Ian Jackson wrote:
> > to the bad run which prompted your intervention, and the good run
> > which resulted ?
>
> [2] for the bad one.
shed out of the
cache) during the subsequent push.
I am somewhat tempted to say that export-dsc should be forbidden with
split brain quilt modes.
Ian.
--
Ian JacksonThese opinions are my own.
If I emailed you from an address @fyvzl.net or @evade.org.uk, that is
a private address which bypasses my fierce spamfilter.
ation information from britney from yesterday and
today, all is fine except for git's an artificially inflated urgency.
Ian.
[1] https://ci.debian.net/data/autopkgtest/testing/amd64/d/dgit/491815/log.gz
--
Ian JacksonThese opinions are my own.
If I emailed you from an address @fyvz
r (or, even, representable by)
dpkg-source. There's a bug about that too.
Ian.
--
Ian JacksonThese opinions are my own.
If I emailed you from an address @fyvzl.net or @evade.org.uk, that is
a private address which bypasses my fierce spamfilter.
Sean Whitton writes ("Re: Want dgit setup-vcs-git [and 1 more messages]"):
> On Thu, Jun 21 2018, Ian Jackson wrote:
> > Sean, would you care to comment on the name of the command and the
> > manpage ? The way it isn't like other setup-* things (in that
>
901 - 1000 of 3535 matches
Mail list logo
