On Tue, 04 Feb 2025 00:21:25 +0100 Sven Geuer wrote:
> I observered the same flaw with other packages in the past, vanishing
> eventually for unknown reasons.
The same meanwhile happened to the exifprobe package, no complains from
vcswatch anymore.
> Any idea what's going wrong
Package: qa.debian.org
Severity: wishlist
User: qa.debian@packages.debian.org
Usertags: vcswatch
Dear vcswatch developers,
could you please extend vcswatch to check if the upstream and
pristine-tar branches (if they exist) are aligned with the main branch
and its changelog?
For instance
Package: qa.debian.org
Severity: normal
Dear Maintainer,
for package exifprobe vcswatch asserts
Error: fatal: missing blob object '5688c95c719bd0ccef0e904ef4366410356308f4'
error: remote did not send all necessary objects
while a fresh git clone well contains this object:
$ g
Your message dated Thu, 9 Jan 2025 22:01:21 +0100
with message-id
and subject line Re: Bug#1092608: vcswatch wrongly claims dpkg Git repository
is > 500 MiB
has caused the Debian Bug report #1092608,
regarding vcswatch wrongly claims dpkg Git repository is > 500 MiB
to be marked as done.
Your message dated Thu, 9 Jan 2025 22:01:21 +0100
with message-id
and subject line Re: Bug#1092608: vcswatch wrongly claims dpkg Git repository
is > 500 MiB
has caused the Debian Bug report #1092608,
regarding qa.debian.org: vcswatch fails to update dpkg repo due to size limit
to be marked
Package: qa.debian.org
According to <https://qa.debian.org/cgi-bin/vcswatch?package=dpkg>:
Error: Repository size 561401856 exceeds 500 MiB, blocking it
But a fresh clone of the cited repository is only 102 MiB, or 63 MiB
without a working copy:
wraith:/tmp$ git clone https://git.dp
Package: qa.debian.org
Severity: wishlist
User: qa.debian@packages.debian.org
Usertags: vcswatch
vcswatch should monitor the backports archives and report (at least) if
the tag associated with a bpo upload is missing.
It would be nice if vcswatch could also monitor the DEP-14
release
Your message dated Sat, 30 Nov 2024 10:25:45 +0100
with message-id
and subject line Re: Bug#1088731: vcswatch confusion: e2fsprogs does not
exceeds 500 MiB
has caused the Debian Bug report #1088731,
regarding vcswatch confusion: e2fsprogs does not exceeds 500 MiB
to be marked as done.
This
Package: qa.debian.org
Severity: normal
Per
https://qa.debian.org/cgi-bin/vcswatch?package=e2fsprogs
The e2fsprogs repository is blocked because:
Repository size 537583616 exceeds 500 MiB, blocking it
However, this is very clearly not the case:
% git clone https://git.kernel.org/pub
Your message dated Tue, 17 Sep 2024 23:02:56 +0200
with message-id <9dabe50827c31a536498a36ac8dee147776dc097.ca...@debian.org>
and subject line Re: vcswatch: fails to query salsa, impacts DDPO and
tracker.d.o
has caused the Debian Bug report #1081461,
regarding vcswatch: fails to query
Package: qa.debian.org
Severity: important
Dear Maintainer,
vcswatch is failing to query salsa.d.o for several days by now.
Here is, e.g., the error message for the package vpnc:
https://salsa.debian.org/api/v4/projects/pkg-security-team%2Fvpnc
API request failed: 401 Unauthorized at
Hi,
The following error is occurring.
See https://qa.debian.org/cgi-bin/vcswatch?package=celluloid
--
Cheers,
Leandro Cunha
signature.asc
Description: Binary data
Hi!
I think I've broken vcswatch :(
I was updating jupyter-server-mathjax, but probably messed up a commit
along the line. Anyway, somehow, vcswatch now says:
Error: fatal: missing blob object 'ae757feb8182c5c3d2934dee63504ea0a87f02ac'
error: remote did not send all neces
Package: qa.debian.org
Severity: normal
Hi!
It seems vcswatch is having some issues with the dpkg git repo, and it
complains that it is too big, exceeding 500 MiB.
From https://qa.debian.org/cgi-bin/vcswatch?package=dpkg:
,--
ERROR:
Git: https://git.dpkg.org/git/dpkg/dpkg.git
Branch: main
Path
Package: qa.debian.org
Severity: wishlist
Hi,
I've switched off TLS1.2 on my git server (to see what would be
brocken), one of them is VCSWatch:
Error: fatal: unable to access
'https://git.progress-linux.org/users/daniel.baumann/debian/packages/aio-eapi/':
gnutls_handshake(
Your message dated Thu, 21 Mar 2024 11:28:38 +
with message-id <65fc19e6a9c68_20a82994c-...@godard.mail>
and subject line Bug#1067217 fixed in qa.debian.org
has caused the Debian Bug report #1067217,
regarding qa.debian.org: vcswatch fails with "remote: warning: unable to find
Package: qa.debian.org
Severity: normal
This does not seem to be a transient error:
Error: remote: warning: unable to find all commit-graph files
https://qa.debian.org/cgi-bin/vcswatch?package=acpi-call
https://qa.debian.org/cgi-bin/vcswatch?package=nvidia-support
Andreas
Hi!
On Thu, 2024-03-07 at 14:34:39 +0100, Christoph Berg wrote:
> > It looks like this broke for remotes that do not support filtering
> > (yet?). The attached completely untested patch might make this work
> > again. Affecting at least git.hadrons.org and git.dpkg.org, but there
> > might be othe
> It looks like this broke for remotes that do not support filtering
> (yet?). The attached completely untested patch might make this work
> again. Affecting at least git.hadrons.org and git.dpkg.org, but there
> might be others too.
Thanks for spotting that, the patch seems to work.
Re-running t
Hi!
On Wed, 2024-03-06 at 14:15:03 +, Debian Bug Tracking System wrote:
>
> vcswatch: Use --filter blob:none
>
> Required blobs (changelog, control, upstream metadata) are fetched by
> git on demand.
On 3/7/24 3:50 AM, Boyuan Yang wrote:
I am not sure where vcswatch is running on
quantz, see: https://db.debian.org/machines.cgi
search for "qa" in the Description column.
, but obviously the machine
now has a full disk that makes vcswatch fail:
See https://qa.debian.org/cgi-bi
Hi,
I am not sure where vcswatch is running on, but obviously the machine
now has a full disk that makes vcswatch fail:
See https://qa.debian.org/cgi-bin/vcswatch?package=rspamd :
Last scan: 2024-03-03 20:52:07+00
Error: fatal: Unable to create
'/srv/scratch/qa.debian.org/vcswatch/r/r
Your message dated Wed, 06 Mar 2024 14:13:20 +
with message-id <65e87a00c054a_2433439607...@godard.mail>
and subject line Bug#1032623 fixed in qa.debian.org
has caused the Debian Bug report #1032623,
regarding vcswatch: should not raise error on repos > 1GiB in size
to be marked as do
Package: qa.debian.org
Followup-For: Bug #1029068
Hi,
Just noticed at least one of my packages¹ got also hit by this one:
Error: fatal: Unable to create temporary file
'/srv/scratch/qa.debian.org/vcswatch/c/claws-mail/objects/pack/tmp_idx_XX':
No space left on device fatal:
Your message dated Sat, 06 Jan 2024 22:31:14 +
with message-id <6599d4b262eb_59f94c-...@godard.mail>
and subject line Bug#1060115 fixed in qa.debian.org
has caused the Debian Bug report #1060115,
regarding vcswatch: look for changelog in debian/experimental branch
to be marked as done.
Re: Martin
> + my @branch_list = qw(HEAD debian debian/master debian/sid
> debian/latest debian/experimental master);
Applied, thanks!
Christoph
Package: qa.debian.org
Severity: wishlist
Tags: patch
X-Debbugs-Cc: m...@debian.org
Useful for packages only in debian/experimental.
diff --git a/data/vcswatch/vcswatch b/data/vcswatch/vcswatch
index bb20d156..2d8491cc 100755
--- a/data/vcswatch/vcswatch
+++ b/data/vcswatch/vcswatch
@@ -330,7
Your message dated Tue, 5 Sep 2023 00:10:20 +0200
with message-id
and subject line Re: Bug#1050657: UDD/vcswatch importer: add support for
ci_status, ci_url, merge_requests
has caused the Debian Bug report #1050657,
regarding UDD/vcswatch importer: add support for ci_status, ci_url
Package: qa.debian.org
User: qa.debian@packages.debian.org
Usertags: udd
Hi,
vcswatch knows about the last pipeline status, and the number of merge
requests, in /srv/udd.debian.org/mirrors/qa.debian.org-vcswatch.
See ci_url, ci_status, merge_requests.
but UDD does not import that data.
It
Re: Gábor Németh
> I see.
>
> > trycmd ('git', 'clone', '--quiet', '--bare', '--mirror', '--depth',
> > '50', '--no-single-branch', '--template', '', $url, $pkgdir);
> >
> > depth=1 wouldn't work since we use "git describe".
>
> Wouldn't then adding [.., '--filter', 'blob:none', ..] suffice for t
I see.
> trycmd ('git', 'clone', '--quiet', '--bare', '--mirror', '--depth',
> '50', '--no-single-branch', '--template', '', $url, $pkgdir);
>
> depth=1 wouldn't work since we use "git describe".
Wouldn't then adding [.., '--filter', 'blob:none', ..] suffice for the
needs of vncwatch?
BR,
Gábor
Re: Gábor Németh
> and my question if this behavior can be changed on d.o side by doing
> shallow / blobless `git clone`? Either of these only produces a 200MB
> checkout for the package I'm working with.
Hi Gábor,
we are already doing that to some extent:
trycmd ('git', 'clone', '--quiet', '--
and that issue should be resolved (asap).
> Specifically on the kernel, vcswatch is of limited use anyway since
> there are several packages using the repo, but only one of them
> ("linux") contains a debian/changelog where the package name matches.
> On "linux", I had the
nning out of disk space several times in
> the past years, and blacklisting the biggest repos helped a lot.
> (There are some that are even > 4 GB.)
Couldn't vcswatch move off quantz to a separate VM on Debian infra, or
to a VM on AWS? Or maybe separate the control part (on quantz) an
t repos helped a lot.
(There are some that are even > 4 GB.)
Specifically on the kernel, vcswatch is of limited use anyway since
there are several packages using the repo, but only one of them
("linux") contains a debian/changelog where the package name matches.
On "linux",
retitle 1032623 vcswatch: should not raise error on repos > 1GiB in size
thanks
My understanding is that the restriction on 1Gb in size for repositories was
recently introduced and is intentional (due to disk constraints on the host
where vcswatch runs?).
Processing commands for cont...@bugs.debian.org:
> retitle 1032623 vcswatch: should not raise error on repos > 1GiB in size
Bug #1032623 [qa.debian.org] lintian-brush: vcswatch should not raise error on
repos > 1GiB in size
Changed Bug title to 'vcswatch: should not raise error o
Processing commands for cont...@bugs.debian.org:
> reassign 1032623 qa.debian.org
Bug #1032623 [lintian-brush] lintian-brush: vcswatch should not raise error on
repos > 1GiB in size
Bug reassigned from package 'lintian-brush' to 'qa.debian.org'.
No longer marked as
re is a high-priority action
needed item because vcswatch "Failed to analyze the VCS repository".
The reason: "Repository size 1135554560 exceeds 1 GiB, blocking it"
So I queried `udd` to see what it had to tell:
udd=> select vcs, url, status, error from vcswatch
Hi,
On https://tracker.debian.org/pkg/linux I see a high-priority 'action needed'
item triggered by vcswatch as it failed to analyse the VCS repository.
But the kernel repo just happens to be big, so I highly doubt that anything
could be done about it.
I'm guessing that the 1G
Package: qa.debian.org
Severity: important
Dear Maintainer,
the vcswatch service currently fails to run:
"Error: fatal: Unable to create
'/srv/scratch/qa.debian.org/vcswatch/c/cm-super/shallow.lock': No
space left on device"
this hits a few of my git repositories. Please
Thanks!
It would indeed be great if we could also extract the authors into a proper
JSON structure so downstream consumers like the janitor don't have to implement
that themselves for each VCS type.
On 8 August 2022 17:14:44 CEST, Christoph Berg wrote:
>Re: To Jelmer Vernooij
>> Does that work
Re: To Jelmer Vernooij
> Does that work? I guess we could try extracting the authors
> (committers?) into a proper json structure if that helps.
>
> Helmut was approaching me about extracting even more fields from git,
> Maintainer, Uploaders, Homepage, updated Vcs info, debian/watch, and
> expose
Your message dated Fri, 29 Jul 2022 14:49:17 +0200
with message-id
and subject line Re: Bug#1016203: vcswatch: check debian/latest branch as well
(DEP-14)
has caused the Debian Bug report #1016203,
regarding vcswatch: check debian/latest branch as well (DEP-14)
to be marked as done.
This means
Processing control commands:
> tags -1 + patch
Bug #1016203 [qa.debian.org] vcswatch: check debian/latest branch as well
(DEP-14)
Added tag(s) patch.
--
1016203: https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=1016203
Debian Bug Tracking System
Contact ow...@bugs.debian.org with problems
Control: tags -1 + patch
A patch to fix this problem can be found at
https://salsa.debian.org/qa/qa/-/merge_requests/38
--
Gioele Barabucci
Package: qa.debian.org
Dear vcswatch maintainers,
could you please add support for repositories that use `debian/latest`
as the branch for their packaging work?
DEP-14 states:
> In Debian this means that uploads to unstable and experimental should
> be prepared either in the `debian/
Re: Jelmer Vernooij
> It would be great if vcswatch could track who the committers were of the
> commits have not yet made it into the archive.
Hi Jelmer,
sorry for the latish reply.
We are already extracting the not-yet-uploaded part of the git log
which is part of https://qa.debian.or
Package: qa.debian.org
Severity: wishlist
X-Debbugs-Cc: m...@debian.org
It would be great if vcswatch could track who the committers were of the
commits have not yet made it into the archive.
I'm curious about this as a way to track the number of janitor commits that
have not yet been upl
Your message dated Fri, 18 Jun 2021 09:41:36 +
with message-id <60cc6a507bd8f_4dd72ac6b0ac65c41...@godard.mail>
and subject line Bug#989531 fixed in qa.debian.org
has caused the Debian Bug report #989531,
regarding vcswatch: Error: fatal: dumb http transport does not support s
Your message dated Sat, 12 Jun 2021 16:17:13 +0100
with message-id
and subject line Implemented
has caused the Debian Bug report #977174,
regarding vcswatch: web hook for changed repositories
to be marked as done.
This means that you claim that the problem has been dealt with.
If this is not the
Package: qa.debian.org
Severity: normal
X-Debbugs-Cc: t...@mirbsd.de
https://qa.debian.org/cgi-bin/vcswatch?package=campania and others have:
Error: fatal: dumb http transport does not support shallow capabilities
changes to the packaging repository URLs.
> Ah ok.
>
> > For salsa repositories specifically, it may be possible to improve the
> > reponsiveness of vcswatch by installing a server-side hook that
> > notifies vcswatch whenever a salsa repository has changed.
>
> That&#x
ally, it may be possible to improve the
> reponsiveness of vcswatch by installing a server-side hook that
> notifies vcswatch whenever a salsa repository has changed.
That's actually already in place. We can proxy that to you I think,
and/or write the info to some place. See
https://sa
On Thu, Dec 17, 2020 at 05:40:31PM +0100, Christoph Berg wrote:
> Re: Jelmer Vernooij
> > It would be great if vcswatch could call a webhook whenever a packaging
> > repository changed.
> currently repository information is extracted from the Sources files
> in unstable
Re: Jelmer Vernooij
> It would be great if vcswatch could call a webhook whenever a packaging
> repository changed.
Hi,
currently repository information is extracted from the Sources files
in unstable and experimental, i.e. vcswatch doesn't follow changes in
Vcs-* inside the reposit
Package: qa.debian.org
Severity: wishlist
It would be great if vcswatch could call a webhook whenever a packaging
repository changed.
This could be triggered whenever repositories are found to be different
at poll time, but e.g. for salsa this could be triggered by a server side hook
(https
Processing control commands:
> reassign -1 python-asteval
Bug #959165 [qa.debian.org] udd vcswatch: 429 "Retry later" or "could not read
Username" from Salsa
Bug reassigned from package 'qa.debian.org' to 'python-asteval'.
Ignoring request to alte
Control: reassign -1 python-asteval
Re: Rebecca N. Palmer 2020-04-30
> spyder, python-asteval and beignet don't have a cgit in Vcs-Git. Hence,
> while I agree it's a bug that apertium-bel does, this alone doesn't explain
> either of the above messages.
Salsa was down for two days, you can speed
Processing commands for cont...@bugs.debian.org:
> clone 959165 -1
Bug #959165 [qa.debian.org] udd vcswatch: 429 "Retry later" or "could not read
Username" from Salsa
Bug 959165 cloned as bug 959167
> reassign -1 apertium-bel
Bug #959167 [qa.debian.org] udd vcswatch: 4
Processing control commands:
> clone -1
Unknown command or malformed arguments to command.
> reassign -1 qa.debian.org
Bug #959165 [apertium-bel] udd vcswatch: 429 "Retry later" or "could not read
Username" from Salsa
Bug reassigned from package 'apertium-be
Control: reassign -1 apertium-bel
Re: Rebecca N. Palmer 2020-04-30
> Package: qa.debian.org
> User: qa.debian@packages.debian.org
> Usertags: udd
>
> Some packages' vcswatch pages have the error
>
> fatal: could not read Username for 'https://salsa.debian.or
Processing control commands:
> reassign -1 apertium-bel
Bug #959165 [qa.debian.org] udd vcswatch: 429 "Retry later" or "could not read
Username" from Salsa
Bug reassigned from package 'qa.debian.org' to 'apertium-bel'.
Ignoring request to alter found
Package: qa.debian.org
User: qa.debian@packages.debian.org
Usertags: udd
Some packages' vcswatch pages have the error
fatal: could not read Username for 'https://salsa.debian.org': No such
device or address
(e.g. apertium-bel, python-asteval) or
remote: Retry later f
Package: qa.debian.org
Severity: wishlist
For tools that rely on vcswatch to know what vcs repository to process,
it would be great if vcswatch could store:
* the latest revision (e.g. commit SHA for Git, revision number for
SVN, etc)
* When that revision was created
-- System Information
the available result.
Regards,
Patrice
On Fri, 9 Aug 2019 11:26:36 +0300 =?UTF-8?B?T3R0byBLZWvDpGzDpGluZW4=?= <
o...@debian.org> wrote:
> Package: qa.debian.org
> Severity: wishlist
>
> The page https://qa.debian.org/cgi-bin/vcswatch currently lists stats
> about what V
Package: qa.debian.org
Severity: wishlist
The page https://qa.debian.org/cgi-bin/vcswatch currently lists stats
about what VCS packages are using.
Could you please extend this check to also include a graph on how many
packages have 'salsa.debian.org' in their VCS field?
This should
Processing control commands:
> reassign -1 gcc-8
Bug #912405 [qa.debian.org] vcswatch cannot analyze repositories maintained on
a branch
Bug reassigned from package 'qa.debian.org' to 'gcc-8'.
Ignoring request to alter found versions of bug #912405 to the same values
Control: reassign -1 gcc-8
Control: retitle -1 Use Vcs-Git: -b gcc-8-debian
Re: Matthias Klose 2018-10-31 <9d472e23-3a14-a992-7467-0b9b7c32b...@ubuntu.com>
> Package: qa.debian.org
>
> as seen on
> https://qa.debian.org/cgi-bin/vcswatch?package=gcc-8
> vcswatch cannot
Package: qa.debian.org
as seen on
https://qa.debian.org/cgi-bin/vcswatch?package=gcc-8
vcswatch cannot analyze repositories maintained on a branch
Your message dated Thu, 13 Sep 2018 09:50:03 +0200
with message-id <20180913075003.ga24...@home.ouaza.com>
and subject line Re: Bug#906663: vcswatch: should not flag an UNRELEASED
changelog entry as "ready for upload"
has caused the Debian Bug report #906663,
regarding vcswatch:
Package: qa.debian.org
Severity: normal
Hi!
I'm afraid that the tool that displays changelogs from git breaks non-ASCII
characters. It's not a matter of web server configuration as http headers
are ok:
Content-Type: text/html; charset=utf-8
Seen on https://qa.debian.org/cgi-bin/vcswat
x-y+1~unrel after uploading x-y?
that's indeed nicer than a .0 version.
> 2) only generate this changelog entry with the first change, i.e.
> leave git unmodified until the first change happens.
That would mean that all people committing would have to remember that.
I don't thin
emind that there's pending
changes. I think this is the major issue here. It's a maintainer
decision if pending changes warrant an upload, not an automated
system's one.
I don't think we can easily tell this case from various other edge
cases apart in vcswatch. If there's a new changelog entry, or commits
since the last tag, it needs to raise "NEW" or "UNREL".
Christoph
On Sun, Aug 19, 2018 at 08:48:02PM +0200, Christoph Berg wrote:
Re: Marc Haber 2018-08-19
<153468174721.19286.15758469883614485339.report...@fan.zugschlus.de>
> > I could work in a branch, but that would mean additional rules for team
> > maintained packages which I'd rather avoid. I mean, there
Processing control commands:
> reassign -1 tracker.debian.org
Bug #906663 [qa.debian.org] vcswatch: should not flag an UNRELEASED changelog
entry as "ready for upload"
Bug reassigned from package 'qa.debian.org' to 'tracker.debian.org'.
Ignoring request to alter
Control: reassign -1 tracker.debian.org
Re: Marc Haber 2018-08-19
<153468174721.19286.15758469883614485339.report...@fan.zugschlus.de>
> vcswatch makes this "VCS has unreleased changes", which is correct. The
> tracker, however, makes this "A new version is available
number.
Please also note the UNRELEASED distribution and the NOT YET RELEASED
changelog entry which will stay the top changelog entry until the
package is eventually uploaded.
vcswatch makes this "VCS has unreleased changes", which is correct. The
tracker, however, makes this "A n
Your message dated Sat, 09 Jun 2018 15:54:14 +
with message-id <5b1bf826897d_3402ab695c990d471...@godard.mail>
and subject line Bug #897625 in qa.debian.org fixed
has caused the Debian Bug report #897625,
regarding vcswatch: shallow clone error with dump http git transport
to be marked a
Your message dated Sun, 06 May 2018 19:01:28 +
with message-id <5aef5108326ed_38c2ab250f970cc23...@godard.mail>
and subject line Bug #897624 in qa.debian.org fixed
has caused the Debian Bug report #897624,
regarding vcswatch: no indication of where/how bugreports are to be sent
to be mar
Package: qa.debian.org
User: qa.debian@packages.debian.org
Usertags: vcswatch
Individual vcswatch pages as linked from DDPO do not have any indication
of where to report bugs to, e.g.
https://qa.debian.org/cgi-bin/vcswatch?package=musescore-general-soundfont
only shows “To report a problem
Package: qa.debian.org
User: qa.debian@packages.debian.org
Usertags: vcswatch
https://qa.debian.org/cgi-bin/vcswatch?package=musescore-general-soundfont
lists this error:
Error: Cloning into bare repository
'/srv/scratch/qa.debian.org/vcswatch/m/musescore-general-soundfont'...
f
On Mon, Apr 30, 2018 at 10:18 PM, Thorsten Glaser wrote:
> a bug report against what?
Package: qa.debian.org
User: qa.debian@packages.debian.org
Usertags: vcswatch
> There’s no indication where bugreports should go, except this list.
Please file a second bug report about that :)
-
Hi Paul,
> > fatal: dumb http transport does not support shallow capabilities
> >
> > Just unshallow then ☺
>
> Please file a bug about this so it isn't forgotten.
a bug report against what? Surely not against myon… the page says:
“To report a problem with the QA web site, e-mail
debian-qa@list
On Sat, Apr 21, 2018 at 5:57 AM, Thorsten Glaser
> fatal: dumb http transport does not support shallow capabilities
>
> Just unshallow then ☺
Please file a bug about this so it isn't forgotten.
--
bye,
pabs
https://wiki.debian.org/PaulWise
https://qa.debian.org/cgi-bin/vcswatch?package=musescore-general-soundfont
Error: Cloning into bare repository
'/srv/scratch/qa.debian.org/vcswatch/m/musescore-general-soundfont'...
fatal: dumb http transport does not support shallow capabilities
Just unshallow then ☺
Your message dated Fri, 19 Jan 2018 17:12:13 +0100
with message-id <20180119161213.ga1...@msg.df7cb.de>
and subject line Re: Bug#784907: vcswatch: Fails to parse Git HEAD if multiple
branches point at HEAD
has caused the Debian Bug report #784907,
regarding vcswatch: Fails to parse Git H
Your message dated Thu, 4 Jan 2018 11:54:23 -0500
with message-id <20180104165423.y6r7orikuutd5...@zeta.ryanak.ca>
and subject line Re: Bug#886334: vcswatch: use the debian-branch specified in
gbp.conf for Git
has caused the Debian Bug report #886334,
regarding vcswatch: use the debian-
Processing control commands:
> tag -1 moreinfo
Bug #886334 [qa.debian.org] vcswatch: use the debian-branch specified in
gbp.conf for Git
Added tag(s) moreinfo.
--
886334: https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=886334
Debian Bug Tracking System
Contact ow...@bugs.debian.org w
erex sources specifies that the debian-branch is
> vprerex, and this is the branch vcswatch should be checking.
This would require unpacking the source package, something that clearly
is very expensive to do for all packages in the archive, whereas Vcs-*
fields are available in the package index.
Package: qa.debian.org
Severity: normal
For git repositories, vcswatch should check the configuration file
debian/gbp.conf (if it exists) in the source package to determine what branch to
use as its packaging branch. For example, vcswatch reports
Error: changelog package prerex does not
Hey,
currenty vcswatch is unhappy about the status for owncloud-client [0].
The issue is, that I pushed the current version to the experimental branch and
vcswatch is not getting this. But I have made commen tags (debian/)
for the versions in experimental, so vcswatch could detect them and be
Your message dated Sun, 3 Dec 2017 15:12:03 +0100
with message-id <20171203141203.ga8...@home.ouaza.com>
and subject line Re: tracker.debian.org: add support for the vcswatch service
has caused the Debian Bug report #773294,
regarding tracker.debian.org: add support for the vcswatch service
on=None/False for no compression)
> - or the user relies on compression="auto" (default value) in which
> case it should just rely on the filename extension that we should pass
> along in some way.
>
> In the case of something retrieved through the web, it might be
On Wed, 22 Nov 2017, Pierre-Elliott Bécue wrote:
> > Adding tests afterwards is hard and you are likely to miss some important
> > tests. And it's just not fun to code tests separately when you already
> > have something working... it will be hard to stick to it in the long run.
>
> I agree on the
On Wed, 2017-11-22 at 15:44 +0100, Pierre-Elliott Bécue wrote:
> Would you like a QA link for each package?
Yes, it acts as promotion for the service so more folks know about it.
--
bye,
pabs
https://wiki.debian.org/PaulWise
signature.asc
Description: This is a digitally signed message part
On Wed, 22 Nov 2017, Pierre-Elliott Bécue wrote:
> > - either the user is explicit (compression="gzip" attribute, or
> > compression=None/False for no compression)
> > - or the user relies on compression="auto" (default value) in which
> > case it should just rely on the filename extension that
iding no argument for compression.
Which one would you prefer?
> Also it seems to me that the correct API for generic decompression should
> rely on getting a file object as input and providing a file object as
> output... as that's what you want to be able to process really big files
&
ys.fr/pkg/python-aiosmtpd
>
> This seems to be missing the QA link, but the patch you provided
> doesn't seem to conditionalise on the status and the package is
> definitely present within the vcswatch data.
I only show QA link when status is not OK. So no QA there. Would you like
n some way.
In the case of something retrieved through the web, it might be
interesting to check the Content-Type to infer the compression too:
$ HEAD https://qa.debian.org/data/vcswatch/vcswatch.json.gz |grep Content-Type
Content-Type: application/x-gzip
X-Content-Type-Options: nosniff
(But it m
1 - 100 of 145 matches
Mail list logo
