[Pkg-privacy-maintainers] New torbrowser-launcher upstream version is out Was: Bug#803171: this only affects one profile in complaint mode

Holger Levsen holger at layer-acht.org
Mon Jul 11 13:18:00 UTC 2016


Hi u,

On Sun, Jul 10, 2016 at 08:59:00PM +0000, u wrote:
> Packaging is ready and tested! I'll push a bit later.

I'm a bit surprised about not using pristine-tar anymore. Did we agree
on that on the list?

Then, as a minor nitpick, I wouldn't have done f8b9b90a1 as a revert,
but rather as a simple file removal and normal commit. And then, I
would have mentioned this in debian/changelog: "removed
$description/$path patch as it has been merged upstream". (And that's
actually a less minor comment…)

> > Just be warned:
> > https://jenkins.debian.net/view/torbrowser/job/torbrowser-launcher_test_on_unstable_amd64_from_git_branch_upstream_master/317/console
> > looks like it is broken:
> > 
> > Traceback (most recent call last):
> >   File "/usr/lib/python2.7/dist-packages/torbrowser_launcher/settings.py", line 219, in install
> >     self.save()
> >   File "/usr/lib/python2.7/dist-packages/torbrowser_launcher/settings.py", line 240, in save
> >     self.common.settings['force_en-US'] = self.force_en_checkbox.get_active()
> > AttributeError: Settings instance has no attribute 'force_en_checkbox'
> > 
> > That build is based on f79d4c42fc3fbf492ba9b602195bf7ca7c8909d9
> > which is Micah's latest commit, verifyable as signed tag v0.2.5,
> > in which the current debian/sid branch has been merged into and
> > build on current sid.
> 
> Thanks for pointing this out.
> 
> I do see the correct "en" checkbox in the package which i just built.
> However, I found another little bug
> https://github.com/micahflee/torbrowser-launcher/issues/240 but i don't
> think it's a blocker for a new upstream release, as fixing the apparmor
> issue seems much more important to me than this.

I'm not sure I want to upload with this bug present, as this will break
torbrowser for people for whom it's not broken at the moment (with a
quite standard usecase).

I also think the tests on https://jenkins.debian.net/view/torbrowser/
mimic "normal (yet simple) behaviour" and as such, shouldn't be broken.

Finally fixing apparmor is really nice, but not like this I'd say. Those
who are using apparmor have been able to workaround the bugs for the
last half year, I'm sure they can continue to do that for some little
more time.

Knowingly breaking new stuff is bad.


-- 
cheers,
	Holger
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 811 bytes
Desc: Digital signature
URL: <http://lists.alioth.debian.org/pipermail/pkg-privacy-maintainers/attachments/20160711/9c0b267c/attachment.sig>


More information about the Pkg-privacy-maintainers mailing list