On Tue, Jan 24, 2017 at 05:51:22AM +0100, Andreas Beckmann wrote:
> On 2017-01-22 22:46, Michael Stapelberg wrote:
> > Thanks for the good tips! Answers inline:
> Acked-By: Andreas Beckmann
ah :) (see below)
> > If the patch works for you, could you merge it and apply it on
>
Thanks! Please let me know once this is in place.
On Tue, Jan 24, 2017 at 5:51 AM, Andreas Beckmann wrote:
> On 2017-01-22 22:46, Michael Stapelberg wrote:
>> Thanks for the good tips! Answers inline:
>
> Acked-By: Andreas Beckmann
>
>> If the patch works for
On 2017-01-22 22:46, Michael Stapelberg wrote:
> Thanks for the good tips! Answers inline:
Acked-By: Andreas Beckmann
> If the patch works for you, could you merge it and apply it on
> piuparts.d.o please? Thank you!
Holger, we should merge this for 0.76 (not 0.75) and
Hi,
thanks for your work on this. It looks pretty good already.
On Sun, Jan 22, 2017 at 03:41:22PM +0100, Andreas Beckmann wrote:
> I don't think exporting more than one file is feasible with a
> master-slave setup.
while Michael has shown it's feasible, I wonder if this doesn't cause
too much
https://manpages.debian.org covers oldstable, oldstable-backports,
stable, stable-backports, testing, unstable and experimental (main and
contrib each). If all of the above were possible, that’d be awesome.
Otherwise, we’ll take what we can get :).
On Sun, Jan 22, 2017 at 11:02 PM, Andreas
On 2017-01-22 22:46, Michael Stapelberg wrote:
> If the patch works for you, could you merge it and apply it on
> piuparts.d.o please? Thank you!
which distro(s) would you be interested in being tested with these scrpts?
Andreas
Thanks for the good tips! Answers inline:
On Sun, Jan 22, 2017 at 8:59 PM, Andreas Beckmann wrote:
> On 2017-01-22 19:02, Michael Stapelberg wrote:
>> After going a bit further down the road, I realized that the
>> before/after tarballs of /var/lib/dpkg/alternatives are not as
On 2017-01-22 19:02, Michael Stapelberg wrote:
> After going a bit further down the road, I realized that the
> before/after tarballs of /var/lib/dpkg/alternatives are not as useful
> as I had assumed they would be: when installing non-trivial packages
> (with a bunch of dependencies), they
After going a bit further down the road, I realized that the
before/after tarballs of /var/lib/dpkg/alternatives are not as useful
as I had assumed they would be: when installing non-trivial packages
(with a bunch of dependencies), they contain changes caused by a
number of different packages,
On Sun, Jan 22, 2017 at 3:41 PM, Andreas Beckmann wrote:
> On 2017-01-22 14:43, Michael Stapelberg wrote:
>> Thanks for the review! Answers inline:
>
> I'm thinking about making this more generally useful.
>
> For the piuparts command you want something like
>
> piuparts ...
On 2017-01-22 14:43, Michael Stapelberg wrote:
> Thanks for the review! Answers inline:
I'm thinking about making this more generally useful.
For the piuparts command you want something like
piuparts ... --copy-file-from-chroot
I don't think exporting more than one file is feasible with a
Thanks for the review! Answers inline:
On Sun, Jan 15, 2017 at 10:51 PM, Holger Levsen wrote:
> Hi Michael,
>
> On Wed, Jan 11, 2017 at 11:34:52PM +0100, Michael Stapelberg wrote:
>> Attached you can find a first stab at implementing this feature. I
>> introduced a new
On Thu, Jan 12, 2017 at 04:58:02AM -0800, Michael Stapelberg wrote:
> Thinking some more about it, I think it would be good to export not
> only the alternatives _after_ package installation, but before _and_
> after package installation. That way, consumers can compare the
> differences which the
Hi Michael,
On Wed, Jan 11, 2017 at 11:34:52PM +0100, Michael Stapelberg wrote:
> Attached you can find a first stab at implementing this feature. I
> introduced a new protocol message to transfer base64-encoded arbitrary
> binary data.
the patch looks very nice. thanks for adding all the
Thinking some more about it, I think it would be good to export not
only the alternatives _after_ package installation, but before _and_
after package installation. That way, consumers can compare the
differences which the package caused, or look at the end result,
whichever fits their use case
On Wed, Jan 11, 2017 at 12:55 PM, Andreas Beckmann wrote:
> On 2017-01-11 12:44, Michael Stapelberg wrote:
>> Thanks for the suggestion. I considered this, but I’m worried this
>> might blow up the logfiles quite a bit for some packages.
>
> How much "logfile" data would that be?
On 2017-01-11 12:44, Michael Stapelberg wrote:
> Thanks for the suggestion. I considered this, but I’m worried this
> might blow up the logfiles quite a bit for some packages.
How much "logfile" data would that be? Installing texlive-full in sid
with --install-recommends produces 460 kb,
Thanks for the suggestion. I considered this, but I’m worried this
might blow up the logfiles quite a bit for some packages.
With regards to the suites, I’m interested in oldstable,
oldstable-backports, stable, stable-backports, testing, unstable :).
On Wed, Jan 11, 2017 at 3:33 AM, Andreas
On Wed, Jan 11, 2017 at 03:07:36AM -0800, Michael Stapelberg wrote:
> Sure, the suite can go into the path name :). One question: is it
> necessary for the piuparts status to be reflected by the path? I.e.,
> for my purposes, I don’t really care about whether the other checks
> succeeded, I’m only
Would it be sufficient if that information is just dumped into the
logfile? In that case a new custom script (maybe only active for a few
suites) might be sufficient.
Andreas
Thanks for the quick reply. Answers inline:
On Wed, Jan 11, 2017 at 2:06 AM, Holger Levsen wrote:
> Hi Michael,
>
> thanks for your bug report, this seems like a viable idea to me.
>
> On Wed, Jan 11, 2017 at 09:27:56AM +0100, Michael Stapelberg wrote:
>> …followed by
Hi Michael,
thanks for your bug report, this seems like a viable idea to me.
On Wed, Jan 11, 2017 at 09:27:56AM +0100, Michael Stapelberg wrote:
> …followed by making /tmp/export.tar.bz2 available as
> ___alternatives.tar.bz2, e.g.
> stretch_vim_2:8.0.0134-1_alternatives.tar.bz2.
except that
Source: piuparts
Severity: wishlist
Thanks for your work on piuparts!
I’m working on manpages.debian.org and recently stumbled across an
interesting problem: many packages use slave alternatives for manpages
(think “vi.1” being linked either to “vim.1” or “nvi.1”).
Unfortunately, alternatives
23 matches
Mail list logo