[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Bug#689147: unblock: gajim/0.15.1-1



On 12/28/2012 12:05 AM, intrigeri wrote:
Hi,

Yann Leboulanger wrote (27 Dec 2012 22:02:54 GMT) :
On 12/27/2012 10:48 PM, intrigeri wrote:
I'm absolutely not sure what is the best thing to do now:

   1. unblock the embedded python-gnupg copy to the "current copy of
      Wheezy's python-gnupg + small change that supposedly improves
      things": take the risk to see a regression in gajim due to changes
      brought by the library update;
Have you seen the diff?
I haven't: it's quite big, and most big changes have some potential
for regression during freeze time. I'm not saying the probability is
high, I'm just stating that a risk does exist, so that the release
team can take it into account when they make a decision.

I'm talking about gnupg diff that I attached in my previous mail. it's a 5 line diff or something like that.
But without it, Gajim can traceback, that is a fact.
I'm sorry I missed this important piece of information.

Where was I supposed to learn about it? (Not a rhetorical question,
I've genuinely searched, and failed to find it in the unblock
request -related set of messages. I guess it might be #670243 that is
related to GnuPG support, but it's unclear to me if that one was fixed
by the modifications made to the embedded pythong-gnupg copy, or by
the upgrade thereof.)

I would not have modified it if that didn't fix a bug in the library. It does not handle all messages that GPG can raise. So I added some that Gajim users already encoutered. See https://trac.gajim.org/ticket/7151
I'm sorry about that. Please note the fix to this specific bug was ACK'd by a Release Team member mid-October, so it could have been pretty smoothly fixed in Wheezy, had it not been bundled with a bunch of other changes that were less easy to decide upon, by requiring additional information or other changes from your side. I'm sorry the Release Team is overwhelmed with unblock requests, so their delays in replying to this bug report were quite long sometimes: every back'n'forth round-trip then takes time, so the best way to ensure such an unblock request is treated quickly is to only include changes that are evidently freeze-compliant, and document them very well at unblock request time, when this not done in debian/changelog yet. I hope it may help next time! :) Cheers,

I don't blame release team nor anyone else. I know they do their best.
I don't understand if you ask me to fix only bugs reported in Gajim or all known bugs in Gajim. I did the second by making this 0.15.1 package, thinking it was the better way to do. It's a minor release, so only a bugfix release. I also asked release team if that was a good idea BEFORE doing the package and got a positive reply.

I don't think there will be a next time. After this package will be accepted, I'll find another maintainer. But I finish what I started before, and that's another problem.


Reply to: