Hi,
as was reported on bugs/non-random_nick_in_Pidgin,
it looks like the XXX_NICK_XXX bug was not completely fixed in
testing/devel.
One of us had an ISO that reproduced the bug on every boot,
both in a VM and on bare metal.
I've myself seen it happen at least once a few days ago.
OTOH I could not reproduce it today with various builds from "testing"
and other branches, either in a VM or on bare metal.
So I'm not sure how critical / release blocker I consider that bug.
I've pushed a commit (674164f, cherry-picked into testing) that should
improve the general robustness of the nickname generation script by
using quoting more consistently; it may fix the issue at hand, as well
as it may not. If it does not, I'm almost clueless.
The only thing (I can think of) that could explain why the nick
replacement would not happen is that the script is run in a "set -e"
environment (which is actually the case), and generate_nick() somehow
returns a non-zero exit code. I've re-read the 201-pidgin script and
I fail to see what could possibly cause this, apart of
/usr/local/bin/lc.py itself returning a non-zero exit code.
So, anyone with shell script and/or Python skills, please review one
or both of these scripts (in testing/devel) with fresh eyes!
Cheers,
--
intrigeri <intrigeri@???>
| GnuPG key @ https://gaffer.ptitcanardnoir.org/intrigeri/intrigeri.asc
| OTR fingerprint @ https://gaffer.ptitcanardnoir.org/intrigeri/otr.asc
| Then we'll come from the shadows.
