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

Re: resetting forwarded addresses (Re: Processed: [bts-link] source package gcc-3.4)



Le Sam 6 Mai 2006 02:18, Matthias Klose a écrit :
> Please STOP resetting the forwarded address 

I do not reset, I follow duplicates, which is a most wanted feature

> and revert the changes you already did.

I never do such a thing.

sometimes it looks like so, but it's because atm, I generate commands 
for each single bugs on the BTS, and do not take merged bugs into 
account, meaning that when I run into a merged bug again, my commands 
have already been applied to it.

I will deal with that in a near future, but except for the useless 
control@ load, I do not see a lot of harm here either.

> We do loose information in the upstream BTS: it 
> becomes more tedious to track the reverse direction (upstream ->
> Debian), unless you add information to the upstream bug report as
> well. I.e. you have to search now _every_ duplicate report to find
> the one which was reported in Debian. That's insane when a report has
> a large number of duplicates.

Reusing your tender words: it feels more insane to follow a long chain 
of dupes on the remote bugtracker to see if anything has changed. 
Meaning that a user that reported a GCC bug had a lot of work to do to 
watch what was going on with his bug. It's now one click away.

I don't get why do you need the bugzilla => debian bug map ? before 
bts-link, I can see/undertand why. But now, to track changes, you just 
have to let bts-link do its job.

If you need it, for bts-link I have a map of debian-bugs/forwards, I can 
put it online, and the remote bug to debian #nnn is just a matter of 
grep. If the gcc bugtracker also has a mail interface (meaning that 
mailing nnnnn@bugs.gcc.gnu.org or alike works), then adding the 
informations you ask can be easily done, and since it's a good thing to 
propagate that information along the duplicates chains, not only GCC 
bugzilla would benefit from that. I would be more than happy to 
implement it.

bts-link is a project that aim to remove load from the developers, and I 
will do my best to make bts-link match and help in the ways of work of 
any developper.

> I don't mind adding information to the BTS, but removing information
> from it is just insane.

again I do not do such a thing.


Side Note:

    If I minded answering you, it's because I think debian has a lot to
  win with bts-link, and that I have to make it the best possible,
  whatever it costs. But given that I really try to be in adequation to
  what fellow developpers want here, open to the discussion, and to any
  change of behaviour of bts-link[1], it really cost me to answer to you
  nicely. I hope we can have more... friendly discussions from now on,
  on how to make bts-link better suits your needs.

There is no problems, only solutions.

best regards,

 [1] http://lists.debian.org/debian-devel/2006/05/msg00078.html
     and the rest of the thread.

-- 
·O·  Pierre Habouzit
··O                                                madcoder@debian.org
OOO                                                http://www.madism.org

Attachment: pgplOQX_9DdBS.pgp
Description: PGP signature


Reply to: