PQM Work This Summer

[John Arbash Meinel]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Daniel Mark Watkins wrote:
| Hi all,
|
| As some of you may know, I'm going to be doing some work on PQM over
| the summer, as part of what I've been describing to people as the
| Canonical Summer of Code (because it's a convenient shorthand).
|
| Anyhow, this email is to ask for suggestions as to what people want me
| to work on this summer.  Already on my list are XMLRPC submissions and
| ripping out the crufty VCS stuff.  We did brainstorm this at the London
| sprint[0], but I'm looking for what amongst those ideas people consider
| most important.
|
| On a largely unrelated note, I'd also like to apologise for my complete
| absence over the past couple of months (or so).  I've had exams which
| have completely monopolised my time (with no good reason, as my results
| weren't particularly helped by it).  I should resume my involvement in
| the community from now on (as term finished yesterday), and will
| attempt to get the patches which I've been sitting on for about 4
| versions polished up and submitted.
|
|
| Cheers,
|
| Dan

My personal order of precedence

1) revision_id based, so that when I submit a branch, I can keep working
on follow up patches, without worrying if PQM will try to merge the
wrong thing.

2) XMLRPC requests, so you know right away if PQM has your submission,
rather than switching over and refreshing a page over and over.

3) Attributing work, such as by setting the --author value. Though we
have to decide who actually gets the credit (I'm submitting someone
else's work, etc.)

4) Unicode commit messages... technically we already have them, they
just look like crap going through email. Basically, PQM treats
everything as raw UTF-8, ignoring how things are actually encoded for
email. But pqm-submit does it just fine (I've merged a lot of stuff for
Lukáš Lalinský, even if I can't actually *type* his name.)

5) Queue manipulation would be nice, but not strictly necessary. Mostly
so that when I'm a release manager, I can force a precedence for getting
my release patches merged, rather than bzr.dev patches.

6) This should probably be higher, but documentation on how to set up
PQM, and a nice "apt-get install bzr-pqm" would probably be a good
thing. (Note there is already a collision with 'bzr-pqm' being the
plugin, and I don't know if 'pqm' is overly generic.) I think people
would *like* support for it with other VCSes, but that doesn't have to
be your problem.

7) It would be nice if, as part of the submission, PQM could do a quick
check to make sure it has access to the branch. Then again 'pqm-submit'
already does this, so it is unlikely pqm would say something different,
and it does mean you don't have to wait for PQM to actually download all
of your changes.

John
=:->
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (Cygwin)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iEYEARECAAYFAkhman8ACgkQJdeBCYSNAAMDHgCggDO44WxZJLkPri48+d0v27OV
eyIAoIitySQsza+FtK+lKT7ubmnJVzde
=S0q0
-----END PGP SIGNATURE-----