gtg-gsoc team mailing list archive

[Paul Natsuo Kishimoto <mail@xxxxxxxxxxxxxxxxxxx>]

Hi all,

  I just tried the following:

$ bzr branch lp:~gtg-user/gtg/liblarch_migration
Branched 870 revision(s).
$ cd liblarch_migration/
$ bzr merge lp:~khaeru/gtg/liblarch_migration
+N  GTG/tests/test_tree.py.OTHER
 M  GTG/gtk/browser/tagtree.py
 M  GTG/tests/test_liblarch.py
 M  GTG/tools/liblarch/__init__.py
 M  GTG/tools/liblarch/filteredtree.py
 M  GTG/tools/liblarch/filters_bank.py
 M  GTG/tools/liblarch/tree.py
Text conflict in GTG/tests/test_liblarch.py
Contents conflict in GTG/tests/test_tree.py
Text conflict in GTG/tools/liblarch/__init__.py
Text conflict in GTG/tools/liblarch/filteredtree.py
4 conflicts encountered.

...basically it seems that the branch Lionel and Luca worked on during
GUADEC, and the branch I'm working on, have been diverging.

  I've sent some longer e-mails as well as an ODF spreadsheet trying to
explain the changes I was making in my branch, but haven't got any
responses. Lionel and I had a discussion on Thursday about conventions
for object attribute access, but I didn't get a sense of whether my
edits, as a whole, were welcome or not. No one else has said anything.
Combined with the fact that the ~gtg-user branch sidesteps my own, this
makes me a bit worried.

  Normally I wouldn't mention this at all, but it's the THIRD such
situation. The first two were the new date class
(https://code.launchpad.net/~gtg-contributors/gtg/new-date-class/+merge/28009) and the second set of code-layout changes (https://code.launchpad.net/~gtg-contributors/gtg/code-layout-2/+merge/30136).

  In one of those cases, I've got an "approve" from Bryce (2 weeks ago)
but no one has committed the code; in the other one we are in
disagreement and someone is needed to broker a compromise.

  But in both cases my code is "in limbo". I am left not knowing whether
I can use it as a basis for future work. In the code-layout situation,
other changes were made to trunk that partially duplicate what I did;
this represents wasted effort.

  I'm worried the same thing will happen to my liblarch improvements. I
was hoping to use those changes as a basis for replicating the
server-side trees on the client-side; but if they will be "in limbo" as
well, there's no basis to start that replication, and I am back to
square one.

  So, two requests: first, could the three branches above please get
some input? Can they be merged (for liblarch, I mean merged with
gtg-user)? Do I give up on them, rewrite them, ask for more reviews?
Will it help if I explain again why they are necessary?

  Second, how to avoid these situations in the future? Should I join
gtg-developers, like Luca, so I can merge my own changes? (Lionel
earlier said the GTG GSoC policy should be to have students propose
merges for mentors to approve, so I haven't tried to do this.) Should I
bug people even more persistently to review code? Something else? Your
guidance on this point would be very much appreciated.

Cheers,
-- 
Paul Kishimoto
SM candidate (2012), Technology & Policy Program
Massachusetts Institute of Technology

http://paul.kishimoto.name — +19053029315

Attachment: signature.asc
Description: This is a digitally signed message part