Skip to main content

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index] [List Home]
[eclipse-dev] Planning Meeting Notes - May 27, 2009

  As we discussed last week, I won't be sending out further status
  messages until after R3.5 ships since, at this point, they all
  start to look like "triage, testing, doc".

  The weekly calls are still happening, and if anything interesting
  is decided, we will send out a note afterwards. Similarly, if you
  have interesting status to report, feel free to send it directly
  to the list.

 Discussion Topics

- Carolyn MacLeod believes we should be have 2 more test days prior
  to M7 in future Eclipse End Game plans"
  - 'I feel that 1 test day is barely enough to scratch the surface
     of the testing I want to get done. I think that M7 is late
     enough in the cycle to be pretty stable, yet early enough to
     catch big problems before they become "last-minute big problems".'
  - What does everyone else think?

- Testing UI:
  - Do we plan to extend our testing environment and add any UI
    testing tools? We can create such tests, however they can be
    run only locally. If we don't plan to use any existing framework,
    do we plan to create our own tool plug-in to help creating such
    tests and avoid duplicating code among components?


Performance (for RC2):
- 3.5 perf tests:
  - need to generate them again as not all results were available
  - suspicious results on windows box baseline runs
 - verification for build
 I20090522-1710 (aka RC2):
    - no new bug:
    - blocking root bug 270824:
      - 20 resolved (verified/wontfix)
      - 2 unresolved:
        - bug 270797 (Platform/UI): any progress?
        - bug 276319 (Platform/Res): open but still RC2!?

JDT APT status:
- newsgroups, conf calls, inbox, Bugzilla
- copyright fixes
- Bug 270754 concerns processing of files whose superclasses are
 annotated with inherited annotations; to fix it we may need to parse
 additional types.  Bug 262791 concerns performance problems in large
 projects, caused by parsing too many types.  Doh.  Addressing these two
 bugs will take some clever thinking.  Per discussion in today's JDT
 call, this will be a priority for 3.5.1

Platform Workspace:
- final fixes for RC2
- testing Platform Workspace components using RC2 candidates
- reviewing/fixing documentation
- bug 276319. The fix is ready. The regression is hardly noticeable for
  regular users:
  - see
  - That's why I would like to release it to 3.5.1 to avoid any
    unexpected complication at the end of 3.5

Back to the top