| [eclipse-dev] Planning Meeting Notes - Nov 25, 2009 |
-------------------
Discussion Topics
-------------------
Performance:
- When should I do a verification on all scenarios (i.e. non-fingerprint ones):
- before M4, at the beginning/end of each milestone, ...?
- For 3.4 and 3.5 version, I did it during M5 cycle (which usually is the
'performance' milestone...)
--------
Status
--------
Debug:
- continued review of breakpoint view enhancements (bug 238956)
- released API for sending JDWP commands and refreshing Java debug model
state (bug 290285)
- received contribution for breakpoint actions from CDT (bug 295658)
- bug fixing
PDE:
- continued target definition enhancements in branch (bug 28769)
- validate build.properties with project output folder settings:
(bug 253950)
- synchronize build.properties with project specific compiler compliance
settings (bug 154292)
- investigating API tooling performance improvements
- added support to filter reference patterns from API use reports
(bug 295481)
- bug fixing
Ant:
- bug fixing (293696, 295152)
JDT UI:
- working on adopting new org.junit bundle with version 4.x in order to
be able to run JUnit 4 tests with the Eclipse Test Framework
- improved 'Open Implementation' hyperlink/action
- started to work on passing a context when creating an ImportRewrite
- started to investigated whether we can add a link to a clean up in
the problem hover
- reviewed several patches
- fixed some minor bugs
- inbox tracking
- 1 day vacation
Platform Text and JDT Text:
- started to work on reducing the memory used by the spelling engine:
- some initial ideas ready but nothing committed yet
- created a performance test to make sure we don't hamper performance
- reviewed several patches
- fixed some minor bugs
- inbox tracking
- 1 day vacation
JDT Core:
- finalized the patch for 288211 (APT memory issue)
- bug fixes and triage
- investigating a performance regression on Linux boxes
- vacation
- continued work on the jsr 308 in the JSR_308 branch
Performance:
- 3.5 perf tests:
- verification for M20091118-0800:
- no results for Windows machine
- bug 296003 (Platform/Runtime): [perfs] New regression on 'Core/UI
Startup' test
- 3.6 perf tests:
- verification for last week builds (until N20091121-2000):
- bug 295992 (JDT/Core): [perfs] Big regression on 'Build JDT/Core
Project' test for Linux machines
- bug 295998 (JDT/UI): [perfs] All JDT/UI tests were dramatically
slowed down on Linux machines
Rel. Eng.:
- dicussions with Equinox and JDT teams on implementing JUnit 4 in the
test framework, along with more test builds
- test builds to switch 3.5.x and 3.6 builds checkout code using ssh
instead of pserver:
- https://bugs.eclipse.org/bugs/show_bug.cgi?id=294900
- discussions with Pascal and Oleg on how to improve the way that test
results are reported:
- https://bugs.eclipse.org/bugs/show_bug.cgi?id=295926
- Olivier has contributed a patch to improve test results reporting:
- https://bugs.eclipse.org/bugs/show_bug.cgi?id=296009
- discussions with foundation regarding test infrastructure at eclipse:
- https://bugs.eclipse.org/bugs/show_bug.cgi?id=294330
User Assistance:
- commited contribution for quick print enhancements
- looking at ui forms layout bugs
Workspace:
- Bug 259643 - LocalFileNatives.internalSetFileInfo only sets exec
bit for owner (fixed and testing):
- Solaris Sparc, Linux x86 and x86_64, Mac filesystem libs rebuilt
- new filesystem fragments for Solarix SPARC v9, AIX PPC and PPC64 added
- Bug 236169 - Use Synchronize View for Apply Patch in progress
- fixing and polishing new Resources API and UI (Filters, Groups etc.)
- backporting fixes to 3.5.2
- inbox tracking
- the Platform Workspace 3.6 dev plan available under:
- http://www.eclipse.org/eclipse/platform-team/team3.6/plan.php
SWT:
- new API for bidi segments contributed:
- https://bugs.eclipse.org/bugs/show_bug.cgi?id=241482
- investigating minimum width for table columns:
- https://bugs.eclipse.org/bugs/show_bug.cgi?id=293570
- working on open document API (all platforms, for associating file
types with eclipse)
- new script compiles swt's 64-bit implementations hourly to detect
released pointer size problems
- CSS styling work: making CTabFolder skinnable
- IA2 work continues
- WebKitGTK investigation continues:
-> rough implementation for most Browser functionality done,
BrowserFunction is still TODO