You are currently viewing a snapshot of taken on April 21, 2008. Most of this content is highly out of date (some pages haven't been updated since the project began in 1998) and exists for historical purposes only. If there are any pages on this archive site that you think should be added back to, please file a bug.

status update

maintained by Tim Rowley <>

Last Updated Friday August 11, 2000

This status update page is updated every weekend. To get updates and news throughout the week you are invited to check out mozillaZine, a site devoted to Mozilla advocacy.

Previous Updates

Friends of the Tree

  • Nikhil Bhatla and Ben Goodger for their work on the Classic theme - Henri Sivonen
  • Tomi Leppikangas for hacking on linux performance (see 26502) - Christopher Blizzard
  • Dean Tessman for many bug fixes in menus - Mike Pinkerton
Module Updates
MailNews Performance
August 5
Submitted by Pratik Dhupia <>

MailNews Performance test results for Win32, Linux 6.1 and MAC OS have been updated and can be viewed at under Quality Assurance. Tested on comm build 2000-08-02-04, M18.

August 7
Submitted by Steve Elmer <>


  • Still about 100 nsbeta3 candidates to evaluate
  • Lots of time spent on beta2 last week

Vacations & Sabbaticals!

  • JeffT 7/10-9/5
  • SSpitzer 7/5-8/26
  • CHuang 7/24-8/11
  • Bienvenu 7/31-8/11
  • RHP 8/18-8/27


Profile Manager

  • The usual birth pangs for activation


  • Resolved Bug List (15)
  • highlights -The CCK Beta2 is finally ready to be pushed out of the door.
  • lowlights - Still couldn't get infozip into the builds.
August 7
Submitted by Peter Trudelle <>


  • The XPToolkit team resolved 96  bugs in the last week, fixing 45  of  these, including  33 '+'.  For details, see our resolved bug list.


  • Mike Pinkerton (pinkerton)
    Fixed 8 bugs, including:
    • added ability to distinguish src from dest during drags (39326)
    • fixed link dragging not firing if near one of the edges of the link (38359)
    • dragging of internet shortcuts from 4.x/mozilla/IE into navigator works (37412) (help from
    • truncate long mac menu items better (29569) and refactor a lot of ugly i18n code there
    • tooltips now go away on mouseDown (39078)
    • fix up mouse location wrt translucent drag image on macos (40890)
    • applied some menu patches from
    • mac menu code now looks for "collapsed" in addition to "hidden" attributes (46001)
    • drags on win32 now properly respect modifier keys (47195)
  • Chris Saari (saari)
    • -Fixed a few nsbeta3+ bugs, trying to land 44503, 43309, 28508, 35394, 13973, 35553 fixes tonight. Yes, I have fixes for all of those, I just need to test them to death, especially joki's changes. Plus there will be a bunch of non-nsbeta3 bugs fixed by those checkins.
  • Daniel Matejka (danm)
    • 15472 Reinvoking Messenger after minimizing won't reopen Messenger
    • 17311 dimensions should be inner size not outer size
    • 31516unable to create offscreen window OR unmapped XUL window
    • z-ordering and modal features need security che
    • 43470 crash closing new account wizard with confused text field
    • 46556 Crash after change theme [@ DocumentViewerImpl::MakeWindow]
    • 46699 Crash logging into geocities [@ nsWalletlibService::OnEndDoc
    • 46979 Crash when pressing save button on wallet interview html for
  • Stuart Parmenter (pavlov)
    • Profiled x11 shm code and found that it was not providing any speedup, so I have removed it. bug 32736
    • Looked in to Mozilla supposidly leaking shared memory segments, but found it to actually be the reporters X server, and newer X servers don't have this problem. bug 42942
    • I have done numerous quantify runs on various things that seem to be really slow on Linux and I have begun investigating the data. bug 26502
  • David Hyatt  (hyatt)
    • Make disabled work for XUL widgets (30644)
    • Crasher in nsXBLBinding::InitClass (46849)
    • Column resizing in tree view (41357)
    • Infinite loops in tree reflow (46862 and 47346)
    • Sped up loading of text areas and input fields (42451)
    • Sped up Mark as Read performance (17470)
    • Stubbed out asynchronous loading mechanism for remote XBL (42320)
    • Made XBL work with the chrome cache (46129)?
  • Dan Rosen (dr)
  • Brian Ryner (bryner)
    Fixed  many bugs:
    • 46757 (crash collapsing classic skin menubar)
    • 41409 (make unix filepicker start in home directory)
    • 45808 (linux focus problem)
    • 42052 (tweak menuitem width in menulists to improve linux filepicker appearance)
    • 46545 (discontinuous tree selection on Mac)
    • 46115 (removed old tree widget)
    • 23140 (refixed to not cause menu slowdown)
    • 47536 (column hiding regression from tree column resizer landing)
    • 40256 (get rid of nonfunctional history list in linux filepicker)
    Some cleanup of old/unused files in the tree



  • pinkerton:
    • 43428 [nsbeta2-][nsbeta3+]crash, nsbeta2, nsbeta3Tree auto-scrolling crashes in
  • saari:
    •   If the above plan for checking in works (ha!) I'll have 5 nsbeta3+ bugs left. There is one on my untargeted list that should be + as well, so let's call it 6 bugs to work on. 42443 is also on mjudge's nsbeta3+ list, but will be a pain. 24525 sounds like it might be a pain. 39655 scares me as that code isn't my cup of tea and I know it is a timing related style reflow bug. Oy.?
  • hyatt:
    • Finish up 46129
    • Work on chrome registry default (46330)
  • pavlov:
    • continue performance work. bug 26502
  • danm:
    • 47491 (who noticed a 100% reproducible crash and then checked in anyway?)
  • dr:
    • 36532: Mozilla hangs on the splash screen when using UNC to launch 34301: Closing Composer closes the browser
  • bryner:
    • 42109 [nsbeta3+]crash, nsbeta3crash at shift-clicking a title in the messages list
    • 42345 [nsbeta2-][nsbeta3+]nsbeta2, nsbeta3, regression, relnote2Window position no longer remembered on exit (localstore pro


  • None

Beta Stoppers

  • None


  • Network and associated resources seem to be having an inordinate amount of downtime recently.  Every time this happens, we are dead in the water.



  • Brian Ryner's last day at the Mountain View office is Tuesday, August 8. He'll be flying back to Illinois on 8/10, and hopefully be back online by Monday 8/14. We're extending his employment part time, working remotely for the next few months.
August 7
Submitted by Gagan Saksena <>


  • High nsbeta3 count. Trying to keep them low.
  • To help us ship a more robust Cache and Imglib investigation is underway.

Dogfood and Risks

  • None

Beta Stoppers

  • Quite a phew... Hopefully today's triaging excercise would leave us with roughly 5 per head.


  • neeti
    • Fixed bug 40449
    • Worked with Pam on investigating bug 30852
    • Bug triage of cache bugs. Bugs closed 31424, 46745, 46827, 45658, 44472, 46514, 46031, 46048, 45394, 44766, 44785, 47266, 45076, 46512.
  • pnunn
    • Worked on nsbeta2+ reload bug. Proved bug existed in trunk not m17branch. The culprit was a bug in nsDocViewer introduced into trunk just after branch created.
    • Worked with Syd on getting imgcache logging info for Footprint/Performance meeting.
    • Imgcache meeting:  Opened 7 bugs for tracking. Most are assigned to me.
    • Bugs Closed: bug#47411, bug#74793, bug#46360,  bug#47158,  bug#47031,  bug#46340, bug#46535, bug#46340, bug#42788,  bug#43808
  • gordon
    • Triaged bugs for nsbeta3
    • Investigating potential cache modifications
    • Helped Wan-Teh update his build environment on Mac
  • gagan
    • Bugs resolved: Bugscape 1800.
    • Coordinated a meeting to identify, fix and address cache problems with mozilla. See tracking bug 46708.
    • Coordinated a similar meeting for imagelib.
    • Triaging bugs for the whole team.


  • ruslan
    • no status from ruslan this week.
  • gordon
    • missed a good part of Thursday due to various medical appointments (for my kids)


  • None.
August 7
Submitted by Nisheeth Ranjan <>


  • The initial pass through Heikki, Joki, and Harish's beta 3 nominated  bug lists are done.  Johnny and Nisheeth's bug lists will be attacked early this week.


  • Peter Van der Beken will join us from Belgium for two weeks starting August 14th.  Among other things, he'll hack on XSL in Mozilla, attend the Mozilla Developer Meeting, and check out the Silicon Valley.
  • Tom is on sabbatical till August 25th.


  • Harish Dhurvasula (harishd)
    • Made SCRIPTs to work on strict DTD
    • Have *a* fix to enable NOSCRIPT for strict DTD.
    • Triaged Clayton's bug list (5 bugs).
    • Triaged Rickg's bug list ( 8 bugs ).
    • Attended the seamonkey leads meeting while Nisheeth was away to the XML Query face to face.
    • Fixed two nsbeta2+ bugs that popped up in the middle of the week.
    • Worked on converting CToken cache to be arena based.
    • Participated in nsbeta3 nominated bug triage.
  • Heikki Toivonen (heikki)
    • Was able to actually fix some of joki's bugs, in addition to some of my own and resolved some others: 46117 (WORKSFORME), 45589 (WORKSFORME), 46042 (INVALID), 38399 (FIXED), 40889 (FIXED), 44211 (FIXED). Attached fixes to 38647 and 29473. Futured 10 bugs.
    • Got in contact with W3C XML Linking group, although couldn't get fully in yet because Clayton needs to confirm and he is on vacation.
    • Triaged my nsbeta3 nominated bugs as well as joki's, and then some...
    • Fixed 41599, invalidated 47477.
  • Johnny Stenback (jst)
    • Fixed nsbeta2+ bug 45019. Mozilla didn't display images in a document that was fully created with document.write();. This broke
    • Fixed nsbeta2+ bug 28572. The docshell was fireing the script unload handler too early in some cases and in some other cases the unload handler was fired even if the document wasn't unloaded.
    • Closed nsbeta2+ bug 30627 since it wasn't reproduceable any more. I have no idea who fixed this nor when it was fixed
    • Acted (together with Harish) as virtual Nisheeth while he was attending a W3C XML Query F2F meeting
    • Worked on resolving some sidebar related bugs
    • Attended the DOM face to face meeting in Seattle from July 31 to August 2.
    • Sat in on triage sessions for Harish, Joki, and Heikki's beta 3 nominated bug lists.
    • Checked in a fix for a nsbeta2+ bug on Vidur's behalf.
  • Nisheeth Ranjan (nisheeth)
    • Fixed nsbeta2+ bug 45814 -'s customization page was coming up blank.
    • Checked in fix for nsbeta2+ bug 47010 on behalf of dougt - the lock icon wasn't showing up on https:// urls.
    • Partially fixed nsbeta2+ bug 18321 which covered a bunch of session history related issues.
    • Fixed nsbeta2+ bugscape bug 1231 - the back button did not work in the activation dialog.
    • Checked in patches from Justin Smith ( that fix the windows build of Transformiix, the XSLT processor checked into Mozilla.
    • Attended the XML Query face to face meeting in Seattle on July 25th and 26th.
    • Attended the .NET meeting on Tuesday, August 1st.
    • Met with the CPD W3C reps and discussed the current status and future plans of each working group.
    • Conducted triage of Heikki, Harish and Joki's beta 3 nominated bugs.
    • Completed the code review of Peter Van der Beken's patches to the XSL hooks in Mozilla.

Plans for next week

  • Harish Dhurvasula (harishd)
    • Do some analysis on CToken bloat with the new arena cache.
    • Checkin fixes for nsbeta3+ bugs.
  • Heikki Toivonen (heikki)
    • Check in all nsbeta3+ approved bugs with fixes in them.
    • Participate in the XML Linking group telecon on August 10th.
    • Kill at least one nsbeta3+ bug without a fix (yet).
  • Johnny Stenback (jst)
    • Bug triage of beta 3 nominated bugs.
    • Start work on fixing nsbeta3+ bugs.
  • Nisheeth Ranjan (nisheeth)
    • Complete the triage of the team's beta 3 nominated bugs
    • Try and finish the triage of the team's untargeted bugs.


  • None.


  • None.
August 7
Submitted by Warrin Harris <>


  • Architecture/Embedding bug triage status:
  • Seriously concerned about the amount of work necessary to satisfy both the seamonkey and embedding project goals. We will continue the triaging effort this week.


  • too numerous to mention

Engineering Status

  • waterson
    • Spent some time looking at a crasher in RDF.
    • Decided to submarine and investigate some of the RDF performance bugs and nsbeta3+ bugs.
    • Fixed 46964, and 46616
    • Implemented 15% speedup for "mark all read"
    • Suggested further improvements that mailnews team can do to make this faster.
    • Implemented batching API for RDF observer  (These are all still awaiting code review. Will check in this week.)
    • Partial review for IBM bi-directional text changes. They have a bit of work to do to clean up these patches.
    • Examined editor typing performance with kin & jfrancis.  Trying to determine if we can remove this code which will dramatically speed up typing in large documents by  reducing reflow and repaint.
    • Attended several bug triage parties. Still hung-over.
  • vidur
    • no status
  • rayw
    • Spent most of last week at W3C DOM meetings.
    • Reviewed and edited parts of various specs and proposals and worked towards a resolution on several XPCOM or unidentified bugs.
    • Spent a little time reviewing the APIs I modified for XPCOM. Need to get the modified XPCOM APIs, together with an XPCOM work schedule for nsbeta3 and beyond posted onto XPCOM, after coordinating some of the issues a little better, e.g.
      • the discussion regarding terminating and reinitializing XPCOM is not approaching any good resolution and could be destabilizing for nsbeta3.
  • waldemar
    • Caught up after coming back from vacation.
    • Planned JS2 compiler implementation.
    • Investigated font mapping problems with Navigator 6 that are messing up the JS2 web pages.
  • rpotts
    • no status (1 more week of vacation)
  • jband
    • Fixed bugs (?)
  • warren
    • Spent a good chunk of time with bug 46777, stream API cleanup. Interestingly, after making these changes, I can eliminate an extra memory copy in RDF (bug 11232), and another when writing out network requests.
    • Worked a bit on how to determine if we're in a low-memory condition on linux (bug 44352)
    • Worked on reducing hash table memory requirements by making string keys a little smarter (bug 46711)
    • Spent time working with Edward on memory leaks.
  • dprice
    • Finished making changes to -- part of the jar file packaging process.
    • On vacation until 8/14
  • scc
    • Wrote the functions Vidur needed for his DOM conversion work, nsReadableUtils.h, and .cpp
    • Spent most of this week on vacation
  • kandrot
    • Still learning his way around our memory tools.
August 11
Submitted by Michael Kaply <>

We have moved the tinderboxes to the SeaMonkey-Ports page. Hopefully this will mean that get noticed more by people checking in code.

We are in the process of obtaining space for our nightly build - we should have them up by the end of next week.

We are also packaging an M17 build for inclusion at

August 11
Submitted by Michael Kaply <>

We have created a branch where the Bidi code lives right now - IBMBIDI_20000809_BRANCH.

We are hoping to land the Bidi code on the trunk next week, turned off by default.

Please feel free to checkout and build the code and let us know what you think.

Once we have landed the code on the trunk, we will make some binaries available.

August 11
Submitted by Dan Mosedale <>

Recent Work

  • Cleanup: fixed Mac warnings; got rid of C-style casts; replaced some libc includes and function calls with their NSPR / nsCRT equivalents.
  • Added ldap_abandon()-like functionality to the XPCOM wrapper, and changed the nsLDAPChannel code to implement request cancellation (eg pressing the stop button in the browser) using that functionality.
  • Created an LDAP PRlogModuleInfo (gLDAPLogModule) in DEBUG builds and migrated most instances of PR_fprintf(STDERR) bracked by DEBUG_dmose conditionals to use PR_LOG instead. Should release builds have this too?
  • Moved LDAP callbacks to execute on the nsLDAPConnection thread by default rather than the UI thread. Did this by changing nsLDAPConnection to execute the nsILDAPMessageListener callbacks directly rather than through an nsISupports proxy. Callers can still create a proxy manually and pass it in, if they want the callbacks to happen on some other thread. Changed the datasource and channel objects to run correctly on the nsLDAPConnection thread, in the hopes that this would help avoid UI stalls. This didn't make any perceptible improvement, but it still is probably a good thing, and may help more once event system work by blizzard and/or dougt gets checked in. Where the channel code gets called back is governed by the INVOKE_LDAP_CALLBACKS_ON_MAIN_THREAD define in nsLDAPChannel.h

In progress

  • Major fixes to the code that cleans up at the end of a search and/or when a search has been cancelled. Most visible effect: once this code lands, the throbber should always stop spinning at the end of a search, rather than sometimes continuing on.


  • Add timeouts to XPCOM wrapper as well as the channel and datasource.
  • Fixes to the nsLDAPConnection cleanup code.
  • More RDF datasource work, once the necessary JS component-loader security hacking lands.

Previous Updates