status update

maintained by Tim Rowley <tor@cs.brown.edu>

Last Updated Thursday May 9, 2002

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

Previous Updates

Module Updates
Necko/Imagelib/XPCOM
May 8
Submitted by Gagan Saksena <gagan@netscape.com>

Summary

  • Pipelining is now available on the 1.0 branch! You will see most benefits on the narrowband connection. Check it out by turning on these prefs--
    • user_pref("network.http.pipelining", true);
    • user_pref("network.http.proxy.pipelining", true);
  • Bugfixes and triaging continues.

Highlights

  • darin
    • landed HTTP/1.1 pipelining patch on the 1.0 branch (default disabled)
    • landed patch for bug 138877 "Browser crashes if HREF contains entity #9619;" on trunk and branch
    • landed patch for bug 141061 "XMLHttpRequest allows reading of local files" on trunk
    • landed patch for bug 124465 "freeze various necko APIs for mozilla 1.0"
    • worked with jag to figure out a plan for freezing strings
    • ran some performance tests on gordon's disk cache streams patch
    • reviewed patches
    • heavy triaging
  • dougt
    • Fixed a the indexed to html stream converter so that top level directories don't fail.
    • Lots of work on ftp uploading. The RFC's are terrible at defining exactly what to do in certian situations. The basic problem is determining which path to pass as the parameter to STOR. Is it pwd, cwd, the url directory path, or some logicial concatination of these? I would like to be able to avoid retries while uploading too.
    • Fixed the "Publish takes 99% CPU" bug.
    • Spend some time working up a solution for the DHCP/DNS woes that 'nix users have. I really can't believe that the dns resolver isn't reinitalized when the settings change. This should really all be under the hood. Linux users, check out 117628.
    • Researched a potential security hole 132657.
  • gordon
    • Started getting performance numbers for Data Cache Block patch (bug 81724). The patch doesn't seem to make much difference on Linux, or high performance Macs. Lower performance Macs show a 4%-8% performance improvement. I'm still working on getting performance numbers for Windows.
  • pavlov
    • The fix for bug 129953 was landed on the 1.0 branch for RC2.
    • Was finally able to close bug 117061 after much arguing with matic.
  • neeti
  • gagan

Lowlights

Plans for next week

  • darin
    • fix more bugs
    • triage bugs
  • pavlov
    • Review Paper's giant animated gif patch for bug 85595.

Issues

Previous Updates