Hi!
Here’s the weekly summary of Qt related changes to WebKit trunk. Big changes include Yael’s patch for WebSockets support, the beginnings of QtScript on top of JavaScriptCore’s C API, Maemo 5 tweaks and layout test fixes:
- Janne added the necessary meta-data to make QtWebKit play nicely with Symbian backups (34077).
- I did some code cleanups in RenderThemeQt and fixed a bug with combo boxes not showing up in Maemo5 (34088).
- Holger fixed a regression in the JavaScript prompt handling (30914).
- Jedrzej landed the first files for building QtScript on top of JavaScriptCore’s C API (32565).
- Diego added history support to the Qt DRT, we now pass the
http/tests/historylayout tests! (34167) - Daniel fixed a bug with the height of button elements (29564).
- Kent fixed support for ES5 style introspection with QMetaObject methods (34087).
- Yael implemented the Qt part of WebSocket support, we now pass
websocket/testsin the layout tests (34180). - Diego fixed more worker layout tests by adding support for counting worker threads in the Qt DRT (34221).
- Holger found a neat way to speed up the conversion from KURL to QUrl (33873).
- Trond fixed an endless loop in QWebPage printing (r53997).
- Kenneth fixed incorrect fonts on comboboxes on Maemo5 and Symbian (r53999).
- Andreas upstreamed Ralf and Robert’s kinetic scrolling support for QWebView using
QAbstractKineticScroller(34267). - Benjamin implemented support for the
display()method in the Qt DRT (34258). - Oswald speed up the conversion between
WebCore::StringandQStringby avoidingQString::fromUtf16()(r54060). - Kenneth landed a patch to disable auto-uppercasing and text prediction for password input fields (r54064).
- Kenneth also continued to clean up the QtLauncher for a future merge with QGVLauncher
- Andreas and Kenneth submitted tweaks to the look’n'feel of QtLauncher on Maemo5.
Possibly related posts:
7 comments
Hi! There is an open bug in the Qt BugTracker regarding the incompatibility of rand_s() and win2000. (http://bugreports.qt.nokia.com/browse/QTBUG-6483) The report contains also the (one line!) patch to fix it. The corresponding bug in the WebKit tracker (https://bugs.webkit.org/show_bug.cgi?id=17032) is set to prio P4 Minor and is’nt assigned to anybody.
Fixing this tiny bug would allow us to use the standard Qt binaries on win2000 without having to recompile them.
Is there a chance to get this fixed by 4.6.2?
Greetings Jens.
Is it just me, or do those QtWebKit “blog” entries look awfully like some manager told you guys to “get out there and mingle with the community – for example, do a regular blog!” — yet you completely misunderstood the idea of a blog?
Folks, what you write is a frigging CHANGE LOG, not a blog. If I want to read a change log, I can get it from git. If all you have for us are one-line change log entries, by all means, stop it. All this achieves is spamming the RSS feed of everybody is interested — for no purpose whatsoever.
If you have interesting stuff, write it. “Interesting” as in “multiple lines, also known as prose text”. If you don’t, better not write anyting than something that might as well be machine-generated. The other teams working on Qt got the concept right. You don’t.
No offense.
@jstockhausen: Please try to refrain from posting your favorite bugs in comments, I know it seems like a good idea because one feels like you’re getting straight to developers but you’re not; use the task tracker to post comments/ping people working on it to that that. It’s impossible to manage bugs from blogs (especially as more and more people do that) and awfully confusing to everyone who doesn’t care about that bug.
@HuHa: It’s just you. Of course one really should point out the irony of you, an anonymous individual, accusing one of the people who were the foundation of the KDE community and the voice of the Free Software community for ages at tt of not understanding the community. Also, guy, what you describe is an article not a blog. If all you have for us are bitter long winded rants, by all means, stop it. All this achieves is spamming the comments section of an interesting blog – for no purpose whatsoever. No to mention that you seem to be confused about the meaning of the expression “no offense”; it doesn’t work if you’re deliberately offending someone.
@Simon: Loving the lists, keep them coming! QtWebKit is coming together very nicely.
Absolute agree with Zack. This weekly short-yet-fascinating summary of QtWebKit hacking activities is great!
QtWebKit is great and getting better. What I still miss is graphics “quality” (at least on Mac).
There are font errors (spacing between character wrong), rounded corners are are ugly, box-shadows are not shadows, etc.
Are there plans/projects that target these issues?
(I’m posting these here because it’s not about bugs, but about development plans)
Andrea
@zack: Sorry. I hadn’t intended to do the bug management in comments in this blog entry (I have added comment in the tracker a while ago). On the other hand new features in WebKit are nice but this special bug causes us quit a lot of additional work. So my idea was to ask if there is a chance that someone addresses this issue as there are already fixes in the comments.
Jens
@jstockhausen
I think the problem is that nobody is suggesting a patch on https://bugs.webkit.org/show_bug.cgi?id=17032
Since nobody does, webkit developers assume this bug is not important.
Since you know how to fix it, the best way to make that move is to add your patch: https://bugs.webkit.org/attachment.cgi?bugid=17032&action=enter
And put it for review (flag “?” on review and commit queue). More info on how to contribute: https://trac.webkit.org/wiki/QtWebKitContrib
@Andrea
Have you reported a test case on the bug tracker of WebKit? I personally don’t have those font problems on Mac on the websites I use regularly.
Comments on this entry are closed.