Zero-sized fonts need to render as nothing and take no space. Hyatt got this working on other platforms but it seems that it is not yet working for Qt given that bot failures show fast/text/font-size-zero.html failing.
A consequence of this is that Qt almost certainly fails acid3 currently.
It also fails in GTK, skipping the test to check the issue.
void QFont::setPixelSize(int pixelSize) { if (pixelSize <= 0) { qWarning("QFont::setPixelSize: Pixel size <= 0 (%d)", pixelSize); return; }
*** Bug 49851 has been marked as a duplicate of this bug. ***
In addition to a Qt API change permitting zero pixel-size fonts we need to fix 49879 to allow this to pass.
See http://bugreports.qt.nokia.com/browse/QTBUG-15485
Note we also need to do this in DRT: globalSettings->setFontSize(QWebSettings::MinimumFontSize, 0); globalSettings->setFontSize(QWebSettings::MinimumLogicalFontSize, 0);
Created attachment 74595 [details] Patch
(In reply to comment #8) > Created an attachment (id=74595) [details] > Patch A Qt API change is undesirable according to the Qt Jira and gitorious comments and this can be fixed in QtWebKit anyway.
I asked Jiang about this and he said: "I'm afraid there is no reliable way in Qt now to verify if a font family exists. Can you please create a bug report in http://bugreports.qt.nokia.com/ ? The workaround has to be platform specific." See http://bugreports.qt.nokia.com/browse/QTBUG-15575 Should we fix it for now in QtWebKit using the attached patch?
(In reply to comment #10) > I asked Jiang about this and he said: > > "I'm afraid there is no reliable way in Qt now to verify if a font family exists. Can you please create a bug report in http://bugreports.qt.nokia.com/ ? The workaround has to be platform specific." > > See http://bugreports.qt.nokia.com/browse/QTBUG-15575 > > Should we fix it for now in QtWebKit using the attached patch? Whoops, wrong font bug!
Andreas, can you have a look at this?
Comment on attachment 74595 [details] Patch LGTM, thanks for fixing this Robert!
Comment on attachment 74595 [details] Patch Clearing flags on attachment: 74595 Committed r73341: <http://trac.webkit.org/changeset/73341>
All reviewed patches have been landed. Closing bug.
Test has since been updated to test for font width so it still fails in qt. Patch to fix the failure at https://bugs.webkit.org/show_bug.cgi?id=50539
(In reply to comment #16) > Test has since been updated to test for font width so it still fails in qt. > > Patch to fix the failure at https://bugs.webkit.org/show_bug.cgi?id=50539 Next time please don't leave the bot red for a day. :S
(In reply to comment #17) > (In reply to comment #16) > > Test has since been updated to test for font width so it still fails in qt. > > > > Patch to fix the failure at https://bugs.webkit.org/show_bug.cgi?id=50539 > > Next time please don't leave the bot red for a day. :S Yeah sorry, should I have just re-skipped the test while waiting for the follow-up review? I figured the shortest route to fixing was just to correct the issue with a follow-up patch.
(In reply to comment #18) > Yeah sorry, should I have just re-skipped the test while waiting for the follow-up review? I figured the shortest route to fixing was just to correct the issue with a follow-up patch. Not a big problem. But we like always green bots, because you can notice sooner a green vs. red change than a 1 vs. 2 failing tests change.