⚠ Archived content — this site is no longer maintained.   Current WebKit documentation is at docs.webkit.org.

Timeline



Oct 10, 2005:

11:55 PM Changeset in webkit [10812] by ggaren
  • 2 edits in trunk/WebCore
  • Fixed <rdar://problem/4245682> Regression: CrashTracer: 923 crashes in Safari at com.apple.WebCore: KJS::Window::clear + 132

Reviewed by mjs.

No test case because this crash isn't reliably reproducible.

  • khtml/ecma/kjs_window.cpp: (KJS::Window::Window): Initialize m_returnValueSlot to 0, so that we know not to write to it until it's explicitly set to a valid address.
8:45 PM Changeset in webkit [10811] by darin
  • 3 edits in trunk/LayoutTests

The old results were incorrectly clipped. This always worked fine in the
browser, but didn't draw correctly in DumpRenderTree until the bug was fixed.

  • fast/clip/014-expected.checksum: Updated.
  • fast/clip/014-expected.png: Updated.
7:49 PM Changeset in webkit [10810] by darin
  • 7 edits in trunk/WebCore

Reviewed by Maciej.

The first half of the fix alone broke DumpRenderTree. Better now.

  • kwq/KWQWidget.h: Made paint non-virtual (it had no reason to be virtual in KWQ) and added an isPainting function.
  • kwq/KWQWidget.mm: (QWidget::QWidget): Added painting boolean (actually a count). (QWidget::~QWidget): Check that painting is 0 when the widget is destroyed. (QWidget::paint): Use SPI to draw in a way that works no matter what context we are in. (QWidget::isPainting): Return true if inside the paint function.
  • kwq/KWQTextArea.h: Remove inDrawingMachinery field since we can use isPainting now.
  • kwq/KWQTextArea.mm: (-[KWQTextAreaTextView textView:shouldDrawInsertionPointInRect:color:turnedOn:]): Use isPainting() instead of inDrawingMachinery. inDrawingMachinery required that we use displayRectIgnoringOpacity: to paint, which we no longer do.
  • kwq/KWQTextField.h: More of the same.
  • kwq/KWQTextField.mm: (-[KWQTextFieldController textView:shouldDrawInsertionPointInRect:color:turnedOn:]): Ditto.
4:48 PM Changeset in webkit [10809] by eseidel
  • 5 edits in trunk/WebCore

Bug #: 5320
Submitted by: darin
Reviewed by: eseidel

Mem leak fixes in xslt code.
​http://bugzilla.opendarwin.org/show_bug.cgi?id=5320

  • khtml/xml/dom_docimpl.cpp: (DocumentImpl::applyXSLTransform): (DocumentImpl::setTransformSourceDocument):
  • khtml/xml/xml_tokenizer.cpp: (khtml::parseXMLDocumentFragment):
  • khtml/xsl/xslt_processorimpl.cpp: (DOM::XSLTProcessorImpl::XSLTProcessorImpl): (DOM::XSLTProcessorImpl::~XSLTProcessorImpl): (DOM::stylesheetLoadFunc): (DOM::XSLTProcessorImpl::transformDocument): (DOM::XSLTProcessorImpl::documentFromXMLDocPtr):
  • khtml/xsl/xslt_processorimpl.h: (DOM::XSLTProcessorImpl::stylesheet): (DOM::XSLTProcessorImpl::sourceDocument):
4:32 PM Changeset in webkit [12207]
  • 1 copy in branches/Safari-2-0-branch/JavaScriptGlue/Info.plist

This commit was manufactured by cvs2svn to create branch
'Safari-2-0-branch'.

4:32 PM Changeset in webkit [12206] by darin
  • 16 edits in trunk/JavaScriptGlue

Reviewed by Eric.

  • converted tabs to spaces, NULL to 0, added some (void), got rid of prefixes in headers (reserved for compiler/library)
  • Info.plist:
  • JSBase.cpp:
  • JSBase.h:
  • JSObject.cpp:
  • JSObject.h:
  • JSRun.cpp:
  • JSRun.h:
  • JSUtils.cpp:
  • JSUtils.h:
  • JSValueWrapper.cpp:
  • JSValueWrapper.h:
  • JavaScriptGlue.cpp:
  • JavaScriptGlue.h:
  • UserObjectImp.cpp:
  • UserObjectImp.h:
11:54 AM Changeset in webkit [12205] by darin
  • 2 edits
    1 add in trunk/JavaScriptGlue
  • .cvsignore: Added. Ignore Makefile.in and Makefile.
12:16 AM Changeset in webkit [10808] by hyatt
  • 1 edit in trunk/WebCore/khtml/rendering/bidi.cpp

Bug #:
Submitted by:
Reviewed by:

12:15 AM Changeset in webkit [10807] by hyatt
  • 1 edit in branches/Safari-2-0-branch/WebCore/khtml/rendering/bidi.cpp

Bug #:
Submitted by:
Reviewed by:

Oct 9, 2005:

11:35 PM Changeset in webkit [10806] by darin
  • 5 edits in trunk/WebKitTools
  • Scripts/check-dom-results: Dump ".xhtml" for tests in the xhtml directory.
  • Scripts/cvs-apply: Handle added/deleted files in the current directory.
  • Scripts/cvs-unapply: Ditto.
  • Scripts/run-webkit-tests: Add a few more false-positive leaks.
8:50 PM Changeset in webkit [10805] by hyatt
  • 2 edits in /

Land fix for bugzilla bug 4974, make sure to only move below floats in line layout if white-space is normal.
One case was using !isPre and including nowrap incorrectly as a result. Patch from koivisto.

Reviewed by hyatt

  • khtml/rendering/bidi.cpp: (khtml::RenderBlock::findNextLineBreak):
6:28 PM Changeset in webkit [10804] by darin
  • 2 edits in trunk/WebKit

Reviewed by Maciej.

This was actually a Tiger regression. When AppKit added a new code path for rendering
NSView, our special hack for doing layout when we draw didn't work any more. So we
were able to fix this without adding any API.

  • WebView.subproj/WebHTMLView.m: (-[WebHTMLView _recursiveDisplayRectIgnoringOpacity:inContext:topView:]): Added. Does the same thing that other _recursiveDisplay methods do.
4:15 PM Changeset in webkit [10803] by thatcher
  • 1 edit in trunk/WebKitTools/DumpRenderTree/TextInputController.m

Fixes the following build error when building Deployment:
DumpRenderTree/TextInputController.m: In function '-[TextInputController textInput]':
DumpRenderTree/TextInputController.m:87: warning: class 'NSView <WebDocumentView>' does not implement the 'NSTextInput' protocol

3:57 PM Changeset in webkit [10802] by darin
  • 1 edit in trunk/JavaScriptCore/ChangeLog

Updated change log to mention one more bug fixed.

3:56 PM Changeset in webkit [10801] by darin
  • 5 edits
    2 adds in trunk

JavaScriptCore:

Reviewed by Maciej; some changes done after review.

Retested all tests to be sure nothing broke; added layout test for bug 5280.

  • kjs/config.h: Removed TIME_WITH_SYS_TIME define. Also set HAVE_SYS_TIMEB_H for the APPLE case (the latter is accurate but irrelevant).
  • kjs/date_object.h: Reformatted. Removed unnecessary include of "function_object.h". Moved declarations of helper classes and functions into the cpp file.
  • kjs/date_object.cpp: Removed code at top to define macros to use CoreFoundation instead of POSIX date functions. (KJS::styleFromArgString): Tweaked to return early instead of using a variable. (KJS::formatLocaleDate): Tweaked to check for undefined rather than checking argument count. (KJS::formatDate): Made parameter const. (KJS::formatDateUTCVariant): Ditto. (KJS::formatTime): Ditto. (KJS::DateProtoFuncImp::callAsFunction): Use gmtime_r and localtime_r instead of gmtime and localtime. (KJS::DateObjectImp::callAsFunction): Use localtime_r instead of localtime. (KJS::ymdhmsToSeconds): Renamed from ymdhms_to_seconds. Changed computation to avoid possible overflow if year is an extremely large or small number. (KJS::makeTime): Removed code to move large month numbers from tm_mon to tm_year; this was to accomodate CFGregorianDate, which is no longer used (and didn't handle negative values). (KJS::parseDate): Renamed from KRFCDate_parseDate; changed to return a value in milliseconds rather than in seconds. Reformatted the code. Changed to use UTF8String() instead of ascii(), since ascii() is not thread safe. Changed some variables back from int to long to avoid trouble if the result of strtol does not fit in an int (64-bit issue only).

LayoutTests:

  • fast/js/date-negative-setmonth-expected.txt: Added.
  • fast/js/date-negative-setmonth.html: Added.
Note: See TracTimeline for information about the timeline view.