⚠ Archived content — this site is no longer maintained.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

Timeline



Jun 13, 2006:

10:54 PM Changeset in webkit [14860] by ggaren
  • 6 edits
    2 adds in trunk

LayoutTests:

Test for createHTMLDocument.

  • fast/dom/implementation-createHTMLDocument-expected.txt: Added.
  • fast/dom/implementation-createHTMLDocument.html: Added.

WebCore:

Reviewed by Maciej.


  • Fixed <rdar://problem/4562192> Creating a new DOMHTMLDocument appears challenging.


  • Added createHTMLDocument to Objc bindings, fixed up implementation to comply with the DOM 2 Candidate Recommendation in which is was defined. (It never made the final spec.) Also removed use of deprecatedString, to avoid unnecessary killing of puppies.
  • WebCore.xcodeproj/project.pbxproj:
  • bindings/objc/DOM.mm: (-[DOMImplementation createHTMLDocument:]):
  • bindings/objc/DOMPrivate.h:
  • dom/DOMImplementation.cpp: (WebCore::DOMImplementation::createDocument): (WebCore::DOMImplementation::createHTMLDocument):
10:03 PM Changeset in webkit [14859] by mjs
  • 2 edits
    1 add in trunk/LayoutTests
  • fixed up results of the layout test I committed for 4583892, they were guaranteed to fail
  • fast/events/no-window-load-expected.txt:
  • fast/events/resources/greenbox.png: Added.
10:01 PM Changeset in webkit [14858] by ggaren
  • 1 edit
    1 add in trunk/LayoutTests

RS by MacDome. Added missing results for this test.

  • fast/css/xml-stylesheet-pi-not-in-prolog-expected.txt: Added.
6:35 PM Changeset in webkit [14857] by aliceli1
  • 2 edits
    2 adds in trunk

Reviewed by Adele.

fixed <rdar://problem/4457902> HTMLSelectElement.remove() can't handle an option obj being passed to it, results in hang on ​http://www.ibc-solar.de/www_ibc/fst_solarmonitoring.jsp

  • bindings/js/kjs_html.cpp: (KJS::HTMLElementFunction::callAsFunction): support this by converting the argument to an options object before passing to remove()
6:33 PM Changeset in webkit [14856] by darin
  • 5 edits
    2 adds in trunk

LayoutTests:

Reviewed by Geoff.

  • test for <rdar://problem/4585333> Changing location for weather on yahoo.com home page redirects to another page
  • fast/events/event-listener-html-non-html-confusion-expected.txt: Added.
  • fast/events/event-listener-html-non-html-confusion.html: Added.

WebCore:

Reviewed by Geoff.

  • fix <rdar://problem/4585333> Changing location for weather on yahoo.com home page redirects to another page

This patch fixes a bug where the event listener cache does not distinguish
HTML and non-HTML listeners. Incorrect behavior where stopPropagation also
prevented default masked a case of this bug on the yahoo.com home page until
we fixed bug 5180 on 2005-10-03.

Test: fast/events/event-listener-html-non-html-confusion.html

  • bindings/js/kjs_window.h: Add additional listener maps for HTML event listeners.
  • bindings/js/kjs_window.cpp: (KJS::Window::~Window): Go through the additional maps when clearing the window object pointer in event listeners. (KJS::Window::getJSEventListener): Look in the HTML or non-HTML map depending on the argument passed. (KJS::Window::getJSUnprotectedEventListener): Ditto.
  • bindings/js/kjs_events.cpp: (KJS::JSUnprotectedEventListener::JSUnprotectedEventListener): Add to either the HTML or non-HTML map depending on the argument passed. (KJS::JSUnprotectedEventListener::~JSUnprotectedEventListener): Remove from either the HTML or non-HTML map depending on whether the HTML flag is set. (KJS::JSEventListener::JSEventListener): More of the same. (KJS::JSEventListener::~JSEventListener): Ditto. (KJS::JSLazyEventListener::parseCode): Same thing here. In a lazy event listener there is not a listener at construction time, thus the code here to put the listener into a map needs the HTML vs. non-HTML logic.
6:25 PM Changeset in webkit [14855] by mjs
  • 3 edits
    2 adds in trunk

LayoutTests:

Reviewed by Geoff.


  • test case for <rdar://problem/4583892> 10.4.7 regression: Hang occurs when attempting to load search results at mapquest.com
  • fast/events/no-window-load-expected.txt: Added.
  • fast/events/no-window-load.html: Added.

WebCore:

Reviewed by Darin.


<rdar://problem/4583892> 10.4.7 regression: Hang occurs when attempting to load search results at mapquest.com

  • dom/EventTargetNode.cpp: (WebCore::EventTargetNode::dispatchGenericEvent): Don't allow "load" events to propagate up to the window. We need this quirk to avoid site hangs, because they depend on an old Mozilla bug.
6:11 PM Changeset in webkit [14854] by thatcher
  • 1 delete in tags/WebCore-418.7/WebCore

Bad tag.

5:56 PM Changeset in webkit [14853] by thatcher
  • 1 copy in tags/WebCore-418.17/WebCore

New tag.

5:56 PM Changeset in webkit [14852] by thatcher
  • 1 delete in tags/WebCore-418.17/WebCore

Bad tag.

5:53 PM Changeset in webkit [14851] by thatcher
  • 1 copy in tags/WebCore-418.7/WebCore

New tag.

5:53 PM Changeset in webkit [14850] by thatcher
  • 1 delete in tags/WebCore-418.7/WebCore

Bad tag.

5:52 PM Changeset in webkit [14849] by thatcher
  • 1 edit in branches/Safari-2-0-branch/WebCore/ChangeLog

Versioning

5:46 PM Changeset in webkit [14848] by ggaren
  • 4 edits in branches/Safari-2-0-branch/WebCore

Patch by Darin, tweaked by me, reviewed by Harrison.

  • Fixed <rdar://problem/4584598> 10.4.7 regression: Changing location for weather on yahoo.com home page redirects to another page


We need to keep separate caches for HTML and non-HTML event listeners
so that, if you add one of each to the same node, the two don't clobber
each other.

  • khtml/ecma/kjs_events.cpp: (JSUnprotectedEventListener::JSUnprotectedEventListener): (JSUnprotectedEventListener::~JSUnprotectedEventListener): (JSEventListener::JSEventListener): (JSEventListener::~JSEventListener): (JSLazyEventListener::parseCode):
  • khtml/ecma/kjs_window.cpp: (Window::~Window): (Window::getJSEventListener): (Window::getJSUnprotectedEventListener):
  • khtml/ecma/kjs_window.h:
4:19 PM Changeset in webkit [14847] by antti
  • 7 edits
    12 adds in trunk

LayoutTests:

Reviewed by Hyatt

​http://bugzilla.opendarwin.org/show_bug.cgi?id=9314

  • fast/block/positioning/relative-overflow-block-expected.checksum: Added.
  • fast/block/positioning/relative-overflow-block-expected.png: Added.
  • fast/block/positioning/relative-overflow-block-expected.txt: Added.
  • fast/block/positioning/relative-overflow-block.html: Added.
  • fast/block/positioning/relative-overflow-replaced-expected.checksum: Added.
  • fast/block/positioning/relative-overflow-replaced-expected.png: Added.
  • fast/block/positioning/relative-overflow-replaced-expected.txt: Added.
  • fast/block/positioning/relative-overflow-replaced-float-expected.checksum: Added.
  • fast/block/positioning/relative-overflow-replaced-float-expected.png: Added.
  • fast/block/positioning/relative-overflow-replaced-float-expected.txt: Added.
  • fast/block/positioning/relative-overflow-replaced-float.html: Added.
  • fast/block/positioning/relative-overflow-replaced.html: Added.

WebCore:

Reviewed by Hyatt.

​http://bugzilla.opendarwin.org/show_bug.cgi?id=9314
Relative positioned block size doesnt update root layer size

  • take relative positioning into account in leftmost/rightmost/lowestPosition()
  • ignore zero width/height boxes in leftmost/rightmost/lowestPosition()
  • split relativePositionOffset() to x and y functions


  • rendering/RenderBox.cpp: (WebCore::RenderBox::absolutePosition): (WebCore::RenderBox::relativePositionOffsetX): (WebCore::RenderBox::relativePositionOffsetY): (WebCore::RenderBox::lowestPosition): (WebCore::RenderBox::rightmostPosition): (WebCore::RenderBox::leftmostPosition):
  • rendering/RenderBox.h:
  • rendering/RenderFlow.cpp: (WebCore::RenderFlow::lowestPosition): (WebCore::RenderFlow::rightmostPosition): (WebCore::RenderFlow::leftmostPosition):
  • rendering/RenderLayer.cpp: (WebCore::RenderLayer::updateLayerPosition):
  • rendering/RenderObject.cpp: (WebCore::RenderObject::offsetLeft): (WebCore::RenderObject::offsetTop):
3:20 PM Changeset in webkit [14846] by andersca
  • 4 edits
    2 adds in trunk

WebCore:

2006-06-13 Anders Carlsson <​acarlsson@apple.com>

Reviewed by Dave Hyatt.

​http://bugzilla.opendarwin.org/show_bug.cgi?id=9427
xml-stylesheet processing instructions outside of the prolog should have no effect.

  • dom/ProcessingInstruction.cpp: (WebCore::ProcessingInstruction::checkStyleSheet): Remove todo comment.


  • dom/xml_tokenizer.cpp: (WebCore::XMLTokenizer::processingInstruction): Only check for style sheet if the root element hasn't yet been encountered.

LayoutTests:

2006-06-13 Anders Carlsson <​acarlsson@apple.com>

Reviewed by Dave Hyatt.

​http://bugzilla.opendarwin.org/show_bug.cgi?id=9427


  • fast/css/resources/xml-stylesheet-pi-not-in-prolog.css: Added.
  • fast/css/xml-stylesheet-pi-not-in-prolog.xml: Added.
3:19 PM Changeset in webkit [14845] by sullivan
  • 2 edits in trunk/WebKit

Reviewed by Tim O.


  • fixed <rdar://problem/4498606> REGRESSION (417.8-420+): 3 missing items (but extra separators) in context menu in Mail message body
  • WebView/WebView.m: (-[WebView _menuForElement:defaultItems:]): Add special-case hackery to recover from this SPI -> API mismatch.
2:58 PM Changeset in webkit [14844] by tomernic
  • 1 edit in trunk/WebKit/ChangeLog

Clarified my previous ChangeLog entry.

2:56 PM Changeset in webkit [14843] by tomernic
  • 2 edits in trunk/WebKit

Reviewed by Anders.

Fixed a recently-introduced assertion failure when handling 404 errors.

  • WebView/WebDataSource.m: (-[WebDataSource _handleFallbackContent]): Use the -[WebFrame _bridge] instead of -[WebDataSource _bridge]. The former is not valid until the data source has been committed, as is the case when the resource fails to load. The latter is safe to call at any time. This broke last night with Maciej's change to WebFrameResourceLoader. The old code used to call -[WebFrame _bridge].
1:17 PM Changeset in webkit [14842] by thatcher
  • 1 copy in tags/WebCore-418.17/WebCore

Tag for WebCore 418.17 (part 2 of 2.)

1:17 PM Changeset in webkit [14841] by thatcher
  • 1 add in tags/WebCore-418.17

Tag for WebCore 418.17 (part 1 of 2.)

1:15 PM Changeset in webkit [14840] by thatcher
  • 2 edits in branches/Safari-2-0-branch/WebCore

Versioning

12:56 PM Changeset in webkit [14839] by thatcher
  • 2 edits in branches/Safari-2-0-branch/WebCore

Reviewed by Harrison, landed by Tim H.

Handle window events for capture phase, except load events, this quirk is needed because
Mozilla used to not do the bubble or capture phase for image load events and sites mistakenly
put capturing load event listeners on the window

  • khtml/xml/dom_nodeimpl.cpp: (NodeImpl::dispatchGenericEvent):
6:37 AM Changeset in webkit [14838] by andersca
  • 6 edits in trunk

WebCore:

2006-06-13 Anders Carlsson <​acarlsson@apple.com>

Reviewed by Geoff.

​http://bugzilla.opendarwin.org/show_bug.cgi?id=9406
REGRESSION: fix for bug 9390 broke two layout tests


  • loader/PluginDocument.cpp: (WebCore::PluginTokenizer::writeRawData): Call finished() after setting up the document structure so we'll emit onload events.

WebKit:

2006-06-13 Anders Carlsson <​acarlsson@apple.com>

Reviewed by Geoff.

​http://bugzilla.opendarwin.org/show_bug.cgi?id=9406
REGRESSION: fix for bug 9390 broke two layout tests

  • Plugins/WebPluginDatabase.h: (-[WebPluginDatabase isMIMETypeRegistered:]): Add new function isMIMETypeRegistered which returns whether a given MIME type has a plugin registered.
  • Plugins/WebPluginDatabase.m: (-[WebPluginDatabase init]): Init set of registered MIME types.


(-[WebPluginDatabase refresh]):
Add and remove MIME types from the set of registered MIME types when registering and unregistering
plugin MIME types.


(-[WebPluginDatabase dealloc]):
Release set of registered MIME types.


  • WebCoreSupport/WebFrameBridge.m: (-[WebFrameBridge determineObjectFromMIMEType:URL:]): Use isMIMETypeRegistered here.

Jun 12, 2006:

11:37 PM Changeset in webkit [14837] by ggaren
  • 1 edit
    1 add in trunk/WebCore

build fix -- forgot to svn add this file


  • ForwardingHeaders/kjs/SavedBuiltins.h: Added.
11:27 PM Changeset in webkit [14836] by ggaren
  • 2 edits in trunk/JavaScriptCore

build fix


  • bindings/objc/WebScriptObject.mm: (+[WebScriptObject throwException:]): Restore assignment I accidentally deleted in previous commit
11:10 PM Changeset in webkit [14835] by beidson
  • 2 edits in trunk/WebCore

Reviewed by Maciej.

Fixed a bug in append(char) and append(UChar) where our intended copy-on-write semantics was ignored!

  • platform/String.cpp: (WebCore::String::append):
11:08 PM Changeset in webkit [14834] by ggaren
  • 29 edits
    2 copies
    1 move
    4 adds in trunk

JavaScriptCore:

Reviewed by TimO, Maciej.


  • Merged InterpreterImp code into Interpreter, which implements all interpreter functionality now. This is part of my continuing quest to create an external notion of JS "execution context" that is unified and simple -- something to replace the mix of Context, ContextImp, ExecState, Interpreter, InterpreterImp, and JSRun.


All tests pass. Leaks test has not regressed from its baseline ~207 leaks
with ~3460 leaked nodes.

  • JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.vcproj:
  • JavaScriptCore.xcodeproj/project.pbxproj:
  • bindings/NP_jsobject.cpp:
  • bindings/objc/WebScriptObject.mm: (+[WebScriptObject throwException:]):
  • bindings/runtime_root.cpp:
  • bindings/runtime_root.h:
  • kjs/Context.cpp: (KJS::Context::Context):
  • kjs/ExecState.cpp: Added. (KJS::ExecState::lexicalInterpreter):
  • kjs/ExecState.h: Added. (KJS::ExecState::dynamicInterpreter):
  • kjs/SavedBuiltins.h: Added.
  • kjs/bool_object.cpp: (BooleanPrototype::BooleanPrototype):
  • kjs/collector.cpp: (KJS::Collector::collect): (KJS::Collector::numInterpreters):
  • kjs/context.h:
  • kjs/debugger.cpp: (Debugger::attach): (Debugger::detach):
  • kjs/function.cpp: (KJS::FunctionImp::callAsFunction): (KJS::GlobalFuncImp::callAsFunction):
  • kjs/function_object.cpp: (FunctionObjectImp::construct):
  • kjs/internal.cpp:
  • kjs/internal.h:
  • kjs/interpreter.cpp: (KJS::interpreterMap): (KJS::Interpreter::Interpreter): (KJS::Interpreter::init): (KJS::Interpreter::~Interpreter): (KJS::Interpreter::globalObject): (KJS::Interpreter::initGlobalObject): (KJS::Interpreter::globalExec): (KJS::Interpreter::checkSyntax): (KJS::Interpreter::evaluate): (KJS::Interpreter::builtinObject): (KJS::Interpreter::builtinFunction): (KJS::Interpreter::builtinArray): (KJS::Interpreter::builtinBoolean): (KJS::Interpreter::builtinString): (KJS::Interpreter::builtinNumber): (KJS::Interpreter::builtinDate): (KJS::Interpreter::builtinRegExp): (KJS::Interpreter::builtinError): (KJS::Interpreter::builtinObjectPrototype): (KJS::Interpreter::builtinFunctionPrototype): (KJS::Interpreter::builtinArrayPrototype): (KJS::Interpreter::builtinBooleanPrototype): (KJS::Interpreter::builtinStringPrototype): (KJS::Interpreter::builtinNumberPrototype): (KJS::Interpreter::builtinDatePrototype): (KJS::Interpreter::builtinRegExpPrototype): (KJS::Interpreter::builtinErrorPrototype): (KJS::Interpreter::builtinEvalError): (KJS::Interpreter::builtinRangeError): (KJS::Interpreter::builtinReferenceError): (KJS::Interpreter::builtinSyntaxError): (KJS::Interpreter::builtinTypeError): (KJS::Interpreter::builtinURIError): (KJS::Interpreter::builtinEvalErrorPrototype): (KJS::Interpreter::builtinRangeErrorPrototype): (KJS::Interpreter::builtinReferenceErrorPrototype): (KJS::Interpreter::builtinSyntaxErrorPrototype): (KJS::Interpreter::builtinTypeErrorPrototype): (KJS::Interpreter::builtinURIErrorPrototype): (KJS::Interpreter::mark): (KJS::Interpreter::interpreterWithGlobalObject): (KJS::Interpreter::saveBuiltins): (KJS::Interpreter::restoreBuiltins):
  • kjs/interpreter.h: (KJS::Interpreter::setCompatMode): (KJS::Interpreter::compatMode): (KJS::Interpreter::firstInterpreter): (KJS::Interpreter::nextInterpreter): (KJS::Interpreter::prevInterpreter): (KJS::Interpreter::debugger): (KJS::Interpreter::setDebugger): (KJS::Interpreter::setContext): (KJS::Interpreter::context):
  • kjs/nodes.cpp: (StatementNode::hitStatement): (RegExpNode::evaluate):
  • kjs/protect.h:

JavaScriptGlue:

Reviewed by TimO, Maciej.


  • JSGlue part of merging InterpreterImp into Interpreter.


Since there's now a kjs forwarding header, I moved all forwarding
headers into a "ForwardingHeaders" directory, like in WebCore.

  • ForwardingHeaders: Added.
  • ForwardingHeaders/kjs: Added.
  • ForwardingHeaders/kjs/ExecState.h: Added.
  • ForwardingHeaders/wtf: Added.
  • JavaScriptGlue.xcodeproj/project.pbxproj:
  • config.h:
  • wtf: Removed.
  • wtf/AlwaysInline.h: Removed.
  • wtf/Assertions.h: Removed.
  • wtf/FastMalloc.h: Removed.
  • wtf/Forward.h: Removed.
  • wtf/HashCountedSet.h: Removed.
  • wtf/HashSet.h: Removed.
  • wtf/Noncopyable.h: Removed.
  • wtf/OwnArrayPtr.h: Removed.
  • wtf/OwnPtr.h: Removed.
  • wtf/PassRefPtr.h: Removed.
  • wtf/Platform.h: Removed.
  • wtf/RefPtr.h: Removed.
  • wtf/Vector.h: Removed.

WebCore:

Reviewed by TimO, Maciej.

  • WebCore part of merging InterpreterImp into Interpreter. No test because there's no behavior change.


A substantive change here is that ScriptInterpreter::mark must now chain to
Interpreter::mark, since Interpreter needs to mark the things that
InterpreterImp used to mark.


  • WebCore.xcodeproj/project.pbxproj:
  • bindings/js/kjs_binding.cpp:
  • bindings/js/kjs_window.cpp: (KJS::Window::put):
  • bridge/mac/WebCoreFrameBridge.mm:
  • bridge/mac/WebCoreScriptDebugger.mm: (-[WebCoreScriptCallFrame evaluateWebScript:]):
  • kwq/KWQPageState.mm:
10:09 PM Changeset in webkit [14833] by mjs
  • 3 edits in trunk/WebKit
  • fix for cocoa exception (whoops)
  • WebView/WebView.m: (+[WebView _generatedMIMETypeForURLScheme:]): put this back
  • WebView/WebDataSource.m: (+[WebDataSource _generatedMIMETypeForURLScheme:]): call WebView
8:36 PM Changeset in webkit [14832] by mjs
  • 8 edits in trunk/WebKit

Reviewed by Darin.


  • remove use of WebView and related from WebMainResourceLoader
  • WebKit.xcodeproj/project.pbxproj:
  • WebView/WebDataSource.m: (+[WebDataSource _generatedMIMETypeForURLScheme:]): (+[WebDataSource _representationExistsForURLScheme:]): (+[WebDataSource _canShowMIMEType:]): (-[WebDataSource _handleFallbackContent]): (-[WebDataSource _decidePolicyForMIMEType:decisionListener:]):
  • WebView/WebDataSourceInternal.h:
  • WebView/WebFrame.m: (-[WebFrame _isMainFrame]):
  • WebView/WebFrameInternal.h:
  • WebView/WebMainResourceLoader.m: (-[WebMainResourceLoader willSendRequest:redirectResponse:]): (-[WebMainResourceLoader continueAfterContentPolicy:response:]): (-[WebMainResourceLoader checkContentPolicyForResponse:]): (-[WebMainResourceLoader loadWithRequestNow:]):
  • WebView/WebView.m:
6:04 PM Changeset in webkit [14831] by beidson
  • 8 edits in trunk/WebCore

Reviewed by Maciej.

Changed String::ascii() to return a Vector<char> instead of const char*
This allows us to use it "regularly" as the returned vector will destruct
and not leak memory like the previous approach.
We can now do a String.ascii().data() to get a char* buffer instead of
String.deprecatedString().ascii(). It doesn't improve the style much but
dumping the memory-leak issue is a plus.

  • icon/IconDatabase.cpp: (WebCore::IconDatabase::open): (WebCore::IconDatabase::clearDatabase):
  • icon/SQLDatabase.cpp: (SQLDatabase::open):
  • icon/SQLStatement.cpp: (WebCore::SQLStatement::prepare): (WebCore::SQLStatement::step): (WebCore::SQLStatement::returnTextResults): (WebCore::SQLStatement::returnTextResults16): (WebCore::SQLStatement::returnIntResults): (WebCore::SQLStatement::returnInt64Results): (WebCore::SQLStatement::returnDoubleResults): In addition to the changes to the string classes, changed my database code over to the new preferred method.
  • platform/PlatformString.h:
  • platform/String.cpp: (WebCore::String::ascii):
  • platform/StringImpl.cpp: (WebCore::StringImpl::ascii):
  • platform/StringImpl.h:
5:27 PM Changeset in webkit [14830] by tomernic
  • 2 edits in trunk/WebKit

Reviewed by Maciej.

<rdar://problem/4526052> intermittent assertion failure in -[WebBasePluginPackage dealloc] running layout tests

  • Plugins/WebNetscapePluginPackage.m: (-[WebNetscapePluginPackage _initWithPath:]): Renamed from -initWithPath:. Instead of releasing/deallocating self on error, return NO. (-[WebNetscapePluginPackage initWithPath:]): Call the new -_initWithPath:. If it returns NO, unload the plugin package before deallocating it.
4:16 PM Changeset in webkit [14829] by thatcher
  • 1 copy in tags/WebCore-418.16/WebCore

Tag for WebCore 418.16 (part 2 of 2.)

4:16 PM Changeset in webkit [14828] by thatcher
  • 1 add in tags/WebCore-418.16

Tag for WebCore 418.16 (part 1 of 2.)

4:03 PM Changeset in webkit [14827] by hyatt
  • 17 edits in trunk/WebCore

Add selection foreground color on Win32.

4:02 PM Changeset in webkit [14826] by sullivan
  • 2 edits in trunk/WebCore

Reviewed by Darin Adler.


  • page/Frame.cpp: (WebCore::Frame::markAllMatchesForText): Do a "fake" paint here so that the rectangles for the text matches will have been computed by the time this method returns.
3:43 PM Changeset in webkit [14825] by beidson
  • 4 edits in trunk/WebCore

Reviewed by Levi and Tim Omernick.

-Added a skeleton sqlite3 icon database file to IconDatabase
-Added functionality to validate and recreate this icon.db file
-Fixed some buggys in SQLDatabase.cpp

  • icon/IconDatabase.cpp: (WebCore::IconDatabase::open): (WebCore::IconDatabase::isValidDatabase): (WebCore::IconDatabase::clearDatabase): (WebCore::IconDatabase::recreateDatabase):
  • icon/IconDatabase.h:
  • icon/SQLStatement.cpp: (WebCore::SQLStatement::columnCount): (WebCore::SQLStatement::getColumnName): (WebCore::SQLStatement::getColumnName16): (WebCore::SQLStatement::getColumnText): (WebCore::SQLStatement::getColumnText16): (WebCore::SQLStatement::getColumnDouble): (WebCore::SQLStatement::getColumnInt): (WebCore::SQLStatement::getColumnInt64): (WebCore::SQLStatement::getColumnBlob): -Added checks to make sure we had a valid working sqlite3_statement as the sqlite3_* function calls weren't as error-tolerant as documentation advertised (maybe differences between the 3.3 docs I looked at and the 3.1.3 version installed on OSX)
1:54 PM Changeset in webkit [14824] by thatcher
  • 2 edits in branches/Safari-2-0-branch/WebCore

Versioning

1:53 PM Changeset in webkit [14823] by thatcher
  • 2 edits in branches/Safari-2-0-branch/WebCore

Reviewed by Maciej, Harrison.

Fixes <rdar://problem/4582172> frameset load events don't fire (can't login into Gmail account or see map at 511.org)

Framesets need to register their load, beforeunload and unload events with the window.

  • khtml/html/html_baseimpl.cpp: (HTMLFrameSetElementImpl::parseHTMLAttribute):
12:53 PM Changeset in webkit [14822] by hyatt
  • 3 edits in trunk/WebCore/platform

Fix Win32 build bustage.

12:02 PM Changeset in webkit [14821] by ggaren
  • 10 edits in trunk/JavaScriptCore

Reviewed by Maciej.


  • Have *.lut.h files #include lookup.h to eliminate surprising header include order dependency.
  • DerivedSources.make:
  • kjs/array_object.cpp:
  • kjs/date_object.cpp:
  • kjs/date_object.h: (KJS::DateProtoFunc::):
  • kjs/lexer.cpp:
  • kjs/math_object.cpp:
  • kjs/number_object.cpp:
  • kjs/regexp_object.cpp:
  • kjs/string_object.cpp:
11:07 AM Changeset in webkit [14820] by beidson
  • 7 edits
    3 adds in trunk/WebCore

Reviewed by Levi.

-Added SQLite helper wrappers to ease use of SQLite in the IconDatabase.
-Changed the base IconDatabase over to this new framework.

  • WebCore.xcodeproj/project.pbxproj:
  • bridge/mac/WebCoreIconDatabaseBridge.mm: (-[WebCoreIconDatabaseBridge closeSharedDatabase]):

-By popular request, removed an annoying log message I'd accidentally left in

  • icon/IconDatabase.cpp: (WebCore::IconDatabase::IconDatabase): (WebCore::IconDatabase::open): (WebCore::IconDatabase::close): (WebCore::IconDatabase::~IconDatabase):
  • icon/IconDatabase.h: (WebCore::IconDatabase::isOpen):

-Changed over IconDatabase from direct sqlite3_* calls to the new SQLDatabase calls

  • icon/SQLDatabase.cpp: Added. (SQLDatabase::SQLDatabase): (SQLDatabase::open): (SQLDatabase::close): (SQLDatabase::executeCommand): (SQLDatabase::tableExists):
  • icon/SQLDatabase.h: Added. (WebCore::SQLDatabase::isOpen): (WebCore::SQLDatabase::getPath): (WebCore::SQLDatabase::lastError): (WebCore::SQLDatabase::lastErrorMsg): (WebCore::SQLStatement::isPrepared): (WebCore::SQLStatement::lastError): (WebCore::SQLStatement::lastErrorMsg):
  • icon/SQLStatement.cpp: Added. (WebCore::SQLStatement::SQLStatement): (WebCore::SQLStatement::~SQLStatement): (WebCore::SQLStatement::prepare): (WebCore::SQLStatement::step): (WebCore::SQLStatement::finalize): (WebCore::SQLStatement::reset): (WebCore::SQLStatement::executeCommand): (WebCore::SQLStatement::bindBlob): (WebCore::SQLStatement::bindText): (WebCore::SQLStatement::columnCount): (WebCore::SQLStatement::getColumnName): (WebCore::SQLStatement::getColumnName16): (WebCore::SQLStatement::getColumnText): (WebCore::SQLStatement::getColumnText16): (WebCore::SQLStatement::getColumnDouble): (WebCore::SQLStatement::getColumnInt): (WebCore::SQLStatement::getColumnInt64): (WebCore::SQLStatement::getColumnBlob): (WebCore::SQLStatement::returnTextResults): (WebCore::SQLStatement::returnTextResults16): (WebCore::SQLStatement::returnIntResults): (WebCore::SQLStatement::returnInt64Results): (WebCore::SQLStatement::returnDoubleResults):

-Initial checking of SQLDatabase framework

  • platform/PlatformString.h:
  • platform/String.cpp: (WebCore::String::String):

-Added an explicit UChar* constructor to our string class as much of sqlite3's UTF16 handling is based on
null-terminated UTF16 which we didn't yet support.

Note: See TracTimeline for information about the timeline view.