[webkit-dev] Setting up Xcode for debugging webkit

[ying lcs]

I am using Xcode 3.1.2 on Mac 10.5.5.

I try to setup XCode to debug webkit by following this:
http://webkit.org/building/debug.html


Active SDK: Use ProjectSettings
Active Configuration: Debug
Active Target: JavaScriptCore
Active Executable: MyExecutable

For step #4, I enter '/Applications/Safari.app'

Set a breakpoint at line 40 of Parser.cpp:
void Parser::parse(JSGlobalData* globalData, int* errLine, UString* errMsg);

Then, I did 'Build and Debug'.

An instance of Safai get started.

And then I load 'www.cnn.com' (a site which has JavaScript).

But the breakpoint does not get break. I expect it should break there
since it needs to parse the Javascript files.

Can you please tell me what am I missing?

Thank you for any pointers.
___
webkit-dev mailing list
webkit-dev@lists.webkit.org
http://lists.webkit.org/mailman/listinfo.cgi/webkit-dev

Re: [webkit-dev] Propose some changes to WebKit implementation of HTML5 workers in order to run them in Chrome's worker process

[Jian Li]

Based on the initial refactoring of WorkerMessagingProxy class, I've updated
the abstractions as the following:

// A proxy to talk to the worker context.
// All calls to any methods should come from worker object thread.class
WorkerContextProxyBase {
public:
// Creator.
static WorkerContextProxy* create(const String& scriptUrl, Worker*
worker);

// Terminates the worker context.
virtual void terminateWorkerContext() = 0;

// Posts the message to the worker context.
virtual void postMessageToWorkerContext(const String& message) = 0;

// Has any pending activity?
virtual bool hasPendingActivity() const = 0;
};

// A proxy to talk to the worker object.
// All calls to any methods should come from worker context thread.
class WorkerObjectProxyBase {
public:
// Posts the message to the worker object.
virtual void postMessageToWorkerObject(const String& message) = 0;

// Posts the exception to the worker object.
virtual void postExceptionToWorkerObject(const String& errorMessage, int
lineNumber, const String& sourceURL) = 0;

// Reports if there is pending activity in the worker context.
virtual void reportPendingActivity(bool hasPendingActivity) = 0;

// Called when the worker context is destroyed.
virtual void workerContextDestroyed() = 0;
};




On Mon, Feb 2, 2009 at 6:10 PM, Jian Li  wrote:

> The current WebKit implementation of HTML5 workers is based on cross-thread
> communication in single process. To make it work for Chrome, we need to run
> the workers in Chrome's worker process. Hence, we propose the following
> changes:
>
> MessagingWorkerProxy acts as the gateway between Worker and WorkerContext in
> order to exchange messages. This does not work well across process. To
> support both cross-thread and cross-process models, we propose splitting
> MessagingWorkerProxy into two pieces: WorkerContextProxy and
> WorkerObjectProxy.
>
> For WebKit, MessagingWorkerProxy can implement both WorkerContextProxy
>  and WorkerObjectProxy, that communicate with each other via cross-thread
> task posting. Some changes need also to be made in DOM objects in order to
> put them in the right scope. For example, MessageWorkerTask.performTask can
> be moved to Worker object and MessageWorkerContextTask.performTask can be
> moved to WorkerContext object.
>
> For Chrome, the implementation of WorkerContextProxy uses IPC to
> communicate with the implementation of WorkerObjectProxy at the other end.
>
> Worker <=> WorkerContextProxy <=> ... <=> WorkParentProxy <=>
> WorkerContext
>
>   WebKit: \--- in main thread  ---/   CTC \-- in worker thread
>  -/
>   (CTC: Cross-thread communication by posting task)
>
>   Chrome: \- in renderer process -/   IPC \-- in worker process
> -/
>   (IPC: Inter-process communication)
>
>
> The abstractions of WorkerContextProxy and WorkerObjectProxy are defined
> as the following:
>
> // A proxy to talk to the worker context.
> // All calls to any methods should come from main thread.
> class WorkerContextProxy {
> public:
>   // Creator.
>   static WorkerContextProxy* create();
>
>   // Starts the worker context. Returns 0 if successful, error code
> otherwise.
>   virtual int startContext(const String& scriptUrl, Worker* worker) = 0;
>
>   // Terminates the worker context.
>   virtual void terminateContext() = 0;
>
>   // Posts the message to the worker context.
>   virtual void postMessageToContext(const String& message) = 0;
>
>   // Has any pending activity?
>   virtual bool hasPendingActivity() = 0;
> };
>
> // A proxy to talk to the worker object.
> // All calls to any methods should come from worker thread.
> class WorkerObjectProxy {
> public:
>   // Runs the script with the specified url.
>   virtual bool run(const String& scriptURL) = 0;
>
>   // Posts the message to the worker object.
>   virtual void postMessageToObject(const String& message) = 0;
>
>   // Posts the exception to the worker object.
>   virtual void postExceptionToObject(const String& errorMessage, int
> lineNumber, const String& sourceURL) = 0;
> };
>
>
>
___
webkit-dev mailing list
webkit-dev@lists.webkit.org
http://lists.webkit.org/mailman/listinfo.cgi/webkit-dev

Re: [webkit-dev] open jpeg failed,but ok if this jpeg on local

[RyanYee/叶晔]

I have tested WebKit-r39293,this jpeg open ok,but those more bigger still
failed.
I think maybe my memory too smaller.

2009/2/3 RyanYee/叶晔 

> 1.I had tried to open jpeg :
> ./GtkLauncher
> http://farm4.static.flickr.com/3218/2936385481_d4cb8fd4ff.jpg?v=0
>
> failed with these message:
> UNIMPLEMENTED:
> (WebKit/gtk/WebCoreSupport/FrameLoaderClientGtk.cpp:559 virtual void
> WebKit::FrameLoaderClient::dispatchDidFirstLayout())
>
> (GtkLauncher:1168): Gdk-DirectFB-WARNING **: gdk_window_set_keep_above()
> not implemented.
>
>
> (GtkLauncher:1168): Gdk-DirectFB-WARNING **: gdk_window_set_keep_below()
> not implemented.
>
> UNIMPLEMENTED:
> (WebKit/gtk/WebCoreSupport/FrameLoaderClientGtk.cpp:600 virtual bool
> WebKit::FrameLoaderClient::canShowMIMEType(const WebCore::String&) const)
> UNIMPLEMENTED:
> (WebKit/gtk/WebCoreSupport/FrameLoaderClientGtk.cpp:513 virtual void
> WebKit::FrameLoaderClient::dispatchWillClose())
> UNIMPLEMENTED:
> (WebKit/gtk/WebCoreSupport/FrameLoaderClientGtk.cpp:441 virtual void
> WebKit::FrameLoaderClient::makeRepresentation(WebCore::DocumentLoader*))
> UNIMPLEMENTED:
> (WebKit/gtk/WebCoreSupport/FrameLoaderClientGtk.cpp:756 virtual void
> WebKit::FrameLoaderClient::updateGlobalHistory(const WebCore::KURL&))
>
> (GtkLauncher:1168): Gtk-CRITICAL **: gtk_entry_set_text: assertion
> `GTK_IS_ENTRY (entry)' failed
> UNIMPLEMENTED:
> (WebKit/gtk/WebCoreSupport/FrameLoaderClientGtk.cpp:648 virtual void
> WebKit::FrameLoaderClient::dispatchDidReceiveContentLength(WebCore::DocumentLoader*,
> long unsigned int, int))
> (!) [ 1168:   10.120] --> Caught signal 7 (at 0x2092f000, non-existent
> physical address) <--
>
> 2.then I tried another one,some one smaller
> ./GtkLauncher http://otho.douban.com/icon/u1293270-9.jpg
>
> ok with these message:
> UNIMPLEMENTED:
> (WebKit/gtk/WebCoreSupport/FrameLoaderClientGtk.cpp:559 virtual void
> WebKit::FrameLoaderClient::dispatchDidFirstLayout())
>
> (GtkLauncher:1201): Gdk-DirectFB-WARNING **: gdk_window_set_keep_above()
> not implemented.
>
>
> (GtkLauncher:1201): Gdk-DirectFB-WARNING **: gdk_window_set_keep_below()
> not implemented.
>
> UNIMPLEMENTED:
> (WebKit/gtk/WebCoreSupport/FrameLoaderClientGtk.cpp:600 virtual bool
> WebKit::FrameLoaderClient::canShowMIMEType(const WebCore::String&) const)
> UNIMPLEMENTED:
> (WebKit/gtk/WebCoreSupport/FrameLoaderClientGtk.cpp:513 virtual void
> WebKit::FrameLoaderClient::dispatchWillClose())
> UNIMPLEMENTED:
> (WebKit/gtk/WebCoreSupport/FrameLoaderClientGtk.cpp:441 virtual void
> WebKit::FrameLoaderClient::makeRepresentation(WebCore::DocumentLoader*))
> UNIMPLEMENTED:
> (WebKit/gtk/WebCoreSupport/FrameLoaderClientGtk.cpp:756 virtual void
> WebKit::FrameLoaderClient::updateGlobalHistory(const WebCore::KURL&))
>
> (GtkLauncher:1201): Gtk-CRITICAL **: gtk_entry_set_text: assertion
> `GTK_IS_ENTRY (entry)' failed
> UNIMPLEMENTED:
> (WebKit/gtk/WebCoreSupport/FrameLoaderClientGtk.cpp:648 virtual void
> WebKit::FrameLoaderClient::dispatchDidReceiveContentLength(WebCore::DocumentLoader*,
> long unsigned int, int))
> UNIMPLEMENTED:
> (WebCore/platform/gtk/LocalizedStringsGtk.cpp:326 WebCore::String
> WebCore::imageTitle(const WebCore::String&, const WebCore::IntSize&))
> UNIMPLEMENTED:
> (WebKit/gtk/WebCoreSupport/FrameLoaderClientGtk.cpp:488 virtual void
> WebKit::FrameLoaderClient::dispatchDidHandleOnloadEvents())
> UNIMPLEMENTED:
> (WebKit/gtk/WebCoreSupport/FrameLoaderClientGtk.cpp:554 virtual void
> WebKit::FrameLoaderClient::dispatchDidFinishDocumentLoad())
> UNIMPLEMENTED:
> (WebKit/gtk/WebCoreSupport/FrameLoaderClientGtk.cpp:574 virtual void
> WebKit::FrameLoaderClient::dispatchDidLoadMainResource(WebCore::DocumentLoader*))
> UNIMPLEMENTED:
> (WebKit/gtk/WebCoreSupport/FrameLoaderClientGtk.cpp:653 virtual void
> WebKit::FrameLoaderClient::dispatchDidFinishLoading(WebCore::DocumentLoader*,
> long unsigned int))
>
> 3.so I downloaded the big one to local,then write a 1.html ( added  src="file:///home/1.jpg"  point to this picture>)
> ./GtkLauncher file:///home/1.html
>
> then ok with these message like step 2
> UNIMPLEMENTED:
> (WebKit/gtk/WebCoreSupport/FrameLoaderClientGtk.cpp:559 virtual void
> WebKit::FrameLoaderClient::dispatchDidFirstLayout())
>
> (GtkLauncher:1210): Gdk-DirectFB-WARNING **: gdk_window_set_keep_above()
> not implemented.
>
>
> (GtkLauncher:1210): Gdk-DirectFB-WARNING **: gdk_window_set_keep_below()
> not implemented.
>
> UNIMPLEMENTED:
> (WebKit/gtk/WebCoreSupport/FrameLoaderClientGtk.cpp:600 virtual bool
> WebKit::FrameLoaderClient::canShowMIMEType(const WebCore::String&) const)
> UNIMPLEMENTED:
> (WebKit/gtk/WebCoreSupport/FrameLoaderClientGtk.cpp:513 virtual void
> WebKit::FrameLoaderClient::dispatchWillClose())
> UNIMPLEMENTED:
> (WebKit/gtk/WebCoreSupport/FrameLoaderClientGtk.cpp:441 virtual void
> WebKit::FrameLoaderClient::makeRepresentation(WebCore::DocumentLoader*))
> UNIMPLEMENTED:
> (WebKit/gtk/WebCoreSupport/FrameLoaderClientGtk.cpp:756 virtual void
> WebKit::FrameLoaderClient::u

Re: [webkit-dev] li with position outside combined with img float left problem

[David Kilzer]

Please file a bug on .  Thanks!

Dave




From: Magnus Rembold 
To: webkit-dev@lists.webkit.org
Sent: Thursday, February 5, 2009 11:41:58 AM
Subject: [webkit-dev] li with position outside combined with img float left 
problem


A complete description of the problem with source HTML and inline CSS together 
with images can be found here:
http://www.munterbund.de/problems/090205/li_outside_and_img_float_problem.html

My question is: this is obviously not as is should be. Do the "standards" 
require that this is displayed this way? Or is it only a part of the rendering, 
that's considered "not important"? Or will it be fixed in some way? Everybody 
that I asked told me to format  in a completely awkward way with background 
pictures and so on. This shouldn't be necessary or am I wrong?

Thanks kindly for your replys, comments or suggestions!
Best regards,
Magnus



Magnus RemboldZÜRCHER HOCHSCHULE DER KÜNSTE
Dozent ITZurich University of the Arts

Fon +41 43 446 3256Ausstellungsstrasse 60
Fax +41 43 446 4587CH-8031 Zürich
http://iad.zhdk.chSwitzerland___
webkit-dev mailing list
webkit-dev@lists.webkit.org
http://lists.webkit.org/mailman/listinfo.cgi/webkit-dev

Re: [webkit-dev] JSObjectSetProperty & JSObjectHasProperty stuck in an infinite loop

[Laurent Calburtin]

>>Please file a bug report @ bugs.webkit.or

done, bug# 23771

thanks,
Laurent
___
webkit-dev mailing list
webkit-dev@lists.webkit.org
http://lists.webkit.org/mailman/listinfo.cgi/webkit-dev

[webkit-dev] li with position outside combined with img float left problem

[Magnus Rembold]

Hi All!

I think there is a problem with the alignment of  elements when  
they are besides a floated element (say a  with float: left;  
width: auto; and an  inside). They seem to orient themselves with  
the beginning of the text on the margin that is set for the floating  
element. What they should do is align themselves with their bullets at  
the margin. Especially if list-style-position: outside;is set.


A complete description of the problem with source HTML and inline CSS  
together with images can be found here:

http://www.munterbund.de/problems/090205/li_outside_and_img_float_problem.html

My question is: this is obviously not as is should be. Do the  
"standards" require that this is displayed this way? Or is it only a  
part of the rendering, that's considered "not important"? Or will it  
be fixed in some way? Everybody that I asked told me to format  in  
a completely awkward way with background pictures and so on. This  
shouldn't be necessary or am I wrong?


Thanks kindly for your replys, comments or suggestions!
Best regards,
Magnus



Magnus Rembold  ZÜRCHER HOCHSCHULE DER KÜNSTE
Dozent IT   Zurich University of 
the Arts

Fon +41 43 446 3256 Ausstellungsstrasse 60
Fax +41 43 446 4587 CH-8031 Zürich
http://iad.zhdk.ch  Switzerland




smime.p7s
Description: S/MIME cryptographic signature
___
webkit-dev mailing list
webkit-dev@lists.webkit.org
http://lists.webkit.org/mailman/listinfo.cgi/webkit-dev

Re: [webkit-dev] JSObjectSetProperty & JSObjectHasProperty stuck in an infinite loop

[Geoffrey Garen]

Hi Laurent.

The API shouldn't have any subtle dependency on  
kJSClassAttributeNoAutomaticPrototype. Please file a bug report @  
bugs.webkit.or, and attach a source file that we can compile in order  
to reproduce this bug.


Thanks,
Geoff

On Feb 4, 2009, at 1:55 PM, Laurent Calburtin wrote:


Hi,

With the trunk version of javascriptcore, JSObjectHasProperty is  
stuck in an infinite loop with following code:


   JSClassDefinition globalObjectClassDefinition =  
kJSClassDefinitionEmpty;
   globalObjectClassDefinition.staticFunctions =  
globalObject_staticFunctions; // not null
   JSClassRef globalObjectClass =  
JSClassCreate( &globalObjectClassDefinition);
   JSGlobalContextRef context = JSGlobalContextCreateInGroup(NULL,  
globalObjectClass);

   JSStringRef propName = JSStringCreateWithUTF8CString("name");
   JSObjectHasProperty( context, JSContextGetGlobalObject(context),  
propName);


By comparing my code to testapi.c I found that adding the attribute  
kJSClassAttributeNoAutomaticPrototype to my global class definition  
solved the problem.


Is this a bug or an expected behavior?
If kJSClassAttributeNoAutomaticPrototype is mandatory in such a  
situation, may I suggest to state it in the header file?
Are there other situations I should use  
kJSClassAttributeNoAutomaticPrototype?


Note that JSObjectSetProperty is also stuck without the attribute  
because it internally calls the same code as JSObjectHasProperty.
Also note that JSObjectHasProperty from MacOSX10.5 webkit framework  
doesn't get stuck.


Laurent
___
webkit-dev mailing list
webkit-dev@lists.webkit.org
http://lists.webkit.org/mailman/listinfo.cgi/webkit-dev


___
webkit-dev mailing list
webkit-dev@lists.webkit.org
http://lists.webkit.org/mailman/listinfo.cgi/webkit-dev

Re: [webkit-dev] webkit parser

[Shariq Rizvi]

On Thu, Feb 5, 2009 at 6:13 AM, abhas saroha  wrote:

> hi all,
>  Qt port of webkit lists bison as a prerequisite, i wanted to know is the
> dom parser created with bison, and is it also the case with other ports of
> webkit? If not then where does the parser code reside and is it webkit port
> independent?
> thanx in advance.
>


Look at WebCore/html/HTMLTokenizer.cpp and HTMLParser.cpp

It is a custom parser, and I don't see it using a context-free grammar or
Bison. Also does not look port-dependent.




>
> Abhas Saroha
>
> ___
> webkit-dev mailing list
> webkit-dev@lists.webkit.org
> http://lists.webkit.org/mailman/listinfo.cgi/webkit-dev
>
>
___
webkit-dev mailing list
webkit-dev@lists.webkit.org
http://lists.webkit.org/mailman/listinfo.cgi/webkit-dev

[webkit-dev] Create widgets with WebKit/Gtk

[Massimo]

Hi all,

 

I successfully ported WebKit on both MIPS and x86 embedded architectures on
top of Gtk/DirectFB on Linux.

 

I'm now looking for information about how creating very simple widgets with
WebKit/Gtk (weather forecasts, clock, etc.). 

 

Does WebKit implement some custom architecture to build widgets and/or
support some feature from W3C Widgets working draft?

 

Can someone please point me out some references (documentation, examples,
etc.) about this topic?

 

Thanks in advance.

Massimo

 

___
webkit-dev mailing list
webkit-dev@lists.webkit.org
http://lists.webkit.org/mailman/listinfo.cgi/webkit-dev

[webkit-dev] webkit parser

[abhas saroha]

hi all,
 Qt port of webkit lists bison as a prerequisite, i wanted to know is the
dom parser created with bison, and is it also the case with other ports of
webkit? If not then where does the parser code reside and is it webkit port
independent?
thanx in advance.

Abhas Saroha
___
webkit-dev mailing list
webkit-dev@lists.webkit.org
http://lists.webkit.org/mailman/listinfo.cgi/webkit-dev

[webkit-dev] Error Handling in webkit

[Nitin Walke]

Hi all
How can I do error handling in webkit? Suppose I have given wrong URL then
how can I show proper message? How can I handle this situation?Can you give
sample example?

Thanks in advance..

Regards,
Nitin Walke
___
webkit-dev mailing list
webkit-dev@lists.webkit.org
http://lists.webkit.org/mailman/listinfo.cgi/webkit-dev