From rbruehlmann at nexsystems.ch Sat Oct 1 01:55:58 2011
From: rbruehlmann at nexsystems.ch (=?Windows-1252?Q?Reto_Br=FChlmann?=)
Date: Sat, 1 Oct 2011 08:55:58 +0000
Subject: [webkit-help] Strange behaviour on scrollable DIV's
Message-ID: <1AE66644725714469FA084A626A54B481BF2E942@MAILBOX.nexsystems.ch>
Hi @ all...
First, sorry for my bad english ? i know, thats very hard to understand my spelling, but i?ll do my best ? ;-)
First on the specs?
XOOM (Android 3.2), Mobile Application based on webkit with HTML5 ?
I have a
that?s hold the calendar and is fixed in height and width and so on, its scrollable in height, if more entrys are in the div as the height is. Now, when I scroll it to the bottom, because I have a lot of entrys in it, which entrys are tablerows (td?s) with a onclick function on it, and click the last item or items that are not shown on default (in relation to the height of the div) it will every time click on the default visible (in relation to the height of the div) td?s.
But this behavior is only on XOOM, on Google Chrome it works like it should ? I have no clue, what I make wrong or is it a bug? Do I forget something? Is this a well known problem?
Thanks for your help?
Cheers
Reto
P.S Here an example?
AGENDA
foreach($returnvalue_ as $aid => $value_) { ?>
= nl2br($value_[title]); ?>
} ?>
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
From julien.chaffraix at gmail.com Sun Oct 2 11:44:38 2011
From: julien.chaffraix at gmail.com (Julien Chaffraix)
Date: Sun, 2 Oct 2011 11:44:38 -0700
Subject: [webkit-help] error of building webkit with directfb
In-Reply-To: <1a85427.763.132bd4ca592.Coremail.wenmhappy@126.com>
References: <1a85427.763.132bd4ca592.Coremail.wenmhappy@126.com>
Message-ID:
> ?? I have successfully built (cross-compile) webkit-r59711 with --gtk (x11
> default) on embed platform.? But now there are errors when i build it with
> --gtk --with-target=directfb.? In that case, webkit will depends on
> libgdk-x11-2.0.so and libgdk-directfb-2.0.so, they depend respectively on
> libcairo( with xlib) and libcairo( with directfb).
If you don't disable plugins (NPAPI) then WebKit has a dependency on X
on Linux. Have you tried compiling with --no-netscape-plugin? (I am
assuming that you are using build-webkit here if not you will need to
find the equivalent switch)
Hope it helps,
Julien
From yamazaki at fusenetwork.co.jp Mon Oct 3 02:43:10 2011
From: yamazaki at fusenetwork.co.jp (Kunio Yamazaki)
Date: Mon, 03 Oct 2011 18:43:10 +0900
Subject: [webkit-help] I need help in building webkit wincairo
Message-ID: <4E8983AE.6070605@fusenetwork.co.jp>
Hi Everyone,
I am trying to build the wincairo version with the source of webkit r96399.
Development environment is [ Window + cygwin + vs2005 ES(Sp1) ]
However, the error same as a release and debug build was outputted.
WKCACFViewWindow.obj : error LNK2001: external symbol "__imp__WKCACFViewUpdate" is unresolved.
WKCACFViewWindow.obj : error LNK2001: external symbol "__imp__WKCACFViewDraw" is unresolved.
WKCACFViewWindow.obj : error LNK2001: external symbol "__imp__WKCACFViewDrawIntoDC" is unresolved.
E:\WebKit\WebKitBuild\Debug_Cairo_CFLite\bin\WebKit_debug.dll : fatal error LNK1120:
How can I fix that errors?
Yours sincerely
Kunio
From aroben at apple.com Mon Oct 3 08:19:29 2011
From: aroben at apple.com (Adam Roben)
Date: Mon, 03 Oct 2011 11:19:29 -0400
Subject: [webkit-help] I need help in building webkit wincairo
In-Reply-To: <4E8983AE.6070605@fusenetwork.co.jp>
References: <4E8983AE.6070605@fusenetwork.co.jp>
Message-ID: <0C3C8F1B-19F3-4AF7-84B9-83DD9F2C8F9E@apple.com>
On Oct 3, 2011, at 5:43 AM, Kunio Yamazaki wrote:
> I am trying to build the wincairo version with the source of webkit r96399.
> Development environment is [ Window + cygwin + vs2005 ES(Sp1) ]
>
> However, the error same as a release and debug build was outputted.
>
> WKCACFViewWindow.obj : error LNK2001: external symbol "__imp__WKCACFViewUpdate" is unresolved.
> WKCACFViewWindow.obj : error LNK2001: external symbol "__imp__WKCACFViewDraw" is unresolved.
> WKCACFViewWindow.obj : error LNK2001: external symbol "__imp__WKCACFViewDrawIntoDC" is unresolved.
> E:\WebKit\WebKitBuild\Debug_Cairo_CFLite\bin\WebKit_debug.dll : fatal error LNK1120:
>
> How can I fix that errors?
Something is causing the WinCairo build to try to link against WebKitQuartzCoreAdditions.lib. I don't think the WinCairo port should be using that library. Brent Fulgham might have some ideas as to why this is happening.
-Adam
From bfulgham at gmail.com Mon Oct 3 12:18:38 2011
From: bfulgham at gmail.com (Brent Fulgham)
Date: Mon, 3 Oct 2011 12:18:38 -0700
Subject: [webkit-help] I need help in building webkit wincairo
In-Reply-To: <4E8983AE.6070605@fusenetwork.co.jp>
References: <4E8983AE.6070605@fusenetwork.co.jp>
Message-ID:
Hi Kunio,
On Mon, Oct 3, 2011 at 2:43 AM, Kunio Yamazaki
wrote:
> I am trying to build the wincairo version with the source of webkit r96399.
> Development environment is [ Window + cygwin + vs2005 ES(Sp1) ]
According to the build bot, that revision should be fine.
(http://build.webkit.org/builders/WinCairo%20Debug%20%28Build%29/builds/11498).
I'm guessing that one of the following is the problem:
1. Are you building from within Visual Studio? If so, please be sure
you are using either the "Release_Cairo_CFLite" or
"Debug_Cairo_CFLite" targets. I have accidentally built with the wrong
flavor on more than one occasion! :-)
2. If you are using "build-webkit", please make sure you are passing
the "--wincairo" command line argument to be sure it builds properly.
Can you post a full build log somewhere I could view it?
Thanks,
-Brent
From yamazaki at fusenetwork.co.jp Mon Oct 3 18:38:12 2011
From: yamazaki at fusenetwork.co.jp (Kunio Yamazaki)
Date: Tue, 04 Oct 2011 10:38:12 +0900
Subject: [webkit-help] I need help in building webkit wincairo
In-Reply-To:
References: <4E8983AE.6070605@fusenetwork.co.jp>
Message-ID: <4E8A6384.7040903@fusenetwork.co.jp>
Hi Brent,
(2011/10/04 4:18), Brent Fulgham wrote:
> I'm guessing that one of the following is the problem:
>
> 1. Are you building from within Visual Studio?
No, I am building the wincairo port at the command line.
> 2. If you are using "build-webkit", please make sure you are passing
> the "--wincairo" command line argument to be sure it builds properly.
>
> Can you post a full build log somewhere I could view it?
Thank you Brent.
Build log is carried. But the following sites are Japanese although I am sorry.
http://wiki.livedoor.jp/noppo/d/wincairo%20build%20error%20%282011/10/03%29
And in order to build at a command line, the following commands are executed in cygwin.
Although this is the procedure currently continued from last year, has it made a mistake in some?
$ cd /cygdrive/e
$ svn checkout --revision 96399 http://svn.webkit.org/repository/webkit/trunk WebKit
$ WebKit/Tools/Scripts/update-webkit-auxiliary-libs
$ WebKit/Tools/Scripts/update-webkit-support-libs
$ WebKit/Tools/Scripts/update-webkit-wincairo-libs
$ WebKit/Tools/Scripts/set-webkit-configuration --debug --32-bit
$ WebKit/Tools/Scripts/build-webkit --debug --wincairo
Checking Last-Modified date of WinCairoRequirements.zip...
Current WinCairoRequirements is up to date
Building results into: /cygdrive/e/WebKit/WebKitBuild
WEBKITOUTPUTDIR is set to: E:\WebKit\WebKitBuild
WEBKITLIBRARIESDIR is set to: E:\WebKit\WebKitLibraries\win
/cygdrive/c/Program Files/Microsoft Visual Studio 8/Common7/IDE/VCExpress.exe win\WebKit.vcproj\WebKit.sln /build Debug_Cairo_CFLite
Found 0 Visual Studio Express Build Logs:
===== BUILD FAILED ======
Please ensure you have run Tools/Scripts/update-webkit to install dependencies.
You can view build errors by checking the BuildLog.htm files located at:
/cygdrive/e/WebKit/WebKitBuild/obj//.
Thanks,
-Kunio
From vizshala at gmail.com Mon Oct 3 19:48:12 2011
From: vizshala at gmail.com (vizshala)
Date: Tue, 4 Oct 2011 10:48:12 +0800
Subject: [webkit-help] how to get formatted console message from inspector
without displaying frontend
Message-ID:
Dear All,
I'm trying to get the formatted console message from inside WebCore.
At the first, I added a function called formattedMessage to Console domain
defined in inspector.json.
Then, I modify ConsoleView.js to allow ConsoleMessage be sent back to
InspectorConsoleAgent through this customized function.
The net effect is that I can not receive the formatted message unless the
inspector frontend is shown.
If there is anyway to get the formatted console message without showing the
inspector frontend?
Thanks.
With Regards,
vizshala at gmail.com
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
From bfulgham at gmail.com Mon Oct 3 21:22:28 2011
From: bfulgham at gmail.com (Brent Fulgham)
Date: Mon, 3 Oct 2011 21:22:28 -0700
Subject: [webkit-help] I need help in building webkit wincairo
In-Reply-To: <4E8A6384.7040903@fusenetwork.co.jp>
References: <4E8983AE.6070605@fusenetwork.co.jp>
<4E8A6384.7040903@fusenetwork.co.jp>
Message-ID: <591B73CB-3460-417A-9078-A9118B1772E5@gmail.com>
Hi Kunio,
On Oct 3, 2011, at 6:38 PM, Kunio Yamazaki wrote:
>> 2. If you are using "build-webkit", please make sure you are passing
>> the "--wincairo" command line argument to be sure it builds properly.
>>
>> Can you post a full build log somewhere I could view it?
>
> Thank you Brent.
> Build log is carried. But the following sites are Japanese although I am sorry.
> http://wiki.livedoor.jp/noppo/d/wincairo%20build%20error%20%282011/10/03%29
It looks like the build is proceeding using the correct target, so I'm not sure what's going on.
The only thing I can think of at the moment is:
> $ cd /cygdrive/e
> $ svn checkout --revision 96399 http://svn.webkit.org/repository/webkit/trunk WebKit
Is /cygdrive/e your HOME location? I have only really had success building in a subdirectory of the Cygwin 'HOME' directory.
> $ WebKit/Tools/Scripts/set-webkit-configuration --debug --32-bit
> $ WebKit/Tools/Scripts/build-webkit --debug --wincairo
I think you want the "wincairo" flag here as well:
> $ WebKit/Tools/Scripts/set-webkit-configuration --debug --32-bit --wincairo
I'm worried that the build failures are somehow related to the Japanese Windows localization. Are there any error messages in the earlier sections of the build (e.g., while building WTF/JavaScriptCore)?
> Checking Last-Modified date of WinCairoRequirements.zip...
> Current WinCairoRequirements is up to date
> Building results into: /cygdrive/e/WebKit/WebKitBuild
> WEBKITOUTPUTDIR is set to: E:\WebKit\WebKitBuild
> WEBKITLIBRARIESDIR is set to: E:\WebKit\WebKitLibraries\win
> /cygdrive/c/Program Files/Microsoft Visual Studio 8/Common7/IDE/VCExpress.exe win\WebKit.vcproj\WebKit.sln /build Debug_Cairo_CFLite
> Found 0 Visual Studio Express Build Logs:
This looks exactly right. I don't understand why this is causing trouble.
Where is your WEBKITLIBRARIESDIR pointed? Some of the flags that control the build are stored in this directory. If the environment variable points to an 'old' location, it might be causing the build to use the wrong flags.
Thanks,
-Brent
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
From tech.tantrik at gmail.com Tue Oct 4 00:00:15 2011
From: tech.tantrik at gmail.com (Techie Tantrik)
Date: Tue, 4 Oct 2011 12:30:15 +0530
Subject: [webkit-help] qtWebKit CSS Trasformations
Message-ID:
Hi,
I have qtwebkit running on ARM based platform.
I am facing a strange problem. When run normal pages with JScript movement,
it works fine.
But the same if I use CSS transformations, it is not working.
I am using only one transformation *"**$("#images,
#captions").css("-webkit-transform","translate(0px, 0px)");"* in my HTML
page, it just hangs. There is no further movement.
Instead of CSS transformation, i tried with Just an alert message. But even
that its hanging. (Even if it is not executing the CSS transformation)
To check whats happening, i compiled webcore with trace functionality and
last few call trace is given below:
ENT WebCore::HTMLScriptElement::dispatchErrorEvent
ENT WebCore::HTMLScriptElement::dispatchErrorEvent
EXT WebCore::HTMLScriptElement::dispatchErrorEvent
EXT WebCore::HTMLScriptElement::dispatchErrorEvent
EXT WebCore::HTMLScriptElement::dispatchErrorEvent
ENT WebCore::CachedResource::removeClient
ENT WebCore::CachedResource::removeClient
ENT WebCore::CachedResource::removeClient
ENT WebCore::CachedResource::removeClient
ENT WebCore::CachedResource::removeClient
ENT WTF::HashTableIterator
ENT std::pair
ENT std::pair
EXT std::pair
ENT std::pair
ENT std::pair
ENT std::pair
ENT std::pair
EXT std::pair
EXT std::pair
EXT std::pair
EXT std::pair
ENT std::pair
EXT std::pair
ENT std::pair
ENT std::pair
EXT std::pair
EXT std::pair
ENT std::pair
ENT std::pair
EXT std::pair
ENT std::pair
EXT std::pair
EXT std::pair
ENT std::pair
EXT std::pair
ENT std::pair
EXT std::pair
ENT std::pair
ENT std::pair
EXT std::pair
EXT std::pair
EXT std::pair
ENT WTF::HashTable
ENT WTF::HashTable
ENT WTF::HashTable
ENT WTF::HashTable
EXT WTF::HashTable
EXT WTF::HashTable
EXT WTF::HashTable
EXT WTF::HashTable
EXT WTF::HashTableIterator
EXT WebCore::CachedResource::removeClient
ENT WebCore::CachedResource::removeClient
EXT WebCore::CachedResource::removeClient
EXT WebCore::CachedResource::removeClient
EXT WebCore::CachedResource::removeClient
ENT WTF::HashCountedSet
ENT WTF::HashCountedSet
ENT WTF::HashCountedSet
ENT WTF::HashCountedSet
ENT WTF::HashTable
ENT WTF::HashTable
ENT WTF::HashTable
ENT WTF::HashTable
EXT WTF::HashTable
EXT WTF::HashTable
EXT WTF::HashTable
EXT WTF::HashTable
EXT WTF::HashCountedSet
ENT WTF::HashCountedSet
EXT WTF::HashCountedSet
EXT WTF::HashCountedSet
EXT WTF::HashCountedSet
ENT WTF::HashCountedSet
ENT WTF::HashTableIterator
ENT WTF::HashTableIterator
ENT WTF::HashTableIterator
EXT WTF::HashTableIterator
EXT WTF::HashTableIterator
EXT WTF::HashTableIterator
EXT WTF::HashCountedSet
ENT WTF::HashTableIteratorAdapter
ENT WTF::HashTableIteratorAdapter
ENT WTF::HashTableIteratorAdapter
ENT WTF::HashTableIteratorAdapter
ENT WTF::HashTableIteratorAdapter
EXT WTF::HashTableIteratorAdapter
EXT WTF::HashTableIteratorAdapter
EXT WTF::HashTableIteratorAdapter
EXT WTF::HashTableIteratorAdapter
EXT WTF::HashTableIteratorAdapter
ENT WTF::HashCountedSet
After this I dont see any call. So does any body has any idea what might be
the problem?
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
From stanislas.selle at gmail.com Tue Oct 4 05:19:27 2011
From: stanislas.selle at gmail.com (Stanislas Selle)
Date: Tue, 4 Oct 2011 14:19:27 +0200
Subject: [webkit-help] Call a javascript by the WebKitView GTK api
Message-ID:
Hello , how can i call a Javascript code , or fire a alert for instance , by
using the webkit port in GTK ?
i found webkit_dom_document_exec_command
but i didn't find a documentation about it , and i don't know to use it...
any help is welcome.
--
Stanislas
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
From yamazaki at fusenetwork.co.jp Tue Oct 4 17:34:27 2011
From: yamazaki at fusenetwork.co.jp (Kunio Yamazaki)
Date: Wed, 05 Oct 2011 09:34:27 +0900
Subject: [webkit-help] I need help in building webkit wincairo
In-Reply-To: <591B73CB-3460-417A-9078-A9118B1772E5@gmail.com>
References: <4E8983AE.6070605@fusenetwork.co.jp>
<4E8A6384.7040903@fusenetwork.co.jp>
<591B73CB-3460-417A-9078-A9118B1772E5@gmail.com>
Message-ID: <4E8BA613.4050901@fusenetwork.co.jp>
Hi Brent,
Thank you for your reply.
(2011/10/04 13:22), Brent Fulgham wrote:
>> $ cd /cygdrive/e
>> $ svn checkout --revision 96399 http://svn.webkit.org/repository/webkit/trunk WebKit
> Is /cygdrive/e your HOME location?
> I'm worried that the build failures are somehow related to the Japanese Windows localization.
/cygdrive/e is my HOME location.
webkit 533+ build, there was no problem.
> Are there any error messages in the earlier sections of the build
> (e.g., while building WTF/JavaScriptCore)?
No, the message had neither an error nor warning.
> I don't understand why this is causing trouble.
Really, it is regrettable.
The external symbol which incidentally became an error,
It seems that it is defined as WebKitQuatzCoreAdditions.lib
Would you teach around here, if some are understood?
Thanks,
-Kunio
From vsbhogle at gmail.com Tue Oct 4 23:12:52 2011
From: vsbhogle at gmail.com (Sumedh)
Date: Wed, 5 Oct 2011 11:42:52 +0530
Subject: [webkit-help] IWebDocumentView and IWebDocumentRepresentation for
Windows Apple Build
Message-ID:
Hi,
I am using WebKit apple port for windows, i have extended the WinLauncher
project. I wanted to write my own protocol handler for particular mime type,
lets say "image/". I have implemented IWebDocumentView and
IWebDocumentRepresentation. As required, i am passing the pointer to the
objects of the respective classes to the RegisterViewClass method of
IWebView object, now this method returns me E_NOTIMPL, which means the
method is not implemented, Do i need to implement IWebView also? Is there
anything extra to be done to write my own handler for a particular mime
type.
Meanwhile, my aim is to load all the resources from the memory, Idea is
something like this, all web pages, resources (js, images etc) resides in
some database, to be particular SQLITE. I want to read this resources and
pages from the blob, and render it on the WebView, from the memory. For
Internet Explorer, i have written similar kind of thing, implementing
IInternetProtocol, thanks to Igor. (
http://msdn.microsoft.com/en-us/library/aa767883%28v=vs.85%29.aspx). Am i
following the right approach for the WebKit.
Sample Code:
//Create WebView Instance
hr = WebKitCreateInstance(CLSID_WebView, 0, IID_IWebView,
(void**)&myWebView);
//Create IWebDocumentView instance
myWebDocumentView = new WinLauncherWebDocumentView();
myWebDocumentView->AddRef();
//Create IWebDocumentRepresentation Instance
myWebDocumentRepresentation = new
WinLauncherWebDocumentRepresentation();
myWebDocumentRepresentation->AddRef();
//Register the class
hr = myWebView->registerViewClass(myWebDocumentView,
myWebDocumentRepresentation, L"image/");
In this case hr is E_NOTIMPL
EIN? (Please thell me if any extra Information Needed)
Dhanyawaad (Thanks)
--
Namaskar,
Sumedh S Bhogle
Mumbai | Mob. 09969085122
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
From hazyboy at 126.com Wed Oct 5 02:14:52 2011
From: hazyboy at 126.com (=?GBK?B?z/K2qA==?=)
Date: Wed, 5 Oct 2011 17:14:52 +0800 (CST)
Subject: [webkit-help] how to breakpoint to debug for webkit project.
In-Reply-To: <32a88ba7.2b4b.132d3508f69.Coremail.hazyboy@126.com>
References: <32a88ba7.2b4b.132d3508f69.Coremail.hazyboy@126.com>
Message-ID: <7581b75b.2c88.132d35d3170.Coremail.hazyboy@126.com>
my os is : mac 10.7,
and webkit source is : R9**** version.
I Can debug breakpoint JavascriptCore, webcore, webkit2 for attach to safari webProcess, but only can't breakpoint debug in webkit project.
please tell me how to breakpoint debug in webkit project, thanks .
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
From stanislas.selle at gmail.com Wed Oct 5 04:08:42 2011
From: stanislas.selle at gmail.com (Stanislas Selle)
Date: Wed, 5 Oct 2011 13:08:42 +0200
Subject: [webkit-help] Call a javascript by the WebKitView GTK api
In-Reply-To:
References:
Message-ID:
i found the function :
webkit_web_view_execute_script (webkitwebview, gchar*)
2011/10/4 Stanislas Selle
> Hello , how can i call a Javascript code , or fire a alert for instance ,
> by using the webkit port in GTK ?
>
> i found webkit_dom_document_exec_command
>
> but i didn't find a documentation about it , and i don't know to use it...
>
> any help is welcome.
>
> --
> Stanislas
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
From haertel at equinux.de Wed Oct 5 04:54:46 2011
From: haertel at equinux.de (=?iso-8859-1?Q?Simon_H=E4rtel?=)
Date: Wed, 5 Oct 2011 13:54:46 +0200
Subject: [webkit-help] WebKit for multiple Mac OS versions
Message-ID:
Hello,
We are considering to ship a custom build or nightly build of WebKit with our application, which runs on Mac OS X 10.5 and later, to take advantage of the latest improvements to HTML editing (especially removal of Apple style spans). We noticed that the nightly builds come in different versions for each major OS version. Is it possible to get/build a WebKit which runs on all Mac OS versions starting with 10.5?
Thanks in advance!
Simon Haertel
From TELarson at west.com Wed Oct 5 06:27:08 2011
From: TELarson at west.com (Larson, Timothy E.)
Date: Wed, 5 Oct 2011 08:27:08 -0500
Subject: [webkit-help] WebKit for multiple Mac OS versions
In-Reply-To:
References:
Message-ID: <226316B3E1F749498E28ACA66321D5BA02B69D2DC5@oma00cexmbx03.corp.westworlds.com>
> We are considering to ship a custom build or nightly build of WebKit with
> our application, which runs on Mac OS X 10.5 and later, to take advantage
> of the latest improvements to HTML editing (especially removal of Apple
> style spans). We noticed that the nightly builds come in different
> versions for each major OS version. Is it possible to get/build a WebKit
> which runs on all Mac OS versions starting with 10.5?
Not speaking for WebKit, but I'd assume that the one for 10.5 would run on later versions as well. The tendency is to add more features as function calls to the OS, not remove them (which would break existing apps).
Tim
From aroben at apple.com Wed Oct 5 06:34:33 2011
From: aroben at apple.com (Adam Roben)
Date: Wed, 05 Oct 2011 09:34:33 -0400
Subject: [webkit-help] IWebDocumentView and IWebDocumentRepresentation
for Windows Apple Build
In-Reply-To:
References:
Message-ID:
On Oct 5, 2011, at 2:12 AM, Sumedh wrote:
> Hi,
>
> I am using WebKit apple port for windows, i have extended the WinLauncher project. I wanted to write my own protocol handler for particular mime type, lets say "image/". I have implemented IWebDocumentView and IWebDocumentRepresentation. As required, i am passing the pointer to the objects of the respective classes to the RegisterViewClass method of IWebView object, now this method returns me E_NOTIMPL, which means the method is not implemented, Do i need to implement IWebView also? Is there anything extra to be done to write my own handler for a particular mime type.
>
> Meanwhile, my aim is to load all the resources from the memory, Idea is something like this, all web pages, resources (js, images etc) resides in some database, to be particular SQLITE. I want to read this resources and pages from the blob, and render it on the WebView, from the memory. For Internet Explorer, i have written similar kind of thing, implementing IInternetProtocol, thanks to Igor. (http://msdn.microsoft.com/en-us/library/aa767883%28v=vs.85%29.aspx). Am i following the right approach for the WebKit.
>
> Sample Code:
>
> //Create WebView Instance
> hr = WebKitCreateInstance(CLSID_WebView, 0, IID_IWebView, (void**)&myWebView);
>
> //Create IWebDocumentView instance
> myWebDocumentView = new WinLauncherWebDocumentView();
> myWebDocumentView->AddRef();
>
> //Create IWebDocumentRepresentation Instance
> myWebDocumentRepresentation = new WinLauncherWebDocumentRepresentation();
> myWebDocumentRepresentation->AddRef();
>
> //Register the class
> hr = myWebView->registerViewClass(myWebDocumentView, myWebDocumentRepresentation, L"image/");
>
> In this case hr is E_NOTIMPL
>
> EIN? (Please thell me if any extra Information Needed)
The IWebView::registerViewClass function has never been implemented in WebKit. (See .)
If all you want to do is provide custom data for certain resources and have WebKit render the resulting data for you, you might want to investigate using IWebResourceLoadDelegate::willSendRequest to modify the URLs of those resources to use a custom URL scheme, and then implement a handler for your URL scheme that provides the desired data. URL handlers are a feature of whatever networking library you're using, not of WebKit itself, so the mechanism for implementing one depends on which networking library you're using.
If you really need to provide a custom view, you could investigate IWebUIDelegatePrivate::embeddedViewWithArguments and IWebEmbeddedView. This lets you supply a custom HWND to be used for rendering an or element.
-Adam
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
From aroben at apple.com Wed Oct 5 06:35:12 2011
From: aroben at apple.com (Adam Roben)
Date: Wed, 05 Oct 2011 09:35:12 -0400
Subject: [webkit-help] how to breakpoint to debug for webkit project.
In-Reply-To: <7581b75b.2c88.132d35d3170.Coremail.hazyboy@126.com>
References: <32a88ba7.2b4b.132d3508f69.Coremail.hazyboy@126.com>
<7581b75b.2c88.132d35d3170.Coremail.hazyboy@126.com>
Message-ID:
On Oct 5, 2011, at 5:14 AM, ?? wrote:
> my os is : mac 10.7,
> and webkit source is : R9**** version.
>
> I Can debug breakpoint JavascriptCore, webcore, webkit2 for attach to safari webProcess, but only can't breakpoint debug in webkit project.
>
> please tell me how to breakpoint debug in webkit project, thanks .
Safari on OS X Lion doesn't use WebKit for rendering web pages, it uses WebKit2. So it's not surprising that your breakpoints in WebKit aren't being hit; the code is not being called!
-Adam
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
From aroben at apple.com Wed Oct 5 06:37:41 2011
From: aroben at apple.com (Adam Roben)
Date: Wed, 05 Oct 2011 09:37:41 -0400
Subject: [webkit-help] WebKit for multiple Mac OS versions
In-Reply-To:
References:
Message-ID: <048331AA-9C38-4048-BFAA-DAF40C3E67AB@apple.com>
On Oct 5, 2011, at 7:54 AM, Simon H?rtel wrote:
> We are considering to ship a custom build or nightly build of WebKit with our application, which runs on Mac OS X 10.5 and later, to take advantage of the latest improvements to HTML editing (especially removal of Apple style spans). We noticed that the nightly builds come in different versions for each major OS version. Is it possible to get/build a WebKit which runs on all Mac OS versions starting with 10.5?
Apple's Mac port of WebKit decides at compile-time what OS features to use. If you wanted to do something like you describe, you'd have to change it to decide a run-time instead. That would be quite a large project.
You'd also need to be sure that your custom build of WebKit doesn't introduce any naming conflicts with the system version (e.g., duplicate class names, etc.). That is also a challenging task.
-Adam
From rniwa at webkit.org Wed Oct 5 10:53:55 2011
From: rniwa at webkit.org (Ryosuke Niwa)
Date: Wed, 5 Oct 2011 10:53:55 -0700
Subject: [webkit-help] WebKit for multiple Mac OS versions
Message-ID:
As far as I know Chrmium Mac uses the same binary for 10.5 and 10.6 so you
might be interested in looking into that.
Also, if you're shipping ToT WebKit, just be aware that Apple-style-span
removal has introduced a regression bug
https://bugs.webkit.org/show_bug.cgi?id=68168, which I'm hoping to fix this
week.
- Ryosuke
On Oct 5, 2011 4:54 AM, "Simon H?rtel" wrote:
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
From yamazaki at fusenetwork.co.jp Thu Oct 6 18:35:37 2011
From: yamazaki at fusenetwork.co.jp (Kunio Yamazaki)
Date: Fri, 07 Oct 2011 10:35:37 +0900
Subject: [webkit-help] I need help in building webkit wincairo
In-Reply-To: <4E8E5637.9070807@fusenetwork.co.jp>
References: <4E8E5637.9070807@fusenetwork.co.jp>
Message-ID: <4E8E5769.7020501@fusenetwork.co.jp>
Hi Adam,
(2011/10/04 0:19), Adam Roben wrote:
> On Oct 3, 2011, at 5:43 AM, Kunio Yamazaki wrote:
>
> Something is causing the WinCairo build to try to link against WebKitQuartzCoreAdditions.lib.
The place which was being investigated about the link error of WebKit wincario build,
In /Source/WebKit2/DerivedSources.make,
If ./include/WebKitQuartzCoreAdditions/WebKitQuartzCoreAdditionsBase.h is contained in WebKitSupportLibrary,
#define HAVE_WKQCA 1 is defined as HeaderDetection.h.
The newest WebKitSupportLibrary was checked and WebKitQuartzCoreAdditionsBase.h was contained.
It checked that "#define HAVE_WKQCA 1" was defined with HeaderDetection.h.
Then, WebKitSupportLibrary which was being used in the past checked,
WebKitQuartzCoreAdditionsBase.h is not contained, and HeaderDetection.h was empty.
Building at this time was successful.
Although I think that this caused the link error, how is it?
Then, It builds by deleting WebKitQuartzCoreAdditionsBase.h.
About a result, I think that I will report later.
Thanks,
-Kunio
From bfulgham at gmail.com Thu Oct 6 21:27:38 2011
From: bfulgham at gmail.com (Brent Fulgham)
Date: Thu, 6 Oct 2011 21:27:38 -0700
Subject: [webkit-help] I need help in building webkit wincairo
In-Reply-To: <4E8E5769.7020501@fusenetwork.co.jp>
References: <4E8E5637.9070807@fusenetwork.co.jp>
<4E8E5769.7020501@fusenetwork.co.jp>
Message-ID: <07EC2411-DCA9-428E-9BEC-EC6C6CEBD37C@gmail.com>
Hi,
Thanks for figuring this out!
> (2011/10/04 0:19), Adam Roben wrote:
>> On Oct 3, 2011, at 5:43 AM, Kunio Yamazaki wrote:
>>
>> Something is causing the WinCairo build to try to link against WebKitQuartzCoreAdditions.lib.
>
> The place which was being investigated about the link error of WebKit wincario build,
> In /Source/WebKit2/DerivedSources.make,
> If ./include/WebKitQuartzCoreAdditions/WebKitQuartzCoreAdditionsBase.h is contained in WebKitSupportLibrary,
> #define HAVE_WKQCA 1 is defined as HeaderDetection.h.
>
> The newest WebKitSupportLibrary was checked and WebKitQuartzCoreAdditionsBase.h was contained.
> It checked that "#define HAVE_WKQCA 1" was defined with HeaderDetection.h.
I've filed a bug (https://bugs.webkit.org/show_bug.cgi?id=69600) about this. I was not seeing the error because I hadn't been good about staying up to date with the Apple webkit support libraries.
When I cleared out everything on my build machine today, I started getting the error, too.
You can obviously work around this issue by renaming that file.
Thanks,
-Brent
From yamazaki at fusenetwork.co.jp Thu Oct 6 21:46:16 2011
From: yamazaki at fusenetwork.co.jp (Kunio Yamazaki)
Date: Fri, 07 Oct 2011 13:46:16 +0900
Subject: [webkit-help] I need help in building webkit wincairo
In-Reply-To: <07EC2411-DCA9-428E-9BEC-EC6C6CEBD37C@gmail.com>
References: <4E8E5637.9070807@fusenetwork.co.jp>
<4E8E5769.7020501@fusenetwork.co.jp>
<07EC2411-DCA9-428E-9BEC-EC6C6CEBD37C@gmail.com>
Message-ID: <4E8E8418.4070403@fusenetwork.co.jp>
Hi Brent,
(2011/10/07 13:27), Brent Fulgham wrote:
> Thanks for figuring this out!
I am glad to could you help.
>> The newest WebKitSupportLibrary was checked and WebKitQuartzCoreAdditionsBase.h was contained.
>> It checked that "#define HAVE_WKQCA 1" was defined with HeaderDetection.h.
>
> I've filed a bug (https://bugs.webkit.org/show_bug.cgi?id=69600) about this. I was not seeing the error because I hadn't been good about staying up to date with the Apple webkit support libraries.
> When I cleared out everything on my build machine today, I started getting the error, too.
>
> You can obviously work around this issue by renaming that file.
Yes,
I renamed WebKitQuartzCoreAdditionsBase.h to _WebKitQuartzCoreAdditionsBase.h
and built WebKit wincairo.
As a result, it succeeded in construction!
Thanks to people which gave advice.
-Kunio
From sanachan.y at gmail.com Fri Oct 7 08:41:17 2011
From: sanachan.y at gmail.com (Y.Sanachan)
Date: Sat, 08 Oct 2011 00:41:17 +0900
Subject: [webkit-help] Which SYNTAX_ERR should be thrown from
WorkerContext::importScripts()
Message-ID: <4E8F1D9D.4020208@gmail.com>
Hi there,
In WebCore::WorkerContext::importScripts() method,
it will throw an exception code of SYNTAX_ERR from the 1st for loop.
When ENABLE_FILE_SYSTEM=0,
I met the compile error as like 'SYNTAX_ERR is not defined'.
Can I include dom/ExceptionCode.h instead of the header files relating
to FILE_SYSTEM API ?
OR, should it be set to 1 when compile WorkerContext.cpp ?
Thanks in advance,
Sanachan.
From justin at sweaky.com Fri Oct 7 19:00:39 2011
From: justin at sweaky.com (Justin Thomas)
Date: Fri, 07 Oct 2011 19:00:39 -0700
Subject: [webkit-help] Linker issues with jsc
Message-ID: <4E8FAEC7.90403@sweaky.com>
Hi All,
I have been working on getting JavaScriptCore building and running but
trying to use a different setup. My goal, is to get it building on
Windows, either through visual studio or cygwin (or both) using
glib/gobject libs instead of libicuxx libs and without core foundation.
Currently, I have been able to get it pretty far, I've branched my own
projects and vsprops, removed the libicu references and added in the
glib references/includes into the vsprops, defined ENABLE_GLIB_SUPPORT,
removed the icu headers/code files from the vs project file and added in
the glib/gobject ones. I've defined in unicodeglib.h UChar and UChar32
as guint16 and guint32 (not sure what is was before, maybe I just
circled back around). I have also defined JSChar as unsigned short (even
wchar_t), after fixing a few compile errors I've managed to get it to
begin linking, but am stuck on some (probably quite trivial) problems,
build output:
3>Linking...
3>JSByteArray.obj : warning LNK4197: export
'?s_info at JSByteArray@JSC@@2UClassInfo at 2@B' specified multiple times;
using first specification
3> Creating library
G:\webkit-latest\WebKitBuild\Debug\lib\JavaScriptCore.lib and object
G:\webkit-latest\WebKitBuild\Debug\lib\JavaScriptCore.exp
3>JavaScriptCore.exp : error LNK2001: unresolved external symbol
"public: __thiscall JSC::UString::UString(wchar_t const *)"
(??0UString at JSC@@QAE at PB_W@Z)
3>JavaScriptCore.exp : error LNK2001: unresolved external symbol
"public: __thiscall JSC::UString::UString(wchar_t const *,unsigned int)"
(??0UString at JSC@@QAE at PB_WI@Z)
3>JavaScriptCore.exp : error LNK2001: unresolved external symbol
"public: void __thiscall WTF::StringBuilder::append(wchar_t const
*,unsigned int)" (?append at StringBuilder@WTF@@QAEXPB_WI at Z)
3>JavaScriptCore.exp : error LNK2001: unresolved external symbol
"public: enum WTF::Collator::Result __thiscall
WTF::Collator::collate(wchar_t const *,unsigned int,wchar_t const
*,unsigned int)const " (?collate at Collator@WTF@@QBE?AW4Result at 12@PB_WI0I at Z)
3>JavaScriptCore.exp : error LNK2001: unresolved external symbol "enum
WTF::Unicode::ConversionResult __cdecl
WTF::Unicode::convertUTF16ToUTF8(wchar_t const * *,wchar_t const *,char
* *,char *,bool)"
(?convertUTF16ToUTF8 at Unicode@WTF@@YA?AW4ConversionResult at 12@PAPB_WPB_WPAPADPAD_N at Z)
3>JavaScriptCore.exp : error LNK2001: unresolved external symbol "enum
WTF::Unicode::ConversionResult __cdecl
WTF::Unicode::convertUTF8ToUTF16(char const * *,char const *,wchar_t *
*,wchar_t *,bool)"
(?convertUTF8ToUTF16 at Unicode@WTF@@YA?AW4ConversionResult at 12@PAPBDPBDPAPA_WPA_W_N at Z)
3>JavaScriptCore.exp : error LNK2001: unresolved external symbol "bool
__cdecl WTF::Unicode::equalUTF16WithUTF8(wchar_t const *,wchar_t const
*,char const *,char const *)"
(?equalUTF16WithUTF8 at Unicode@WTF@@YA_NPB_W0PBD1 at Z)
3>JavaScriptCore.exp : error LNK2001: unresolved external symbol
"public: unsigned int __thiscall
WTF::DecimalNumber::toStringDecimal(wchar_t *,unsigned int)const "
(?toStringDecimal at DecimalNumber@WTF@@QBEIPA_WI at Z)
3>JavaScriptCore.exp : error LNK2001: unresolved external symbol
"public: unsigned int __thiscall
WTF::DecimalNumber::toStringExponential(wchar_t *,unsigned int)const "
(?toStringExponential at DecimalNumber@WTF@@QBEIPA_WI at Z)
3>G:\webkit-latest\WebKitBuild\Debug\bin\JavaScriptCore.dll : fatal
error LNK1120: 9 unresolved externals
Where wchar_t is specified, in the code those are UChar's, which are not
defined as wchar_t (I have it defined as unsigned shorts, even maybe vsc
is just assuming since it is ushort that it's the same as wchar_t?),
I've been at this for a couple days now trying to figure out why it's
looking for these signatures. I've also tried configuring wchar as
built-in type and not with no difference.
As a side note, which I don't think is related but might be, while
building for vs it seems nmake would hang while generating the
KeywordLookup and RexExJit tables, so I ended up just running those
myself through cygwin (just as DerivedSources.make does) and copy them
over during each clean, a bit wonky but again right now I'm just trying
to see if I can get a viable build going using just glib.
Any help, pointers, etc.. would be appreciated.
Thanks,
-JT
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
From wenmhappy at 126.com Sat Oct 8 01:43:43 2011
From: wenmhappy at 126.com (wenmhappy)
Date: Sat, 8 Oct 2011 16:43:43 +0800 (CST)
Subject: [webkit-help] build error of webkit-r29506
Message-ID: <49a3d196.169ed.132e2b3bfd3.Coremail.wenmhappy@126.com>
Hi everybody,
I have successfully built(cross compile) webkit-r59711 on embeded platform with script below.
export BUILD_WEBKIT_ARGS="--host=sh4-linux --build=i686-pc-linux-gnu \
--prefix=/opt/STM/STLinux-2.3/devkit/sh4/target/usr "
./WebKitTools/Scripts/build-webkit --gtk --release --makeargs="2> buildlog"
but failed to build webkit-r29506 with script below.
./WebKitTools/Scripts/build-webkit --gtk --release --host=sh4-linux --build=i686-pc-linux-gnu --prefix=/opt/STM/STLinux-2.3/devkit/sh4/target/usr --with-target=directfb
error is:
Can't exec "qmake": No such file or directory at /home/wen/WebKit-r29506/WebKitTools/Scripts/webkitdirs.pm line 693.
Use of uninitialized value $mkspec in scalar chomp at /home/wen/WebKit-r29506/WebKitTools/Scripts/webkitdirs.pm line 693.
Calling 'qmake CONFIG+=gtk-port CONFIG-=qt -r OUTPUT_DIR=/home/wen/WebKit-r29506/WebKitBuild/Release /home/wen/WebKit-r29506/WebKit.pro CONFIG+=release CONFIG-=debug' in /home/wen/WebKit-r29506/WebKitBuild/Release
Can't exec "qmake": No such file or directory at /home/wen/WebKit-r29506/WebKitTools/Scripts/webkitdirs.pm line 765.
Failed to setup build environment using qmake!
I found there are different between two build-webkit scripts, i don`t know how to fix it.
Thanks for help.
Best regards
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
From wenmhappy at 126.com Sat Oct 8 21:34:33 2011
From: wenmhappy at 126.com (wenmhappy)
Date: Sun, 9 Oct 2011 12:34:33 +0800 (CST)
Subject: [webkit-help] Problem about DerivedSources of WebKit
Message-ID: <681d82ee.468f.132e6f5fdd6.Coremail.wenmhappy@126.com>
Hi everybody,
I wonder that why are there so many derived sourece code. How are they generated automatically? Could they be modified?
Best regards
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
From vsbhogle at gmail.com Sun Oct 9 22:42:29 2011
From: vsbhogle at gmail.com (Sumedh)
Date: Mon, 10 Oct 2011 11:12:29 +0530
Subject: [webkit-help] IWebDocumentView and IWebDocumentRepresentation
for Windows Apple Build -> Implementing url handler for CFNetwork
Message-ID:
Hi,
Thanks Roben for clearing the concept.
We want to provide custom data and then let webkit render it on the view. I
guess we are using CFNetwork dll for networking. Can you guys please guide
me how to implement url handler for custom scheme. It will be great help, if
you direct me somewhere.
Dhnayawaad(Thanks)
On Wed, Oct 5, 2011 at 7:04 PM, Adam Roben wrote:
>
> On Oct 5, 2011, at 2:12 AM, Sumedh wrote:
>
> Hi,
>
> I am using WebKit apple port for windows, i have extended the WinLauncher
> project. I wanted to write my own protocol handler for particular mime type,
> lets say "image/". I have implemented IWebDocumentView and
> IWebDocumentRepresentation. As required, i am passing the pointer to the
> objects of the respective classes to the RegisterViewClass method of
> IWebView object, now this method returns me E_NOTIMPL, which means the
> method is not implemented, Do i need to implement IWebView also? Is there
> anything extra to be done to write my own handler for a particular mime
> type.
>
> Meanwhile, my aim is to load all the resources from the memory, Idea is
> something like this, all web pages, resources (js, images etc) resides in
> some database, to be particular SQLITE. I want to read this resources and
> pages from the blob, and render it on the WebView, from the memory. For
> Internet Explorer, i have written similar kind of thing, implementing
> IInternetProtocol, thanks to Igor. (
> http://msdn.microsoft.com/en-us/library/aa767883%28v=vs.85%29.aspx). Am i
> following the right approach for the WebKit.
>
> Sample Code:
>
> //Create WebView Instance
> hr = WebKitCreateInstance(CLSID_WebView, 0, IID_IWebView,
> (void**)&myWebView);
>
> //Create IWebDocumentView instance
> myWebDocumentView = new WinLauncherWebDocumentView();
> myWebDocumentView->AddRef();
>
> //Create IWebDocumentRepresentation Instance
> myWebDocumentRepresentation = new
> WinLauncherWebDocumentRepresentation();
> myWebDocumentRepresentation->AddRef();
>
> //Register the class
> hr = myWebView->registerViewClass(myWebDocumentView,
> myWebDocumentRepresentation, L"image/");
>
> In this case hr is E_NOTIMPL
>
> EIN? (Please thell me if any extra Information Needed)
>
>
> The IWebView::registerViewClass function has never been implemented in
> WebKit. (See <
> http://trac.webkit.org/browser/trunk/Source/WebKit/win/WebView.cpp?rev=96549#L3489
> >.)
>
> If all you want to do is provide custom data for certain resources and have
> WebKit render the resulting data for you, you might want to investigate
> using IWebResourceLoadDelegate::willSendRequest to modify the URLs of those
> resources to use a custom URL scheme, and then implement a handler for your
> URL scheme that provides the desired data. URL handlers are a feature of
> whatever networking library you're using, not of WebKit itself, so the
> mechanism for implementing one depends on which networking library you're
> using.
>
> If you really need to provide a custom view, you could investigate
> IWebUIDelegatePrivate::embeddedViewWithArguments and IWebEmbeddedView. This
> lets you supply a custom HWND to be used for rendering an or
> element.
>
> -Adam
>
>
--
Namaskar,
Sumedh S Bhogle
Mumbai | Mob. 09969085122
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
From just2contribute at gmail.com Tue Oct 11 01:33:42 2011
From: just2contribute at gmail.com (just2 contribute)
Date: Tue, 11 Oct 2011 14:03:42 +0530
Subject: [webkit-help] Reg: Can anybody clarify me the procedure for
assigning a bug to myself !
In-Reply-To:
References:
Message-ID:
Hi,
I am a newbie to webkit and just understanding the whole developer
eco-system.
Is it like the bugs in the *NEW* state are those ones that are not been
confirmed as real bugs in webkit?
How and when it gets assigned to a particular assignee and who assigns it?
B'coz even after having an bugzilla account I am unable to modify the
statuses, in case I want to assign a bug to myself.
I went thru' the page, http://www.webkit.org/quality/lifecycle.html but
still I am not very clear on these.
-
Justin
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
From ssseintr2 at gmail.com Tue Oct 11 01:59:53 2011
From: ssseintr2 at gmail.com (Vicky Tux)
Date: Tue, 11 Oct 2011 14:29:53 +0530
Subject: [webkit-help] Reg. Plugin Loading Issue...
Message-ID:
hi all,
any one have solution for https://bugs.webkit.org/show_bug.cgi?id=68072
regards,
vicky.
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
From justin at sweaky.com Tue Oct 11 02:35:24 2011
From: justin at sweaky.com (Justin Thomas)
Date: Tue, 11 Oct 2011 02:35:24 -0700
Subject: [webkit-help] Reg. Plugin Loading Issue...
In-Reply-To:
References:
Message-ID: <4E940DDC.4000702@sweaky.com>
If your intentions are having it not interfere with layout, can you not
just get away with an html/css workaround?
1. set width/height to 0 (or even 1x1) and visibility:hidden
2. set position absolute with negative z-index and visibility
3. embed in a iframe
or any combination of the above...
-JT
On 10/11/2011 1:59 AM, Vicky Tux wrote:
> hi all,
>
> any one have solution for https://bugs.webkit.org/show_bug.cgi?id=68072
>
> regards,
> vicky.
>
>
>
> _______________________________________________
> webkit-help mailing list
> webkit-help at lists.webkit.org
> http://lists.webkit.org/mailman/listinfo.cgi/webkit-help
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
From leanderbb at gmail.com Tue Oct 11 10:09:27 2011
From: leanderbb at gmail.com (Leander Bessa)
Date: Tue, 11 Oct 2011 18:09:27 +0100
Subject: [webkit-help] Help removing -Werror flags from the chromium port
Message-ID:
Hello,
I'm currently trying to build chromium's port of webkit on Fedora 15 (64
bit) and the build keeps stopping due to warnings being treated as errors.
Is there any way to instruct the build-webkit script to remove the flag
"-Werror" from the final build?
- Leander
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
From Ballabas.Gabor at stud.u-szeged.hu Tue Oct 11 10:28:59 2011
From: Ballabas.Gabor at stud.u-szeged.hu (Ballabas.Gabor at stud.u-szeged.hu)
Date: Tue, 11 Oct 2011 19:28:59 +0200
Subject: [webkit-help] Help removing -Werror flags from the chromium port
In-Reply-To:
References:
Message-ID: <20111011192859.rsis3thx8g8s4k8o@www.stud.u-szeged.hu>
Quoting Leander Bessa :
> Hello,
>
> I'm currently trying to build chromium's port of webkit on Fedora 15 (64
> bit) and the build keeps stopping due to warnings being treated as errors.
>
> Is there any way to instruct the build-webkit script to remove the flag
> "-Werror" from the final build?
>
>
> - Leander
>
Hi,
You can use gyp to turn off this behavior like this:
./build/gyp_chromium -Dwerror=
(There is nothing behind the = sign.)
Best regards,
Gabor
----------------------------------------------------------------
This message was sent using IMP, the Internet Messaging Program.
From rniwa at webkit.org Tue Oct 11 10:59:34 2011
From: rniwa at webkit.org (Ryosuke Niwa)
Date: Tue, 11 Oct 2011 10:59:34 -0700
Subject: [webkit-help] Reg: Can anybody clarify me the procedure for
assigning a bug to myself !
In-Reply-To:
References:
Message-ID:
On Tue, Oct 11, 2011 at 1:33 AM, just2 contribute wrote:
>
> I am a newbie to webkit and just understanding the whole developer
> eco-system.
> Is it like the bugs in the *NEW* state are those ones that are not been
> confirmed as real bugs in webkit?
>
NEW bugs are ones that have been confirmed to be bugs. UNCONFIRMED are bugs
that have not been confirmed as bugs.
> How and when it gets assigned to a particular assignee and who assigns it?
> B'coz even after having an bugzilla account I am unable to modify the
> statuses, in case I want to assign a bug to myself.
>
In general, you assign a bug to yourself if you're working on the bug. You
can't modify bugs until you become a committer so you should ask someone
else to assign it to you. But you can still work on a bug regardless of
whether it's assigned to you or not.
- Ryosuke
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
From per.bothner at oracle.com Tue Oct 11 17:40:20 2011
From: per.bothner at oracle.com (Per Bothner)
Date: Tue, 11 Oct 2011 17:40:20 -0700
Subject: [webkit-help] how to get ExecState from ScriptExecutionContext
Message-ID: <4E94E1F4.3070800@oracle.com>
In our WebNode implementation (part of JavaFX) we have a class
JavaEventListener, which extends WebCore::EventListener.
The virtual handleEvent method takes a ScriptExecutionContext*
and an Event*, as in the parent class. This methods needs to map the
WebKit Event to a Java Event.
The Java Event object needs access to the ExecState. (The reason is
that the Java Event object has methods like getSrcElement, which
return a Java Node object, and for convenience this Node object also
has access to the corresponding JavaScript object. The various
JavaScript operations make use of JSGlobalContextRef which is the
same as ExecState.)
So the question is there an approved way to get from a
ScriptExecutionContext
to an ExecState/CallFrame/JSGlobalContextRef? I've looked through the
source code, but haven't found the solution.
--
--Per Bothner
per.bothner at oracle.com per at bothner.com http://per.bothner.com/
From stephenmattison at gmail.com Tue Oct 11 18:39:19 2011
From: stephenmattison at gmail.com (Stephen Mattison)
Date: Tue, 11 Oct 2011 18:39:19 -0700
Subject: [webkit-help] Background PNG decoding in WebKit
In-Reply-To:
References:
Message-ID:
The problem I'm having is that it does not appear that PNGs are decoded in
the background in Webkit on iOS. Right now what I have is essentially a
ScrollView. I have a div the size of the viewport with another div that
scales with the content. I am scrolling the content div using a transition
and translate3d to achieve GPU acceleration. The scrolling looks great.
However the content is composed of images and they cannot all fit in memory
at once so I am loading them on demand. The problem arises when I set the
background-image css property to a locally stored PNG image and then attempt
to scroll. I get a noticeable delay/jerkiness/hiccup as the image is
decoded. Is there something I should be doing differently?
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
From ssseintr2 at gmail.com Wed Oct 12 00:06:19 2011
From: ssseintr2 at gmail.com (Vicky Tux)
Date: Wed, 12 Oct 2011 12:36:19 +0530
Subject: [webkit-help] webkit-help Digest, Vol 28, Issue 12
In-Reply-To:
References:
Message-ID:
Hi,
Sorry for not explaining my context.
Acutally I'm not a Web Developer. We are using webkit as part of
our middle-ware. Some of the web applications authored to hide the plugin
from layout as like below,
Also those plugins are scripted from Java Script. In the above case webkit
is doesn't initialize plugins if it's(or parent's) style is display : none.
Currently webkit's plugin initialization is based on RenderTree not DOMTree.
To fix this issue I thinking to implement a RenderNone class(as like
RenderBlock,RenderInline,..etc) which paints nothing but just acts as a
container for child elements.
But i don't have any idea of RenderTree. Currently implementing RenderNone
based on RenderBlock & RenderInline.
Will it work..? any suggestions..?
On Tue, Oct 11, 2011 at 7:30 PM, Justin Thomas wrote:
>
> If your intentions are having it not interfere with layout, can you not
> just get away with an html/css workaround?
>
> 1. set width/height to 0 (or even 1x1) and visibility:hidden
> 2. set position absolute with negative z-index and visibility
> 3. embed in a iframe
>
> or any combination of the above...
>
> -JT
>
> On 10/11/2011 1:59 AM, Vicky Tux wrote:
> > hi all,
> >
> > any one have solution for https://bugs.webkit.org/show_bug.cgi?id=68072
> >
> > regards,
> > vicky.
> >
> >
> >
> > _______________________________________________
> > webkit-help mailing list
> > webkit-help at lists.webkit.org
> > http://lists.webkit.org/mailman/listinfo.cgi/webkit-help
> -------------- next part --------------
> An HTML attachment was scrubbed...
> URL: <
> http://lists.webkit.org/pipermail/webkit-help/attachments/20111011/07d93d76/attachment-0001.html
> >
>
> ------------------------------
>
> _______________________________________________
> webkit-help mailing list
> webkit-help at lists.webkit.org
> http://lists.webkit.org/mailman/listinfo.cgi/webkit-help
>
>
> End of webkit-help Digest, Vol 28, Issue 12
> *******************************************
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
From zoltan at webkit.org Wed Oct 12 03:01:48 2011
From: zoltan at webkit.org (Zoltan Horvath)
Date: Wed, 12 Oct 2011 12:01:48 +0200
Subject: [webkit-help] Background PNG decoding in WebKit
In-Reply-To:
References:
Message-ID:
On Wed, 12 Oct 2011 03:39:19 +0200, Stephen Mattison
wrote:
> The problem I'm having is that it does not appear that PNGs are decoded
> in
> the background in Webkit on iOS. Right now what I have is essentially a
> ScrollView. I have a div the size of the viewport with another div that
> scales with the content. I am scrolling the content div using a
> transition
> and translate3d to achieve GPU acceleration. The scrolling looks great.
> However the content is composed of images and they cannot all fit in
> memory
> at once so I am loading them on demand. The problem arises when I set the
> background-image css property to a locally stored PNG image and then
> attempt
> to scroll. I get a noticeable delay/jerkiness/hiccup as the image is
> decoded. Is there something I should be doing differently?
Hey!
WebKit is decoding images in the mainthread when there is new chunk of
image and it hasn't decoded yet, thus it is imaginable that it blocks your
main thread for 'longer' time.
I have a sketchy implementation for Qt to support imagedecoding in a
(currently 1) separate thread.
If you are interested check: https://bugs.webkit.org/show_bug.cgi?id=69515
Regards,
Zoltan
From just2contribute at gmail.com Wed Oct 12 03:04:00 2011
From: just2contribute at gmail.com (just2 contribute)
Date: Wed, 12 Oct 2011 15:34:00 +0530
Subject: [webkit-help] Reg: Build error in WebCore
Message-ID:
Hi,
I am getting the following error while building webcore on windows platform.
I guess some header file is missing !
6>..\platform\text\cf\HyphenationCF.cpp(43) : error C3861:
'CFStringIsHyphenationAvailableForLocale': identifier not found
May I know which is the library containing the definition of this?
--
Justin
From AYanov at luxoft.com Wed Oct 12 03:28:39 2011
From: AYanov at luxoft.com (Yanov, Andrey)
Date: Wed, 12 Oct 2011 14:28:39 +0400
Subject: [webkit-help] Reg: Build error in WebCore
In-Reply-To:
References:
Message-ID:
Looks like MAC CoreFoundation framework:
http://developer.apple.com/library/mac/#releasenotes/CoreFoundation/CoreFoundation.html
-----Original Message-----
From: webkit-help-bounces at lists.webkit.org [mailto:webkit-help-bounces at lists.webkit.org] On Behalf Of just2 contribute
Sent: Wednesday, October 12, 2011 1:04 PM
To: webkit-help at lists.webkit.org
Subject: [webkit-help] Reg: Build error in WebCore
Hi,
I am getting the following error while building webcore on windows platform.
I guess some header file is missing !
6>..\platform\text\cf\HyphenationCF.cpp(43) : error C3861:
'CFStringIsHyphenationAvailableForLocale': identifier not found
May I know which is the library containing the definition of this?
--
Justin
_______________________________________________
webkit-help mailing list
webkit-help at lists.webkit.org
http://lists.webkit.org/mailman/listinfo.cgi/webkit-help
From aroben at apple.com Wed Oct 12 04:36:34 2011
From: aroben at apple.com (Adam Roben)
Date: Wed, 12 Oct 2011 07:36:34 -0400
Subject: [webkit-help] webkit-help Digest, Vol 28, Issue 12
In-Reply-To:
References:
Message-ID: <39493C9A-7631-4A69-8473-C5848B3385C6@apple.com>
On Oct 12, 2011, at 3:06 AM, Vicky Tux wrote:
> Acutally I'm not a Web Developer. We are using webkit as part of our middle-ware. Some of the web applications authored to hide the plugin from layout as like below,
>
>
>
>
>
>
> Also those plugins are scripted from Java Script. In the above case webkit is doesn't initialize plugins if it's(or parent's) style is display : none.
>
> Currently webkit's plugin initialization is based on RenderTree not DOMTree. To fix this issue I thinking to implement a RenderNone class(as like RenderBlock,RenderInline,..etc) which paints nothing but just acts as a container for child elements.
>
> But i don't have any idea of RenderTree. Currently implementing RenderNone based on RenderBlock & RenderInline.
>
> Will it work..? any suggestions..?
Do any other browsers do this? If not, this seems like a potential compatibility issue.
If you want to hide plugins visually bug still have them run I'd suggest using visibility:hidden instead of display:none.
-Adam
From aroben at apple.com Wed Oct 12 04:42:04 2011
From: aroben at apple.com (Adam Roben)
Date: Wed, 12 Oct 2011 07:42:04 -0400
Subject: [webkit-help] Reg: Build error in WebCore
In-Reply-To:
References:
Message-ID: <4D3ABE5C-A3F0-4422-BDB7-CBEE8D08DFAC@apple.com>
On Oct 12, 2011, at 6:04 AM, just2 contribute wrote:
> I am getting the following error while building webcore on windows platform.
>
> I guess some header file is missing !
>
>
> 6>..\platform\text\cf\HyphenationCF.cpp(43) : error C3861:
> 'CFStringIsHyphenationAvailableForLocale': identifier not found
>
> May I know which is the library containing the definition of this?
That function is part of CoreFoundation's CFString API. CoreFoundation headers are included in WebKitSupportLibrary, which http://www.webkit.org/building/checkout.html tells you how to install.
-Adam
From julien.chaffraix at gmail.com Wed Oct 12 08:24:09 2011
From: julien.chaffraix at gmail.com (Julien Chaffraix)
Date: Wed, 12 Oct 2011 08:24:09 -0700
Subject: [webkit-help] build error of webkit-r29506
In-Reply-To: <49a3d196.169ed.132e2b3bfd3.Coremail.wenmhappy@126.com>
References: <49a3d196.169ed.132e2b3bfd3.Coremail.wenmhappy@126.com>
Message-ID:
> ??? I have successfully built(cross compile)?webkit-r59711 on embeded
> platform with script below.
[snip]
> but failed to build webkit-r29506 with script below.
r29506 is 3 years old. Apart from the (security) risk of using such an
ancient revision, it looks like the build script is not doing what it
should:
> ./WebKitTools/Scripts/build-webkit --gtk --release --host=sh4-linux
> --build=i686-pc-linux-gnu
> --prefix=/opt/STM/STLinux-2.3/devkit/sh4/target/usr --with-target=directfb
> error is:
> Can't exec "qmake": No such file or directory at
> /home/wen/WebKit-r29506/WebKitTools/Scripts/webkitdirs.pm line 693.
gtk does not use qmake (IIRC it used to in the beginning of the gtk
port but then switched to the auto-tools).
> I? found there are different between two build-webkit scripts,?? i don`t
> know how to fix it.
I would advise against using such an old revision but if you do, you
are on your own.
Hope it helps,
Julien
From julien.chaffraix at gmail.com Wed Oct 12 09:44:17 2011
From: julien.chaffraix at gmail.com (Julien Chaffraix)
Date: Wed, 12 Oct 2011 09:44:17 -0700
Subject: [webkit-help] Which SYNTAX_ERR should be thrown from
WorkerContext::importScripts()
In-Reply-To: <4E8F1D9D.4020208@gmail.com>
References: <4E8F1D9D.4020208@gmail.com>
Message-ID:
> In WebCore::WorkerContext::importScripts() method,
> it will throw an exception code of SYNTAX_ERR from the 1st for loop.
>
> When ENABLE_FILE_SYSTEM=0,
> I met the compile error as like 'SYNTAX_ERR is not defined'.
> Can I include dom/ExceptionCode.h instead of the header files relating
> to FILE_SYSTEM API ?
This is the right solution. Care to file a bug and attach a patch to
solve the problem for everyone?
> OR, should it be set to 1 when compile WorkerContext.cpp ?
No, Workers and Filesystem API are orthogonal.
Thanks,
Julien
From ssseintr2 at gmail.com Wed Oct 12 10:16:32 2011
From: ssseintr2 at gmail.com (Vicky Tux)
Date: Wed, 12 Oct 2011 22:46:32 +0530
Subject: [webkit-help] Reg. Plugin Loading issue when "display: none".
Message-ID:
>Do any other browsers do this? If not, this seems like a potential
compatibility issue.
Yes. Some of the browsers like ANT & Netfront supports plugin loading.with
display:none.
>If you want to hide plugins visually bug still have them run I'd suggest
using visibility:hidden instead of >display:none.
Actually I'm not a web developer. Some web applications are authored to load
plugins with "display:none" to act as a audio player by implementing it's
GUI using HTML/JS/CSS.
-Vicky
On Wed, Oct 12, 2011 at 5:06 PM, Adam Roben wrote:
> On Oct 12, 2011, at 3:06 AM, Vicky Tux wrote:
>
> > Acutally I'm not a Web Developer. We are using webkit as part of our
> middle-ware. Some of the web applications authored to hide the plugin from
> layout as like below,
> >
> >
> >
> >
> >
> >
> > Also those plugins are scripted from Java Script. In the above case
> webkit is doesn't initialize plugins if it's(or parent's) style is display :
> none.
> >
> > Currently webkit's plugin initialization is based on RenderTree not
> DOMTree. To fix this issue I thinking to implement a RenderNone class(as
> like RenderBlock,RenderInline,..etc) which paints nothing but just acts as a
> container for child elements.
> >
> > But i don't have any idea of RenderTree. Currently implementing
> RenderNone based on RenderBlock & RenderInline.
> >
> > Will it work..? any suggestions..?
>
> Do any other browsers do this? If not, this seems like a potential
> compatibility issue.
>
> If you want to hide plugins visually bug still have them run I'd suggest
> using visibility:hidden instead of display:none.
>
> -Adam
>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
From stephenmattison at gmail.com Wed Oct 12 13:53:25 2011
From: stephenmattison at gmail.com (Stephen Mattison)
Date: Wed, 12 Oct 2011 13:53:25 -0700
Subject: [webkit-help] Background PNG decoding in WebKit
In-Reply-To:
References:
Message-ID:
Hey Zoltan thanks for the input! Is there anything I can do to make sure
that the decoding always happens in the WebThread? Ie not adding to the DOM
or creating a dom fragment or something?
On Wed, Oct 12, 2011 at 3:01 AM, Zoltan Horvath wrote:
> On Wed, 12 Oct 2011 03:39:19 +0200, Stephen Mattison <
> stephenmattison at gmail.com> wrote:
>
> The problem I'm having is that it does not appear that PNGs are decoded in
>> the background in Webkit on iOS. Right now what I have is essentially a
>> ScrollView. I have a div the size of the viewport with another div that
>> scales with the content. I am scrolling the content div using a transition
>> and translate3d to achieve GPU acceleration. The scrolling looks great.
>> However the content is composed of images and they cannot all fit in
>> memory
>> at once so I am loading them on demand. The problem arises when I set the
>> background-image css property to a locally stored PNG image and then
>> attempt
>> to scroll. I get a noticeable delay/jerkiness/hiccup as the image is
>> decoded. Is there something I should be doing differently?
>>
>
> Hey!
>
> WebKit is decoding images in the mainthread when there is new chunk of
> image and it hasn't decoded yet, thus it is imaginable that it blocks your
> main thread for 'longer' time.
>
> I have a sketchy implementation for Qt to support imagedecoding in a
> (currently 1) separate thread.
> If you are interested check: https://bugs.webkit.org/show_**
> bug.cgi?id=69515
>
> Regards,
> Zoltan
> ______________________________**_________________
> webkit-help mailing list
> webkit-help at lists.webkit.org
> http://lists.webkit.org/**mailman/listinfo.cgi/webkit-**help
>
--
Stephen Mattison
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
From leanderbb at gmail.com Wed Oct 12 14:11:23 2011
From: leanderbb at gmail.com (Leander Bessa)
Date: Wed, 12 Oct 2011 22:11:23 +0100
Subject: [webkit-help] Minimum required for embedding
Message-ID:
Hi,
I've been trying to embed webkit into my application in order to render
HTML5 + CSS3 pages. I need a port that supports the css 3D transform and is
cross platform in nature. As far as i could tell, only the chromium port
meets these requirements.
I have managed to compile the chromium port, but i can't figure out which of
the files are required for embedding or how to embed webkit itself. After
the build process, i checked the output folder and i only see executables
for testing purposes.
Am i doing this right or am i forgetting something?
Regards,
Leander
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
From rniwa at webkit.org Wed Oct 12 14:21:41 2011
From: rniwa at webkit.org (Ryosuke Niwa)
Date: Wed, 12 Oct 2011 14:21:41 -0700
Subject: [webkit-help] Minimum required for embedding
In-Reply-To:
References:
Message-ID:
WebKit's Chromium port doesn't come up with a browser / launcher. It only
provides you of the minimum binary to run layout tests. As far as I know,
there is no embedding layer or API for Chromium yet.
At any rate, if you're interested in embedding Chromium code, then you
probably want to ask it on chromium-dev or chromium-discuss instead.
- Ryosuke
On Wed, Oct 12, 2011 at 2:11 PM, Leander Bessa wrote:
> Hi,
>
> I've been trying to embed webkit into my application in order to render
> HTML5 + CSS3 pages. I need a port that supports the css 3D transform and is
> cross platform in nature. As far as i could tell, only the chromium port
> meets these requirements.
>
> I have managed to compile the chromium port, but i can't figure out which
> of the files are required for embedding or how to embed webkit itself. After
> the build process, i checked the output folder and i only see executables
> for testing purposes.
>
> Am i doing this right or am i forgetting something?
>
> Regards,
>
> Leander
>
>
>
> _______________________________________________
> webkit-help mailing list
> webkit-help at lists.webkit.org
> http://lists.webkit.org/mailman/listinfo.cgi/webkit-help
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
From ariya.hidayat at gmail.com Wed Oct 12 14:25:47 2011
From: ariya.hidayat at gmail.com (Ariya Hidayat)
Date: Wed, 12 Oct 2011 14:25:47 -0700
Subject: [webkit-help] Minimum required for embedding
In-Reply-To:
References:
Message-ID:
> I've been trying to embed webkit into my application in order to render
> HTML5 + CSS3 pages. I need a port that supports the css 3D transform and is
> cross platform in nature. As far as i could tell, only the chromium port
> meets these requirements.
In case you miss it, QtWebKit also fulfills that requirements.
--
Ariya Hidayat, http://ariya.ofilabs.com
From leanderbb at gmail.com Wed Oct 12 14:34:36 2011
From: leanderbb at gmail.com (Leander Bessa)
Date: Wed, 12 Oct 2011 22:34:36 +0100
Subject: [webkit-help] Minimum required for embedding
In-Reply-To:
References:
Message-ID:
@Ryosuke Niwa Will do, thank you for the information.
@Ariya Hidayat I'll have to check the Qt port again. Thanks for the heads
up.
Regards,
Leander
On Wed, Oct 12, 2011 at 10:25 PM, Ariya Hidayat wrote:
> > I've been trying to embed webkit into my application in order to render
> > HTML5 + CSS3 pages. I need a port that supports the css 3D transform and
> is
> > cross platform in nature. As far as i could tell, only the chromium port
> > meets these requirements.
>
> In case you miss it, QtWebKit also fulfills that requirements.
>
>
>
>
> --
> Ariya Hidayat, http://ariya.ofilabs.com
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
From dpranke at chromium.org Wed Oct 12 14:45:47 2011
From: dpranke at chromium.org (Dirk Pranke)
Date: Wed, 12 Oct 2011 14:45:47 -0700
Subject: [webkit-help] Minimum required for embedding
In-Reply-To:
References:
Message-ID:
Ryosuke's answer is correct. The chromium team is working on something
called the "content API" that will eventually be useful for an
embedder. There is also the Chromium Embedded Framework, which is a
third-party project built on top of Chromium.
But, chromium-dev or chromium-discuss is indeed a better place for
such questions.
-- Dirk
On Wed, Oct 12, 2011 at 2:21 PM, Ryosuke Niwa wrote:
> WebKit's Chromium port doesn't come up with a browser / launcher. It only
> provides you of the minimum binary to run layout tests. As far as I know,
> there is no embedding layer or API for Chromium yet.
> At any rate, if you're interested in embedding Chromium code, then you
> probably want to ask it on chromium-dev or chromium-discuss instead.
> - Ryosuke
> On Wed, Oct 12, 2011 at 2:11 PM, Leander Bessa wrote:
>>
>> Hi,
>> I've been trying to embed webkit into my application in order to render
>> HTML5 + CSS3 pages. I need a port that supports the css 3D transform and is
>> cross platform in nature. As far as i could tell, only the chromium port
>> meets these requirements.
>> I have managed to compile the chromium port, but i can't figure out which
>> of the files are required for embedding or how to embed webkit itself.?After
>> the build process, i checked the output folder and i only see executables
>> for testing purposes.
>> Am i doing this right or am i forgetting something?
>> Regards,
>> Leander
>>
>>
>> _______________________________________________
>> webkit-help mailing list
>> webkit-help at lists.webkit.org
>> http://lists.webkit.org/mailman/listinfo.cgi/webkit-help
>>
>
>
> _______________________________________________
> webkit-help mailing list
> webkit-help at lists.webkit.org
> http://lists.webkit.org/mailman/listinfo.cgi/webkit-help
>
>
From jaebok.lee at gmail.com Wed Oct 12 17:34:41 2011
From: jaebok.lee at gmail.com (JaeBok Lee)
Date: Thu, 13 Oct 2011 09:34:41 +0900
Subject: [webkit-help] how to get ExecState from ScriptExecutionContext
In-Reply-To: <4E94E1F4.3070800@oracle.com>
References: <4E94E1F4.3070800@oracle.com>
Message-ID:
I think you can find the answer in JSEventListener::handleEvent method.
JSDOMGlobalObject* globalObject = toJSDOMGlobalObject(scriptExecutionContext,
> m_isolatedWorld.get());
> ...
> ExecState* exec = globalObject->globalExec();
2011/10/12 Per Bothner
> In our WebNode implementation (part of JavaFX) we have a class
> JavaEventListener, which extends WebCore::EventListener.
> The virtual handleEvent method takes a ScriptExecutionContext*
> and an Event*, as in the parent class. This methods needs to map the
> WebKit Event to a Java Event.
>
> The Java Event object needs access to the ExecState. (The reason is
> that the Java Event object has methods like getSrcElement, which
> return a Java Node object, and for convenience this Node object also
> has access to the corresponding JavaScript object. The various
> JavaScript operations make use of JSGlobalContextRef which is the
> same as ExecState.)
>
> So the question is there an approved way to get from a
> ScriptExecutionContext
> to an ExecState/CallFrame/**JSGlobalContextRef? I've looked through the
> source code, but haven't found the solution.
> --
> --Per Bothner
> per.bothner at oracle.com per at bothner.com http://per.bothner.com/
> ______________________________**_________________
> webkit-help mailing list
> webkit-help at lists.webkit.org
> http://lists.webkit.org/**mailman/listinfo.cgi/webkit-**help
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
From just2contribute at gmail.com Wed Oct 12 22:56:02 2011
From: just2contribute at gmail.com (just2 contribute)
Date: Thu, 13 Oct 2011 11:26:02 +0530
Subject: [webkit-help] Reg: Issue while running Safari on Windows using the
latest Webkit Trunk!
Message-ID:
Hi,
I tried running Safari 5.0.4 on Windows using the latest webkit trunk, and I
got the error that the CoreFoundation.dll is missing, *
CFStringIsHyphenationAvailableForLocale()*.
Hence should I use the latest Safari 5.1 or are there any other alternative,
b'coz on Safari 5.1 I heard that there are some issues with the ICU
libraries.... :)
--
Justin
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
From zoltan at webkit.org Wed Oct 12 23:25:42 2011
From: zoltan at webkit.org (Zoltan Horvath)
Date: Thu, 13 Oct 2011 08:25:42 +0200
Subject: [webkit-help] Background PNG decoding in WebKit
In-Reply-To:
References:
Message-ID:
On Wed, 12 Oct 2011 22:53:25 +0200, Stephen Mattison
wrote:
> Hey Zoltan thanks for the input! Is there anything I can do to make sure
> that the decoding always happens in the WebThread? Ie not adding to the
> DOM
> or creating a dom fragment or something?
Currently, there is no WebThread in WebKit. If you are using WebKit2 then
there will be separated process for UI things and for Web things, what
increases UI's responsiveness, but with the current implementation
imagedecoding can block the loading a bit.
Zoltan
> On Wed, Oct 12, 2011 at 3:01 AM, Zoltan Horvath
> wrote:
>
>> On Wed, 12 Oct 2011 03:39:19 +0200, Stephen Mattison <
>> stephenmattison at gmail.com> wrote:
>>
>> The problem I'm having is that it does not appear that PNGs are
>> decoded in
>>> the background in Webkit on iOS. Right now what I have is essentially a
>>> ScrollView. I have a div the size of the viewport with another div that
>>> scales with the content. I am scrolling the content div using a
>>> transition
>>> and translate3d to achieve GPU acceleration. The scrolling looks great.
>>> However the content is composed of images and they cannot all fit in
>>> memory
>>> at once so I am loading them on demand. The problem arises when I set
>>> the
>>> background-image css property to a locally stored PNG image and then
>>> attempt
>>> to scroll. I get a noticeable delay/jerkiness/hiccup as the image is
>>> decoded. Is there something I should be doing differently?
>>>
>>
>> Hey!
>>
>> WebKit is decoding images in the mainthread when there is new chunk of
>> image and it hasn't decoded yet, thus it is imaginable that it blocks
>> your
>> main thread for 'longer' time.
>>
>> I have a sketchy implementation for Qt to support imagedecoding in a
>> (currently 1) separate thread.
>> If you are interested check: https://bugs.webkit.org/show_**
>> bug.cgi?id=69515
>>
>> Regards,
>> Zoltan
>> ______________________________**_________________
>> webkit-help mailing list
>> webkit-help at lists.webkit.org
>> http://lists.webkit.org/**mailman/listinfo.cgi/webkit-**help
>>
>
>
>
--
Using Opera's revolutionary email client: http://www.opera.com/mail/
From per.bothner at oracle.com Thu Oct 13 00:17:11 2011
From: per.bothner at oracle.com (Per Bothner)
Date: Thu, 13 Oct 2011 00:17:11 -0700
Subject: [webkit-help] how to get ExecState from ScriptExecutionContext
In-Reply-To:
References: <4E94E1F4.3070800@oracle.com>
Message-ID: <4E969077.1020304@oracle.com>
On 10/12/2011 05:34 PM, JaeBok Lee wrote:
> I think you can find the answer in JSEventListener::handleEvent method.
>
> JSDOMGlobalObject* globalObject =
> toJSDOMGlobalObject(scriptExecutionContext, m_isolatedWorld.get());
> ...
> ExecState* exec = globalObject->globalExec();
Thanks - that got me unstuck. I didn't actually do it this way, but
looking at JSEventListener got me heading in the right direction.
(It turns out a JSEventListener is only created in places where I have
a DOM node, so I already have the ExecState I can stash.)
--
--Per Bothner
per.bothner at oracle.com per at bothner.com http://per.bothner.com/
From leanderbb at gmail.com Thu Oct 13 04:20:18 2011
From: leanderbb at gmail.com (Leander Bessa)
Date: Thu, 13 Oct 2011 12:20:18 +0100
Subject: [webkit-help] Removing -Werror flag from QtWebkit port
Message-ID:
Hello,
I'm having some difficulties removing the -Werror flag from the QtWebkit
port. I'm using Fedora 15 (64bit) with GCC 4.6.1. With the -Werror flag i
always end up with the following error:
../../../Source/JavaScriptCore/wtf/NullPtr.h:50:1: error: identifier
?nullptr? will become a keyword in C++0x [-Werror=c++0x-compat]
cc1plus: all warnings being treated as errors
I've tried passing the following arguments in order to remove the flag, but
they did not work:
--qmakearg="CXXFLAGS-=-Werror CFLAGS-=-Werror"
Any suggestions?
Regards,
Leander
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
From oszi at inf.u-szeged.hu Thu Oct 13 04:23:57 2011
From: oszi at inf.u-szeged.hu (Osztrogonac Csaba)
Date: Thu, 13 Oct 2011 13:23:57 +0200
Subject: [webkit-help] Removing -Werror flag from QtWebkit port
In-Reply-To:
References:
Message-ID: <4E96CA4D.1000803@inf.u-szeged.hu>
Hi,
You should remove the following line from WebKit.pri to disable Werror:
isEqual(QT_ARCH,x86_64)|isEqual(QT_ARCH,i386): QMAKE_CXXFLAGS += -Werror
br,
Ossy
Leander Bessa ?rta:
> Hello,
>
> I'm having some difficulties removing the -Werror flag from the QtWebkit
> port. I'm using Fedora 15 (64bit) with GCC 4.6.1. With the -Werror flag
> i always end up with the following error:
>
> ../../../Source/JavaScriptCore/wtf/NullPtr.h:50:1: error: identifier
> 'nullptr' will become a keyword in C++0x [-Werror=c++0x-compat]
> cc1plus: all warnings being treated as errors
>
> I've tried passing the following arguments in order to remove the flag,
> but they did not work:
>
> --qmakearg="CXXFLAGS-=-Werror CFLAGS-=-Werror"
>
>
> Any suggestions?
>
>
> Regards,
>
> Leander
From leanderbb at gmail.com Thu Oct 13 04:37:11 2011
From: leanderbb at gmail.com (Leander Bessa)
Date: Thu, 13 Oct 2011 12:37:11 +0100
Subject: [webkit-help] Removing -Werror flag from QtWebkit port
In-Reply-To: <4E96CA4D.1000803@inf.u-szeged.hu>
References:
<4E96CA4D.1000803@inf.u-szeged.hu>
Message-ID:
That did it, thank you.
Regards,
Leander
2011/10/13 Osztrogonac Csaba
> Hi,
>
> You should remove the following line from WebKit.pri to disable Werror:
> isEqual(QT_ARCH,x86_64)|**isEqual(QT_ARCH,i386): QMAKE_CXXFLAGS += -Werror
>
> br,
> Ossy
>
> Leander Bessa ?rta:
>
> Hello,
>>
>> I'm having some difficulties removing the -Werror flag from the QtWebkit
>> port. I'm using Fedora 15 (64bit) with GCC 4.6.1. With the -Werror flag i
>> always end up with the following error:
>>
>> ../../../Source/**JavaScriptCore/wtf/NullPtr.h:**50:1: error: identifier
>> 'nullptr' will become a keyword in C++0x [-Werror=c++0x-compat]
>> cc1plus: all warnings being treated as errors
>>
>> I've tried passing the following arguments in order to remove the flag,
>> but they did not work:
>>
>> --qmakearg="CXXFLAGS-=-Werror CFLAGS-=-Werror"
>>
>> Any suggestions?
>>
>>
>> Regards,
>>
>> Leander
>>
> ______________________________**_________________
> webkit-help mailing list
> webkit-help at lists.webkit.org
> http://lists.webkit.org/**mailman/listinfo.cgi/webkit-**help
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
From aroben at apple.com Thu Oct 13 05:40:23 2011
From: aroben at apple.com (Adam Roben)
Date: Thu, 13 Oct 2011 08:40:23 -0400
Subject: [webkit-help] Reg: Issue while running Safari on Windows using
the latest Webkit Trunk!
In-Reply-To:
References:
Message-ID: <433CCEEE-B41C-4E29-B3A3-D4F60D26CB5B@apple.com>
On Oct 13, 2011, at 1:56 AM, just2 contribute wrote:
> I tried running Safari 5.0.4 on Windows using the latest webkit trunk, and I got the error that the CoreFoundation.dll is missing, CFStringIsHyphenationAvailableForLocale().
>
> Hence should I use the latest Safari 5.1
Yes. WebKit trunk is compatible with Safari 5.1 and newer.
-Adam
From ssseintr2 at gmail.com Thu Oct 13 08:28:50 2011
From: ssseintr2 at gmail.com (Vicky Tux)
Date: Thu, 13 Oct 2011 20:58:50 +0530
Subject: [webkit-help] Reg. Plugin Loading issue when "display: none".
In-Reply-To:
References:
Message-ID:
>Do any other browsers do this? If not, this seems like a potential
compatibility issue.
Yes. Some of the browsers like ANT & Netfront supports plugin loading.with
display:none.
>If you want to hide plugins visually bug still have them run I'd suggest
using visibility:hidden instead of >display:none.
Actually I'm not a web developer. Some web applications are authored to load
plugins with "display:none" to act as a audio player by implementing it's
GUI using HTML/JS/CSS.
-Vicky
On Wed, Oct 12, 2011 at 5:06 PM, Adam Roben wrote:
> On Oct 12, 2011, at 3:06 AM, Vicky Tux wrote:
>
> > Acutally I'm not a Web Developer. We are using webkit as part of our
> middle-ware. Some of the web applications authored to hide the plugin from
> layout as like below,
> >
> >
> >
> >
> >
> >
> > Also those plugins are scripted from Java Script. In the above case
> webkit is doesn't initialize plugins if it's(or parent's) style is display :
> none.
> >
> > Currently webkit's plugin initialization is based on RenderTree not
> DOMTree. To fix this issue I thinking to implement a RenderNone class(as
> like RenderBlock,RenderInline,..etc) which paints nothing but just acts as a
> container for child elements.
> >
> > But i don't have any idea of RenderTree. Currently implementing
> RenderNone based on RenderBlock & RenderInline.
> >
> > Will it work..? any suggestions..?
>
> Do any other browsers do this? If not, this seems like a potential
> compatibility issue.
>
> If you want to hide plugins visually bug still have them run I'd suggest
> using visibility:hidden instead of display:none.
>
> -Adam
>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
From sanachan.y at gmail.com Thu Oct 13 08:58:15 2011
From: sanachan.y at gmail.com (Y.Sanachan)
Date: Fri, 14 Oct 2011 00:58:15 +0900
Subject: [webkit-help] Which SYNTAX_ERR should be thrown from
WorkerContext::importScripts()
In-Reply-To:
References: <4E8F1D9D.4020208@gmail.com>
Message-ID: <4E970A97.9060809@gmail.com>
Thanks for your help.
I will register this problem to bugs.webkit.org with a patch.
Thank you for your time !
Sanachan.
(2011/10/13 1:44), Julien Chaffraix wrote:
>> In WebCore::WorkerContext::importScripts() method,
>> it will throw an exception code of SYNTAX_ERR from the 1st for loop.
>>
>> When ENABLE_FILE_SYSTEM=0,
>> I met the compile error as like 'SYNTAX_ERR is not defined'.
>> Can I include dom/ExceptionCode.h instead of the header files relating
>> to FILE_SYSTEM API ?
> This is the right solution. Care to file a bug and attach a patch to
> solve the problem for everyone?
>
>> OR, should it be set to 1 when compile WorkerContext.cpp ?
> No, Workers and Filesystem API are orthogonal.
>
> Thanks,
> Julien
>
From stephenmattison at gmail.com Thu Oct 13 10:16:14 2011
From: stephenmattison at gmail.com (Stephen Mattison)
Date: Thu, 13 Oct 2011 10:16:14 -0700
Subject: [webkit-help] Background PNG decoding in WebKit
In-Reply-To:
References:
Message-ID:
Thanks for the reply Zoltan. What I am seeing on iOS is that sometimes the
decoding task runs in the main thread sometimes in the web thread. Is this
something that Apple is doing?
Cheers,
Stephen
On Wed, Oct 12, 2011 at 11:25 PM, Zoltan Horvath wrote:
> On Wed, 12 Oct 2011 22:53:25 +0200, Stephen Mattison <
> stephenmattison at gmail.com> wrote:
>
> Hey Zoltan thanks for the input! Is there anything I can do to make sure
>> that the decoding always happens in the WebThread? Ie not adding to the
>> DOM
>> or creating a dom fragment or something?
>>
>
> Currently, there is no WebThread in WebKit. If you are using WebKit2 then
> there will be separated process for UI things and for Web things, what
> increases UI's responsiveness, but with the current implementation
> imagedecoding can block the loading a bit.
>
> Zoltan
>
> On Wed, Oct 12, 2011 at 3:01 AM, Zoltan Horvath
>> wrote:
>>
>> On Wed, 12 Oct 2011 03:39:19 +0200, Stephen Mattison <
>>> stephenmattison at gmail.com> wrote:
>>>
>>> The problem I'm having is that it does not appear that PNGs are decoded
>>> in
>>>
>>>> the background in Webkit on iOS. Right now what I have is essentially a
>>>> ScrollView. I have a div the size of the viewport with another div that
>>>> scales with the content. I am scrolling the content div using a
>>>> transition
>>>> and translate3d to achieve GPU acceleration. The scrolling looks great.
>>>> However the content is composed of images and they cannot all fit in
>>>> memory
>>>> at once so I am loading them on demand. The problem arises when I set
>>>> the
>>>> background-image css property to a locally stored PNG image and then
>>>> attempt
>>>> to scroll. I get a noticeable delay/jerkiness/hiccup as the image is
>>>> decoded. Is there something I should be doing differently?
>>>>
>>>>
>>> Hey!
>>>
>>> WebKit is decoding images in the mainthread when there is new chunk of
>>> image and it hasn't decoded yet, thus it is imaginable that it blocks
>>> your
>>> main thread for 'longer' time.
>>>
>>> I have a sketchy implementation for Qt to support imagedecoding in a
>>> (currently 1) separate thread.
>>> If you are interested check: https://bugs.webkit.org/show_****
>>> bug.cgi?id=69515
>>> >
>>>
>>> Regards,
>>> Zoltan
>>> ______________________________****_________________
>>> webkit-help mailing list
>>> webkit-help at lists.webkit.org
>>> http://lists.webkit.org/****mailman/listinfo.cgi/webkit-****help
>>>
>>> >
>>>
>>>
>>
>>
>>
>
> --
> Using Opera's revolutionary email client: http://www.opera.com/mail/
>
--
Stephen Mattison
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
From simon.fraser at apple.com Thu Oct 13 10:55:41 2011
From: simon.fraser at apple.com (Simon Fraser)
Date: Thu, 13 Oct 2011 10:55:41 -0700
Subject: [webkit-help] Background PNG decoding in WebKit
In-Reply-To:
References:
Message-ID:
This isn't the right place to discuss iOS behaviors. Your best bet is to file a bug at http://bugreporter.apple.com describing the behavior you see. Be sure to include a test case.
Simon
On Oct 13, 2011, at 10:16 AM, Stephen Mattison wrote:
> Thanks for the reply Zoltan. What I am seeing on iOS is that sometimes the decoding task runs in the main thread sometimes in the web thread. Is this something that Apple is doing?
>
> Cheers,
> Stephen
>
> On Wed, Oct 12, 2011 at 11:25 PM, Zoltan Horvath wrote:
> On Wed, 12 Oct 2011 22:53:25 +0200, Stephen Mattison wrote:
>
> Hey Zoltan thanks for the input! Is there anything I can do to make sure
> that the decoding always happens in the WebThread? Ie not adding to the DOM
> or creating a dom fragment or something?
>
> Currently, there is no WebThread in WebKit. If you are using WebKit2 then there will be separated process for UI things and for Web things, what increases UI's responsiveness, but with the current implementation imagedecoding can block the loading a bit.
>
> Zoltan
>
> On Wed, Oct 12, 2011 at 3:01 AM, Zoltan Horvath wrote:
>
> On Wed, 12 Oct 2011 03:39:19 +0200, Stephen Mattison <
> stephenmattison at gmail.com> wrote:
>
> The problem I'm having is that it does not appear that PNGs are decoded in
> the background in Webkit on iOS. Right now what I have is essentially a
> ScrollView. I have a div the size of the viewport with another div that
> scales with the content. I am scrolling the content div using a transition
> and translate3d to achieve GPU acceleration. The scrolling looks great.
> However the content is composed of images and they cannot all fit in
> memory
> at once so I am loading them on demand. The problem arises when I set the
> background-image css property to a locally stored PNG image and then
> attempt
> to scroll. I get a noticeable delay/jerkiness/hiccup as the image is
> decoded. Is there something I should be doing differently?
>
>
> Hey!
>
> WebKit is decoding images in the mainthread when there is new chunk of
> image and it hasn't decoded yet, thus it is imaginable that it blocks your
> main thread for 'longer' time.
>
> I have a sketchy implementation for Qt to support imagedecoding in a
> (currently 1) separate thread.
> If you are interested check: https://bugs.webkit.org/show_**
> bug.cgi?id=69515
>
> Regards,
> Zoltan
> ______________________________**_________________
> webkit-help mailing list
> webkit-help at lists.webkit.org
> http://lists.webkit.org/**mailman/listinfo.cgi/webkit-**help
>
>
>
>
>
>
> --
> Using Opera's revolutionary email client: http://www.opera.com/mail/
>
>
>
> --
> Stephen Mattison
> _______________________________________________
> webkit-help mailing list
> webkit-help at lists.webkit.org
> http://lists.webkit.org/mailman/listinfo.cgi/webkit-help
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
From huyanke at gmail.com Thu Oct 13 11:00:05 2011
From: huyanke at gmail.com (Yanke (Yon) Hu)
Date: Thu, 13 Oct 2011 11:00:05 -0700
Subject: [webkit-help] Is WebKit Gtk+ port support mobile DOM?
Message-ID:
hi guys,
I just build webkit gtk+ port under ubuntu. When I launch GtkLauncher, I
find it doesn't support any mobile DOM like touchstart, touchend, touchmove,
etc.
Previously, I also build a webkit qt port under WIn 7. QtWebKit has a
fancybrowser example, and fancybrowser support these mobile DOM
like touchstart, touchend, touchmove, etc.
Is there any open source webkit gtk+ browser supporting mobile DOM?
Thanks very much, and have a great day!!
Yanke
--
Yanke Hu
phone: (469) 644-8799
email: huyanke at gmail.com
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
From ariya.hidayat at gmail.com Thu Oct 13 15:54:01 2011
From: ariya.hidayat at gmail.com (Ariya Hidayat)
Date: Thu, 13 Oct 2011 15:54:01 -0700
Subject: [webkit-help] Minimum required for embedding
In-Reply-To:
References:
Message-ID:
> @Ariya Hidayat I'll have to check the Qt port again. Thanks for the heads
> up.
Make sure you enable accelerated compositing:
http://labs.qt.nokia.com/2010/05/17/qtwebkit-now-accelerates-css-animations-3d-transforms/
--
Ariya Hidayat, http://ariya.ofilabs.com
From zhangyu_allen at qq.com Thu Oct 13 19:27:51 2011
From: zhangyu_allen at qq.com (=?ISO-8859-1?B?emhhbmd5dQ==?=)
Date: Fri, 14 Oct 2011 10:27:51 +0800
Subject: [webkit-help] how can i add extend JSObjct in webkit?
Message-ID:
can you tell me the method or send me some reference materials?
thank you very much.
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
From bkardell at gmail.com Fri Oct 14 06:37:41 2011
From: bkardell at gmail.com (Brian Kardell)
Date: Fri, 14 Oct 2011 09:37:41 -0400
Subject: [webkit-help] Help Getting Started / locating what I am looking for.
Message-ID:
In general, I am starting to wade through webkit code but there is a lot and
not a lot of docs (at least that I am finding), so any pointer to file
in webkit source or docs would be very greatly appreciated.
Specifically, I am interested in a deeper understanding of how webkit
determines which elements to test selectors against (not which
selectors to test
or any of those ptimizations, just literally which elements)? Does it
just walk the
subtree any time a DOM modification happens and collect for retesting
or is it smarter than that?
Thanks.
From eric.hennigan at uci.edu Fri Oct 14 09:34:01 2011
From: eric.hennigan at uci.edu (Eric Hennigan)
Date: Fri, 14 Oct 2011 09:34:01 -0700
Subject: [webkit-help] JSC call stack layout details
In-Reply-To:
References:
Message-ID:
Hello fellow hackers,
? ?I've been working on JSC (interpreter only, not JIT) for a month
or so now, and so I'm somewhat familiar with the internals. However,
I've reached a point where I feel that some external commentary (from
those who've been around longer) would be beneficial. I was wondering
if anyone on this list could spare the time to answer the following
questions about how JSC handles the call stack. I will give a brief
overview of my current understanding before asking the questions. If
anything I say is in some way misguided, please correct it.
? ?My current understanding is that JSC::Interpreter contains a
RegisterFile, which can be thought of as an array of bytes. Calling a
function is done in a 3 step fashion: push arguments, place call
frame, call. Returning can simply pop both the arguments and frame by
restoring the stack pointer. The bytecompiler keeps track of the
number of local and temporary values necessary to execute the
function.
? ?Question 1.
? ?Is there a safety mechanism for accessing registers from the
callFrame? and if so, how does it relate to
Interpreter::slideRegisterWindowForCall(...)? I've noticed that
CallFrame::r(int) simply returns an offset from the callframe's base
address. There is no check that this offset might be beyond the end of
the RegisterFile. I also notice that there is no check for the total
amount of space (arguments + callframe + locals + temps) that a
function will use before the frame is setup with
Interpreter::slideRegisterWindowForCall(...). However, there is a
check for the amount of space that the callframe will occupy.
? ?Question 2.
? ?Since JavaScript has first class function objects and support
continuations, how is the state of a function preserved? Does the
(arguments + callframe + locals + temps) get copied out int heap, as a
JSActivation? It didn't look like this was the case. I'm curious
because continuations mean you end up with a cactus stack, which does
not play well with the linear sequence of callframes I described
above.
? ?Question 3.
? ?My pet project requires that each activation record hold a small
stack-like data structure, which is manipulated at runtime. This
structure must reside with the activation record. That is, if the
function is a continuation, the data structure, and its state, must be
preserved across calls. I currently perform an analysis during parsing
which allows me to know the maximum number of entries in the stack
(just as we can know the number of locals, temps and scopes).
? ?Option 1. Because this data structure works similarly to temporary
variables, I was considering reserving the appropriate amount of space
in the RegisterFile when the callframe is setup.
? ?Option 2. Alternatively, I was considering hanging off the data
structure as a separate entity. I can easily add another fixed entry
to the CallFrame (alongside the argument count, caller frame, callee,
scope chain, return address and code block). This slot can then point
to my data structure. I worry a bit over memory management though. I'd
want to de-alloc my data structure when the call frame becomes
inactive. I'm ok with inheriting JSObject, and placing it in the
garbage collected heap.
? ?I thank you in advance for your time and patience,
--
Eric Hennigan
From ariya.hidayat at gmail.com Fri Oct 14 12:58:44 2011
From: ariya.hidayat at gmail.com (Ariya Hidayat)
Date: Fri, 14 Oct 2011 12:58:44 -0700
Subject: [webkit-help] Help Getting Started / locating what I am looking
for.
In-Reply-To:
References:
Message-ID:
Maybe you can start from SelectorQuery::execute() function and trace
it further from that starting point.
--
Ariya Hidayat, http://ariya.ofilabs.com
From justin at sweaky.com Fri Oct 14 19:48:59 2011
From: justin at sweaky.com (Justin Thomas)
Date: Fri, 14 Oct 2011 19:48:59 -0700
Subject: [webkit-help] JavaScriptCore and threading
In-Reply-To:
References:
Message-ID: <4E98F49B.8060608@sweaky.com>
Hi All,
I'm been trying to create a new context in a new thread using
JSGlobalContextCreateInGroup but I can't seem to get around an "access
violation exception while writing to...", is this not the correct way to
setup mutlithreading? I assumed that for each thread created a separate
context should be used. I am passing in the same group that was created
from the main thread each time I try to create a new global context.
Some code:
// Main thread
mMainGroup = JSContextGroupCreate();
mMainContext = JSGlobalContextCreateInGroup(mMainGroup, NULL);
// Secondary thread
mOtherContext = JSGlobalContextCreateInGroup(mMainGroup, NULL); // <-
fails here
When looking through it is failing at the assertion in
/RefCounted.h:53/, ASSERT(m_verifier.isSafeToUse()), which looks like
the verifier is set to SingleThreadVerificationMode, I'm thinking it's
pretty obvious that needs to be changed, but not sure how to go about
setting a new verification mode?
I'm using the latest webkit built on my windows box.
Thanks,
-JT
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
From julien.chaffraix at gmail.com Fri Oct 14 20:44:35 2011
From: julien.chaffraix at gmail.com (Julien Chaffraix)
Date: Fri, 14 Oct 2011 20:44:35 -0700
Subject: [webkit-help] Problem about DerivedSources of WebKit
In-Reply-To: <681d82ee.468f.132e6f5fdd6.Coremail.wenmhappy@126.com>
References: <681d82ee.468f.132e6f5fdd6.Coremail.wenmhappy@126.com>
Message-ID:
> ?I wonder that why are there so many derived sourece code.
Simple, because it is easier to auto-generate them - ie to maintain a
script that does the generation - than to manually maintain the files
themselves.
>? How are they generated automatically?
That depends on the files you are interested in. We have several generators:
* make_names.pl for the element factories
* makevalues.pl and makeprop.pl for the CSS values and properties
* generate-bindings.pl (and friends) for the bindings generation.
A _precise_ question would get a better answer: what are you trying to do?
> Could they be modified?
Usually you don't want to modify the auto-generated files as anything
that triggers a generation would overwrite your changes. The scripts
are modified fairly often and that's the preferred way.
Hope it helps,
Julien
From julien.chaffraix at gmail.com Fri Oct 14 21:07:18 2011
From: julien.chaffraix at gmail.com (Julien Chaffraix)
Date: Fri, 14 Oct 2011 21:07:18 -0700
Subject: [webkit-help] how can i add extend JSObjct in webkit?
In-Reply-To:
References:
Message-ID:
Providing your port would be helpful as there is several ways to do
what you want. As you are *just* asking about JSC, the answer is
port-agnostic.
> can you tell me the method or send me some reference materials?
This is something that has been asked over and over in the past years.
One of the most relevant results as I had to do the query [1].
Some inline answer: the JSC API located here [2] or Apple has a more
readable version if you use a compatible version [3]. The other option
is to study the bindings in WebKit (after building it) or an
application that uses the API (see [1] for some examples).
Thanks,
Julien
[1] http://lists.macosforge.org/pipermail/webkit-dev/2009-July/008689.html
[2] http://trac.webkit.org/browser/trunk/Source/JavaScriptCore/API
[3] http://developer.apple.com/library/mac/#documentation/Carbon/Reference/WebKit_JavaScriptCore_Ref/
From noreply at googlegroups.com Fri Oct 14 21:21:50 2011
From: noreply at googlegroups.com (safynour/safwatsalm (Google Groups))
Date: Sat, 15 Oct 2011 04:21:50 +0000
Subject: [webkit-help] Google Groups Invitation: Group
safynour-Globalfinancial
Message-ID: <20111015042152.D18B2A8A82A@lists.macosforge.org>
safynour/safwatsalm safwateconomic at gmail.com has invited you to join the
Group safynour-Globalfinancial group with this message:
we hope communication between us and the peoples of the world in various world.
Here is the group's description:
An dependent group that wants to attract and develop the best people in our
industry ,wherever they are from and wherever they are. we provide
---------------------- Google Groups Information ----------------------
You can accept this invitation by clicking the following URL:
http://groups.google.com/group/group-safynour-globalfinancial/sub?s=hak8DRQAAAB8YRRyjDVQNunUWCzwqLPZ-F7cqcY-y_yzAHrrTpivXg&hl=en
--------------------- If This Message Is Unwanted ---------------------
If you feel that this message is abuse, please inform the Google Groups staff
by using the URL below.
http://groups.google.com/groups/abuse?invite=YgAAAOvY4PtHAAAAGFgr4aAAAAAAAESG4K4GkIva8qd7AKC8Dyrinw4&hl=en
From noreply at googlegroups.com Fri Oct 14 23:55:10 2011
From: noreply at googlegroups.com (safynour/safwatsalm (Google Groups))
Date: Sat, 15 Oct 2011 06:55:10 +0000
Subject: [webkit-help] Google Groups Invitation: safynouradveristement
Message-ID: <20111015065511.9387BA8B4A0@lists.macosforge.org>
safynour/safwatsalm safwateconomic at gmail.com has invited you to join the
safynouradveristement group with this message:
we can connect through social media
Here is the group's description:
????? / ???? ??? ???? / ???????? / ??? ???? / ???? / ??? ??????? / ????? ??? ?
? ? ??? ??? ??????? / ????? ? ???? ??? ???? ? ?????? / ????? ?? ??? ??? ???
?????? ????? ??????? ????? ??? ?????? ??? ( ????? - ???? ?????? -??? ?? -???
?? ? ???? ??? -?????? ??? ??????? ? - ???? ? ??? ? ??? - )
---------------------- Google Groups Information ----------------------
You can accept this invitation by clicking the following URL:
http://groups.google.com/group/safynouradveristement/sub?s=aomLzBQAAABuYwILB4StxH38WAnrUpr9toyLDoKl1m2SyNb9xyqRUA&hl=en
--------------------- If This Message Is Unwanted ---------------------
If you feel that this message is abuse, please inform the Google Groups staff
by using the URL below.
http://groups.google.com/groups/abuse?invite=YgAAABBGeAeeAAAAGFgr4aAAAAAAAESIYq4GkIva8qd7AKC8Dyrinw4&hl=en
From a.j.w at me.com Sat Oct 15 12:43:28 2011
From: a.j.w at me.com (Andrew Wood)
Date: Sat, 15 Oct 2011 20:43:28 +0100
Subject: [webkit-help] Best way to block tags
Message-ID: <4E99E260.2070104@me.com>
Whats the best way to prevent file upload form fields from working. Im
developing a browser for a public kiosk so I want to prevent this sort
of thing.
Thanks
Andrew
From sanjeetpratapsingh at gmail.com Mon Oct 17 05:05:18 2011
From: sanjeetpratapsingh at gmail.com (Sanjeet Pratap Singh)
Date: Mon, 17 Oct 2011 17:35:18 +0530
Subject: [webkit-help] NeedHelp_In_SpatialNavigation
Message-ID:
Hi All,
I am a beginner to WebKit, looking into "*Spatial-Navigation*" feature of
WebKit.
I am looking further for some optimisations in the same.
I find there "*findFocusCandidateInContainer() "* in which we start
from first child of "container" and iterate until the we get the focus
candidate.
My question is,"Are we checking the whole DOM tree to find the closest node
or we have a small area(container) accordingly?"
What does the function*
"container=scrollableEnclosingBoxOrParentFrameForNodeInDirection()"* do?
Waiting for the reply.
Thanks & Regards,
Sanjeet
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
From eric.hennigan at uci.edu Mon Oct 17 17:19:45 2011
From: eric.hennigan at uci.edu (Eric Hennigan)
Date: Mon, 17 Oct 2011 17:19:45 -0700
Subject: [webkit-help] HTML and JavaScript reflection layer
Message-ID:
Hey fellow hackers,
I'v noticed that in the interface descriptions of html objects, not
every field is marked with a Reflect. For example,
WebCore/html/HTMLImageElement.idl does not mark the width and height
attributes as Reflect. However, name, hspace and vspace are marked
Reflect.
I'm just wondering, why the inconsistency? why not just Reflect on
every attribute?
--
Eric Hennigan
From wenmhappy at 126.com Mon Oct 17 19:11:43 2011
From: wenmhappy at 126.com (wenmhappy)
Date: Tue, 18 Oct 2011 10:11:43 +0800 (CST)
Subject: [webkit-help] How to cross-compile wetkit-r59711 with qt?
Message-ID: <195b8ba.9833.13314cc740a.Coremail.wenmhappy@126.com>
Hi everyone,
I have built(cross-compile) webkit with gtk using script below.
export PKG_CONFIG_PATH="$SDKROOT/usr/lib/pkgconfig":$PKG_CONFIG_PATH
export PATH=`pwd`::$PATH
export BUILD_WEBKIT_ARGS="--host=sh4-linux --build=i686-pc-linux-gnu \
--prefix=/opt/STM/STLinux-2.3/devkit/sh4/target/usr \
LDFLAGS=-L$LIBTOOL_FILE_DIR \
--with-target=directfb"
./WebKitTools/Scripts/build-webkit --gtk --release --makeargs="2> buildlog"
But failed when building webkit with qt like this:
QMAKE_DIR=/usr/lib/qt4/bin
export PKG_CONFIG_PATH="$SDKROOT/usr/lib/pkgconfig":$PKG_CONFIG_PATH
export PATH=`pwd`:$QMAKE_DIR:$PATH
export BUILD_WEBKIT_ARGS="--host=sh4-linux --build=i686-pc-linux-gnu \
--prefix=/opt/STM/STLinux-2.3/devkit/sh4/target/usr \
LDFLAGS=-L$LIBTOOL_FILE_DIR \
--with-target=directfb"
./WebKitTools/Scripts/build-webkit --qt --release --makeargs="2> buildlog"
details is below:
[root at lhb WebKit-r59711]# ./cc-webkit
Generating derived sources
Calling 'qmake -r --host=sh4-linux --build=i686-pc-linux-gnu LDFLAGS=-L/opt/STM/STLinux-2.3/devkit/sh4/target/usr/lib --with-target=directfb -r /home/wen/WebKit-r59711/DerivedSources.pro -o Makefile.DerivedSources' in /home/wen/WebKit-r59711/WebKitBuild/Release
***Unknown option --host=sh4-linux
Usage: qmake [mode] [options] [files]
QMake has two modes, one mode for generating project files based on
some heuristics, and the other for generating makefiles. Normally you
shouldn't need to specify a mode, as makefile generation is the default
mode for qmake, but you may use this to test qmake on an existing project
Mode:
-project Put qmake into project file generation mode
In this mode qmake interprets files as files to
be built,
defaults to *.c; *.ui; *.y; *.l; *.ts; *.qrc; *.h; *.hpp; *.hh; *.hxx; *.H; *.cpp; *.cc; *.cxx; *.C
-makefile Put qmake into makefile generation mode (default)
In this mode qmake interprets files as project files to
be processed, if skipped qmake will try to find a project
file in your current working directory
Warnings Options:
-Wnone Turn off all warnings
-Wall Turn on all warnings
-Wparser Turn on parser warnings
-Wlogic Turn on logic warnings
Options:
* You can place any variable assignment in options and it will be *
* processed as if it was in [files]. These assignments will be parsed *
* before [files]. *
-o file Write output to file
-unix Run in unix mode
-win32 Run in win32 mode
-macx Run in Mac OS X mode
-d Increase debug level
-t templ Overrides TEMPLATE as templ
-tp prefix Overrides TEMPLATE so that prefix is prefixed into the value
-help This help
-v Version information
-after All variable assignments after this will be
parsed after [files]
-norecursive Don't do a recursive search
-recursive Do a recursive search
-cache file Use file as cache [makefile mode only]
-spec spec Use spec as QMAKESPEC [makefile mode only]
-nocache Don't use a cache file [makefile mode only]
-nodepend Don't generate dependencies [makefile mode only]
-nomoc Don't generate moc targets [makefile mode only]
-nopwd Don't look for files in pwd [project mode only]
Failed while running qmake to generate derived sources!
I installed qt4-qmake-4.3.2-3.i386.rpm and compat-libstdc++-33-3.2.3-68.i686.rpm on fedora14, there are libs of qt on embedded platform(STLinux). I don`t know where is problem. I think maybe script is wrong.
How to fix it?
Thanks for any help.
Best regards
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
From julien.chaffraix at gmail.com Tue Oct 18 08:48:12 2011
From: julien.chaffraix at gmail.com (Julien Chaffraix)
Date: Tue, 18 Oct 2011 08:48:12 -0700
Subject: [webkit-help] HTML and JavaScript reflection layer
In-Reply-To: