July 2006 - webkit-dev - lists.webkit.org

Debugging a (JavaScript) "Maximum call stack size exceeded" exception
by Giulio Cesare Solaroli 08 Jul '06

08 Jul '06

Hi everybody, I have a question loosely related to WebKit, but I hope someone could share some useful knowledge. I am working on a JavaScript framework. Testing the framework with FireFox everything works as expected. Using Safari I get a "Maximum call stack size exceded". I have investigated this issue and find out that FireFox JavaScript interpreter has a much deeper (1000) stack size than Safari (100). Using FireBug (a FireFox extension to degub JavaScript) I have tried to get the full stack-trace of the execution in the point where presumably Safari was hitting the stack limit. To my surprise, the stack trace I got was only 65 level deep. I have tried to run the latest nightly build with Drosera attached, but no useful information is provided in my case. Is there a way in Safari to get a full stack-trace of the execution of a JavaScript in the event of an exception like the one I am facing? Thanks for your attention. Best regards, Giulio Cesare Solaroli

2 2

Crash reports which show extensions enabled: what to do?
by Joost 'AlthA' de Valk 07 Jul '06

07 Jul '06

Hi all, there seems to be some discussion going on what to do with bugs when they have crash reports with them, which show enabled extensions. My view on this is VERY simple: I close each and every such bug as INVALID. We have neither the time or the resources to do anything else, and i'd like to keep confirming bugs as easy as possible. Now I know I have no "official" right to decide on this, but I hope you will all agree with me and do the same. There's been some hard work the last few days to get the UNCONFIRMED list under 100 again, we have succeeded in that, and have 67 unconfirmed bugs at this moment. I want to thank everybody who's been doing that work, it's the first step in getting our bug reporters to feel that their bugs are being looked at and taken care of, which is very important IMHO. Regards, Joost

3 2

Layout tests
by Ronan Meneu 06 Jul '06

06 Jul '06

Hello, I would like to compile and run layout tests on a linux machine. I have several ways to follow, but i am not sure of the best one : * I have tried to compile DumpRenderTree (objective-c code). Using GNUstep, i can resolve most of the missing headers and objects. I still have few compilation errors. Do you think it's worth going on, or is it to close to some Apple tools to get it running on a linux platform ? * I have compiled the DumpRenderTree tool in the .vcproj, modified it to build a FrameGdk instead of a FrameWin. Purpose of this tool i think is to just dump render tree. It does not have some functionalities of the objective-c code like "image diff". It works on linux, but layer size is always 0x0. Is this the DumpRenderTree tool used on win32 platform ? * Another solution is to port objective-c code in C++. It seems to me that it is quite complex. Has anyone tried to do this ? Have you got some advice to realize it ? Thanks for your help, -- Ronan Meneu Origyn Web Browser for Embedded Systems Team Senior Software Engineer

3 5

Upgrade Cairo to Cairo 1.2.0?
by Justin Haygood 05 Jul '06

05 Jul '06

Any chance we can upgrade Cairo to 1.2.0 soonish? Changes: * New PDF backend * Many optimizations. On Win32, Cairo now uses GDI over pixman every chance it can get, and includes new Font APIs. * Many overall performance and optimizations to speed up WebKit on Windows (and whatever else uses the Cairo rendering backend)

2 1