ES Team Update (June 30)

[Shiki Okasaka]

Hi team,

This week I've submitted a Firefox 3 plugin with which we can control
DOM objects in Firefox 3 from ES. The attached screen shot shows we
can draw in the Firefox 3 canvas from esjs, like we can do so with our
own canvas services running on an actual PC (QEMU) and on a pseudo
frame buffer.

Note a new standard Web IDL draft has been published from W3C two weeks ago:

http://dev.w3.org/2006/webapi/Binding4DOM/

Our plan is to modify the esidl to support this new Web IDL, and test
the esidl outputs with Firefox 3 using the plugin. Shortly, I will
post how to build the ES Firefox 3 plugin.

Cheers,

- Shiki

[Santosh vattam]

Hi all,

Finally!! The exams are over! Back to work!
--
Cheers!

Santosh G Vattam

[Chis Dan]

Hi all,

Last week I was on "vacation" :d, because I was tired after all the exams. I did however look over the WebKit source code and made great progress in understanding how the code is organized. This week I will actually start coding and I will begin by writing a basic browser using webkit and adding es to the platform directories with stubs for platform dependent stuff. The main goal will be to get some basic rendering using the pseudo framebuffer. Working in this top-down fashion will allow us to have results before a complete integration of WebKit is achieved.

Regards,
Dan

[Ishibashi Kenichi]

Hi all,

Last week, I submitted a basic support of Variant type into esidl.
As part of the supporting Web IDL preparations, I've switched my
development environment from F8 to F9 and build Firefox3 from source
code on F9.

Sorry that I'm going to make a presentation on domestic conference in
this week, so I think I can't take a time to ES.

--
ISHIBASHI Kenichi

[joao taveira]

Hi,

Last week I spent all of my time running through the networking testsuites with gdb. I'm now feel pretty comfortable with debugging and I'm understanding the conduits framework a lot better from analyzing how packets move through the stack, but I'm a bit confused on what the best way to check for bugs. It seems that fixing bugs is a lot easier than finding them in these cases. Ideally there would be some regression testing tool to stress test our implementation and generate reports on standards compliance, but I'm not quite sure any open source suite is available (actually, one SoC project is implementing this in FreeBSD).

Cheers,
João.