Re: [webkit-dev] Help with a possible Webkit bug UPDATE

2011-09-26 Thread malch


This problem appears to be a duplicate of:

https://bugs.webkit.org/show_bug.cgi?id=61452

I have annotated that bug report with a few additional comments.

Thanks to David Brown for his help with this.
-- 
View this message in context: 
http://old.nabble.com/Help-with-a-possible-Webkit-bug-tp32503920p32503951.html
Sent from the Webkit mailing list archive at Nabble.com.

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


[webkit-dev] Help with a possible Webkit bug

2011-09-25 Thread malch

I have smacked into a problem which appears to have it's root within Webkit.
However, I'm well outside of my familiar territory and am seeking some help
to further understand the issue and formulate a useful bug report, if
appropriate.

Here's the background:

1. I have a web page with a TinyMCE edit control.

2. The document being edited is HTML and includes a lt;IMG SRC...gt; tag.
The referenced image is generated dynamically! The problem does not arise
with working static images but it does with dynamically generated images
and, interestingly, static images that are 404 Not Found.

3. I also have a 0.5 second timer that constantly checks the contents of the
TinyMCE edit control. 

On Webkit browsers (Chrome, Safari) every time I check the contents of the
edit control, the browser makes a new request for the aforementioned image.
That's an issue for me because with multiple users making 2 requests per
second, my server is getting hammered!

It appears I've found a workaround. This involves a change to some options
set when requesting the edit control contents from TinyMCE.

With the right set of options I can get TinyMCE to acquire the contents of
the control via the innerHTML method. This prevents the constant stream of
image requests from client to server.

The problem does arise when TinyMCE attempts to fetch the contents via the
cloneNode() method.

So, it seems to me that cloneNode() is triggering a bunch of unnecessary
HTTP requests and that's probably not a Good Thing!

Anyone have an insight, theories, or able to reproduce something like this?

-- 
View this message in context: 
http://old.nabble.com/Help-with-a-possible-Webkit-bug-tp32503920p32503920.html
Sent from the Webkit mailing list archive at Nabble.com.

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