[prev in list] [next in list] [prev in thread] [next in thread] 

List:       kfm-devel
Subject:    Re: KHTML and onLoad
From:       Tobias Anton <TA () ESC-Electronics ! de>
Date:       2004-09-27 17:18:05
Message-ID: 200409272018.59940.TA () ESC-Electronics ! de
[Download RAW message or body]

On Montag, 27. September 2004 14:03, Sebastien Richard wrote:
> Tobias Anton wrote:
> >In that case, completed() would adequately be emitted on load, but a
> > second later, we'd have to "revoke" it, because then, we're in progress
> > of loading again.
>
> Why shouldn't use completed(pending=true) in this case ?

Because we already emit completed(with no parameter) in that case.
We should have dropped completed(with no parameter) in favor of 
completed(bool) long ago, but I assume that too much software depended on 
that.

What is your problem with the current implementation and what behaviour would 
you expect for the general case, if in a page's javascript code a redirection 
exists that possibly depends on dynamic parameters?

How would you decide whether to emit completed() or completed(bool)?

> Actually, I am using KHTML to render web pages, so I would like to be 
> able to render any page with an onLoad without having to change it before.

Huh? Wasn't your statement before that it did render your page correctly? 

The completed()-signal has little to do with KHTML's rendering. The signal is 
only important for a shell that embeds a KHTMLView to display a "finished 
loading"-message at the Right Time. And the Right Time is when your web page 
and page1.html and page2.html are completely loaded. Page3.html is loaded 
_after_ completion. Does IE, for example, display the "finished"-message 
later than we do?

Cheers
-- Tobias
[prev in list] [next in list] [prev in thread] [next in thread]