Bug 96124

Summary:	onLoad="focus()" is executed even after I've started typing in a field
Product:	[Applications] konqueror	Reporter:	doctorkaeding
Component:	khtml	Assignee:	Konqueror Developers <konq-bugs>
Status:	RESOLVED NOT A BUG
Severity:	normal	CC:	christian.nitschkowski, mers
Priority:	NOR
Version:	3.3.1
Target Milestone:	---
Platform:	Compiled Sources
OS:	Linux
Latest Commit:		Version Fixed In:

Description doctorkaeding 2005-01-01 19:22:36 UTC

Version:           3.3.1 (using KDE KDE 3.3.1)
Installed from:    Compiled From Sources
Compiler:          gcc 3.4.2 
OS:                Linux

This is similar, but not the same as another bug re Yahoo Mail;
and I notice this behavior in Yahoo's login page as well.

onLoad="focus(....)" is being executed by Konqueror even after I have
focused on a text input field with the mouse and started typing.  I get
about halfway through the password, when focus is set on the "ID" field.

I suggest either:
- only execute focus() if focus is not already set
- execute the onLoad stuff *before* the graphics are loaded

Thanks.

Comment 1 Philip Rodrigues 2006-09-01 13:06:23 UTC

Can you provide a minimal testcase that shows this behaviour?

Comment 2 Maksim Orlovich 2006-10-28 23:03:50 UTC

*** Bug 83047 has been marked as a duplicate of this bug. ***

Comment 3 Christian Nitschkowski 2006-10-31 18:59:25 UTC

I don't think this is actually a valid bug.
It's simply a "not nice" design descision by the website's designer.
When calling "onLoad" before receiving the images, some script might get into trouble when it expects the page to be loaded completely.
When you don't call "focus()" after the user had already set the focus manually, I'm sure you will break some scripts as well.
Either way: Some users will be upset.
I guess it's best to stick with the current version as it implements what I'd expect from a web browser from the web designer's view.

Comment 4 illogic-al 2007-01-06 17:02:10 UTC

Is this a bug or intended behavior devs?

Comment 5 Stefan Monov 2007-10-09 10:10:38 UTC

I agree with Christian. Both solutions proposed by the reporter contradict the standards for no good reason other than making a particular site work. Closing as invalid.