48181 - Incorrectly ordered formcontrol array (due to bad HTML)

Status: VERIFIED DUPLICATE of bug 49951

(Core :: DOM: Core & HTML, defect, P3)

Description

[gabriel]

In the attached minimal test case, the parameter values returned do not match
NS4, the parameters are in a different order.

Comment 1

[gabriel]

Attachment: test case

Comment 2

[gabriel]

WIN 95, build 2000080804.

Comment 3

[gabriel]

Also happens under Linux 2000080808.

Comment 4

[gabriel]

Expected results: when you click on the button, an alert should pop up showing
the username entered. If you try under NS4, you will see this is what happens.

Actual results: alert shows 'second' which is the value of the second parameter
in the form, rather than the requested fourth parameter (username).


This is a hugely important bug, as it could well screw up a lot of sites which
use javascript to parse form input.

Comment 5

[gabriel]

cc'ing asa...somebody needs to look at this quick.

Comment 6

[gabriel]

Not sure really what the component should be; trying with dom level 0.

Comment 7

[gabriel]

jst: is this a dom bug ?

(Sorry for all the spam).

Comment 8

[Johnny Stenback (:jst)]

Yup, it's a DOM bug. However, why is this a blocker? Doesn't seem critical at
all, but in stead it seems like a very likely Future candidate, lowering
severity to normal, here's why:

The reason the testcase doesn't work is that it contains invalid HTML, there's a
form tag, then a table tag, but then the form is closed before the table is
closed, ie wrong order.

I.e. the fix for this is to make the HTML valid.

The bug in mozilla is most likely in the fixup code that corrects the invalid
HTML, for some reason the two form controls inside the table end up first in the
document.FORM.param[] array and the three hidden formcontrols end up in the end
of the array.

It's a bug but I need a good reason as to why this is critical, unless I get one
this will be Futured.

Comment 9

[gabriel]

Please accept my apologies, I hadn't noticed the html error (must've been
working too hard :-/ ). I've tested with corrected html and it works fine.

I agree that this is not a serious bug. Removing asa from cc.

Comment 10

[Johnny Stenback (:jst)]

Ok, np, thanks for clearing that up, it is still a bug of course so I'll keep it
on my list but I won't get to it for the first release, so I'm futuring this for
now.

Comment 11

[Johnny Stenback (:jst)]

*** This bug has been marked as a duplicate of 49951 ***

Comment 12

[Phil Schwartau]

Marking Verified -