128420 - [FIX]getComputedStyle should raise NOT_SUPPORTED_ERR

Status: RESOLVED FIXED

(Core :: DOM: CSS Object Model, defect, P3)

Description

[Brian 'netdragon' Bober]

Error: uncaught exception: [Exception... "Component returned failure code:
0x80004003 (NS_ERROR_INVALID_POINTER) [nsIDOMViewCSS.getComputedStyle]" 
nsresult: "0x80004003 (NS_ERROR_INVALID_POINTER)"  location: "JS frame ::
file:///F:/netscape/projects/patchandtest/Bug/kitchensinkerror.html ::
<TOP_LEVEL> :: line 65"  data: no]

Comment 1

[Brian 'netdragon' Bober]

Attachment: Error on GetComputedStyle

The error appeared in the javascript console.

Comment 2

[Christopher Aillon (sabbatical, not receiving bugmail)]

Well, that's because you pass an undefined value for the element value of
getComputedStyle().  Try adding an |alert (typeof elt);| right after you assign
it.  When you call document.getElementById("sink") there is no element with that
ID.  It doesn't get defined until after you run your JS.

Define your element before trying to access it and you'll see that it does
indeed work fine.  Note though that the value will be practically useless to you
though as it will be "auto".  the top property only applies to positioned
elements and thus returns the default value per the spec.

Comment 3

[Brian 'netdragon' Bober]

Attachment: New

You're right.

Comment 4

[Brian 'netdragon' Bober]

I still don't like the exception it returned. It should give something a little
more informative (object not initialized) or something.

Comment 5

[Boris Zbarsky [:bzbarsky]]

Um... there is no object to not be initialized.  There is a reference to an
element and the reference is "null".  It's a null pointer.

That said, getComputedStyle is defined as returning no exceptions (even for
stupid cases like this).  I hate the DOM Style spec.  So we should just be
silently failing, with an assert in debug builds.

Comment 6

[Boris Zbarsky [:bzbarsky]]

to caillon, but feel free to punt to me if you don't want to deal.  :)

Comment 7

[Phil Schwartau]

Setting default QA -

Comment 8

[Christopher Aillon (sabbatical, not receiving bugmail)]

Daniel, should this be fixed?  Or will the new CSS OM allow this (or another?)
exception?  Let me know and I'll whip out a patch.

Comment 9

[Daniel Glazman (:glazou) (not active in Mozilla any more)]

raising issue to CSS WG

Comment 10

[Daniel Glazman (:glazou) (not active in Mozilla any more)]

resolution is "should throw NOT_SUPPORTED_ERR DOMException"

Comment 11

[Christopher Aillon (sabbatical, not receiving bugmail)]

Thanks Daniel.  I'll have a patch tomorrow, about to go to bed.

Comment 12

[Christopher Aillon (sabbatical, not receiving bugmail)]

Attachment: Raise the right exception.

Daniel, could you r=?  Boris, sr=?

Comment 13

[Boris Zbarsky [:bzbarsky]]

Comment on attachment 97210 [details] [diff] [review]
Raise the right exception.

sr=bzbarsky

Comment 14

[Daniel Glazman (:glazou) (not active in Mozilla any more)]

Comment on attachment 97210 [details] [diff] [review]
Raise the right exception.

r=glazman
check that in ! let's make the specs on thursday and be compliant on saturday
:-)

Comment 15

[Christopher Aillon (sabbatical, not receiving bugmail)]

So let it be written.  So let it be done.