43721 - [INVALID]WIDTH and HEIGHT of DIV not inclusive of PADDING and BORDER

Status: VERIFIED INVALID

(Core :: Layout, defect, P3)

Description

[Val Sharp]

From Bugzilla Helper:
User-Agent: Mozilla/4.6 [en-gb]C-CCK-MCD NetscapeOnline.co.uk  (Win98; I)
BuildID:    2000061311

If the WIDTH and/or HEIGHT of a DIV is set explicitly by a style, either inline 
or otherwise, and PADDING and/or BORDER are also defined, then the dimensions of 
the resulting DIV will be WIDTH/HEIGHT plus the PADDING and/or BORDER.

Shouldn't WIDTH be inclusive of PADDING and BORDER?

Reproducible: Always
Steps to Reproduce:
For example:

1.<div style="height:300; width:200; padding:30; border-width:4px; 
border-style:solid"></div> 


Actual Results:  DIV on the screen has actual width of 268px and height of 368px

Expected Results:  DIV should have actual width of 200px and height of 300px - 
ie contents should make room for padding and border.

Comment 1

[Hixie (not reading bugmail)]

Nope, sorry. See http://www.w3.org/TR/REC-CSS2/ for more details.
Note that IE5 gets this wrong.

Comment 2

[Stephen Walker]

verified. using the 2000081308 nightly on win2k

Comment 3

[Asa Dotzler [:asa]]

*** Bug 54201 has been marked as a duplicate of this bug. ***

Comment 4

[post]

If you have a specified with (100%) and you specify a border-width, your element
expands beyond 100%.

Im sorry, it makes absolutely no sense to have
an element expand beyond the specified width.

say I wanted a DIV to take up 100% of a cell, and I wanted a 3px border and 3px
padding. I could not do this because it would over lap the cell border.

please reconsider this behavior. I can forsee allot of problems.

Comment 5

[David Baron :dbaron: (⌚️UTC-5, no longer working on Mozilla)]

You mean 'width: auto' instead of 'width: 100%'.  You could also create an extra
element in the markup.  (And I bet you'd expect it to work the right way for
images!  This way is also more flexible.)

Comment 6

[post]

Attachment: overlapping div

Comment 7

[post]

I'm not quite sure what you mean.
I still do not agree with the behavior, but my opinion doesnt count for much:)
I created the above attachment to display the "problem".

Comment 8

[Val Sharp]

Realised pretty quickly back in June that Mozilla is correct here.
(Mislead by other browsers, and should have read the spec!)

Have since come to believe however that for _percentage_ widths/heights (esp. 
for percentages less than 100%) it does seem to be counter-intuitive, and makes 
precision in layout less easy.

But this is a W3C rather than a Mozilla issue.
Posted, with possibly inane suggestion, to the www-style newsgroup - "Percentage 
values for Height and Width properties" - Mon, 28 Aug 2000) - to no effect.

Comment 9

[Dave Townsend [:mossop]]

*** Bug 361950 has been marked as a duplicate of this bug. ***