13562 - select elements ignore 'size' when parent has padding

Status: VERIFIED DUPLICATE of bug 1939

(Core :: Layout: Form Controls, defect, P3)

Description

[Hixie (not reading bugmail)]

Select elements treat their 'size' attribute very oddly when their parent has
padding. They seem to decide to have a height which is a multiple of the size
and some other figure, and inherit the padding of their parent (!). This is
very odd to look at.

Furthermore, the padding-left is then carried through to the padding-right (!)
forcing, if the padding-left is bigger than the -right, a scroll bar to appear.
A similar oddity occurs with the -top and -bottom paddings.

For example:

   <p style="padding: 2em">
      <select size=4>
         <option>a</option>
         <option>b</option>
         <option>c</option>
         <option>d</option>
         <option>e</option>
         <option>f</option>
         <option>g</option>
      </select>
   </p>

In the above, the select element gets 2em padding between its frame and its
options, and has a huge height. This is visible on the following test pages:

   http://www.bath.ac.uk/%7Epy8ieh/internet/projects/mozilla/selectsize.html
   http://www.bath.ac.uk/%7Epy8ieh/internet/projects/mozilla/optgroup.shtml

Explicitly setting the padding on SELECT or OPTION seems to have little or
no effect.

Comment 1

[karnaze (gone)]

Reassigning to Rod.

Comment 2

[rods (gone)]

I thought there was a bug on this already, but I guess not. I did mention this
in some check in message. Yes, we don't obey padding or borders and it is a bug
problem (code wise) on trying to make this happen.

Basically, if you set the border and padding on a select the dropdown picks
those style settings and messes all the sizing up. So I will keep this bug open
to track this problem, this probably won't be fixed for beta.

Comment 3

[Hixie (not reading bugmail)]

rods: I'm not setting padding on the SELECT. I'm setting it on its parent.

Comment 4

[Hixie (not reading bugmail)]

Yup, there is a bug already open on this: bug 1939. Marking as DUP.

*** This bug has been marked as a duplicate of 1939 ***

Comment 5

[Sebastian Späth]

cleaning up: VERIFY duplicate