446273 - Column rule does not always stretch the full height of the column box.

Status: VERIFIED FIXED

(Core :: Layout, defect)

Description

[philippe (part-time)]

Attachment: testcase (requires Ahem)

If a column isn't filled with content, the rule is not drawn at full height of the column. WebKit always draws the rules at full height of the column, whereas Gecko stops with the text.

http://www.w3.org/TR/css3-multicol/#column4 states
'The length of the column gaps and column rules is equal to the length of the columns'
But also:
'Column rules are only drawn between columns that have content in the normal flow'

This _could_ be interpreted to validate what Gecko does. However, WebKit's implementation is more intuitive.

Comment 1

[philippe (part-time)]

Attachment: screenshot (gecko vs WebKit)

Gecko on top, bottom is WebKit latest build

Gecko/2008071918 Minefield/3.1a1pre

Comment 2

[Bill Gianopoulos [:WG9s]]

WORKSFORME with:

Mozilla/5.0 (X11; U; Linux i686 (x86_64); en-US; rv:1.9.1a1pre) Gecko/2008072004 Firefox/3.01a1pre ID:2008072004

My rendering matches what your screenshot shows for WebKit.

Comment 3

[Bill Gianopoulos [:WG9s]]

(In reply to comment #2)
> WORKSFORME with:
> 
> Mozilla/5.0 (X11; U; Linux i686 (x86_64); en-US; rv:1.9.1a1pre)
> Gecko/2008072004 Firefox/3.01a1pre ID:2008072004
> 
> My rendering matches what your screenshot shows for WebKit.
> 
Oops.  Please ignore above.  Sorry for bugspam.  My rendering matches what you show for gecko. 

Comment 4

[Michael Ventnor]

Intentional. Roc said that he wanted the column rule to be the minimum height of each column frame, so the last column rule changes size with the last column frame.

Comment 5

[j.j.]

Attachment: testcase 2

IMHO this looks obviously wrong

Comment 6

[philippe (part-time)]

(In reply to comment #4)
> Intentional. Roc said that he wanted the column rule to be the minimum height
> of each column frame, so the last column rule changes size with the last column
> frame.
As I said in comment 0, the spec seems to allow it.

Hmm, then we have an inter-operability problem with WebKit. I don't think there are other implementations of the multi-col module out there – or does PrinceXML implement it ?

I'll raise the issue with the CSS-WG.

Comment 7

[Robert O'Callahan (:roc) (email my personal email if necessary)]

(In reply to comment #4)
> Intentional. Roc said that he wanted the column rule to be the minimum height
> of each column frame, so the last column rule changes size with the last column
> frame.

Sorry, I think I was wrong. Let's just change it to match Webkit (and, I think, the spec).

Comment 8

[Michael Ventnor]

Attachment: Patch

Comment 9

[Robert O'Callahan (:roc) (email my personal email if necessary)]

This isn't right. The line should go all the way to the bottom of the column set and not depend on the content height of any particular column.

Comment 10

[Michael Ventnor]

Attachment: Patch 2

Comment 11

[Robert O'Callahan (:roc) (email my personal email if necessary)]

Comment on attachment 330540 [details] [diff] [review]
Patch 2

+  nsRect contentRect = GetContentRect();
+  contentRect -= GetRect().TopLeft();
+  contentRect += aPt;

Write this as one expression.

And of course you'll need tests. Make sure you test an element with padding so we know we're using the right edge.

Comment 12

[Michael Ventnor]

Attachment: Patch 3

With padding tests.

Comment 13

[Dave Camp (:dcamp)]

Pushed to mozilla-central as revision 5c45945bd43b.

Comment 14

[philippe (part-time)]

Roc, Michael: thank you !

verified with Mozilla/5.0 (Macintosh; U; Intel Mac OS X 10.5; en-US; rv:1.9.1a1pre) Gecko/2008072202 Minefield/3.1a1pre