Convert fast/multicol/column-count-with-rules (and related) to reftests.

Convert fast/multicol/column-count-with-rules (and related) to reftests.

 * Make test small enough to fit in content_shell. (Only the baselines for
   the -webkit-writing-mode versions were incorrect, though).
 * Used explicit column-gap to simplify ref.
 * Used even-size column-rule to simplify ref.

R=mstensho@opera.com
BUG=386567
Committed: https://src.chromium.org/viewvc/blink?view=rev&revision=184202

[mstensho (USE GERRIT), 2014-10-20 21:42:31]

Only raised issues in the root directory, but the same issues apply in
vertical-{lr,rl}/

https://codereview.chromium.org/670473002/diff/1/LayoutTests/fast/multicol/co...
File LayoutTests/fast/multicol/column-count-with-rules-expected.html (right):

https://codereview.chromium.org/670473002/diff/1/LayoutTests/fast/multicol/co...
LayoutTests/fast/multicol/column-count-with-rules-expected.html:13: display:
inline-block;
float:left instead of display:inline-block? Then there will be no need to mess
with vertical alignment or worry about whitespace between the elements.

https://codereview.chromium.org/670473002/diff/1/LayoutTests/fast/multicol/co...
LayoutTests/fast/multicol/column-count-with-rules-expected.html:23: position:
absolute;
You could also do the rules as floats, of course, between the "columns". Might
even make them as wide as the coulmn-gap, by setting "margin:0 9px", and then
let the "column" elements have the correct width of 100px, instead of 120px.

https://codereview.chromium.org/670473002/diff/1/LayoutTests/fast/multicol/co...
File LayoutTests/fast/multicol/column-count-with-rules.html (left):

https://codereview.chromium.org/670473002/diff/1/LayoutTests/fast/multicol/co...
LayoutTests/fast/multicol/column-count-with-rules.html:1: <p
style="-webkit-column-count:3; -webkit-column-rule:3px solid black;
column-count:3; column-rule:3px solid black; column-fill:auto; height:100px;
border:10px solid maroon; padding:20px;">
I think we should keep the padding. It's interesting to test that rules are not
painted in the padding area.

[andersr, 2014-10-21 12:33:35]

Updated patch. Also removed mention of padding-removal from description.

PTAL.

https://codereview.chromium.org/670473002/diff/1/LayoutTests/fast/multicol/co...
File LayoutTests/fast/multicol/column-count-with-rules-expected.html (right):

https://codereview.chromium.org/670473002/diff/1/LayoutTests/fast/multicol/co...
LayoutTests/fast/multicol/column-count-with-rules-expected.html:13: display:
inline-block;
On 2014/10/20 21:42:31, mstensho wrote:
> float:left instead of display:inline-block? Then there will be no need to mess
> with vertical alignment or worry about whitespace between the elements.

Done.

https://codereview.chromium.org/670473002/diff/1/LayoutTests/fast/multicol/co...
LayoutTests/fast/multicol/column-count-with-rules-expected.html:23: position:
absolute;
On 2014/10/20 21:42:31, mstensho wrote:
> You could also do the rules as floats, of course, between the "columns". Might
> even make them as wide as the coulmn-gap, by setting "margin:0 9px", and then
> let the "column" elements have the correct width of 100px, instead of 120px.

Yeah, that's much better.

Done.

https://codereview.chromium.org/670473002/diff/1/LayoutTests/fast/multicol/co...
File LayoutTests/fast/multicol/column-count-with-rules.html (left):

https://codereview.chromium.org/670473002/diff/1/LayoutTests/fast/multicol/co...
LayoutTests/fast/multicol/column-count-with-rules.html:1: <p
style="-webkit-column-count:3; -webkit-column-rule:3px solid black;
column-count:3; column-rule:3px solid black; column-fill:auto; height:100px;
border:10px solid maroon; padding:20px;">
On 2014/10/20 21:42:31, mstensho wrote:
> I think we should keep the padding. It's interesting to test that rules are
not
> painted in the padding area.

Done.

[mstensho (USE GERRIT), 2014-10-21 12:39:22]

That second patch set doesn't look sane. It suddenly has a bunch of
column-break-with-balancing stuff.

[andersr, 2014-10-21 12:48:59]

On 2014/10/21 12:39:22, mstensho wrote:
> That second patch set doesn't look sane. It suddenly has a bunch of
> column-break-with-balancing stuff.

Oops. Used the wrong base, sorry.

Fixed now.

[mstensho (USE GERRIT), 2014-10-21 13:00:18]

lgtm

[andersr, 2014-10-22 16:52:42]

The CQ bit was checked by andersr@opera.com

[commit-bot: I haz the power, 2014-10-22 16:53:48]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/670473002/40001

[commit-bot: I haz the power, 2014-10-22 18:00:57]

Committed patchset #3 (id:40001) as 184202