Clear floats that are nested in previous siblings and don't overhang.

Clear floats that are nested in previous siblings and don't overhang.

If margin collapsing involving negative margins moves a child up into a previous
sibling we need to clear any floats it contains no matter how nested they are.

This means that we should keep a record of the lowest enclosed float around so
that siblings with negative margin can clear/avoid it if necessary. We do this already
we just need to be more thorough about it.

There could be performance implications here.

BUG=535381
Committed: https://crrev.com/adab870d3d0f09c1be51b5bc326d7b0d385794dc
Cr-Commit-Position: refs/heads/master@{#353126}

[rhogan, 2015-10-06 19:10:52]

robhogan@gmail.com changed reviewers:
+ leviw@chromium.org

[leviw_travelin_and_unemployed, 2015-10-08 08:40:25]

Code change makes perfect sense. I'm not sure we have a choice on the perf
implications.

LGTM with test nits :)

https://codereview.chromium.org/1393533002/diff/60001/third_party/WebKit/Layo...
File
third_party/WebKit/LayoutTests/fast/block/margin-collapse/clear-nested-float-more-than-one-previous-sibling-away.html
(right):

https://codereview.chromium.org/1393533002/diff/60001/third_party/WebKit/Layo...
third_party/WebKit/LayoutTests/fast/block/margin-collapse/clear-nested-float-more-than-one-previous-sibling-away.html:25:
<div style="position:absolute; width:100px; height:100px; background-color: red;
top:0px; z-index: -1;"></div>
Put this in the stylesheet as well? "#red" or something?

https://codereview.chromium.org/1393533002/diff/60001/third_party/WebKit/Layo...
third_party/WebKit/LayoutTests/fast/block/margin-collapse/clear-nested-float-more-than-one-previous-sibling-away.html:28:
<div id="fr" class="fr"></div>
id="float", no class?

https://codereview.chromium.org/1393533002/diff/60001/third_party/WebKit/Layo...
third_party/WebKit/LayoutTests/fast/block/margin-collapse/clear-nested-float-more-than-one-previous-sibling-away.html:29:
<div id="clearone" class="clear">Test</div>
id clearone isn't used. clear could be an id.

https://codereview.chromium.org/1393533002/diff/60001/third_party/WebKit/Layo...
third_party/WebKit/LayoutTests/fast/block/margin-collapse/clear-nested-float-more-than-one-previous-sibling-away.html:31:
<div id="clear2">Test</div>
no id necessary?

https://codereview.chromium.org/1393533002/diff/60001/third_party/WebKit/Layo...
third_party/WebKit/LayoutTests/fast/block/margin-collapse/clear-nested-float-more-than-one-previous-sibling-away.html:33:
<div id="oh" class="oh" data-offset-y=50></div>
id = test? no class

https://codereview.chromium.org/1393533002/diff/60001/third_party/WebKit/Layo...
third_party/WebKit/LayoutTests/fast/block/margin-collapse/clear-nested-float-more-than-one-previous-sibling-away.html:36:
window.checkLayout("#oh", document.getElementById("test-output"));
#oh could be #test?

[rhogan, 2015-10-08 18:57:11]

The CQ bit was checked by robhogan@gmail.com

[rhogan, 2015-10-08 18:57:11]

The patchset sent to the CQ was uploaded after l-g-t-m from leviw@chromium.org
Link to the patchset: https://codereview.chromium.org/1393533002/#ps80001
(title: "Updated")

[commit-bot: I haz the power, 2015-10-08 18:57:49]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1393533002/80001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1393533002/80001

[commit-bot: I haz the power, 2015-10-08 20:22:07]

Committed patchset #5 (id:80001)

[commit-bot: I haz the power, 2015-10-08 20:23:47]

Patchset 5 (id:??) landed as
https://crrev.com/adab870d3d0f09c1be51b5bc326d7b0d385794dc
Cr-Commit-Position: refs/heads/master@{#353126}