Get the right ascent/descent for the tallest vertically aligned container

Get the right ascent/descent for the tallest vertically aligned container

When we're figuring out the right adjustment to make to the ascent/descent in
a line to fill the full height of the line with vertical align top/bottom we
shouldn't carry forward the results of previous speculative adjustments from
one box on the line to the next.

Also ascent/descent are round()'ed so do the same to the lineHeight we're fitting
to.

BUG=604144
Committed: https://crrev.com/020751ff9d9524fe83cc858edf4d1aea78e305ae
Cr-Commit-Position: refs/heads/master@{#389668}

[rhogan, 2016-04-24 10:49:33]

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

[rhogan, 2016-04-24 10:51:43]

Description was changed from

==========
Get the right ascent/descent for the tallest vertically aligned container

BUG=604144
==========

to

==========
Get the right ascent/descent for the tallest vertically aligned container

When we're figuring out the right adjustment to make to the ascent/descent in
a line to fill the full height of the line with vertical align top/bottom we
shouldn't carry forward the results of previous speculative adjustments from
one box on the line to the next.

Also ascent/descent are round()'ed so do the same to the lineHeight we're
fitting
to.

BUG=604144
==========

[eae, 2016-04-25 00:31:28]

Change itself makes sense and matches how we handle accent and descent. I have
some questions about the tests however.

https://codereview.chromium.org/1910053003/diff/20001/third_party/WebKit/Layo...
File
third_party/WebKit/LayoutTests/fast/inline-block/vertical-align-top-and-bottom-expected.txt
(right):

https://codereview.chromium.org/1910053003/diff/20001/third_party/WebKit/Layo...
third_party/WebKit/LayoutTests/fast/inline-block/vertical-align-top-and-bottom-expected.txt:3:
Expected 98 for clientTop+offsetTop, but got 110, clientTop: 0, + offsetTop:
110.
Checking in FAIL lines for new tests is less than ideal...

https://codereview.chromium.org/1910053003/diff/20001/third_party/WebKit/Layo...
File
third_party/WebKit/LayoutTests/fast/inline-block/vertical-align-top-and-bottom.html
(right):

https://codereview.chromium.org/1910053003/diff/20001/third_party/WebKit/Layo...
third_party/WebKit/LayoutTests/fast/inline-block/vertical-align-top-and-bottom.html:27:
<div id="line" style="display: inline-block;" data-total-y=98>crbug.com/604144:
In an inline formatting context with a mix of top and bottom vertically aligned
inline-blocks use the lowest.</div>
Why 98?

[rhogan, 2016-04-25 20:16:18]

On 2016/04/25 at 00:31:28, eae wrote:
>
https://codereview.chromium.org/1910053003/diff/20001/third_party/WebKit/Layo...
>
third_party/WebKit/LayoutTests/fast/inline-block/vertical-align-top-and-bottom.html:27:
<div id="line" style="display: inline-block;" data-total-y=98>crbug.com/604144:
In an inline formatting context with a mix of top and bottom vertically aligned
inline-blocks use the lowest.</div>
> Why 98?

Whoops - corrected the test case and result.

[eae, 2016-04-26 00:21:44]

LGTM

[eae, 2016-04-26 00:21:51]

The CQ bit was checked by eae@chromium.org

[commit-bot: I haz the power, 2016-04-26 00:22:26]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1910053003/60001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1910053003/60001

[commit-bot: I haz the power, 2016-04-26 02:21:01]

Committed patchset #4 (id:60001)

[commit-bot: I haz the power, 2016-04-26 02:23:20]

Description was changed from

==========
Get the right ascent/descent for the tallest vertically aligned container

When we're figuring out the right adjustment to make to the ascent/descent in
a line to fill the full height of the line with vertical align top/bottom we
shouldn't carry forward the results of previous speculative adjustments from
one box on the line to the next.

Also ascent/descent are round()'ed so do the same to the lineHeight we're
fitting
to.

BUG=604144
==========

to

==========
Get the right ascent/descent for the tallest vertically aligned container

When we're figuring out the right adjustment to make to the ascent/descent in
a line to fill the full height of the line with vertical align top/bottom we
shouldn't carry forward the results of previous speculative adjustments from
one box on the line to the next.

Also ascent/descent are round()'ed so do the same to the lineHeight we're
fitting
to.

BUG=604144

Committed: https://crrev.com/020751ff9d9524fe83cc858edf4d1aea78e305ae
Cr-Commit-Position: refs/heads/master@{#389668}
==========

[commit-bot: I haz the power, 2016-04-26 02:23:21]

Patchset 4 (id:??) landed as
https://crrev.com/020751ff9d9524fe83cc858edf4d1aea78e305ae
Cr-Commit-Position: refs/heads/master@{#389668}