[css-grid] Wipe SizingOperation out

[css-grid] Wipe SizingOperation out

crrev.com/2823183003 was the first step towards removing the SizingOperation
enum which was added in the past as a convenient way to modify the track
sizing algorithm behaviour depending on whether the size was
indefinite/definite but also on whether we were computing the grid intrinsic
size or doing a layout. As mentioned in the CL above that was a mistake
because indefinite sizes do not always correspond to preferred widths
computations.

Now that we have the algorithm in a separate class with different strategies
the existence of that enum is no longer required. We can handle the
remaining code path using SizingOperation just by using the algorithm
strategies.

No new tests required as this does not imply a change in the behaviour.

Review-Url: https://codereview.chromium.org/2842413003
Cr-Commit-Position: refs/heads/master@{#468599}
Committed: https://chromium.googlesource.com/chromium/src/+/8829f12bd0eb33225c2a562fa1d5c9493aadd09a

[svillar, 2017-04-27 15:02:33]

svillar@igalia.com changed reviewers:
+ eae@chromium.org, jfernandez@igalia.com, rego@igalia.com

[svillar, 2017-04-27 15:02:34]

Sending out for review.

[Manuel Rego, 2017-04-27 16:06:49]

Shouldn't this be split in 2 patches?

https://codereview.chromium.org/2842413003/diff/20001/third_party/WebKit/Sour...
File third_party/WebKit/Source/core/layout/GridTrackSizingAlgorithm.cpp (right):

https://codereview.chromium.org/2842413003/diff/20001/third_party/WebKit/Sour...
third_party/WebKit/Source/core/layout/GridTrackSizingAlgorithm.cpp:425:
GridTrackSizingAlgorithmStrategy::ExtentForBaselineAlignment(
This seems unrelated to the SizingOperation removal.

https://codereview.chromium.org/2842413003/diff/20001/third_party/WebKit/Sour...
third_party/WebKit/Source/core/layout/GridTrackSizingAlgorithm.cpp:427: auto
grid = algorithm_.layout_grid_;
Nit: We've GetLayoutGrid() for this.

[jfernandez, 2017-04-28 08:45:01]

For me, the change lgtm but let's agree something about @rego's comment on the
unrelated refactoring. I like it a lot, so if it's not going to be landed now,
please, don't forget to do it ASAP in a new CL.

[Manuel Rego, 2017-04-28 08:59:33]

On 2017/04/28 08:45:01, jfernandez wrote:
> For me, the change lgtm but let's agree something about @rego's comment on the
> unrelated refactoring. I like it a lot, so if it's not going to be landed now,
> please, don't forget to do it ASAP in a new CL.

If you want to land it as part of this CL, please update
the description to talk about it.
I was trying to understand what that change had to do with the SizingOperation
removal, when I realized it was just an independent refactoring.
I don't think creating 2 different CLs is a big deal.

[Manuel Rego, 2017-05-02 09:24:44]

Thanks LGTM too.

[svillar, 2017-05-02 09:25:04]

The CQ bit was checked by svillar@igalia.com

[svillar, 2017-05-02 09:25:04]

The patchset sent to the CQ was uploaded after l-g-t-m from
jfernandez@igalia.com
Link to the patchset: https://codereview.chromium.org/2842413003/#ps40001
(title: "Fix w/o refactoring")

[commit-bot: I haz the power, 2017-05-02 09:25:18]

CQ is trying da patch.

Follow status at:
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[commit-bot: I haz the power, 2017-05-02 11:32:25]

CQ is committing da patch.

Bot data: {"patchset_id": 40001, "attempt_start_ts": 1493717104151720,
"parent_rev": "032e6e69429f3d726bcf71d342752a7f04824533", "commit_rev":
"8829f12bd0eb33225c2a562fa1d5c9493aadd09a"}

[commit-bot: I haz the power, 2017-05-02 11:32:36]

Description was changed from

==========
[css-grid] Wipe SizingOperation out

crrev.com/2823183003 was the first step towards removing the SizingOperation
enum which was added in the past as a convenient way to modify the track
sizing algorithm behaviour depending on whether the size was
indefinite/definite but also on whether we were computing the grid intrinsic
size or doing a layout. As mentioned in the CL above that was a mistake
because indefinite sizes do not always correspond to preferred widths
computations.

Now that we have the algorithm in a separate class with different strategies
the existence of that enum is no longer required. We can handle the
remaining code path using SizingOperation just by using the algorithm
strategies.

No new tests required as this does not imply a change in the behaviour.
==========

to

==========
[css-grid] Wipe SizingOperation out

crrev.com/2823183003 was the first step towards removing the SizingOperation
enum which was added in the past as a convenient way to modify the track
sizing algorithm behaviour depending on whether the size was
indefinite/definite but also on whether we were computing the grid intrinsic
size or doing a layout. As mentioned in the CL above that was a mistake
because indefinite sizes do not always correspond to preferred widths
computations.

Now that we have the algorithm in a separate class with different strategies
the existence of that enum is no longer required. We can handle the
remaining code path using SizingOperation just by using the algorithm
strategies.

No new tests required as this does not imply a change in the behaviour.

Review-Url: https://codereview.chromium.org/2842413003
Cr-Commit-Position: refs/heads/master@{#468599}
Committed:
https://chromium.googlesource.com/chromium/src/+/8829f12bd0eb33225c2a562fa1d5...
==========

[commit-bot: I haz the power, 2017-05-02 11:32:37]

Committed patchset #3 (id:40001) as
https://chromium.googlesource.com/chromium/src/+/8829f12bd0eb33225c2a562fa1d5...