1289200 - Add grid areas to css grid properties exposed to dev tools

Status: RESOLVED FIXED

(Core :: CSS Parsing and Computation, defect)

Description

[Brad Werth [:bradwerth]]

Augment the properties exposed in bug 1241932 to include information on implicit and explicit grid areas.

Comment 1

[Brad Werth [:bradwerth]]

Attachment: gridAreas2.patch

Adds areas property to the grid objects returned by <element>.getGridFragments().

https://treeherder.mozilla.org/#/jobs?repo=try&revision=79379f4496c8

Comment 2

[Brad Werth [:bradwerth]]

Attachment: gridAreas3.patch

Quick update to support Gabe's prototype. Explicit GridAreas get bounding lines, but implicit GridAreas do not (yet).

Comment 3

[Gabriel Luong [:gl] (ΦωΦ)]

Comment on attachment 8782147 [details] [diff] [review]
gridAreas3.patch

Current patch to get the bounding star/end row and column for css grid area has been working great!

Comment 4

[Brad Werth [:bradwerth]]

Attachment: gridAreas4.patch

Adds grid areas, add comments to the Grid.webidl explaining which numbers are 0-indexed and which are 1-indexed. Explicit grid areas get correct boundary lines; implicit grid areas get 0 for all boundary lines until Bug 1297189 resolves that.

I'd like to land this to get the areas and the implicit/explicit state stuff to the dev tools team while working on the implicit boundary lines.

https://treeherder.mozilla.org/#/jobs?repo=try&revision=072ad0d350fe

Comment 5

[Mats Palmgren (inactive)]

Comment on attachment 8783717 [details] [diff] [review]
gridAreas4.patch

r=mats on the layout changes.  You should ask a DOM peer to review the rest.

Comment 6

[Brad Werth [:bradwerth]]

Comment on attachment 8783717 [details] [diff] [review]
gridAreas4.patch

Requesting webidl review from bz as a DOM peer.

Comment 7

[Boris Zbarsky [:bzbarsky]]

Comment on attachment 8783717 [details] [diff] [review]
gridAreas4.patch

This is going to return the implicit areas in some random order, right?  Is that OK?

Is there a reason for the separate SetAreaValues function instead of just passing that stuff to the constructor?  Can any of those values change on a GridArea object?  If not, it would be good to make them const on GridArea and pass them to the constructor.

>+          (i > aTrackInfo->mNumLeadingImplicitTracks +
>+               aTrackInfo->mNumExplicitTracks) ?

Should this be > or >=?  That is, if mNumExplicitTracks and mNumLeadingImplicitTracks are both 0, what should happen for i == 0?  Should it be explicit or implicit?

It's not clear to me why it should be explicit in that situation; the comment seems to say it should, but not _why_...

>+  [Cached, Pure]
>+  readonly attribute sequence<GridArea> areas;

It's actually [Constant], not just [Pure], right?

>+  readonly attribute sequence<DOMString> names;

And this one too?  Also, please put the extended attributes (the stuff in []) right before this line, not way off before the comment.

r=me with those fixed.

Comment 8

[Brad Werth [:bradwerth]]

Attachment: gridAreas5.patch

Per review from bz: Consolidated GridArea setter into constructor, improved comments, and updated GridLine type.

https://treeherder.mozilla.org/#/jobs?repo=try&revision=b263a9513cf7

Comment 9

[Gabriel Luong [:gl] (ΦωΦ)]

https://hg.mozilla.org/integration/fx-team/rev/c98b8c992c2f3dfc4789deab913ca942ae61f304
Bug 1289200 - Adds GridAreas to grid css dev tools API. r=bz, mats

Comment 10

[Ryan VanderMeulen [:RyanVM]]

https://hg.mozilla.org/mozilla-central/rev/c98b8c992c2f