gfx::ImageSkia: Added a note about using this class vs ImageFamily.

gfx::ImageSkia: Added a note about using this class vs ImageFamily.

This is the source of some confusion; ImageSkia has been abused in the
past for storing different logical image sizes, where an ImageFamily is
the appropriate class. This comment mirrors the comment in ImageFamily
explaining when Image/ImageSkia is appropriate vs ImageFamily.

Committed: https://crrev.com/fa117f242ff32e5154023120962f1791595a44f5
Cr-Commit-Position: refs/heads/master@{#361067}

[Matt Giuca, 2015-11-18 01:58:53]

mgiuca@chromium.org changed reviewers:
+ rsesek@chromium.org

[Robert Sesek, 2015-11-21 03:42:32]

LGTM

I think it's hard for people to misuse this given the APIs that return these
objects, but this is a subtle distinction that's worth clarifying.

[Matt Giuca, 2015-11-23 03:08:21]

On 2015/11/21 03:42:32, Robert Sesek wrote:
> I think it's hard for people to misuse this given the APIs that return these
> objects, but this is a subtle distinction that's worth clarifying.

They are still actively being misused by old code (e.g.,
extensions/browser/image_loader.cc LoadImagesAsync) that builds an ImageSkia
full of differently sized bitmaps, each associated with 100P. Then other code
that handles those ImageSkias needs to know that they are "special" ImageSkias
where you have to not trust the scale factor of each bitmap, and instead iterate
over them to find the image of the size you are looking for.

That is why I created ImageFamily in the first place. But I still get people
asking "why don't you just use ImageSkia instead of ImageFamily", so I wanted to
add this comment to point them at.

[Matt Giuca, 2015-11-23 03:08:26]

The CQ bit was checked by mgiuca@chromium.org

[commit-bot: I haz the power, 2015-11-23 03:08:30]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1438603003/1
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1438603003/1

[commit-bot: I haz the power, 2015-11-23 04:16:53]

Committed patchset #1 (id:1)

[commit-bot: I haz the power, 2015-11-23 04:17:32]

Patchset 1 (id:??) landed as
https://crrev.com/fa117f242ff32e5154023120962f1791595a44f5
Cr-Commit-Position: refs/heads/master@{#361067}