Recognize more sRGB gammas

Recognize more sRGB gammas

BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=1975883003

Committed: https://skia.googlesource.com/skia/+/c2e4bb39d4ac987cc011d171c89e18922ac06e75

[msarett, 2016-05-12 21:17:55]

Description was changed from

==========
Recognize more sRGB gammas

BUG=skia:
==========

to

==========
Recognize more sRGB gammas

BUG=skia:
GOLD_TRYBOT_URL=
https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&is...
==========

[msarett, 2016-05-12 21:19:44]

msarett@google.com changed reviewers:
+ scroggo@google.com

[msarett, 2016-05-12 21:19:45]

Of my set of 10k test images, this allows us to recognize 30 more sRGB profiles
that we were previously missing.

[scroggo, 2016-05-13 12:45:25]

lgtm

https://codereview.chromium.org/1975883003/diff/1/src/core/SkColorSpace.cpp
File src/core/SkColorSpace.cpp (right):

https://codereview.chromium.org/1975883003/diff/1/src/core/SkColorSpace.cpp#n...
src/core/SkColorSpace.cpp:343: 3366 == read_big_endian_short((const uint8_t*)
&table[257]) &&
Does still correctly identify images that were recognized by the old code?

https://codereview.chromium.org/1975883003/diff/1/src/core/SkColorSpace.cpp#n...
src/core/SkColorSpace.cpp:364: if (0 == read_big_endian_short((const uint8_t*)
&table[0]) &&
As we add more of these, it makes me wonder if we should lazily create the
SkColorSpace, so clients who won't ever apply it don't need to spend time
reading it and creating it. (That said, the cost is probably cheap compared to
decoding the image.)

[msarett, 2016-05-13 12:51:48]

https://codereview.chromium.org/1975883003/diff/1/src/core/SkColorSpace.cpp
File src/core/SkColorSpace.cpp (right):

https://codereview.chromium.org/1975883003/diff/1/src/core/SkColorSpace.cpp#n...
src/core/SkColorSpace.cpp:343: 3366 == read_big_endian_short((const uint8_t*)
&table[257]) &&
On 2016/05/13 12:45:24, scroggo wrote:
> Does still correctly identify images that were recognized by the old code?

Yes!

https://codereview.chromium.org/1975883003/diff/1/src/core/SkColorSpace.cpp#n...
src/core/SkColorSpace.cpp:364: if (0 == read_big_endian_short((const uint8_t*)
&table[0]) &&
On 2016/05/13 12:45:24, scroggo wrote:
> As we add more of these, it makes me wonder if we should lazily create the
> SkColorSpace, so clients who won't ever apply it don't need to spend time
> reading it and creating it. (That said, the cost is probably cheap compared to
> decoding the image.)

I think this is a really good idea.

[msarett, 2016-05-13 12:52:25]

The CQ bit was checked by msarett@google.com

[commit-bot: I haz the power, 2016-05-13 12:52:27]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1975883003/1
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1975883003/1

[commit-bot: I haz the power, 2016-05-13 13:19:06]

Description was changed from

==========
Recognize more sRGB gammas

BUG=skia:
GOLD_TRYBOT_URL=
https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&is...
==========

to

==========
Recognize more sRGB gammas

BUG=skia:
GOLD_TRYBOT_URL=
https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&is...

Committed:
https://skia.googlesource.com/skia/+/c2e4bb39d4ac987cc011d171c89e18922ac06e75
==========

[commit-bot: I haz the power, 2016-05-13 13:19:07]

Committed patchset #1 (id:1) as
https://skia.googlesource.com/skia/+/c2e4bb39d4ac987cc011d171c89e18922ac06e75