Add checks for bogus profiles to SkColorSpace::NewICC()

Add checks for bogus profiles to SkColorSpace::NewICC()

This shouldn't change anything on Gold.

Definitely think we should have these to deal with whatever crazy
profiles are out in the wild.

I borrowed a few ideas from QCMS and also made some of my own
judgments.

BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2257983003

[msarett, 2016-08-19 19:54:44]

Description was changed from

==========
Add checks for bogus profiles to SkColorSpace::NewICC()

This shouldn't change anything on Gold.

Definitely think we should have these to deal with whatever crazy
profiles are out in the wild.

I borrowed a few ideas from QCMS and also made some of my own
judgments.

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

to

==========
Add checks for bogus profiles to SkColorSpace::NewICC()

This shouldn't change anything on Gold.

Definitely think we should have these to deal with whatever crazy
profiles are out in the wild.

I borrowed a few ideas from QCMS and also made some of my own
judgments.

BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2257983003
==========

[msarett, 2016-08-19 19:59:01]

The CQ bit was checked by msarett@google.com to run a CQ dry run

[commit-bot: I haz the power, 2016-08-19 19:59:04]

Dry run: CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[msarett, 2016-08-19 19:59:30]

msarett@google.com changed reviewers:
+ brianosman@google.com, herb@google.com, mtklein@google.com, reed@google.com

[commit-bot: I haz the power, 2016-08-19 20:27:12]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2016-08-19 20:27:14]

Dry run: This issue passed the CQ dry run.

[reed1, 2016-08-20 22:21:17]

Hmmm, are these checks motivated by tests or fuzzes?

[msarett, 2016-08-22 12:40:43]

On 2016/08/20 22:21:17, reed1 wrote:
> Hmmm, are these checks motivated by tests or fuzzes?

I don't have any real test cases from the web that exercise this code - though I
can try to find some.

A few motivations:
(1) Be more similar to other ICC parsing libraries.
(2) Avoid allocating a ton of memory when parsing some (likely invalid) ICC
profiles.
(3) Fuzzes.

[msarett, 2016-10-13 22:50:07]

PTAL

I've modified this to remove checks for "valid D50 matrices".
We can add these in if they end up being necessary for some
reason...

I think it's important to land some sanity checks for profile
size though?

[msarett, 2016-10-14 17:16:07]

On 2016/10/13 22:50:07, msarett wrote:
> PTAL
> 
> I've modified this to remove checks for "valid D50 matrices".
> We can add these in if they end up being necessary for some
> reason...
> 
> I think it's important to land some sanity checks for profile
> size though?

Closing this for now.  I think it's fine to be less defensive, and to land these
types of things if/when we come across a need to actually have them.

[msarett, 2016-10-14 17:16:16]

Description was changed from

==========
Add checks for bogus profiles to SkColorSpace::NewICC()

This shouldn't change anything on Gold.

Definitely think we should have these to deal with whatever crazy
profiles are out in the wild.

I borrowed a few ideas from QCMS and also made some of my own
judgments.

BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2257983003
==========

to

==========
Add checks for bogus profiles to SkColorSpace::NewICC()

This shouldn't change anything on Gold.

Definitely think we should have these to deal with whatever crazy
profiles are out in the wild.

I borrowed a few ideas from QCMS and also made some of my own
judgments.

BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2257983003
==========