Update obvious deps.

Update obvious deps.

This is a step toward targets declaring their deps in a sane fashion.
This change resolves cycles by forcing core to the root, then everything
in skia_lib pointing toward core as best possible, then everything
outside skia_lib depending on skia_lib for things in skia_lib. This
prevents double definitions where a symbol is provided by both the
skia_lib shared object and and a statically linked component of skia_lib.

R=djsollen@google.com

Committed: https://code.google.com/p/skia/source/detail?r=10231

[bungeman-skia, 2013-07-19 18:15:22]

https://codereview.chromium.org/19823003/diff/2001/gyp/pdf.gyp
File gyp/pdf.gyp (left):

https://codereview.chromium.org/19823003/diff/2001/gyp/pdf.gyp#oldcode13
gyp/pdf.gyp:13: '../include/config',
Note that these are already provided by the dependency on skia_lib.

https://codereview.chromium.org/19823003/diff/2001/gyp/ports.gyp
File gyp/ports.gyp (left):

https://codereview.chromium.org/19823003/diff/2001/gyp/ports.gyp#oldcode12
gyp/ports.gyp:12: '../include/effects',
The rest of these includes indicate potential circular dependencies.

https://codereview.chromium.org/19823003/diff/2001/gyp/utils.gyp
File gyp/utils.gyp (left):

https://codereview.chromium.org/19823003/diff/2001/gyp/utils.gyp#oldcode12
gyp/utils.gyp:12: '../include/effects',
The rest of these includes indicate potential circular dependencies.

[djsollen, 2013-07-19 18:57:15]

lgtm, do you have any ideas on how to handle the circular deps within utils?

[djsollen, 2013-07-19 18:57:42]

I should have said lgtm...pending trybot success

[bungeman-skia, 2013-07-19 22:31:27]

Committed patchset #3 manually as r10217.

[bungeman-skia, 2013-07-22 04:33:34]

Looks like this will now work at Patch Set 5. Patch Set 3 has issues with
targets outside skia_lib depending directly on targets inside skia_lib, which
lead to double defined symbols.

reed: This removes a really old ifdef for Android in
src/animator/SkDisplayEvent.cpp (looks like it's from r27). It looks like this
is no longer needed, causes problems for this change, and defines a bunch of
functions in a really strange place. Just wanted to make sure there wasn't any
real reason for this code anymore.

[djsollen, 2013-07-22 13:24:01]

https://codereview.chromium.org/19823003/diff/20001/src/animator/SkDisplayEve...
File src/animator/SkDisplayEvent.cpp (right):

https://codereview.chromium.org/19823003/diff/20001/src/animator/SkDisplayEve...
src/animator/SkDisplayEvent.cpp:254: #ifdef SK_BUILD_FOR_ANDROID
If nothing is using this code shouldn't we remove it altogether?

[bungeman-skia, 2013-07-22 13:44:32]

https://codereview.chromium.org/19823003/diff/20001/src/animator/SkDisplayEve...
File src/animator/SkDisplayEvent.cpp (right):

https://codereview.chromium.org/19823003/diff/20001/src/animator/SkDisplayEve...
src/animator/SkDisplayEvent.cpp:254: #ifdef SK_BUILD_FOR_ANDROID
On 2013/07/22 13:24:01, djsollen wrote:
> If nothing is using this code shouldn't we remove it altogether?

Done.

[bungeman-skia, 2013-07-22 14:39:49]

Committed patchset #7 manually as r10231.