Initial gyp files for SDL

Initial gyp files for SDL

BUG=skia:

Committed: https://skia.googlesource.com/skia/+/cc92b4a7eca78dcec23d14ccd4ecf5889c35caa1

[joshualitt, 2015-10-28 20:59:23]

joshualitt@google.com changed reviewers:
+ bsalomon@google.com, joshualitt@google.com

[joshualitt, 2015-10-28 20:59:23]

ptal

[joshualitt, 2015-10-28 21:00:11]

Description was changed from

==========
Initial gyp files for SDL

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

to

==========
Initial gyp files for SDL

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

[joshualitt, 2015-10-28 21:00:11]

joshualitt@google.com changed reviewers:
+ mtklein@google.com

[joshualitt, 2015-10-28 21:01:00]

adding Mike.

Re your request to move sdl.gypi into sdl.gyp, unfortunately, GYP has a
limitation where it cannot expand variables inside of 'includes' []
dictionaries, as a result I need to have a separate gyp file so I can expand
base_dir and src_dir

[joshualitt, 2015-10-28 21:01:01]

adding Mike.

Re your request to move sdl.gypi into sdl.gyp, unfortunately, GYP has a
limitation where it cannot expand variables inside of 'includes' []
dictionaries, as a result I need to have a separate gyp file so I can expand
base_dir and src_dir

[mtklein, 2015-10-28 21:12:15]

https://codereview.chromium.org/1411033003/diff/1/gyp/sdl.gyp
File gyp/sdl.gyp (right):

https://codereview.chromium.org/1411033003/diff/1/gyp/sdl.gyp#newcode8
gyp/sdl.gyp:8: 'base_dir%': '../third_party/libsdl',
I don't really get why these are variables?  Are you planning on moving these
around?

https://codereview.chromium.org/1411033003/diff/1/third_party/libsdl/linux/RE...
File third_party/libsdl/linux/README (right):

https://codereview.chromium.org/1411033003/diff/1/third_party/libsdl/linux/RE...
third_party/libsdl/linux/README:1: To generate this file run:
Which file does this generate?  Surely not the README.

https://codereview.chromium.org/1411033003/diff/1/third_party/libsdl/sdl.gypi
File third_party/libsdl/sdl.gypi (right):

https://codereview.chromium.org/1411033003/diff/1/third_party/libsdl/sdl.gypi...
third_party/libsdl/sdl.gypi:20: '<@(sdl_sources)',
Is this the same deal?  Why can't we just write out the sources here?

https://codereview.chromium.org/1411033003/diff/1/third_party/libsdl/sdl.gypi...
third_party/libsdl/sdl.gypi:29: '-Wall'
Do we really care about warnings in SDL?  I'd suggest -w.

https://codereview.chromium.org/1411033003/diff/1/third_party/libsdl/sdl.gypi...
third_party/libsdl/sdl.gypi:34: 'linux/sdl_linux.gypi',
Is this just a factoring preference, or is something making us use a .gypi here?

[joshualitt, 2015-10-29 13:51:40]

https://codereview.chromium.org/1411033003/diff/1/gyp/sdl.gyp
File gyp/sdl.gyp (right):

https://codereview.chromium.org/1411033003/diff/1/gyp/sdl.gyp#newcode8
gyp/sdl.gyp:8: 'base_dir%': '../third_party/libsdl',
On 2015/10/28 21:12:15, mtklein wrote:
> I don't really get why these are variables?  Are you planning on moving these
> around?

Its so you can test with other versions of SDL, such as an SDL outside of Skia.

https://codereview.chromium.org/1411033003/diff/1/third_party/libsdl/linux/RE...
File third_party/libsdl/linux/README (right):

https://codereview.chromium.org/1411033003/diff/1/third_party/libsdl/linux/RE...
third_party/libsdl/linux/README:1: To generate this file run:
On 2015/10/28 21:12:15, mtklein wrote:
> Which file does this generate?  Surely not the README.

Acknowledged.

https://codereview.chromium.org/1411033003/diff/1/third_party/libsdl/sdl.gypi
File third_party/libsdl/sdl.gypi (right):

https://codereview.chromium.org/1411033003/diff/1/third_party/libsdl/sdl.gypi...
third_party/libsdl/sdl.gypi:20: '<@(sdl_sources)',
On 2015/10/28 21:12:15, mtklein wrote:
> Is this the same deal?  Why can't we just write out the sources here?

see below.

https://codereview.chromium.org/1411033003/diff/1/third_party/libsdl/sdl.gypi...
third_party/libsdl/sdl.gypi:29: '-Wall'
On 2015/10/28 21:12:15, mtklein wrote:
> Do we really care about warnings in SDL?  I'd suggest -w.

Acknowledged.

https://codereview.chromium.org/1411033003/diff/1/third_party/libsdl/sdl.gypi...
third_party/libsdl/sdl.gypi:34: 'linux/sdl_linux.gypi',
On 2015/10/28 21:12:15, mtklein wrote:
> Is this just a factoring preference, or is something making us use a .gypi
here?

I prefer smaller gyps which show the core logic, and then dummy gypis with lists
of files.  It just makes it easier for me to work with flatter gyps, as opposed
to heavily nested gyps with long lists of files.

We also do this with the gpu files, so its not unheard of in our codebase.

I'd prefer to leave the separate gpyis for now.  Once I have rolled out on
mac/windows/ios/android, we can revisit.

[mtklein, 2015-10-29 13:57:38]

lgtm

[joshualitt, 2015-10-29 13:58:19]

The CQ bit was checked by joshualitt@google.com

[commit-bot: I haz the power, 2015-10-29 13:58:24]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1411033003/20001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1411033003/20001

[commit-bot: I haz the power, 2015-10-29 14:09:57]

Committed patchset #2 (id:20001) as
https://skia.googlesource.com/skia/+/cc92b4a7eca78dcec23d14ccd4ecf5889c35caa1