Refactor for GPU command buffer PUTSTR code generator.

Refactor for GPU command buffer PUTSTR code generator.

This is a pure code moving around CL.  No logic change at all.

So we add helper functions to pack/unpack strings to/from a bucket, so
the code generator is now much simpler.

This is to get the pack/unpack functions to be used by glGetUniformIndices.

BUG=429053
TEST=gpu_unittests, webgl_conformance
R=sievers@chromium.org

Committed: https://crrev.com/0420f3be94d1ecc2db68dbe436ceec8711ddb106
Cr-Commit-Position: refs/heads/master@{#315118}

[Zhenyao Mo, 2015-02-06 00:07:09]

Patchset #1 (id:1) has been deleted

[Zhenyao Mo, 2015-02-06 00:08:56]

zmo@chromium.org changed reviewers:
+ sievers@chromium.org

[Zhenyao Mo, 2015-02-06 00:10:33]

Daniel: can you review this CL?
Others: FYI (feel free to review).

Please note, this CL is simply moving code from generator to helper function (of
course I should have done it in the first place).

[Zhenyao Mo, 2015-02-06 18:52:38]

Bots all green.  Please take a look.

[no sievers, 2015-02-06 21:01:38]

lgtm

https://codereview.chromium.org/902943002/diff/20001/gpu/command_buffer/build...
File gpu/command_buffer/build_gles2_cmd_buffer.py (right):

https://codereview.chromium.org/902943002/diff/20001/gpu/command_buffer/build...
gpu/command_buffer/build_gles2_cmd_buffer.py:7943: strs.size() > 0 ?
const_cast<const char**>(&strs[0]) : NULL;
Can we somehow assert that these are only NULL if count == 0?
Should it have a test?

[Zhenyao Mo, 2015-02-06 21:53:50]

https://codereview.chromium.org/902943002/diff/20001/gpu/command_buffer/build...
File gpu/command_buffer/build_gles2_cmd_buffer.py (right):

https://codereview.chromium.org/902943002/diff/20001/gpu/command_buffer/build...
gpu/command_buffer/build_gles2_cmd_buffer.py:7943: strs.size() > 0 ?
const_cast<const char**>(&strs[0]) : NULL;
On 2015/02/06 21:01:38, sievers wrote:
> Can we somehow assert that these are only NULL if count == 0?
> Should it have a test?

This is the same as asserting strs.size() == len.size() == count, which I could
do at the end of Bucket::GetAsStrings(), but looking at the function, the code
is apparently doing that.

[Zhenyao Mo, 2015-02-06 21:53:54]

The CQ bit was checked by zmo@chromium.org

[commit-bot: I haz the power, 2015-02-06 21:55:09]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/902943002/20001

[commit-bot: I haz the power, 2015-02-06 21:58:24]

Committed patchset #1 (id:20001)

[commit-bot: I haz the power, 2015-02-06 21:58:54]

Patchset 1 (id:??) landed as
https://crrev.com/0420f3be94d1ecc2db68dbe436ceec8711ddb106
Cr-Commit-Position: refs/heads/master@{#315118}