Docserver: Add @Cache annotation to CompiledFileSystem

Docserver: Add @Cache annotation to CompiledFileSystem
This is the first in a series of CLs intended to reduce the amount of caching
in docserver code. The @Cache annotation is used to indicate that compiled data
should be cached in CompiledFileSystem.

NOTRY=True

Committed: https://crrev.com/ddf432dd4c15f8591a2ff9e8e952c55ce97e7a20
Cr-Commit-Position: refs/heads/master@{#292994}

[ahernandez, 2014-08-29 21:16:39]

ahernandez.miralles@gmail.com changed reviewers:
+ yoz@chromium.org

[ahernandez, 2014-08-29 21:16:40]

Can you please review these changes? I did a small amount of code cleanup as
well since this patch doesn't involve many changes.

https://codereview.chromium.org/521693003/diff/20001/chrome/common/extensions...
File chrome/common/extensions/docs/server2/api_models.py (right):

https://codereview.chromium.org/521693003/diff/20001/chrome/common/extensions...
chrome/common/extensions/docs/server2/api_models.py:66: @Cache
I had to add the @Cache annotation in a couple places where it is assumed that
compiled file system will cache data.

https://codereview.chromium.org/521693003/diff/20001/chrome/common/extensions...
File chrome/common/extensions/docs/server2/compiled_file_system.py (right):

https://codereview.chromium.org/521693003/diff/20001/chrome/common/extensions...
chrome/common/extensions/docs/server2/compiled_file_system.py:21: '''Functions
bound to an object are separate from the unbound
I don't know how to explain this concisely, but the SingleFile annotation wasn't
actually doing it's job in a lot of cases because when a method is annotated,
the unbound version of it is passed to the annotation function, e.g. SingleFile.
But then the bound version is passed to compiled fs, e.g.
c_fs_factory.Create(self._comp_func, ...), so when checking if the compilation
function is in the set of single file functions, the test fails because the
bound method != the unbound method. So I wrote this to ensure that the unbound
version is always grabbed.

[Yoyo Zhou, 2014-08-29 23:29:05]

LGTM

https://chromiumcodereview.appspot.com/521693003/diff/20001/chrome/common/ext...
File chrome/common/extensions/docs/server2/compiled_file_system.py (right):

https://chromiumcodereview.appspot.com/521693003/diff/20001/chrome/common/ext...
chrome/common/extensions/docs/server2/compiled_file_system.py:21: '''Functions
bound to an object are separate from the unbound
On 2014/08/29 21:16:39, ahernandez wrote:
> I don't know how to explain this concisely, but the SingleFile annotation
wasn't
> actually doing it's job in a lot of cases because when a method is annotated,
> the unbound version of it is passed to the annotation function, e.g.
SingleFile.
> But then the bound version is passed to compiled fs, e.g.
> c_fs_factory.Create(self._comp_func, ...), so when checking if the compilation
> function is in the set of single file functions, the test fails because the
> bound method != the unbound method. So I wrote this to ensure that the unbound
> version is always grabbed.

Ok. Clarify 'membership' means in the cache.

[ahernandez, 2014-09-02 18:57:23]

Patchset #2 (id:40001) has been deleted

[ahernandez, 2014-09-02 18:57:41]

The CQ bit was checked by ahernandez.miralles@gmail.com

[commit-bot: I haz the power, 2014-09-02 18:59:49]

CQ is trying da patch. Follow status at

https://chromium-status.appspot.com/cq/ahernandez.miralles@gmail.com/52169300...

[commit-bot: I haz the power, 2014-09-02 20:49:26]

Committed patchset #2 (id:60001) as 312faf7d7a2f96e1bce81a0fea1fbed7e0462b0d

[commit-bot: I haz the power, 2014-09-10 03:20:40]

Patchset 2 (id:??) landed as
https://crrev.com/ddf432dd4c15f8591a2ff9e8e952c55ce97e7a20
Cr-Commit-Position: refs/heads/master@{#292994}