Make json_schema_compiler resolve type names via resolution rules.

Make json_schema_compiler resolve type names via resolution rules.

Instead of sharing a single namespace internally for types, makes
json_schema_compiler treat types as their fully-qualified selfs. This allows
two different namespaces to have types with the same name, and to be able to
reference the others' types by a fully-qualified name. This change also causes
namespaces to not be added twice to their CppTypeGenerators.

BUG=119954
TEST=none


Committed: http://src.chromium.org/viewvc/chrome?view=rev&revision=129251

[miket_OOO, 2012-03-26 17:55:48]

lgtm

http://codereview.chromium.org/9834088/diff/1/tools/json_schema_compiler/cpp_...
File tools/json_schema_compiler/cpp_type_generator.py (right):

http://codereview.chromium.org/9834088/diff/1/tools/json_schema_compiler/cpp_...
tools/json_schema_compiler/cpp_type_generator.py:207: for (type_, namespace) in
self._type_namespaces.items():
This is probably OK, but will you take a quick look at the Python style
guidelines and confirm that foo_ is an acceptable naming convention for a
temporary variable like this? It confused me for a moment (particularly with the
self._type[...] in the same neighborhood). I don't object as long as the style
guide is OK with it.

[asargent_no_longer_on_chrome, 2012-03-26 18:30:06]

LGTM

https://chromiumcodereview.appspot.com/9834088/diff/1/tools/json_schema_compi...
File tools/json_schema_compiler/cpp_type_generator.py (right):

https://chromiumcodereview.appspot.com/9834088/diff/1/tools/json_schema_compi...
tools/json_schema_compiler/cpp_type_generator.py:207: for (type_, namespace) in
self._type_namespaces.items():
On 2012/03/26 17:55:48, miket wrote:
> This is probably OK, but will you take a quick look at the Python style
> guidelines and confirm that foo_ is an acceptable naming convention for a
> temporary variable like this? It confused me for a moment (particularly with
the
> self._type[...] in the same neighborhood). I don't object as long as the style
> guide is OK with it.

A bunch of other code in this file (and potentially others in this dir), so if
it turns out the style guide forbids it we should fix it everywhere. One
possibility is to just use "typename" instead of "type_".

[Garret Kelly, 2012-03-26 18:43:19]

Yep, it's only files in this directory that seem to have used this naming
convention. I've fixed it and will commit. I'll put together another change with
the naming fix in it as well for the rest of this package.

Thanks!

[commit-bot: I haz the power, 2012-03-26 18:43:37]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/gdk@chromium.org/9834088/5002

[Garret Kelly, 2012-03-26 18:46:45]

Oops! Ran my git grep in the wrong place. It's _very_ rarely (4 instances) used
outside of this directory. I still believe it needs fixing.

[commit-bot: I haz the power, 2012-03-26 21:57:03]

Try job failure for 9834088-5002 (retry) on win_rel for step "browser_tests".
It's a second try, previously, step "browser_tests" failed.
http://build.chromium.org/p/tryserver.chromium/buildstatus?builder=win_rel&nu...

[commit-bot: I haz the power, 2012-03-26 22:23:36]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/gdk@chromium.org/9834088/5002

[commit-bot: I haz the power, 2012-03-27 01:47:10]

Try job failure for 9834088-5002 (retry) on win_rel for step "browser_tests".
It's a second try, previously, steps "ui_tests, browser_tests" failed.
http://build.chromium.org/p/tryserver.chromium/buildstatus?builder=win_rel&nu...

[commit-bot: I haz the power, 2012-03-27 17:10:42]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/gdk@chromium.org/9834088/5002

[commit-bot: I haz the power, 2012-03-27 20:03:25]

Change committed as 129251