Reland r130462: Implement FeatureProvider for ExtensionAPI."

Reland r130462: Implement FeatureProvider for ExtensionAPI."

BUG=120069
TBR=mpcomplete@chromium.org

Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=130697

[Aaron Boodman, 2012-04-04 00:46:45]

Matt, the first CL was already reviewed (mostly, I had to make some changes to
make the bots happy). The second one is the changes I made to avoid loading all
the schemas in the browser process.

[Matt Perry, 2012-04-04 16:40:22]

Does this mean that ExtensionAPI will load the full 2-3megs of schema data
twice? It would be unfortunate to add that much memory wastage to the browser
process.

http://codereview.chromium.org/9969136/diff/2001/chrome/browser/extensions/ex...
File chrome/browser/extensions/extension_function_dispatcher.cc (right):

http://codereview.chromium.org/9969136/diff/2001/chrome/browser/extensions/ex...
chrome/browser/extensions/extension_function_dispatcher.cc:97: }
}  // namespace

http://codereview.chromium.org/9969136/diff/2001/chrome/common/extensions/api...
File chrome/common/extensions/api/extension_api.cc (right):

http://codereview.chromium.org/9969136/diff/2001/chrome/common/extensions/api...
chrome/common/extensions/api/extension_api.cc:122: return
g_lazy_instance.Get().api.get();
maybe we should DCHECK that this is the UI thread

[Aaron Boodman, 2012-04-04 18:05:41]

No. The schemas are loaded lazily, and IsPrivileged only loads the ones it
needs.

http://codereview.chromium.org/9969136/diff/2001/chrome/browser/extensions/ex...
File chrome/browser/extensions/extension_function_dispatcher.cc (right):

http://codereview.chromium.org/9969136/diff/2001/chrome/browser/extensions/ex...
chrome/browser/extensions/extension_function_dispatcher.cc:97: }
On 2012/04/04 16:40:22, Matt Perry wrote:
> }  // namespace

Done.

http://codereview.chromium.org/9969136/diff/2001/chrome/common/extensions/api...
File chrome/common/extensions/api/extension_api.cc (right):

http://codereview.chromium.org/9969136/diff/2001/chrome/common/extensions/api...
chrome/common/extensions/api/extension_api.cc:122: return
g_lazy_instance.Get().api.get();
On 2012/04/04 16:40:22, Matt Perry wrote:
> maybe we should DCHECK that this is the UI thread

Good idea. Done.

[commit-bot: I haz the power, 2012-04-04 18:33:56]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/aa@chromium.org/9969136/12001

[Matt Perry, 2012-04-04 18:38:47]

http://codereview.chromium.org/9969136/diff/12001/chrome/common/extensions/ap...
File chrome/common/extensions/api/extension_api.cc (right):

http://codereview.chromium.org/9969136/diff/12001/chrome/common/extensions/ap...
chrome/common/extensions/api/extension_api.cc:248: void
ExtensionAPI::InitDefaultConfiguration() {
This method is called when you create a new ExtensionAPI instance, and it causes
all the JSON files to be read into memory. It looks like that's about 700K. Even
adding that much is a waste when we could just special-case the webRequest API
to avoid instantiating an ExtensionAPI on the IO thread.

[Aaron Boodman, 2012-04-04 18:41:27]

http://codereview.chromium.org/9969136/diff/12001/chrome/common/extensions/ap...
File chrome/common/extensions/api/extension_api.cc (right):

http://codereview.chromium.org/9969136/diff/12001/chrome/common/extensions/ap...
chrome/common/extensions/api/extension_api.cc:248: void
ExtensionAPI::InitDefaultConfiguration() {
On 2012/04/04 18:38:47, Matt Perry wrote:
> This method is called when you create a new ExtensionAPI instance, and it
causes
> all the JSON files to be read into memory. It looks like that's about 700K.
Even
> adding that much is a waste when we could just special-case the webRequest API
> to avoid instantiating an ExtensionAPI on the IO thread.

It doesn't actually do any reading. ReadFromResource is named badly. All it's
returning is a StringPiece. The data is actually read into memory later, in
LoadSchema().

[Matt Perry, 2012-04-04 18:45:10]

lgtm