gdata: Move GDataWapiFeedLoader to a set of new files

gdata: Move GDataWapiFeedLoader to a set of new files

This change reduces ~800 lines from gdata_file_system.cc
Along the way move some functions to gdata_util.h, which
are needed for gdata_file_system.cc and gdata_wapi_feed_loader.cc

BUG=138268
TEST=none; just move code around. should not change the behaviors
TBR=ben@chromium.org 
for chrome_browser.gypi

Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=150027

[satorux1, 2012-08-04 07:42:28]

This is just moving code around. Would love to submit before m22 branch cut.

[achuithb, 2012-08-04 08:02:35]

http://codereview.chromium.org/10834170/diff/2001/chrome/browser/chromeos/gda...
File chrome/browser/chromeos/gdata/gdata_wapi_feed_loader.h (right):

http://codereview.chromium.org/10834170/diff/2001/chrome/browser/chromeos/gda...
chrome/browser/chromeos/gdata/gdata_wapi_feed_loader.h:22: struct
GetDocumentsParams {
Any reason why this is in the header file? I think you should just fwd declare
this and have the definition in the cc.

http://codereview.chromium.org/10834170/diff/2001/chrome/browser/chromeos/gda...
chrome/browser/chromeos/gdata/gdata_wapi_feed_loader.h:52: typedef
base::Callback<void(GetDocumentsParams* params,
Don't have to do it in this patch, but I feel like we should move this into
GDataWapiFeedLoader instead of having it global scope.

[satorux1, 2012-08-04 08:05:28]

http://codereview.chromium.org/10834170/diff/2001/chrome/browser/chromeos/gda...
File chrome/browser/chromeos/gdata/gdata_wapi_feed_loader.h (right):

http://codereview.chromium.org/10834170/diff/2001/chrome/browser/chromeos/gda...
chrome/browser/chromeos/gdata/gdata_wapi_feed_loader.h:22: struct
GetDocumentsParams {
On 2012/08/04 08:02:35, achuith.bhandarkar wrote:
> Any reason why this is in the header file? I think you should just fwd declare
> this and have the definition in the cc.

This has to be here, as this struct is used in gdata_file_system.cc and
gdata_wapi_feed_loader.cc

http://codereview.chromium.org/10834170/diff/2001/chrome/browser/chromeos/gda...
chrome/browser/chromeos/gdata/gdata_wapi_feed_loader.h:52: typedef
base::Callback<void(GetDocumentsParams* params,
On 2012/08/04 08:02:35, achuith.bhandarkar wrote:
> Don't have to do it in this patch, but I feel like we should move this into
> GDataWapiFeedLoader instead of having it global scope.

Not sure if it's better. Nested types result in verbose code in clients (ex.
Foo::Type instead of Type). As we are in 'gdata' namespace, we don't have to be
worried about having non-nested types

[achuithb, 2012-08-04 08:12:24]

lgtm

http://codereview.chromium.org/10834170/diff/2001/chrome/browser/chromeos/gda...
File chrome/browser/chromeos/gdata/gdata_util.h (right):

http://codereview.chromium.org/10834170/diff/2001/chrome/browser/chromeos/gda...
chrome/browser/chromeos/gdata/gdata_util.h:12: #include "base/tracked_objects.h"
I think you can probably fwd declare tracked_objects::Location too similar to
base::SequencedTaskRunner.

[satorux1, 2012-08-04 08:16:33]

http://codereview.chromium.org/10834170/diff/2001/chrome/browser/chromeos/gda...
File chrome/browser/chromeos/gdata/gdata_util.h (right):

http://codereview.chromium.org/10834170/diff/2001/chrome/browser/chromeos/gda...
chrome/browser/chromeos/gdata/gdata_util.h:12: #include "base/tracked_objects.h"
On 2012/08/04 08:12:25, achuith.bhandarkar wrote:
> I think you can probably fwd declare tracked_objects::Location too similar to
> base::SequencedTaskRunner.

Done.