Added unit tests for ClearBrowsingDataController....

Added unit tests for ClearBrowsingDataController.
Fixed a bug in ClearBrowsingDataController: checking Delete Form Data would end up deleting passwords instead.

BUG=12902
TEST=true

[pink (ping after 24hrs), 2009-06-30 18:51:08]

This is AWESOME! A couple of nits to fix:

http://codereview.chromium.org/151095/diff/1/3
File chrome/browser/cocoa/clear_browsing_data_controller.mm (right):

http://codereview.chromium.org/151095/diff/1/3#newcode113
Line 113: remover_->Remove(self.removeMask);
[self removeMask]. dot-notaion isn't allowed by style guide.

http://codereview.chromium.org/151095/diff/1/2
File chrome/browser/cocoa/clear_browsing_data_controller_unittest.mm (right):

http://codereview.chromium.org/151095/diff/1/2#newcode12
Line 12: #include "chrome/browser/profile.h"
can you make sure these are in alphabetical order?

http://codereview.chromium.org/151095/diff/1/2#newcode54
Line 54: TEST_F(ClearBrowsingDataControllerTest, InitialRemoveMask) {
Is there a way to test that nothing bad happens if the removeMask is 0L (ie,
none of the prefs are set)?

http://codereview.chromium.org/151095/diff/1/2#newcode64
Line 64: [controller_ setClearBrowsingHistory: false];
no space after : for parameters.

[Jens, 2009-06-30 22:52:07]

http://codereview.chromium.org/151095/diff/1/3
File chrome/browser/cocoa/clear_browsing_data_controller.mm (right):

http://codereview.chromium.org/151095/diff/1/3#newcode113
Line 113: remover_->Remove(self.removeMask);
On 2009/06/30 18:51:08, pink wrote:
> [self removeMask]. dot-notaion isn't allowed by style guide.

ZOMGWTFBBQ!? It allows @property declarations but not property accessors? I
quit!
...uh, I mean, "Done".

http://codereview.chromium.org/151095/diff/1/2
File chrome/browser/cocoa/clear_browsing_data_controller_unittest.mm (right):

http://codereview.chromium.org/151095/diff/1/2#newcode12
Line 12: #include "chrome/browser/profile.h"
On 2009/06/30 18:51:08, pink wrote:
> can you make sure these are in alphabetical order?

OK, I think these are in order now. (I'm a bit confused, because either I
misunderstand the sort rules, or clear_browsing_data_controller.mm is wrong.)

http://codereview.chromium.org/151095/diff/1/2#newcode54
Line 54: TEST_F(ClearBrowsingDataControllerTest, InitialRemoveMask) {
On 2009/06/30 18:51:08, pink wrote:
> Is there a way to test that nothing bad happens if the removeMask is 0L (ie,
> none of the prefs are set)?

OK, I've added another test that sets all prefs to false and verifies the
removeMask is zero.

http://codereview.chromium.org/151095/diff/1/2#newcode64
Line 64: [controller_ setClearBrowsingHistory: false];
On 2009/06/30 18:51:08, pink wrote:
> no space after : for parameters.

Done.

[pink (ping after 24hrs), 2009-07-01 12:41:20]

LGTM. i'll try to get this landed today, but poke me if i haven't gotten around
to it.

http://codereview.chromium.org/151095/diff/1/3
File chrome/browser/cocoa/clear_browsing_data_controller.mm (right):

http://codereview.chromium.org/151095/diff/1/3#newcode113
Line 113: remover_->Remove(self.removeMask);
It obscures the fact that this is a function call, similar to why we don't allow
operator overloading in the C++ style guide. It also allows things like
self.autorelease, which is wholly nonsensical (even Apple admits that).

http://codereview.chromium.org/151095/diff/1/2
File chrome/browser/cocoa/clear_browsing_data_controller_unittest.mm (right):

http://codereview.chromium.org/151095/diff/1/2#newcode12
Line 12: #include "chrome/browser/profile.h"
You had chrome/common before chrome/browser (which you've now fixed).

http://codereview.chromium.org/151095/diff/1/2#newcode54
Line 54: TEST_F(ClearBrowsingDataControllerTest, InitialRemoveMask) {
I guess I was hoping for a test that called the backend with nothing set to make
sure *it* didn't crash, but that's more a unit test for the back-end, not our UI
controller.

http://codereview.chromium.org/151095/diff/1006/15#newcode87
Line 87: 
these can fit on 1 line.

[pink (ping after 24hrs), 2009-07-01 14:41:02]

landed.