<webview> Implement clear http cache API.

<webview> Implement clear http cache API.

HTTP cache clearing methods have been refactored into a
separate class: StoragePartitionBrowsingDataRemover in
http://crrev.com/1033013003/

One method was exposed in WebCacheManager so we can
clear renderer/ cache for a given render process.

<webview> already has a set of data it can clear, this
CL adds "cache" to that set, similar to chrome.browsingData
API.

BUG=406437
Test=Load a <webview> in a chrome app, e.g.
the browser sample app:
https://github.com/GoogleChrome/chrome-app-samples/tree/master/samples/webview-samples/browser

Open inspector for the webview: from chrome://inspect,
switch to Apps, and inspect the tab that says "Google"
under "Browser sample". Switch to network tab in that inspector so you can monitor http requests.

Visit some page in the <webview>, for my test case,
I visited http://www.google.ca
Now look at the network tab for image requests (cause it
should be much simpler), e.g. nav_logo_195.png
Reload the page couple of times, observer the status code
for the image request, it should say 304 OK.

Now open the app's inspector: Same as above, but inspect "Browser sample"
instead. Call <webview> clear cache api from its console:

document.querySelector('webview').clearData(
  {since: 1},
  {cache: true},
  function() { window.console.log('clear complete'); })

Expect "clear complete" message to show in app's console.
Now reload the page one more time and check the same
image's request, it should say 200 OK instead of 304.

Committed: https://crrev.com/cbff2a79e3aafa333159dbc467efcb8a5dc124f7
Cr-Commit-Position: refs/heads/master@{#322678}

[lazyboy, 2015-03-25 19:16:27]

lazyboy@chromium.org changed reviewers:
+ mkwst@chromium.org

[lazyboy, 2015-03-25 19:16:28]

+mkwst@ for initial review:

chrome/browser/browsing_data/browsing_data_remover.h
chrome/browser/browsing_data/browsing_data_remover.cc
chrome/browser/browsing_data/storage_partition_http_cache_data_remover.h
chrome/browser/browsing_data/storage_partition_http_cache_data_remover.cc
chrome/chrome_browser.gypi

This is bigger piece of the CL so I'll proceed further with
sending it to other OWNERS once you are happy, thanks!

btw, I'm using mock for now in my test (MockWebViewGuestDelegate) as I
can't get caching using EmbeddedTestServer() to write end-to-end test.

[Mike West, 2015-03-26 14:21:31]

Would you mind splitting the BrowsingDataRemover code out into
StoragePartitionHttpCacheDataRemover in a separate CL? That would make this
simpler to evaluate. Basically, that refactoring looks fine (as long as the
existing tests still pass), but it's not really clear to me what else
(anything?) is changing in those ~200 lines.

[lazyboy, 2015-03-26 16:32:38]

On 2015/03/26 14:21:31, Mike West wrote:
> Would you mind splitting the BrowsingDataRemover code out into
> StoragePartitionHttpCacheDataRemover in a separate CL? That would make this
> simpler to evaluate. Basically, that refactoring looks fine (as long as the
> existing tests still pass), but it's not really clear to me what else
> (anything?) is changing in those ~200 lines.

Agreed, I've exported it here: https://codereview.chromium.org/1033013003

1) There is not much changing in those ~200 lines: 1) major change is to use
StoragePartition's
  URLRequestContext instead of Profile's URLRequestContext, main_request_context
and
  media_request_context 2) A base::Callback is used to return from the call.
2) I've run unit_tests:BrowsingDataRemoverTest.* and
browser_tests:BrowsingDataRemoverBrowserTest.*
  all of them seems to pass, will send to trybot to see all tests.

[Mike West, 2015-03-26 16:38:41]

Great. Ping this CL once that one lands (I read it here this afternoon, assuming
you didn't change anything, that one looks fine) (though I don't think you'll
need my review anymore at that point?)

[lazyboy, 2015-03-26 20:05:07]

lazyboy@chromium.org changed reviewers:
+ jochen@chromium.org

[lazyboy, 2015-03-26 20:05:08]

@Mike:
https://codereview.chromium.org/1033013003 has landed and I've updated
this CL (patchset #6). This CL doesn't require browsing_data/* changes anymore
:)
However, if you have any comments on rest of the CL, feel free to add them,
thanks.

+jochen@ for reviewing:
components/web_cache/browser/web_cache_manager.h
components/web_cache/browser/web_cache_manager.cc

I can pull this out to separate CL if you prefer.

[jochen (gone - plz use gerrit), 2015-03-27 14:41:52]

lgtm

[Fady Samuel, 2015-03-27 14:47:37]

fsamuel@chromium.org changed reviewers:
+ fsamuel@chromium.org

[Fady Samuel, 2015-03-27 14:47:38]

https://codereview.chromium.org/1021073002/diff/100001/chrome/browser/guest_v...
File chrome/browser/guest_view/web_view/chrome_web_view_guest_delegate.cc
(right):

https://codereview.chromium.org/1021073002/diff/100001/chrome/browser/guest_v...
chrome/browser/guest_view/web_view/chrome_web_view_guest_delegate.cc:49: //
StoragePartitionHttpCacheDataRemover::ClearData() does not clear that.
Can StoragePartitionHttpCacheDataRemover move to components?

[lazyboy, 2015-03-27 17:34:05]

https://codereview.chromium.org/1021073002/diff/100001/chrome/browser/guest_v...
File chrome/browser/guest_view/web_view/chrome_web_view_guest_delegate.cc
(right):

https://codereview.chromium.org/1021073002/diff/100001/chrome/browser/guest_v...
chrome/browser/guest_view/web_view/chrome_web_view_guest_delegate.cc:49: //
StoragePartitionHttpCacheDataRemover::ClearData() does not clear that.
On 2015/03/27 14:47:38, Fady Samuel wrote:
> Can StoragePartitionHttpCacheDataRemover move to components?

Yes, I've filed http://crbug.com/471287 to do that and added TODO here.

[Fady Samuel, 2015-03-27 20:21:49]

guest_view lgtm

[lazyboy, 2015-03-27 22:46:09]

FYI, patchset #9 has one change since reviewed:
"set<int>({value})" initializer was failing on mac/android. I've changed this to
sth like: "set<int> x; x.insert(value);"

[lazyboy, 2015-03-27 23:34:49]

The CQ bit was checked by lazyboy@chromium.org

[lazyboy, 2015-03-27 23:34:49]

The patchset sent to the CQ was uploaded after l-g-t-m from jochen@chromium.org,
fsamuel@chromium.org
Link to the patchset:
https://chromiumcodereview.appspot.com/1021073002/#ps160001 (title: "fix compile
error on mac/android: std::set initializer list makes them unhappy")

[commit-bot: I haz the power, 2015-03-27 23:35:18]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1021073002/160001

[commit-bot: I haz the power, 2015-03-27 23:39:03]

Committed patchset #9 (id:160001)

[commit-bot: I haz the power, 2015-03-27 23:40:06]

Patchset 9 (id:??) landed as
https://crrev.com/cbff2a79e3aafa333159dbc467efcb8a5dc124f7
Cr-Commit-Position: refs/heads/master@{#322678}