Ensure that window.open requests issued by ChromeFrame carry the correct cook...

Ensure that window.open requests issued by ChromeFrame carry the correct cookies in the outgoing HTTP requests.

To achieve this we no longer issue navigations with the gcf:attach* prefix. We now issue a navigation to the current
page URL with the attach external tab suffix, which indicates that the page is forced into ChromeFrame without making
an actual HTTP request. This ensures that the new IE8 process has access to all HTTP cookies.

We need to patch IInternetProtocol::LockRequest and UnlockRequest to not call the underlying implementations for our
dummy request as this crashes in IE8 in the prebinding code path.

Fixes bug http://b/issue?id=2277519

Added tests for the CanNavigateFullTabMode helper function.

Bug=2277519
Committed: http://src.chromium.org/viewvc/chrome?view=rev&revision=54019

[stoyan, 2010-07-27 19:41:10]

LGTM

http://codereview.chromium.org/3051018/diff/1/5
File chrome_frame/protocol_sink_wrap.cc (right):

http://codereview.chromium.org/3051018/diff/1/5#newcode530
chrome_frame/protocol_sink_wrap.cc:530: // This hack is required because
window.open requests don't have the URL up
window.open requests -> window.open requests from Chrome

http://codereview.chromium.org/3051018/diff/1/5#newcode533
chrome_frame/protocol_sink_wrap.cc:533: // The new page expects to be rendered
in ChromeFrame to preserve the
The new page expects -> The new page must

[amit, 2010-07-27 19:49:39]

Woot! drive by lgtm

http://codereview.chromium.org/3051018/diff/24001/21002
File chrome_frame/chrome_active_document.cc (right):

http://codereview.chromium.org/3051018/diff/24001/21002#newcode997
chrome_frame/chrome_active_document.cc:997: bool allow_gcf_protocol =
!cf_url.attach_to_external_tab() ||
This is nice. Now we need a test to make sure gcf protocol is only supported in
a limited way.

http://codereview.chromium.org/3051018/diff/24001/21008
File chrome_frame/utils.h (right):

http://codereview.chromium.org/3051018/diff/24001/21008#newcode488
chrome_frame/utils.h:488: class ChromeFrameUrl {
nice!

[ananta, 2010-07-27 22:08:48]

http://codereview.chromium.org/3051018/diff/24001/21002
File chrome_frame/chrome_active_document.cc (right):

http://codereview.chromium.org/3051018/diff/24001/21002#newcode997
chrome_frame/chrome_active_document.cc:997: bool allow_gcf_protocol =
!cf_url.attach_to_external_tab() ||
On 2010/07/27 19:49:40, amit wrote:
> This is nice. Now we need a test to make sure gcf protocol is only supported
in
> a limited way.
Added unit tests for the CanNavigateInFullTabMode helper function.

[amit, 2010-07-27 23:15:27]

On Tue, Jul 27, 2010 at 3:08 PM,  <ananta@chromium.org> wrote:
>
> http://codereview.chromium.org/3051018/diff/24001/21002
> File chrome_frame/chrome_active_document.cc (right):
>
> http://codereview.chromium.org/3051018/diff/24001/21002#newcode997
> chrome_frame/chrome_active_document.cc:997: bool allow_gcf_protocol =
> !cf_url.attach_to_external_tab() ||
> On 2010/07/27 19:49:40, amit wrote:
>>
>> This is nice. Now we need a test to make sure gcf protocol is only
>
> supported in
>>
>> a limited way.
>
> Added unit tests for the CanNavigateInFullTabMode helper function.
Awesome! You could log a bug do it in the next CL but what I am
concerned is accidently allowing gcf: to work for more types of URLs
than it is safe. So we need a test covering that.
>
> http://codereview.chromium.org/3051018/show
>