Fix for the RedirectTest.ClientServerServer test.

Fix for the RedirectTest.ClientServerServer test with browser side navigation.

japhet: Please review blink changes.
jam/nasko: Please review everything

This test does the following:
Navigates to a meta tag refreshed page which is refreshed to server 1 via a HTTP
redirect followed by another redirect to server2 and then the final URL.

This works correctly in the regular case as individual redirect messages
are passed back to the renderer (blink) where the list is correctly maintained
and is passed back in the FrameHostMsg_DidCommitProvisionalLoad message with the
url of the referrer which initiated the client side redirect at the head of the list.

With browser side navigation we do pass the server redirect information back to the renderer
in the FrameMsg_CommitNavigation IPC message. When the navigation is processed the url
of the referrer which initated the client side redirect is added after the server side
redirect urls as blink just appends the document url (original referrer) to the redirect
list followed by the destination url. Basically the assumption is that blink is always
informed about server side redirects which is not the case with browser side navigation.

Changes in this patch are as below:
1. Add the document URL for client side redirects to the redirect list in the DocumentLoader ctor. The other changes
are to ensure that the client redirection flag makes it to the DocumentLoader ctor

   This ensures that the redirect list is in the correct order.

BUG=660567
Committed: https://crrev.com/d9062cc1ce0d2fda908c0f8d4e959c12d60d3d90
Cr-Commit-Position: refs/heads/master@{#429207}

[ananta, 2016-10-28 22:29:27]

Description was changed from

==========
PlzNavigate: Fix for the RedirectTest.ClientServerServer test.

This test does the following:
Navigates to a meta tag refreshed page which is refreshed to server 1 via a HTTP
redirect followed by another redirect to server2 and then the final URL.

This works correctly with the non plznavigate case as individual redirect
messages
are passed back to the renderer (blink) where the list is correctly maintained
and is passed back in the FrameHostMsg_DidCommitProvisionalLoad message.

With PlzNavigate we do pass the server redirect information back to the renderer
in the FrameMsg_CommitNavigation IPC message. However the first redirection
which
happens via a client redirect via a meta tag on the page or via JS is not in
this list.

Changes in this patch are as below:
1. Add the referrer URL to the redirect list in NavigationRequest when we
receive a
   navigation request with the PAGE_TRANSITION_CLIENT_REDIRECT transition type.
We also
   need to add the ResourceResponse structure in the redirect_response list
which is also
   part of the RequestNavigationParams structure. These lists have to contain
the same
   number of elements failing which there is a DCHECK in the renderer followed
by a crash.
   To send the ResourceResponse back we need to remember the last response which
was received
   I added a static scoped_refptr to the NavigationRequest class which tracks
this and is
   released in the subsequent navigation request.

2. In Blink when the navigation is processed after we receive the
CommitNavigation request
   there is code which adds the document URL (referrer) to the redirect list
followed by
   the final URL. This breaks the PlzNavigate case because the list now looks
like
   client redirect URL->Server 1->Server 2->Client redirect URL->final URL. I
changed this
   code to add the main document URL (referrer) to the front of the list if
required.
   This does not happen in the PlzNavigate case because this code executes
before the
   other redirects are processed.

BUG=none
==========

to

==========
PlzNavigate: Fix for the RedirectTest.ClientServerServer test.

This test does the following:
Navigates to a meta tag refreshed page which is refreshed to server 1 via a HTTP
redirect followed by another redirect to server2 and then the final URL.

This works correctly with the non plznavigate case as individual redirect
messages
are passed back to the renderer (blink) where the list is correctly maintained
and is passed back in the FrameHostMsg_DidCommitProvisionalLoad message.

With PlzNavigate we do pass the server redirect information back to the renderer
in the FrameMsg_CommitNavigation IPC message. However the first redirection
which
happens via a client redirect via a meta tag on the page or via JS is not in
this list.

Changes in this patch are as below:
1. Add the referrer URL to the redirect list in NavigationRequest when we
receive a
   navigation request with the PAGE_TRANSITION_CLIENT_REDIRECT transition type.
We also
   need to add the ResourceResponse structure in the redirect_response list
which is also
   part of the RequestNavigationParams structure. These lists have to contain
the same
   number of elements failing which there is a DCHECK in the renderer followed
by a crash.
   To send the ResourceResponse back we need to remember the last response which
was received
   I added a static scoped_refptr to the NavigationRequest class which tracks
this and is
   released in the subsequent navigation request.

2. In Blink when the navigation is processed after we receive the
CommitNavigation request
   there is code which adds the document URL (referrer) to the redirect list
followed by
   the final URL. This breaks the PlzNavigate case because the list now looks
like
   client redirect URL->Server 1->Server 2->Client redirect URL->final URL. I
changed this
   code to add the main document URL (referrer) to the front of the list if
required.
   This does not happen in the PlzNavigate case because this code executes
before the
   other redirects are processed.

BUG=none
CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.linux:linux_site_isolation
==========

[ananta, 2016-10-28 22:32:18]

The CQ bit was checked by ananta@chromium.org to run a CQ dry run

[commit-bot: I haz the power, 2016-10-28 22:32:42]

Dry run: CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[ananta, 2016-10-28 22:35:18]

Description was changed from

==========
PlzNavigate: Fix for the RedirectTest.ClientServerServer test.

This test does the following:
Navigates to a meta tag refreshed page which is refreshed to server 1 via a HTTP
redirect followed by another redirect to server2 and then the final URL.

This works correctly with the non plznavigate case as individual redirect
messages
are passed back to the renderer (blink) where the list is correctly maintained
and is passed back in the FrameHostMsg_DidCommitProvisionalLoad message.

With PlzNavigate we do pass the server redirect information back to the renderer
in the FrameMsg_CommitNavigation IPC message. However the first redirection
which
happens via a client redirect via a meta tag on the page or via JS is not in
this list.

Changes in this patch are as below:
1. Add the referrer URL to the redirect list in NavigationRequest when we
receive a
   navigation request with the PAGE_TRANSITION_CLIENT_REDIRECT transition type.
We also
   need to add the ResourceResponse structure in the redirect_response list
which is also
   part of the RequestNavigationParams structure. These lists have to contain
the same
   number of elements failing which there is a DCHECK in the renderer followed
by a crash.
   To send the ResourceResponse back we need to remember the last response which
was received
   I added a static scoped_refptr to the NavigationRequest class which tracks
this and is
   released in the subsequent navigation request.

2. In Blink when the navigation is processed after we receive the
CommitNavigation request
   there is code which adds the document URL (referrer) to the redirect list
followed by
   the final URL. This breaks the PlzNavigate case because the list now looks
like
   client redirect URL->Server 1->Server 2->Client redirect URL->final URL. I
changed this
   code to add the main document URL (referrer) to the front of the list if
required.
   This does not happen in the PlzNavigate case because this code executes
before the
   other redirects are processed.

BUG=none
CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.linux:linux_site_isolation
==========

to

==========
PlzNavigate: Fix for the RedirectTest.ClientServerServer test.

This test does the following:
Navigates to a meta tag refreshed page which is refreshed to server 1 via a HTTP
redirect followed by another redirect to server2 and then the final URL.

This works correctly with the non plznavigate case as individual redirect
messages
are passed back to the renderer (blink) where the list is correctly maintained
and is passed back in the FrameHostMsg_DidCommitProvisionalLoad message.

With PlzNavigate we do pass the server redirect information back to the renderer
in the FrameMsg_CommitNavigation IPC message. However the first redirection
which
happens via a client redirect via a meta tag on the page or via JS is not in
this list.

Changes in this patch are as below:
1. Add the referrer URL to the redirect list in NavigationRequest when we
receive a
   navigation request with the PAGE_TRANSITION_CLIENT_REDIRECT transition type.
We also
   need to add the ResourceResponse structure in the redirect_response list
which is also
   part of the RequestNavigationParams structure. These lists have to contain
the same
   number of elements failing which there is a DCHECK in the renderer followed
by a crash.
   To send the ResourceResponse back we need to remember the last response which
was received
   I added a static scoped_refptr to the NavigationRequest class which tracks
this and is
   released in the subsequent navigation request.

2. In Blink when the navigation is processed after we receive the
CommitNavigation request
   there is code which adds the document URL (referrer) to the redirect list
followed by
   the final URL. This breaks the PlzNavigate case because the list now looks
like
   client redirect URL->Server 1->Server 2->Client redirect URL->final URL. I
changed this
   code to add the main document URL (referrer) to the front of the list if
required.
   This does not happen in the PlzNavigate case because this code executes
before the
   other redirects are processed.

BUG=660567
CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.linux:linux_site_isolation
==========

[ananta, 2016-10-28 22:36:55]

ananta@chromium.org changed reviewers:
+ jam@chromium.org, japhet@chromium.org, nasko@chromium.org

[ananta, 2016-10-28 22:44:49]

Description was changed from

==========
PlzNavigate: Fix for the RedirectTest.ClientServerServer test.

This test does the following:
Navigates to a meta tag refreshed page which is refreshed to server 1 via a HTTP
redirect followed by another redirect to server2 and then the final URL.

This works correctly with the non plznavigate case as individual redirect
messages
are passed back to the renderer (blink) where the list is correctly maintained
and is passed back in the FrameHostMsg_DidCommitProvisionalLoad message.

With PlzNavigate we do pass the server redirect information back to the renderer
in the FrameMsg_CommitNavigation IPC message. However the first redirection
which
happens via a client redirect via a meta tag on the page or via JS is not in
this list.

Changes in this patch are as below:
1. Add the referrer URL to the redirect list in NavigationRequest when we
receive a
   navigation request with the PAGE_TRANSITION_CLIENT_REDIRECT transition type.
We also
   need to add the ResourceResponse structure in the redirect_response list
which is also
   part of the RequestNavigationParams structure. These lists have to contain
the same
   number of elements failing which there is a DCHECK in the renderer followed
by a crash.
   To send the ResourceResponse back we need to remember the last response which
was received
   I added a static scoped_refptr to the NavigationRequest class which tracks
this and is
   released in the subsequent navigation request.

2. In Blink when the navigation is processed after we receive the
CommitNavigation request
   there is code which adds the document URL (referrer) to the redirect list
followed by
   the final URL. This breaks the PlzNavigate case because the list now looks
like
   client redirect URL->Server 1->Server 2->Client redirect URL->final URL. I
changed this
   code to add the main document URL (referrer) to the front of the list if
required.
   This does not happen in the PlzNavigate case because this code executes
before the
   other redirects are processed.

BUG=660567
CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.linux:linux_site_isolation
==========

to

==========
PlzNavigate: Fix for the RedirectTest.ClientServerServer test.

japhet: Please review blink changes.
jam/nasko: Please review everything

This test does the following:
Navigates to a meta tag refreshed page which is refreshed to server 1 via a HTTP
redirect followed by another redirect to server2 and then the final URL.

This works correctly with the non plznavigate case as individual redirect
messages
are passed back to the renderer (blink) where the list is correctly maintained
and is passed back in the FrameHostMsg_DidCommitProvisionalLoad message.

With PlzNavigate we do pass the server redirect information back to the renderer
in the FrameMsg_CommitNavigation IPC message. However the first redirection
which
happens via a client redirect via a meta tag on the page or via JS is not in
this list.

Changes in this patch are as below:
1. Add the referrer URL to the redirect list in NavigationRequest when we
receive a
   navigation request with the PAGE_TRANSITION_CLIENT_REDIRECT transition type.
We also
   need to add the ResourceResponse structure in the redirect_response list
which is also
   part of the RequestNavigationParams structure. These lists have to contain
the same
   number of elements failing which there is a DCHECK in the renderer followed
by a crash.
   To send the ResourceResponse back we need to remember the last response which
was received
   I added a static scoped_refptr to the NavigationRequest class which tracks
this and is
   released in the subsequent navigation request.

2. In Blink when the navigation is processed after we receive the
CommitNavigation request
   there is code which adds the document URL (referrer) to the redirect list
followed by
   the final URL. This breaks the PlzNavigate case because the list now looks
like
   client redirect URL->Server 1->Server 2->Client redirect URL->final URL. I
changed this
   code to add the main document URL (referrer) to the front of the list if
required.
   This does not happen in the PlzNavigate case because this code executes
before the
   other redirects are processed.

BUG=660567
CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.linux:linux_site_isolation
==========

[commit-bot: I haz the power, 2016-10-28 22:48:41]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2016-10-28 22:48:42]

Dry run: Try jobs failed on following builders:
  android_n5x_swarming_rel on master.tryserver.chromium.android (JOB_FAILED,
https://build.chromium.org/p/tryserver.chromium.android/builders/android_n5x_...)

[ananta, 2016-10-29 01:37:14]

The CQ bit was checked by ananta@chromium.org to run a CQ dry run

[commit-bot: I haz the power, 2016-10-29 01:37:27]

Dry run: CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[commit-bot: I haz the power, 2016-10-29 03:46:06]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2016-10-29 03:46:07]

Dry run: Try jobs failed on following builders:
  linux_chromium_rel_ng on master.tryserver.chromium.linux (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.linux/builders/linux_chromium_...)

[ananta, 2016-10-29 14:50:17]

Description was changed from

==========
PlzNavigate: Fix for the RedirectTest.ClientServerServer test.

japhet: Please review blink changes.
jam/nasko: Please review everything

This test does the following:
Navigates to a meta tag refreshed page which is refreshed to server 1 via a HTTP
redirect followed by another redirect to server2 and then the final URL.

This works correctly with the non plznavigate case as individual redirect
messages
are passed back to the renderer (blink) where the list is correctly maintained
and is passed back in the FrameHostMsg_DidCommitProvisionalLoad message.

With PlzNavigate we do pass the server redirect information back to the renderer
in the FrameMsg_CommitNavigation IPC message. However the first redirection
which
happens via a client redirect via a meta tag on the page or via JS is not in
this list.

Changes in this patch are as below:
1. Add the referrer URL to the redirect list in NavigationRequest when we
receive a
   navigation request with the PAGE_TRANSITION_CLIENT_REDIRECT transition type.
We also
   need to add the ResourceResponse structure in the redirect_response list
which is also
   part of the RequestNavigationParams structure. These lists have to contain
the same
   number of elements failing which there is a DCHECK in the renderer followed
by a crash.
   To send the ResourceResponse back we need to remember the last response which
was received
   I added a static scoped_refptr to the NavigationRequest class which tracks
this and is
   released in the subsequent navigation request.

2. In Blink when the navigation is processed after we receive the
CommitNavigation request
   there is code which adds the document URL (referrer) to the redirect list
followed by
   the final URL. This breaks the PlzNavigate case because the list now looks
like
   client redirect URL->Server 1->Server 2->Client redirect URL->final URL. I
changed this
   code to add the main document URL (referrer) to the front of the list if
required.
   This does not happen in the PlzNavigate case because this code executes
before the
   other redirects are processed.

BUG=660567
CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.linux:linux_site_isolation
==========

to

==========
PlzNavigate: Fix for the RedirectTest.ClientServerServer test.

japhet: Please review blink changes.
jam/nasko: Please review everything

This test does the following:
Navigates to a meta tag refreshed page which is refreshed to server 1 via a HTTP
redirect followed by another redirect to server2 and then the final URL.

This works correctly with the non plznavigate case as individual redirect
messages
are passed back to the renderer (blink) where the list is correctly maintained
and is passed back in the FrameHostMsg_DidCommitProvisionalLoad message with the
url of the referrer which initiated the client side redirect at the head of the
list.

With PlzNavigate we do pass the server redirect information back to the renderer
in the FrameMsg_CommitNavigation IPC message. When the navigation is processed
the url
of the referrer which initated the client side redirect is added after the
server side 
redirect urls as blink just appends the document url (original referrer) to the
redirect
list followed by the destination url. Basically the assumption is that blink is
always
informed about server side redirects which is not the case with PlzNavigate.

Changes in this patch are as below:
1. Add the document URL for client side redirects to the head of the redirect
list.
   This ensures that the redirect list is in the correct order. This does not
impact
   the non PlzNavigate case as this code executes prior to the server redirects
getting
   processed.

BUG=660567

CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.linux:linux_site_isolation
==========

[ananta, 2016-10-29 14:51:31]

The CQ bit was checked by ananta@chromium.org to run a CQ dry run

[commit-bot: I haz the power, 2016-10-29 14:51:41]

Dry run: CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[commit-bot: I haz the power, 2016-10-29 14:58:12]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2016-10-29 14:58:13]

Dry run: Try jobs failed on following builders:
  chromium_presubmit on master.tryserver.chromium.linux (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.linux/builders/chromium_presub...)

[ananta, 2016-10-29 18:13:10]

The CQ bit was checked by ananta@chromium.org to run a CQ dry run

[commit-bot: I haz the power, 2016-10-29 18:13:16]

Dry run: CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[commit-bot: I haz the power, 2016-10-29 19:38:02]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2016-10-29 19:38:03]

Dry run: This issue passed the CQ dry run.

[Nate Chapin, 2016-10-31 17:21:11]

lgtm

[nasko, 2016-10-31 18:45:50]

https://codereview.chromium.org/2461043003/diff/80001/third_party/WebKit/Sour...
File third_party/WebKit/Source/core/loader/FrameLoader.cpp (right):

https://codereview.chromium.org/2461043003/diff/80001/third_party/WebKit/Sour...
third_party/WebKit/Source/core/loader/FrameLoader.cpp:1681:
m_provisionalDocumentLoader->prependRedirect(m_frame->document()->url());
Why is there another URL in the redirect chain? The m_provisionalDocumentLoader
is constructed on 1658 and the m_redirectChain should be empty at this point,
which is why just calling appendRedirect should've just worked.

[Nate Chapin, 2016-10-31 19:08:16]

On 2016/10/31 18:45:50, nasko (out until nov 4) wrote:
>
https://codereview.chromium.org/2461043003/diff/80001/third_party/WebKit/Sour...
> File third_party/WebKit/Source/core/loader/FrameLoader.cpp (right):
> 
>
https://codereview.chromium.org/2461043003/diff/80001/third_party/WebKit/Sour...
> third_party/WebKit/Source/core/loader/FrameLoader.cpp:1681:
> m_provisionalDocumentLoader->prependRedirect(m_frame->document()->url());
> Why is there another URL in the redirect chain? The
m_provisionalDocumentLoader
> is constructed on 1658 and the m_redirectChain should be empty at this point,
> which is why just calling appendRedirect should've just worked.

PlzNavigate adds its redirect chain in the didCreateDataSource() callback, which
runs before this point.

Actually, could we just move this to the DocumentLoader constructor? That
removes the need to add a prependRedirect() method, and we have access to the
LocalFrame* there anyway.

[ananta, 2016-10-31 19:11:33]

https://codereview.chromium.org/2461043003/diff/80001/third_party/WebKit/Sour...
File third_party/WebKit/Source/core/loader/FrameLoader.cpp (right):

https://codereview.chromium.org/2461043003/diff/80001/third_party/WebKit/Sour...
third_party/WebKit/Source/core/loader/FrameLoader.cpp:1681:
m_provisionalDocumentLoader->prependRedirect(m_frame->document()->url());
On 2016/10/31 18:45:50, nasko (out until nov 4) wrote:
> Why is there another URL in the redirect chain? The
m_provisionalDocumentLoader
> is constructed on 1658 and the m_redirectChain should be empty at this point,
> which is why just calling appendRedirect should've just worked.

The FrameLoaderClientImpl::createDocumentLoader function internally calls
RenderFrameImpl::didCreateDataSource which adds redirects to the list. It is not
empty as a result.

https://cs.chromium.org/chromium/src/content/renderer/render_frame_impl.cc?q=...

[nasko, 2016-10-31 19:14:57]

Thanks for the clarifications! I defer to Nate on blink code.

[ananta, 2016-10-31 20:16:44]

Description was changed from

==========
PlzNavigate: Fix for the RedirectTest.ClientServerServer test.

japhet: Please review blink changes.
jam/nasko: Please review everything

This test does the following:
Navigates to a meta tag refreshed page which is refreshed to server 1 via a HTTP
redirect followed by another redirect to server2 and then the final URL.

This works correctly with the non plznavigate case as individual redirect
messages
are passed back to the renderer (blink) where the list is correctly maintained
and is passed back in the FrameHostMsg_DidCommitProvisionalLoad message with the
url of the referrer which initiated the client side redirect at the head of the
list.

With PlzNavigate we do pass the server redirect information back to the renderer
in the FrameMsg_CommitNavigation IPC message. When the navigation is processed
the url
of the referrer which initated the client side redirect is added after the
server side 
redirect urls as blink just appends the document url (original referrer) to the
redirect
list followed by the destination url. Basically the assumption is that blink is
always
informed about server side redirects which is not the case with PlzNavigate.

Changes in this patch are as below:
1. Add the document URL for client side redirects to the head of the redirect
list.
   This ensures that the redirect list is in the correct order. This does not
impact
   the non PlzNavigate case as this code executes prior to the server redirects
getting
   processed.

BUG=660567

CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.linux:linux_site_isolation
==========

to

==========
PlzNavigate: Fix for the RedirectTest.ClientServerServer test.

japhet: Please review blink changes.
jam/nasko: Please review everything

This test does the following:
Navigates to a meta tag refreshed page which is refreshed to server 1 via a HTTP
redirect followed by another redirect to server2 and then the final URL.

This works correctly with the non plznavigate case as individual redirect
messages
are passed back to the renderer (blink) where the list is correctly maintained
and is passed back in the FrameHostMsg_DidCommitProvisionalLoad message with the
url of the referrer which initiated the client side redirect at the head of the
list.

With PlzNavigate we do pass the server redirect information back to the renderer
in the FrameMsg_CommitNavigation IPC message. When the navigation is processed
the url
of the referrer which initated the client side redirect is added after the
server side 
redirect urls as blink just appends the document url (original referrer) to the
redirect
list followed by the destination url. Basically the assumption is that blink is
always
informed about server side redirects which is not the case with PlzNavigate.

Changes in this patch are as below:
1. Add the document URL for client side redirects to the redirect list in the
DocumentLoader ctor. The other changes
are to ensure that the client redirection flag makes it to the DocumentLoader
ctor

   This ensures that the redirect list is in the correct order. 

BUG=660567

CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.linux:linux_site_isolation
==========

[ananta, 2016-10-31 20:17:45]

On 2016/10/31 19:08:16, Nate Chapin wrote:
> On 2016/10/31 18:45:50, nasko (out until nov 4) wrote:
> >
>
https://codereview.chromium.org/2461043003/diff/80001/third_party/WebKit/Sour...
> > File third_party/WebKit/Source/core/loader/FrameLoader.cpp (right):
> > 
> >
>
https://codereview.chromium.org/2461043003/diff/80001/third_party/WebKit/Sour...
> > third_party/WebKit/Source/core/loader/FrameLoader.cpp:1681:
> > m_provisionalDocumentLoader->prependRedirect(m_frame->document()->url());
> > Why is there another URL in the redirect chain? The
> m_provisionalDocumentLoader
> > is constructed on 1658 and the m_redirectChain should be empty at this
point,
> > which is why just calling appendRedirect should've just worked.
> 
> PlzNavigate adds its redirect chain in the didCreateDataSource() callback,
which
> runs before this point.
> 
> Actually, could we just move this to the DocumentLoader constructor? That
> removes the need to add a prependRedirect() method, and we have access to the
> LocalFrame* there anyway.

Updated the patch to add the document url to the redirect list for client side
redirects in the DocumentLoader ctor.
PTAL

[ananta, 2016-10-31 20:20:26]

The CQ bit was checked by ananta@chromium.org to run a CQ dry run

[commit-bot: I haz the power, 2016-10-31 20:21:20]

Dry run: CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[commit-bot: I haz the power, 2016-10-31 20:39:05]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2016-10-31 20:39:09]

Dry run: Try jobs failed on following builders:
  linux_chromium_asan_rel_ng on master.tryserver.chromium.linux (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.linux/builders/linux_chromium_...)
  linux_chromium_rel_ng on master.tryserver.chromium.linux (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.linux/builders/linux_chromium_...)

[ananta, 2016-10-31 21:42:28]

The CQ bit was checked by ananta@chromium.org to run a CQ dry run

[commit-bot: I haz the power, 2016-10-31 21:43:17]

Dry run: CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[commit-bot: I haz the power, 2016-10-31 22:03:14]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2016-10-31 22:03:15]

Dry run: Try jobs failed on following builders:
  linux_chromium_chromeos_rel_ng on master.tryserver.chromium.linux (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.linux/builders/linux_chromium_...)
  linux_site_isolation on master.tryserver.chromium.linux (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.linux/builders/linux_site_isol...)

[ananta, 2016-10-31 22:19:27]

The CQ bit was checked by ananta@chromium.org to run a CQ dry run

[commit-bot: I haz the power, 2016-10-31 22:19:54]

Dry run: CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[commit-bot: I haz the power, 2016-10-31 22:37:53]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2016-10-31 22:37:54]

Dry run: Try jobs failed on following builders:
  mac_chromium_rel_ng on master.tryserver.chromium.mac (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.mac/builders/mac_chromium_rel_...)

[ananta, 2016-10-31 22:42:28]

The CQ bit was checked by ananta@chromium.org to run a CQ dry run

[commit-bot: I haz the power, 2016-10-31 22:42:43]

Dry run: CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[Nate Chapin, 2016-10-31 23:06:36]

LGTM with a nit

https://codereview.chromium.org/2461043003/diff/180001/third_party/WebKit/Sou...
File third_party/WebKit/Source/core/loader/FrameLoaderClient.h (right):

https://codereview.chromium.org/2461043003/diff/180001/third_party/WebKit/Sou...
third_party/WebKit/Source/core/loader/FrameLoaderClient.h:174: bool
clientRedirect) = 0;
Nit: here and throughout, pass the enum instead of the bool?

[commit-bot: I haz the power, 2016-11-01 00:44:50]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2016-11-01 00:44:51]

Dry run: Try jobs failed on following builders:
  linux_site_isolation on master.tryserver.chromium.linux (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.linux/builders/linux_site_isol...)

[ananta, 2016-11-01 00:45:58]

The CQ bit was checked by ananta@chromium.org to run a CQ dry run

[commit-bot: I haz the power, 2016-11-01 00:46:34]

Dry run: CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[commit-bot: I haz the power, 2016-11-01 01:03:07]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2016-11-01 01:03:08]

Dry run: Try jobs failed on following builders:
  linux_chromium_asan_rel_ng on master.tryserver.chromium.linux (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.linux/builders/linux_chromium_...)

[ananta, 2016-11-01 01:08:43]

The CQ bit was checked by ananta@chromium.org to run a CQ dry run

[commit-bot: I haz the power, 2016-11-01 01:09:08]

Dry run: CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[commit-bot: I haz the power, 2016-11-01 01:25:12]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2016-11-01 01:25:13]

Dry run: Try jobs failed on following builders:
  linux_chromium_compile_dbg_ng on master.tryserver.chromium.linux (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.linux/builders/linux_chromium_...)
  linux_chromium_rel_ng on master.tryserver.chromium.linux (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.linux/builders/linux_chromium_...)

[ananta, 2016-11-01 01:38:01]

The CQ bit was checked by ananta@chromium.org to run a CQ dry run

[commit-bot: I haz the power, 2016-11-01 01:38:30]

Dry run: CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[commit-bot: I haz the power, 2016-11-01 02:42:10]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2016-11-01 02:42:11]

Dry run: Try jobs failed on following builders:
  linux_site_isolation on master.tryserver.chromium.linux (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.linux/builders/linux_site_isol...)

[ananta, 2016-11-01 02:49:26]

The CQ bit was checked by ananta@chromium.org

[ananta, 2016-11-01 02:49:26]

The patchset sent to the CQ was uploaded after l-g-t-m from japhet@chromium.org
Link to the patchset: https://codereview.chromium.org/2461043003/#ps240001
(title: "More build error fixes")

[commit-bot: I haz the power, 2016-11-01 02:49:45]

CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[ananta, 2016-11-01 02:49:50]

https://codereview.chromium.org/2461043003/diff/180001/third_party/WebKit/Sou...
File third_party/WebKit/Source/core/loader/FrameLoaderClient.h (right):

https://codereview.chromium.org/2461043003/diff/180001/third_party/WebKit/Sou...
third_party/WebKit/Source/core/loader/FrameLoaderClient.h:174: bool
clientRedirect) = 0;
On 2016/10/31 23:06:36, Nate Chapin wrote:
> Nit: here and throughout, pass the enum instead of the bool?

Done.

[commit-bot: I haz the power, 2016-11-01 02:56:37]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2016-11-01 02:56:38]

Try jobs failed on following builders:
  chromium_presubmit on master.tryserver.chromium.linux (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.linux/builders/chromium_presub...)

[ananta, 2016-11-01 13:27:20]

ananta@chromium.org changed reviewers:
+ kbr@chromium.org

[ananta, 2016-11-01 13:27:37]

Description was changed from

==========
PlzNavigate: Fix for the RedirectTest.ClientServerServer test.

japhet: Please review blink changes.
jam/nasko: Please review everything

This test does the following:
Navigates to a meta tag refreshed page which is refreshed to server 1 via a HTTP
redirect followed by another redirect to server2 and then the final URL.

This works correctly with the non plznavigate case as individual redirect
messages
are passed back to the renderer (blink) where the list is correctly maintained
and is passed back in the FrameHostMsg_DidCommitProvisionalLoad message with the
url of the referrer which initiated the client side redirect at the head of the
list.

With PlzNavigate we do pass the server redirect information back to the renderer
in the FrameMsg_CommitNavigation IPC message. When the navigation is processed
the url
of the referrer which initated the client side redirect is added after the
server side 
redirect urls as blink just appends the document url (original referrer) to the
redirect
list followed by the destination url. Basically the assumption is that blink is
always
informed about server side redirects which is not the case with PlzNavigate.

Changes in this patch are as below:
1. Add the document URL for client side redirects to the redirect list in the
DocumentLoader ctor. The other changes
are to ensure that the client redirection flag makes it to the DocumentLoader
ctor

   This ensures that the redirect list is in the correct order. 

BUG=660567

CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.linux:linux_site_isolation
==========

to

==========
PlzNavigate: Fix for the RedirectTest.ClientServerServer test.

japhet: Please review blink changes.
jam/nasko: Please review everything

This test does the following:
Navigates to a meta tag refreshed page which is refreshed to server 1 via a HTTP
redirect followed by another redirect to server2 and then the final URL.

This works correctly with the non plznavigate case as individual redirect
messages
are passed back to the renderer (blink) where the list is correctly maintained
and is passed back in the FrameHostMsg_DidCommitProvisionalLoad message with the
url of the referrer which initiated the client side redirect at the head of the
list.

With PlzNavigate we do pass the server redirect information back to the renderer
in the FrameMsg_CommitNavigation IPC message. When the navigation is processed
the url
of the referrer which initated the client side redirect is added after the
server side 
redirect urls as blink just appends the document url (original referrer) to the
redirect
list followed by the destination url. Basically the assumption is that blink is
always
informed about server side redirects which is not the case with PlzNavigate.

Changes in this patch are as below:
1. Add the document URL for client side redirects to the redirect list in the
DocumentLoader ctor. The other changes
are to ensure that the client redirection flag makes it to the DocumentLoader
ctor

   This ensures that the redirect list is in the correct order. 

BUG=660567
TBR=kbr

CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.linux:linux_site_isolation
==========

[ananta, 2016-11-01 13:28:42]

+kbr for BaseAudioContext.cpp owners. TBR'ing. Will address comments in a
followup.

[ananta, 2016-11-01 13:28:52]

The CQ bit was checked by ananta@chromium.org

[commit-bot: I haz the power, 2016-11-01 13:29:04]

CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[commit-bot: I haz the power, 2016-11-01 15:41:06]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2016-11-01 15:41:07]

Try jobs failed on following builders:
  linux_site_isolation on master.tryserver.chromium.linux (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.linux/builders/linux_site_isol...)

[jam, 2016-11-01 17:19:40]

cool! I'm not that familiar with this code, and I'm not an owner, so I defer to
Nate.

[ananta, 2016-11-01 17:34:14]

The CQ bit was checked by ananta@chromium.org

[commit-bot: I haz the power, 2016-11-01 17:34:48]

CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[commit-bot: I haz the power, 2016-11-01 18:38:12]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2016-11-01 18:38:13]

Try jobs failed on following builders:
  linux_site_isolation on master.tryserver.chromium.linux (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.linux/builders/linux_site_isol...)

[Ken Russell (switch to Gerrit), 2016-11-01 18:47:26]

BaseAudioContextTest lgtm

[ananta, 2016-11-01 19:09:49]

The CQ bit was checked by ananta@chromium.org

[commit-bot: I haz the power, 2016-11-01 19:10:12]

CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[ananta, 2016-11-01 20:27:17]

Description was changed from

==========
PlzNavigate: Fix for the RedirectTest.ClientServerServer test.

japhet: Please review blink changes.
jam/nasko: Please review everything

This test does the following:
Navigates to a meta tag refreshed page which is refreshed to server 1 via a HTTP
redirect followed by another redirect to server2 and then the final URL.

This works correctly with the non plznavigate case as individual redirect
messages
are passed back to the renderer (blink) where the list is correctly maintained
and is passed back in the FrameHostMsg_DidCommitProvisionalLoad message with the
url of the referrer which initiated the client side redirect at the head of the
list.

With PlzNavigate we do pass the server redirect information back to the renderer
in the FrameMsg_CommitNavigation IPC message. When the navigation is processed
the url
of the referrer which initated the client side redirect is added after the
server side 
redirect urls as blink just appends the document url (original referrer) to the
redirect
list followed by the destination url. Basically the assumption is that blink is
always
informed about server side redirects which is not the case with PlzNavigate.

Changes in this patch are as below:
1. Add the document URL for client side redirects to the redirect list in the
DocumentLoader ctor. The other changes
are to ensure that the client redirection flag makes it to the DocumentLoader
ctor

   This ensures that the redirect list is in the correct order. 

BUG=660567
TBR=kbr

CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.linux:linux_site_isolation
==========

to

==========
Fix for the RedirectTest.ClientServerServer test.

japhet: Please review blink changes.
jam/nasko: Please review everything

This test does the following:
Navigates to a meta tag refreshed page which is refreshed to server 1 via a HTTP
redirect followed by another redirect to server2 and then the final URL.

This works correctly with the non plznavigate case as individual redirect
messages
are passed back to the renderer (blink) where the list is correctly maintained
and is passed back in the FrameHostMsg_DidCommitProvisionalLoad message with the
url of the referrer which initiated the client side redirect at the head of the
list.

With PlzNavigate we do pass the server redirect information back to the renderer
in the FrameMsg_CommitNavigation IPC message. When the navigation is processed
the url
of the referrer which initated the client side redirect is added after the
server side 
redirect urls as blink just appends the document url (original referrer) to the
redirect
list followed by the destination url. Basically the assumption is that blink is
always
informed about server side redirects which is not the case with PlzNavigate.

Changes in this patch are as below:
1. Add the document URL for client side redirects to the redirect list in the
DocumentLoader ctor. The other changes
are to ensure that the client redirection flag makes it to the DocumentLoader
ctor

   This ensures that the redirect list is in the correct order. 

BUG=660567
TBR=kbr

CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.linux:linux_site_isolation
==========

[ananta, 2016-11-01 20:27:27]

The CQ bit was unchecked by ananta@chromium.org

[ananta, 2016-11-01 20:27:33]

The CQ bit was checked by ananta@chromium.org

[commit-bot: I haz the power, 2016-11-01 20:27:58]

CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[ananta, 2016-11-01 22:30:03]

Description was changed from

==========
Fix for the RedirectTest.ClientServerServer test.

japhet: Please review blink changes.
jam/nasko: Please review everything

This test does the following:
Navigates to a meta tag refreshed page which is refreshed to server 1 via a HTTP
redirect followed by another redirect to server2 and then the final URL.

This works correctly with the non plznavigate case as individual redirect
messages
are passed back to the renderer (blink) where the list is correctly maintained
and is passed back in the FrameHostMsg_DidCommitProvisionalLoad message with the
url of the referrer which initiated the client side redirect at the head of the
list.

With PlzNavigate we do pass the server redirect information back to the renderer
in the FrameMsg_CommitNavigation IPC message. When the navigation is processed
the url
of the referrer which initated the client side redirect is added after the
server side 
redirect urls as blink just appends the document url (original referrer) to the
redirect
list followed by the destination url. Basically the assumption is that blink is
always
informed about server side redirects which is not the case with PlzNavigate.

Changes in this patch are as below:
1. Add the document URL for client side redirects to the redirect list in the
DocumentLoader ctor. The other changes
are to ensure that the client redirection flag makes it to the DocumentLoader
ctor

   This ensures that the redirect list is in the correct order. 

BUG=660567
TBR=kbr

CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.linux:linux_site_isolation
==========

to

==========
Fix for the RedirectTest.ClientServerServer test with browser side navigation.

japhet: Please review blink changes.
jam/nasko: Please review everything

This test does the following:
Navigates to a meta tag refreshed page which is refreshed to server 1 via a HTTP
redirect followed by another redirect to server2 and then the final URL.

This works correctly with the non plznavigate case as individual redirect
messages
are passed back to the renderer (blink) where the list is correctly maintained
and is passed back in the FrameHostMsg_DidCommitProvisionalLoad message with the
url of the referrer which initiated the client side redirect at the head of the
list.

With PlzNavigate we do pass the server redirect information back to the renderer
in the FrameMsg_CommitNavigation IPC message. When the navigation is processed
the url
of the referrer which initated the client side redirect is added after the
server side 
redirect urls as blink just appends the document url (original referrer) to the
redirect
list followed by the destination url. Basically the assumption is that blink is
always
informed about server side redirects which is not the case with PlzNavigate.

Changes in this patch are as below:
1. Add the document URL for client side redirects to the redirect list in the
DocumentLoader ctor. The other changes
are to ensure that the client redirection flag makes it to the DocumentLoader
ctor

   This ensures that the redirect list is in the correct order. 

BUG=660567
TBR=kbr

CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.linux:linux_site_isolation
==========

[ananta, 2016-11-01 22:30:11]

The CQ bit was unchecked by ananta@chromium.org

[ananta, 2016-11-01 22:30:37]

The CQ bit was checked by ananta@chromium.org

[commit-bot: I haz the power, 2016-11-01 22:31:07]

CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[commit-bot: I haz the power, 2016-11-01 23:48:13]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2016-11-01 23:48:15]

Try jobs failed on following builders:
  linux_site_isolation on master.tryserver.chromium.linux (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.linux/builders/linux_site_isol...)

[ananta, 2016-11-01 23:50:04]

The CQ bit was checked by ananta@chromium.org

[commit-bot: I haz the power, 2016-11-01 23:50:35]

CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[commit-bot: I haz the power, 2016-11-02 01:14:04]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2016-11-02 01:14:06]

Try jobs failed on following builders:
  linux_site_isolation on master.tryserver.chromium.linux (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.linux/builders/linux_site_isol...)

[ananta, 2016-11-02 02:26:42]

Description was changed from

==========
Fix for the RedirectTest.ClientServerServer test with browser side navigation.

japhet: Please review blink changes.
jam/nasko: Please review everything

This test does the following:
Navigates to a meta tag refreshed page which is refreshed to server 1 via a HTTP
redirect followed by another redirect to server2 and then the final URL.

This works correctly with the non plznavigate case as individual redirect
messages
are passed back to the renderer (blink) where the list is correctly maintained
and is passed back in the FrameHostMsg_DidCommitProvisionalLoad message with the
url of the referrer which initiated the client side redirect at the head of the
list.

With PlzNavigate we do pass the server redirect information back to the renderer
in the FrameMsg_CommitNavigation IPC message. When the navigation is processed
the url
of the referrer which initated the client side redirect is added after the
server side 
redirect urls as blink just appends the document url (original referrer) to the
redirect
list followed by the destination url. Basically the assumption is that blink is
always
informed about server side redirects which is not the case with PlzNavigate.

Changes in this patch are as below:
1. Add the document URL for client side redirects to the redirect list in the
DocumentLoader ctor. The other changes
are to ensure that the client redirection flag makes it to the DocumentLoader
ctor

   This ensures that the redirect list is in the correct order. 

BUG=660567
TBR=kbr

CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.linux:linux_site_isolation
==========

to

==========
Fix for the RedirectTest.ClientServerServer test with browser side navigation.

japhet: Please review blink changes.
jam/nasko: Please review everything

This test does the following:
Navigates to a meta tag refreshed page which is refreshed to server 1 via a HTTP
redirect followed by another redirect to server2 and then the final URL.

This works correctly with the non plznavigate case as individual redirect
messages
are passed back to the renderer (blink) where the list is correctly maintained
and is passed back in the FrameHostMsg_DidCommitProvisionalLoad message with the
url of the referrer which initiated the client side redirect at the head of the
list.

With browser side navigation we do pass the server redirect information back to
the renderer
in the FrameMsg_CommitNavigation IPC message. When the navigation is processed
the url
of the referrer which initated the client side redirect is added after the
server side 
redirect urls as blink just appends the document url (original referrer) to the
redirect
list followed by the destination url. Basically the assumption is that blink is
always
informed about server side redirects which is not the case with PlzNavigate.

Changes in this patch are as below:
1. Add the document URL for client side redirects to the redirect list in the
DocumentLoader ctor. The other changes
are to ensure that the client redirection flag makes it to the DocumentLoader
ctor

   This ensures that the redirect list is in the correct order. 

BUG=660567
TBR=kbr

CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.linux:linux_site_isolation
==========

[ananta, 2016-11-02 02:27:14]

Description was changed from

==========
Fix for the RedirectTest.ClientServerServer test with browser side navigation.

japhet: Please review blink changes.
jam/nasko: Please review everything

This test does the following:
Navigates to a meta tag refreshed page which is refreshed to server 1 via a HTTP
redirect followed by another redirect to server2 and then the final URL.

This works correctly with the non plznavigate case as individual redirect
messages
are passed back to the renderer (blink) where the list is correctly maintained
and is passed back in the FrameHostMsg_DidCommitProvisionalLoad message with the
url of the referrer which initiated the client side redirect at the head of the
list.

With browser side navigation we do pass the server redirect information back to
the renderer
in the FrameMsg_CommitNavigation IPC message. When the navigation is processed
the url
of the referrer which initated the client side redirect is added after the
server side 
redirect urls as blink just appends the document url (original referrer) to the
redirect
list followed by the destination url. Basically the assumption is that blink is
always
informed about server side redirects which is not the case with PlzNavigate.

Changes in this patch are as below:
1. Add the document URL for client side redirects to the redirect list in the
DocumentLoader ctor. The other changes
are to ensure that the client redirection flag makes it to the DocumentLoader
ctor

   This ensures that the redirect list is in the correct order. 

BUG=660567
TBR=kbr

CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.linux:linux_site_isolation
==========

to

==========
Fix for the RedirectTest.ClientServerServer test with browser side navigation.

japhet: Please review blink changes.
jam/nasko: Please review everything

This test does the following:
Navigates to a meta tag refreshed page which is refreshed to server 1 via a HTTP
redirect followed by another redirect to server2 and then the final URL.

This works correctly in the regular case as individual redirect messages
are passed back to the renderer (blink) where the list is correctly maintained
and is passed back in the FrameHostMsg_DidCommitProvisionalLoad message with the
url of the referrer which initiated the client side redirect at the head of the
list.

With browser side navigation we do pass the server redirect information back to
the renderer
in the FrameMsg_CommitNavigation IPC message. When the navigation is processed
the url
of the referrer which initated the client side redirect is added after the
server side 
redirect urls as blink just appends the document url (original referrer) to the
redirect
list followed by the destination url. Basically the assumption is that blink is
always
informed about server side redirects which is not the case with PlzNavigate.

Changes in this patch are as below:
1. Add the document URL for client side redirects to the redirect list in the
DocumentLoader ctor. The other changes
are to ensure that the client redirection flag makes it to the DocumentLoader
ctor

   This ensures that the redirect list is in the correct order. 

BUG=660567
TBR=kbr

CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.linux:linux_site_isolation
==========

[ananta, 2016-11-02 02:27:51]

Description was changed from

==========
Fix for the RedirectTest.ClientServerServer test with browser side navigation.

japhet: Please review blink changes.
jam/nasko: Please review everything

This test does the following:
Navigates to a meta tag refreshed page which is refreshed to server 1 via a HTTP
redirect followed by another redirect to server2 and then the final URL.

This works correctly in the regular case as individual redirect messages
are passed back to the renderer (blink) where the list is correctly maintained
and is passed back in the FrameHostMsg_DidCommitProvisionalLoad message with the
url of the referrer which initiated the client side redirect at the head of the
list.

With browser side navigation we do pass the server redirect information back to
the renderer
in the FrameMsg_CommitNavigation IPC message. When the navigation is processed
the url
of the referrer which initated the client side redirect is added after the
server side 
redirect urls as blink just appends the document url (original referrer) to the
redirect
list followed by the destination url. Basically the assumption is that blink is
always
informed about server side redirects which is not the case with PlzNavigate.

Changes in this patch are as below:
1. Add the document URL for client side redirects to the redirect list in the
DocumentLoader ctor. The other changes
are to ensure that the client redirection flag makes it to the DocumentLoader
ctor

   This ensures that the redirect list is in the correct order. 

BUG=660567
TBR=kbr

CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.linux:linux_site_isolation
==========

to

==========
Fix for the RedirectTest.ClientServerServer test with browser side navigation.

japhet: Please review blink changes.
jam/nasko: Please review everything

This test does the following:
Navigates to a meta tag refreshed page which is refreshed to server 1 via a HTTP
redirect followed by another redirect to server2 and then the final URL.

This works correctly in the regular case as individual redirect messages
are passed back to the renderer (blink) where the list is correctly maintained
and is passed back in the FrameHostMsg_DidCommitProvisionalLoad message with the
url of the referrer which initiated the client side redirect at the head of the
list.

With browser side navigation we do pass the server redirect information back to
the renderer
in the FrameMsg_CommitNavigation IPC message. When the navigation is processed
the url
of the referrer which initated the client side redirect is added after the
server side 
redirect urls as blink just appends the document url (original referrer) to the
redirect
list followed by the destination url. Basically the assumption is that blink is
always
informed about server side redirects which is not the case with browser side
navigation.

Changes in this patch are as below:
1. Add the document URL for client side redirects to the redirect list in the
DocumentLoader ctor. The other changes
are to ensure that the client redirection flag makes it to the DocumentLoader
ctor

   This ensures that the redirect list is in the correct order. 

BUG=660567
TBR=kbr

CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.linux:linux_site_isolation
==========

[ananta, 2016-11-02 02:28:09]

The CQ bit was checked by ananta@chromium.org

[commit-bot: I haz the power, 2016-11-02 02:28:24]

CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[ananta, 2016-11-02 02:29:54]

Description was changed from

==========
Fix for the RedirectTest.ClientServerServer test with browser side navigation.

japhet: Please review blink changes.
jam/nasko: Please review everything

This test does the following:
Navigates to a meta tag refreshed page which is refreshed to server 1 via a HTTP
redirect followed by another redirect to server2 and then the final URL.

This works correctly in the regular case as individual redirect messages
are passed back to the renderer (blink) where the list is correctly maintained
and is passed back in the FrameHostMsg_DidCommitProvisionalLoad message with the
url of the referrer which initiated the client side redirect at the head of the
list.

With browser side navigation we do pass the server redirect information back to
the renderer
in the FrameMsg_CommitNavigation IPC message. When the navigation is processed
the url
of the referrer which initated the client side redirect is added after the
server side 
redirect urls as blink just appends the document url (original referrer) to the
redirect
list followed by the destination url. Basically the assumption is that blink is
always
informed about server side redirects which is not the case with browser side
navigation.

Changes in this patch are as below:
1. Add the document URL for client side redirects to the redirect list in the
DocumentLoader ctor. The other changes
are to ensure that the client redirection flag makes it to the DocumentLoader
ctor

   This ensures that the redirect list is in the correct order. 

BUG=660567
TBR=kbr

CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.linux:linux_site_isolation
==========

to

==========
Fix for the RedirectTest.ClientServerServer test with browser side navigation.

japhet: Please review blink changes.
jam/nasko: Please review everything

This test does the following:
Navigates to a meta tag refreshed page which is refreshed to server 1 via a HTTP
redirect followed by another redirect to server2 and then the final URL.

This works correctly in the regular case as individual redirect messages
are passed back to the renderer (blink) where the list is correctly maintained
and is passed back in the FrameHostMsg_DidCommitProvisionalLoad message with the
url of the referrer which initiated the client side redirect at the head of the
list.

With browser side navigation we do pass the server redirect information back to
the renderer
in the FrameMsg_CommitNavigation IPC message. When the navigation is processed
the url
of the referrer which initated the client side redirect is added after the
server side 
redirect urls as blink just appends the document url (original referrer) to the
redirect
list followed by the destination url. Basically the assumption is that blink is
always
informed about server side redirects which is not the case with browser side
navigation.

Changes in this patch are as below:
1. Add the document URL for client side redirects to the redirect list in the
DocumentLoader ctor. The other changes
are to ensure that the client redirection flag makes it to the DocumentLoader
ctor

   This ensures that the redirect list is in the correct order. 

BUG=660567
TBR=kbr

CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.linux
==========

[ananta, 2016-11-02 02:30:09]

The CQ bit was unchecked by ananta@chromium.org

[ananta, 2016-11-02 02:30:22]

The CQ bit was checked by ananta@chromium.org

[commit-bot: I haz the power, 2016-11-02 02:30:37]

CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[commit-bot: I haz the power, 2016-11-02 02:30:39]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2016-11-02 02:30:42]

Your CL can not be processed by CQ because of:
 * Failed to parse additional trybots

[ananta, 2016-11-02 02:43:36]

Description was changed from

==========
Fix for the RedirectTest.ClientServerServer test with browser side navigation.

japhet: Please review blink changes.
jam/nasko: Please review everything

This test does the following:
Navigates to a meta tag refreshed page which is refreshed to server 1 via a HTTP
redirect followed by another redirect to server2 and then the final URL.

This works correctly in the regular case as individual redirect messages
are passed back to the renderer (blink) where the list is correctly maintained
and is passed back in the FrameHostMsg_DidCommitProvisionalLoad message with the
url of the referrer which initiated the client side redirect at the head of the
list.

With browser side navigation we do pass the server redirect information back to
the renderer
in the FrameMsg_CommitNavigation IPC message. When the navigation is processed
the url
of the referrer which initated the client side redirect is added after the
server side 
redirect urls as blink just appends the document url (original referrer) to the
redirect
list followed by the destination url. Basically the assumption is that blink is
always
informed about server side redirects which is not the case with browser side
navigation.

Changes in this patch are as below:
1. Add the document URL for client side redirects to the redirect list in the
DocumentLoader ctor. The other changes
are to ensure that the client redirection flag makes it to the DocumentLoader
ctor

   This ensures that the redirect list is in the correct order. 

BUG=660567
TBR=kbr

CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.linux
==========

to

==========
Fix for the RedirectTest.ClientServerServer test with browser side navigation.

japhet: Please review blink changes.
jam/nasko: Please review everything

This test does the following:
Navigates to a meta tag refreshed page which is refreshed to server 1 via a HTTP
redirect followed by another redirect to server2 and then the final URL.

This works correctly in the regular case as individual redirect messages
are passed back to the renderer (blink) where the list is correctly maintained
and is passed back in the FrameHostMsg_DidCommitProvisionalLoad message with the
url of the referrer which initiated the client side redirect at the head of the
list.

With browser side navigation we do pass the server redirect information back to
the renderer
in the FrameMsg_CommitNavigation IPC message. When the navigation is processed
the url
of the referrer which initated the client side redirect is added after the
server side 
redirect urls as blink just appends the document url (original referrer) to the
redirect
list followed by the destination url. Basically the assumption is that blink is
always
informed about server side redirects which is not the case with browser side
navigation.

Changes in this patch are as below:
1. Add the document URL for client side redirects to the redirect list in the
DocumentLoader ctor. The other changes
are to ensure that the client redirection flag makes it to the DocumentLoader
ctor

   This ensures that the redirect list is in the correct order. 

BUG=660567
==========

[ananta, 2016-11-02 02:43:52]

The CQ bit was checked by ananta@chromium.org

[commit-bot: I haz the power, 2016-11-02 02:44:09]

CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[commit-bot: I haz the power, 2016-11-02 04:22:07]

Description was changed from

==========
Fix for the RedirectTest.ClientServerServer test with browser side navigation.

japhet: Please review blink changes.
jam/nasko: Please review everything

This test does the following:
Navigates to a meta tag refreshed page which is refreshed to server 1 via a HTTP
redirect followed by another redirect to server2 and then the final URL.

This works correctly in the regular case as individual redirect messages
are passed back to the renderer (blink) where the list is correctly maintained
and is passed back in the FrameHostMsg_DidCommitProvisionalLoad message with the
url of the referrer which initiated the client side redirect at the head of the
list.

With browser side navigation we do pass the server redirect information back to
the renderer
in the FrameMsg_CommitNavigation IPC message. When the navigation is processed
the url
of the referrer which initated the client side redirect is added after the
server side 
redirect urls as blink just appends the document url (original referrer) to the
redirect
list followed by the destination url. Basically the assumption is that blink is
always
informed about server side redirects which is not the case with browser side
navigation.

Changes in this patch are as below:
1. Add the document URL for client side redirects to the redirect list in the
DocumentLoader ctor. The other changes
are to ensure that the client redirection flag makes it to the DocumentLoader
ctor

   This ensures that the redirect list is in the correct order. 

BUG=660567
==========

to

==========
Fix for the RedirectTest.ClientServerServer test with browser side navigation.

japhet: Please review blink changes.
jam/nasko: Please review everything

This test does the following:
Navigates to a meta tag refreshed page which is refreshed to server 1 via a HTTP
redirect followed by another redirect to server2 and then the final URL.

This works correctly in the regular case as individual redirect messages
are passed back to the renderer (blink) where the list is correctly maintained
and is passed back in the FrameHostMsg_DidCommitProvisionalLoad message with the
url of the referrer which initiated the client side redirect at the head of the
list.

With browser side navigation we do pass the server redirect information back to
the renderer
in the FrameMsg_CommitNavigation IPC message. When the navigation is processed
the url
of the referrer which initated the client side redirect is added after the
server side 
redirect urls as blink just appends the document url (original referrer) to the
redirect
list followed by the destination url. Basically the assumption is that blink is
always
informed about server side redirects which is not the case with browser side
navigation.

Changes in this patch are as below:
1. Add the document URL for client side redirects to the redirect list in the
DocumentLoader ctor. The other changes
are to ensure that the client redirection flag makes it to the DocumentLoader
ctor

   This ensures that the redirect list is in the correct order. 

BUG=660567
==========

[commit-bot: I haz the power, 2016-11-02 04:22:09]

Committed patchset #13 (id:240001)

[commit-bot: I haz the power, 2016-11-02 04:24:05]

Description was changed from

==========
Fix for the RedirectTest.ClientServerServer test with browser side navigation.

japhet: Please review blink changes.
jam/nasko: Please review everything

This test does the following:
Navigates to a meta tag refreshed page which is refreshed to server 1 via a HTTP
redirect followed by another redirect to server2 and then the final URL.

This works correctly in the regular case as individual redirect messages
are passed back to the renderer (blink) where the list is correctly maintained
and is passed back in the FrameHostMsg_DidCommitProvisionalLoad message with the
url of the referrer which initiated the client side redirect at the head of the
list.

With browser side navigation we do pass the server redirect information back to
the renderer
in the FrameMsg_CommitNavigation IPC message. When the navigation is processed
the url
of the referrer which initated the client side redirect is added after the
server side 
redirect urls as blink just appends the document url (original referrer) to the
redirect
list followed by the destination url. Basically the assumption is that blink is
always
informed about server side redirects which is not the case with browser side
navigation.

Changes in this patch are as below:
1. Add the document URL for client side redirects to the redirect list in the
DocumentLoader ctor. The other changes
are to ensure that the client redirection flag makes it to the DocumentLoader
ctor

   This ensures that the redirect list is in the correct order. 

BUG=660567
==========

to

==========
Fix for the RedirectTest.ClientServerServer test with browser side navigation.

japhet: Please review blink changes.
jam/nasko: Please review everything

This test does the following:
Navigates to a meta tag refreshed page which is refreshed to server 1 via a HTTP
redirect followed by another redirect to server2 and then the final URL.

This works correctly in the regular case as individual redirect messages
are passed back to the renderer (blink) where the list is correctly maintained
and is passed back in the FrameHostMsg_DidCommitProvisionalLoad message with the
url of the referrer which initiated the client side redirect at the head of the
list.

With browser side navigation we do pass the server redirect information back to
the renderer
in the FrameMsg_CommitNavigation IPC message. When the navigation is processed
the url
of the referrer which initated the client side redirect is added after the
server side
redirect urls as blink just appends the document url (original referrer) to the
redirect
list followed by the destination url. Basically the assumption is that blink is
always
informed about server side redirects which is not the case with browser side
navigation.

Changes in this patch are as below:
1. Add the document URL for client side redirects to the redirect list in the
DocumentLoader ctor. The other changes
are to ensure that the client redirection flag makes it to the DocumentLoader
ctor

   This ensures that the redirect list is in the correct order.

BUG=660567

Committed: https://crrev.com/d9062cc1ce0d2fda908c0f8d4e959c12d60d3d90
Cr-Commit-Position: refs/heads/master@{#429207}
==========

[commit-bot: I haz the power, 2016-11-02 04:24:07]

Patchset 13 (id:??) landed as
https://crrev.com/d9062cc1ce0d2fda908c0f8d4e959c12d60d3d90
Cr-Commit-Position: refs/heads/master@{#429207}