Rewrites SerializedNavigationEntry to not have any //content member variables.

Rewrites SerializedNavigationEntry to not have any //content member variables.

This will pave the way for a future CL that completely removes all //content
dependencies from the class.

TBR=ben@google.com
BUG=371476
TEST=No visible impact.

Committed: https://crrev.com/c6d8e740dc97f57293ed336a278ff1934bb361ab
Cr-Commit-Position: refs/heads/master@{#300344}

[rohitrao (ping after 24h), 2014-10-17 15:15:52]

rohitrao@chromium.org changed reviewers:
+ marja@chromium.org, sky@chromium.org

[rohitrao (ping after 24h), 2014-10-17 15:15:53]

rohitrao@chromium.org changed required reviewers:
+ marja@chromium.org

[rohitrao (ping after 24h), 2014-10-17 15:15:54]

+marja as the primary reviewer
+sky as a non-required reviewer (mostly as FYI)
+blundell, sdefresne, and skuhne as FYI

This is step 1 in my master plan.  I've updated SerializedNavigationEntry to not
use content:: member variables at all.  For content::PageState, I'm storing the
encoded std::string representation instead.  For content::Referrer, I'm storing
the constituent parts (url and policy) instead.

I'm pretty confident that I did not affect the pickle format at all with these
changes.  I've annotated where I think this will result in some extra string
copies, but my changes have also removed some string copies as well.

Step 2 will move the remaining content code out of SerializedNavigationEntry
completely and into a set of helper classes.

https://codereview.chromium.org/658073004/diff/40001/components/sessions/seri...
File components/sessions/serialized_navigation_entry.cc (right):

https://codereview.chromium.org/658073004/diff/40001/components/sessions/seri...
components/sessions/serialized_navigation_entry.cc:214: // encoded_page_state_
The PageState was always written to the pickle as an encoded string, so this
should not change the disk representation at all.

https://codereview.chromium.org/658073004/diff/40001/components/sessions/seri...
components/sessions/serialized_navigation_entry.cc:221: // referrer_policy_
I've added referrer_policy_ here to be explicit, but we were always writing the
url and policy to the pickle separately, so this should not change the disk
representation at all.

https://codereview.chromium.org/658073004/diff/40001/components/sessions/seri...
components/sessions/serialized_navigation_entry.cc:511: int
SerializedNavigationEntry::GetDefaultReferrerPolicy() const {
I've tried to move the remaining content-specific logic into these three
functions here.  In a future CL, I will move these functions into a different
class.

https://codereview.chromium.org/658073004/diff/40001/components/sessions/seri...
components/sessions/serialized_navigation_entry.cc:517:
content::PageState::CreateFromEncodedData(encoded_page_state_);
There will be some places where we now do an additional copy of the encoded
pagestate string.  Here, for example.  This is slightly balanced by some places
where the new code avoids a copy that was done previously.

[blundell, 2014-10-17 15:19:27]

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

[blundell, 2014-10-17 15:19:28]

https://codereview.chromium.org/658073004/diff/40001/components/sessions/seri...
File components/sessions/serialized_navigation_entry.cc (right):

https://codereview.chromium.org/658073004/diff/40001/components/sessions/seri...
components/sessions/serialized_navigation_entry.cc:515: std::string
SerializedNavigationEntry::GetSanitizedPageStateForPickle() const {
nit: Add a TODO here like you have in the other two methods.

[marja, 2014-10-20 07:43:16]

lgtm modulo comment

https://codereview.chromium.org/658073004/diff/40001/components/sessions/seri...
File components/sessions/serialized_navigation_entry.h (right):

https://codereview.chromium.org/658073004/diff/40001/components/sessions/seri...
components/sessions/serialized_navigation_entry.h:17: #include
"content/public/common/page_state.h"
These content/ includes can now be removed, right? (Or moved to the .cc)

[rohitrao (ping after 24h), 2014-10-20 13:24:48]

rohitrao@chromium.org changed reviewers:
+ ben@chromium.org, zea@chromium.org

[rohitrao (ping after 24h), 2014-10-20 13:24:49]

rohitrao@chromium.org changed required reviewers:
+ ben@chromium.org, zea@chromium.org

[rohitrao (ping after 24h), 2014-10-20 13:24:50]

+zea for OWNERS approval of chrome/browser/sync

+ben for OWNERS approval of chrome/browser/ui/browser_commands.cc

https://codereview.chromium.org/658073004/diff/40001/components/sessions/seri...
File components/sessions/serialized_navigation_entry.cc (right):

https://codereview.chromium.org/658073004/diff/40001/components/sessions/seri...
components/sessions/serialized_navigation_entry.cc:515: std::string
SerializedNavigationEntry::GetSanitizedPageStateForPickle() const {
On 2014/10/17 15:19:28, blundell wrote:
> nit: Add a TODO here like you have in the other two methods.

Done.

https://codereview.chromium.org/658073004/diff/40001/components/sessions/seri...
File components/sessions/serialized_navigation_entry.h (right):

https://codereview.chromium.org/658073004/diff/40001/components/sessions/seri...
components/sessions/serialized_navigation_entry.h:17: #include
"content/public/common/page_state.h"
On 2014/10/20 07:43:16, marja wrote:
> These content/ includes can now be removed, right? (Or moved to the .cc)

Moved to various .cc files as needed.

[Nicolas Zea, 2014-10-20 20:14:42]

sync lgtm

[rohitrao (ping after 24h), 2014-10-20 21:17:51]

Thanks for the reviews everyone!

Switching Ben to TBR, because the c/b/ui changes are just IWYU fixes.

[rohitrao (ping after 24h), 2014-10-20 21:18:08]

The CQ bit was checked by rohitrao@chromium.org

[commit-bot: I haz the power, 2014-10-20 21:22:03]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/658073004/80001

[commit-bot: I haz the power, 2014-10-20 21:52:11]

Committed patchset #5 (id:80001)

[commit-bot: I haz the power, 2014-10-20 21:52:49]

Patchset 5 (id:??) landed as
https://crrev.com/c6d8e740dc97f57293ed336a278ff1934bb361ab
Cr-Commit-Position: refs/heads/master@{#300344}