Only use new login flow for GmailSmoothPage

Only use new login flow for GmailSmoothPage

Updating the https://mail.google.com/mail/ recording for only top_25_smooth.json
required changing the login mechanism. This unfortunately caused tests using
the old login flow to begin failing. This is a short term patch to only use the
new flow for top_25_smooth.json until a more general solution is developed.

BUG=590735, 590766
Committed: https://crrev.com/184919658089ea1a9e1f66a4799abc2905717e6a
Cr-Commit-Position: refs/heads/master@{#378245}

[flackr, 2016-02-29 16:51:04]

flackr@chromium.org changed reviewers:
+ nednguyen@google.com, rbyers@chromium.org

[flackr, 2016-02-29 16:51:05]

This is really messy :(, but what do you think of this workaround? I've filed a
bug to track coming up with a better solution but short of updating 14 other
recordings, we need some way to use the new login flow for the updated recording
on top_25_smooth.json while still using the old flow on the existing recordings
for other page sets.

[Rick Byers, 2016-02-29 16:58:14]

Thanks Rob.  Some short-term hack like this seems like the right next step to
me.  Lots of bots are failing right now (I should have noticed Friday but there
were serious master issues preventing investigation), so fixing or reverting is
urgent.  You've also poured so much time into getting this _one_ gmail test
updated, that I definitely wouldn't want you to get stuck with trying to update
14 others.

I'm not an expert in the specific code changes here, so I'll leave the detailed
review to Ned.

[nednguyen, 2016-02-29 17:01:26]

On 2016/02/29 16:58:14, Rick Byers wrote:
> Thanks Rob.  Some short-term hack like this seems like the right next step to
> me.  Lots of bots are failing right now (I should have noticed Friday but
there
> were serious master issues preventing investigation), so fixing or reverting
is
> urgent.  You've also poured so much time into getting this _one_ gmail test
> updated, that I definitely wouldn't want you to get stuck with trying to
update
> 14 others.
> 
> I'm not an expert in the specific code changes here, so I'll leave the
detailed
> review to Ned.

lgtm

Thanks Rob. The core bug here is the mapping of recording archive are done
inside page_set instead of page classes :-(

[flackr, 2016-02-29 17:02:58]

Patchset #2 (id:20001) has been deleted

[flackr, 2016-02-29 17:48:24]

The CQ bit was checked by flackr@chromium.org

[commit-bot: I haz the power, 2016-02-29 17:48:32]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1742363002/1
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1742363002/1

[commit-bot: I haz the power, 2016-02-29 19:17:58]

Committed patchset #1 (id:1)

[commit-bot: I haz the power, 2016-02-29 19:20:07]

Description was changed from

==========
Only use new login flow for GmailSmoothPage

Updating the https://mail.google.com/mail/ recording for only top_25_smooth.json
required changing the login mechanism. This unfortunately caused tests using
the old login flow to begin failing. This is a short term patch to only use the
new flow for top_25_smooth.json until a more general solution is developed.

BUG=590735,590766
==========

to

==========
Only use new login flow for GmailSmoothPage

Updating the https://mail.google.com/mail/ recording for only top_25_smooth.json
required changing the login mechanism. This unfortunately caused tests using
the old login flow to begin failing. This is a short term patch to only use the
new flow for top_25_smooth.json until a more general solution is developed.

BUG=590735,590766

Committed: https://crrev.com/184919658089ea1a9e1f66a4799abc2905717e6a
Cr-Commit-Position: refs/heads/master@{#378245}
==========

[commit-bot: I haz the power, 2016-02-29 19:20:08]

Patchset 1 (id:??) landed as
https://crrev.com/184919658089ea1a9e1f66a4799abc2905717e6a
Cr-Commit-Position: refs/heads/master@{#378245}