tools/android/loading: Add support for multiple redirects.

tools/android/loading: Add support for multiple redirects.

This properly identify redirect chains. Before this CL, each link in the
redirect chain would seem to be initiated by the eventual request,
instead of the previous redirect.

Also logs the number of cases where the redirected request doesn't have
the same initiator as the initial request.

Committed: https://crrev.com/57c2ecb994e01eeb2fc46ecd9ebee8ae9cca8d13
Cr-Commit-Position: refs/heads/master@{#371522}

[Benoit L, 2016-01-26 13:06:54]

lizeb@chromium.org changed reviewers:
+ blundell@chromium.org, mattcary@chromium.org

[Benoit L, 2016-01-26 13:06:54]

Hello,

Here is the handling of redirect chains. I don't understand why the initiator
can change, so for now it is logged, and accounted for.

Please take a look, thanks.

[mattcary, 2016-01-26 13:20:26]

lgtm

https://codereview.chromium.org/1633813005/diff/20001/tools/android/loading/r...
File tools/android/loading/request_track.py (right):

https://codereview.chromium.org/1633813005/diff/20001/tools/android/loading/r...
tools/android/loading/request_track.py:238: cls._INCONSISTENT_INITIATORS_KEY]
Call get() instead so we can load previous versions of a trace? 

Doesn't matter too much now but I suspect we're going to continue to extend the
metadata so might be nice to establish a convention.

[blundell, 2016-01-26 13:23:44]

lgtm

https://codereview.chromium.org/1633813005/diff/20001/tools/android/loading/l...
File tools/android/loading/loading_model.py (right):

https://codereview.chromium.org/1633813005/diff/20001/tools/android/loading/l...
tools/android/loading/loading_model.py:249: if 'parser' in annotations:
should this be an elif?

https://codereview.chromium.org/1633813005/diff/20001/tools/android/loading/r...
File tools/android/loading/request_dependencies_lens.py (right):

https://codereview.chromium.org/1633813005/diff/20001/tools/android/loading/r...
tools/android/loading/request_dependencies_lens.py:67: # Redirect suffixes are
added in RequestTrack.
Is this comment still relevant?

[Benoit L, 2016-01-26 13:40:47]

All done (+rebase), thanks!

https://codereview.chromium.org/1633813005/diff/20001/tools/android/loading/l...
File tools/android/loading/loading_model.py (right):

https://codereview.chromium.org/1633813005/diff/20001/tools/android/loading/l...
tools/android/loading/loading_model.py:249: if 'parser' in annotations:
On 2016/01/26 13:23:44, blundell wrote:
> should this be an elif?

Indeed, thanks!
Done.

https://codereview.chromium.org/1633813005/diff/20001/tools/android/loading/r...
File tools/android/loading/request_dependencies_lens.py (right):

https://codereview.chromium.org/1633813005/diff/20001/tools/android/loading/r...
tools/android/loading/request_dependencies_lens.py:67: # Redirect suffixes are
added in RequestTrack.
On 2016/01/26 13:23:44, blundell wrote:
> Is this comment still relevant?

Done.

https://codereview.chromium.org/1633813005/diff/20001/tools/android/loading/r...
File tools/android/loading/request_track.py (right):

https://codereview.chromium.org/1633813005/diff/20001/tools/android/loading/r...
tools/android/loading/request_track.py:238: cls._INCONSISTENT_INITIATORS_KEY]
On 2016/01/26 13:20:26, mattcary wrote:
> Call get() instead so we can load previous versions of a trace? 
> 
> Doesn't matter too much now but I suspect we're going to continue to extend
the
> metadata so might be nice to establish a convention.

Done.

[Benoit L, 2016-01-26 16:05:49]

The CQ bit was checked by lizeb@chromium.org

[Benoit L, 2016-01-26 16:05:49]

The patchset sent to the CQ was uploaded after l-g-t-m from
mattcary@chromium.org, blundell@chromium.org
Link to the patchset: https://codereview.chromium.org/1633813005/#ps60001
(title: "Rebase.")

[commit-bot: I haz the power, 2016-01-26 16:06:14]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1633813005/60001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1633813005/60001

[commit-bot: I haz the power, 2016-01-26 16:24:57]

Committed patchset #4 (id:60001)

[commit-bot: I haz the power, 2016-01-26 16:26:16]

Description was changed from

==========
tools/android/loading: Add support for multiple redirects.

This properly identify redirect chains. Before this CL, each link in the
redirect chain would seem to be initiated by the eventual request,
instead of the previous redirect.

Also logs the number of cases where the redirected request doesn't have
the same initiator as the initial request.
==========

to

==========
tools/android/loading: Add support for multiple redirects.

This properly identify redirect chains. Before this CL, each link in the
redirect chain would seem to be initiated by the eventual request,
instead of the previous redirect.

Also logs the number of cases where the redirected request doesn't have
the same initiator as the initial request.

Committed: https://crrev.com/57c2ecb994e01eeb2fc46ecd9ebee8ae9cca8d13
Cr-Commit-Position: refs/heads/master@{#371522}
==========

[commit-bot: I haz the power, 2016-01-26 16:26:17]

Patchset 4 (id:??) landed as
https://crrev.com/57c2ecb994e01eeb2fc46ecd9ebee8ae9cca8d13
Cr-Commit-Position: refs/heads/master@{#371522}