sandwich: Fix some trivial failures on SWR benchmark runs

sandwich: Fix some trivial failures on SWR benchmark runs

BUG=582080
Committed: https://crrev.com/81bf8788e08a8bdb1d0b68203ae2acb1d3d4fd3d
Cr-Commit-Position: refs/heads/master@{#404645}

[gabadie, 2016-07-11 10:53:48]

gabadie@chromium.org changed reviewers:
+ pasko@chromium.org

[gabadie, 2016-07-11 10:53:49]

Hey Egor,

trivial CL to fixes some failures on the SWR run. PTAL.

[pasko, 2016-07-11 11:26:50]

hm, looks suspicious, but I do not have a complete solution for it, see below

https://codereview.chromium.org/2138713002/diff/1/tools/android/loading/reque...
File tools/android/loading/request_track.py (right):

https://codereview.chromium.org/2138713002/diff/1/tools/android/loading/reque...
tools/android/loading/request_track.py:344: if len(parts) != 2:
was it an empty directive or more than one '='? Is it because of typos on the
web devs side or something legitimate?

https://codereview.chromium.org/2138713002/diff/1/tools/android/loading/reque...
tools/android/loading/request_track.py:393: headers += '{}: {}\x00'.format(
response headers do not have to be valid unicode, trying to decode/encode it,
then setting headers with errors ignored .. look wrong.

can we 'just' concatenate two byte sequences here that read before?

[gabadie, 2016-07-11 12:14:35]

https://codereview.chromium.org/2138713002/diff/1/tools/android/loading/reque...
File tools/android/loading/request_track.py (right):

https://codereview.chromium.org/2138713002/diff/1/tools/android/loading/reque...
tools/android/loading/request_track.py:344: if len(parts) != 2:
On 2016/07/11 11:26:50, pasko wrote:
> was it an empty directive or more than one '='? Is it because of typos on the
> web devs side or something legitimate?

I had a Cache-Control: max-age=stuff1 max-age=stuff2 with the missing `,` in
between, making len(parts) == 3

https://codereview.chromium.org/2138713002/diff/1/tools/android/loading/reque...
tools/android/loading/request_track.py:393: headers += '{}: {}\x00'.format(
On 2016/07/11 11:26:50, pasko wrote:
> response headers do not have to be valid unicode, trying to decode/encode it,
> then setting headers with errors ignored .. look wrong.
> 
> can we 'just' concatenate two byte sequences here that read before?

Good point. Done.

[pasko, 2016-07-11 12:27:09]

lgtm, thank you!

[gabadie, 2016-07-11 12:28:29]

The CQ bit was checked by gabadie@chromium.org

[commit-bot: I haz the power, 2016-07-11 12:28:36]

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, 2016-07-11 13:06:35]

Committed patchset #2 (id:20001)

[commit-bot: I haz the power, 2016-07-11 13:08:12]

Description was changed from

==========
sandwich: Fix some trivial failures on SWR benchmark runs

BUG=582080
==========

to

==========
sandwich: Fix some trivial failures on SWR benchmark runs

BUG=582080

Committed: https://crrev.com/81bf8788e08a8bdb1d0b68203ae2acb1d3d4fd3d
Cr-Commit-Position: refs/heads/master@{#404645}
==========

[commit-bot: I haz the power, 2016-07-11 13:08:12]

Patchset 2 (id:??) landed as
https://crrev.com/81bf8788e08a8bdb1d0b68203ae2acb1d3d4fd3d
Cr-Commit-Position: refs/heads/master@{#404645}