Add the 'cache' datatype to the clear-site-data WPT

Add the 'cache' datatype to the clear-site-data WPT

- Add 'cache' to clear-site-data datatypes
- add() is implemented as fetching a resource from a support server.
- isEmpty() is implemented as fetching the same resource with
  "Cache-Control: only-if-cached" plus special handling for user agents
  that do not support this header.
- Since this is the first datatype where the above two methods are
  asynchronous, it revealed a bug in navigation.html where the add() and
  isEmpty() Promises were executed in parallel rather than in sequence;
  this has been fixed

BUG=607897

[msramek, 2017-05-29 16:21:57]

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

[commit-bot: I haz the power, 2017-05-29 16:22:12]

Dry run: CQ is trying da patch.

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

[msramek, 2017-05-29 16:45:08]

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

[commit-bot: I haz the power, 2017-05-29 16:45:22]

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, 2017-05-29 18:42:12]

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

[commit-bot: I haz the power, 2017-05-29 18:42:13]

Dry run: This issue passed the CQ dry run.

[Mike West, 2017-05-30 07:45:16]

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

[Mike West, 2017-05-30 07:45:17]

Looking good!

https://codereview.chromium.org/2913553004/diff/20001/third_party/WebKit/Layo...
File
third_party/WebKit/LayoutTests/external/wpt/clear-site-data/support/test_utils.js
(right):

https://codereview.chromium.org/2913553004/diff/20001/third_party/WebKit/Layo...
third_party/WebKit/LayoutTests/external/wpt/clear-site-data/support/test_utils.js:10:
var base_url = location.origin + "/clear-site-data/";
Can you add cross-origin tests as well, since I think that's the use case folks
internally are most interested in?

https://codereview.chromium.org/2913553004/diff/20001/third_party/WebKit/Layo...
third_party/WebKit/LayoutTests/external/wpt/clear-site-data/support/test_utils.js:62:
// The server is instructed to return a high "CacheControl: max-age"
Nit: `cache-control`

https://codereview.chromium.org/2913553004/diff/20001/third_party/WebKit/Layo...
third_party/WebKit/LayoutTests/external/wpt/clear-site-data/support/test_utils.js:82:
resolve(true);
Why would you resolve with `true` if the fetch failed?

[msramek, 2017-05-30 08:55:41]

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

[commit-bot: I haz the power, 2017-05-30 08:56:10]

Dry run: CQ is trying da patch.

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

[msramek, 2017-05-30 08:56:48]

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

[commit-bot: I haz the power, 2017-05-30 08:56:55]

Dry run: CQ is trying da patch.

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

[msramek, 2017-05-30 09:08:15]

Patchset #3 (id:40001) has been deleted

[msramek, 2017-05-30 09:10:12]

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

[commit-bot: I haz the power, 2017-05-30 09:10:30]

Dry run: CQ is trying da patch.

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

[msramek, 2017-05-30 09:11:12]

https://codereview.chromium.org/2913553004/diff/20001/third_party/WebKit/Layo...
File
third_party/WebKit/LayoutTests/external/wpt/clear-site-data/support/test_utils.js
(right):

https://codereview.chromium.org/2913553004/diff/20001/third_party/WebKit/Layo...
third_party/WebKit/LayoutTests/external/wpt/clear-site-data/support/test_utils.js:10:
var base_url = location.origin + "/clear-site-data/";
On 2017/05/30 07:45:17, Mike West wrote:
> Can you add cross-origin tests as well, since I think that's the use case
folks
> internally are most interested in?

Of course, but I'd like to do that in a separate CL. This one was intended to
finalize the list of datatypes in TestUtils (which will be retrieved by all
tests).

https://codereview.chromium.org/2913553004/diff/20001/third_party/WebKit/Layo...
third_party/WebKit/LayoutTests/external/wpt/clear-site-data/support/test_utils.js:62:
// The server is instructed to return a high "CacheControl: max-age"
On 2017/05/30 07:45:17, Mike West wrote:
> Nit: `cache-control`

Done. Also capitalized it here and elsewhere for consistency (internal and with
e.g. https://developer.mozilla.org/en-US/docs/Web/HTTP/Headers/Cache-Control).
And verified that the request class in the python server performs "in" and []
operations in a case-insensitive way.

https://codereview.chromium.org/2913553004/diff/20001/third_party/WebKit/Layo...
third_party/WebKit/LayoutTests/external/wpt/clear-site-data/support/test_utils.js:82:
resolve(true);
On 2017/05/30 07:45:17, Mike West wrote:
> Why would you resolve with `true` if the fetch failed?

If the fetch fails, it means that we're dealing with a user agent that supports
"only-if-cached" and didn't find the resource in cache, which means that
"isEmpty()" is true.

[commit-bot: I haz the power, 2017-05-30 10:50:51]

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

[commit-bot: I haz the power, 2017-05-30 10:50:51]

Dry run: This issue passed the CQ dry run.

[Mike West, 2017-05-30 11:01:31]

LGTM.

https://codereview.chromium.org/2913553004/diff/20001/third_party/WebKit/Layo...
File
third_party/WebKit/LayoutTests/external/wpt/clear-site-data/support/test_utils.js
(right):

https://codereview.chromium.org/2913553004/diff/20001/third_party/WebKit/Layo...
third_party/WebKit/LayoutTests/external/wpt/clear-site-data/support/test_utils.js:10:
var base_url = location.origin + "/clear-site-data/";
On 2017/05/30 at 09:11:12, msramek wrote:
> On 2017/05/30 07:45:17, Mike West wrote:
> > Can you add cross-origin tests as well, since I think that's the use case
folks
> > internally are most interested in?
> 
> Of course, but I'd like to do that in a separate CL. This one was intended to
finalize the list of datatypes in TestUtils (which will be retrieved by all
tests).

Sure. Happy to see more CLs. :)

https://codereview.chromium.org/2913553004/diff/20001/third_party/WebKit/Layo...
third_party/WebKit/LayoutTests/external/wpt/clear-site-data/support/test_utils.js:82:
resolve(true);
On 2017/05/30 at 09:11:12, msramek wrote:
> On 2017/05/30 07:45:17, Mike West wrote:
> > Why would you resolve with `true` if the fetch failed?
> 
> If the fetch fails, it means that we're dealing with a user agent that
supports "only-if-cached" and didn't find the resource in cache, which means
that "isEmpty()" is true.

Ah. I see. A comment to that effect might be helpful, as I'd missed it. (Also,
does `only-if-cached` work cross-origin? I don't think it does...)

Have you run the test in Firefox to confirm that it fails?

[msramek, 2017-05-30 12:30:24]

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

[commit-bot: I haz the power, 2017-05-30 12:30:30]

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, 2017-05-30 13:53:06]

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

[commit-bot: I haz the power, 2017-05-30 13:53:07]

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_...)

[msramek, 2017-06-06 23:38:24]

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

[commit-bot: I haz the power, 2017-06-06 23:38:41]

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, 2017-06-07 01:01:17]

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

[commit-bot: I haz the power, 2017-06-07 01:01:18]

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_...)

[msramek, 2017-07-06 16:21:03]

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

[commit-bot: I haz the power, 2017-07-06 16:21:17]

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, 2017-07-06 17:53:30]

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

[commit-bot: I haz the power, 2017-07-06 17:53:31]

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_...)

[msramek, 2017-07-06 19:53:31]

I'm closing this CL, as cache turns out to be currently untestable.

The rest is extracted as https://codereview.chromium.org/2975463002/.

[msramek, 2017-07-11 13:06:40]

Description was changed from

==========
Add the 'cache' datatype to the clear-site-data WPT

- Add 'cache' to clear-site-data datatypes
- add() is implemented as fetching a resource from a support server.
- isEmpty() is implemented as fetching the same resource with
  "Cache-Control: only-if-cached" plus special handling for user agents
  that do not support this header.
- Since this is the first datatype where the above two methods are
  asynchronous, it revealed a bug in navigation.html where the add() and
  isEmpty() Promises were executed in parallel rather than in sequence;
  this has been fixed

BUG=607897
==========

to

==========
Add the 'cache' datatype to the clear-site-data WPT

- Add 'cache' to clear-site-data datatypes
- add() is implemented as fetching a resource from a support server.
- isEmpty() is implemented as fetching the same resource with
  "Cache-Control: only-if-cached" plus special handling for user agents
  that do not support this header.
- Since this is the first datatype where the above two methods are
  asynchronous, it revealed a bug in navigation.html where the add() and
  isEmpty() Promises were executed in parallel rather than in sequence;
  this has been fixed

BUG=607897
==========