Sync Cache API tests w/ web-platform-tests

Sync Cache API tests w/ web-platform-tests

Pull in some of the upstream changes made to the Service Worker Cache API tests.

BUG=476814, 482241, 429832, 564245
Committed: https://crrev.com/641689267fa6fea587b9fa5485ac42dc259fb14d
Cr-Commit-Position: refs/heads/master@{#363010}

[jsbell, 2015-12-01 20:50:49]

jsbell@chromium.org changed reviewers:
+ jkarlin@chromium.org, yhirano@chromium.org

[jsbell, 2015-12-01 20:50:50]

jkarlin@, yhirano@ - Please take a look?

This pulls in some of the changes to the cache storage tests that were made
upstream. There's more work to do here - we have tests on requests with
credentials that were removed upstream but we want to maintain - they should be
pulled out to be SW-specific tests.

https://codereview.chromium.org/1485283002/diff/1/third_party/WebKit/LayoutTe...
File
third_party/WebKit/LayoutTests/http/tests/cachestorage/serviceworker/cache-add-expected.txt
(right):

https://codereview.chromium.org/1485283002/diff/1/third_party/WebKit/LayoutTe...
third_party/WebKit/LayoutTests/http/tests/cachestorage/serviceworker/cache-add-expected.txt:18:
FAIL Cache.addAll called with the same Request object specified twice
promise_test: Unhandled rejection with value: "assert_promise_rejects:
Cache.addAll should throw InvalidStateError if the same request is added twice.
Promise did not reject."
Test assumes that the body will be consumed. This looks wrong upstream if I'm
reading the fetch spec correctly, since request's empty body won't be disturbed.

https://codereview.chromium.org/1485283002/diff/1/third_party/WebKit/LayoutTe...
File
third_party/WebKit/LayoutTests/http/tests/cachestorage/serviceworker/cache-put-expected.txt
(right):

https://codereview.chromium.org/1485283002/diff/1/third_party/WebKit/LayoutTe...
third_party/WebKit/LayoutTests/http/tests/cachestorage/serviceworker/cache-put-expected.txt:22:
FAIL Cache.put with a VARY:* Response promise_test: Unhandled rejection with
value: "assert_promise_rejects: Cache.put should reject VARY:* Responses with a
TypeError. Promise did not reject."
https://crbug.com/476814

https://codereview.chromium.org/1485283002/diff/1/third_party/WebKit/LayoutTe...
File
third_party/WebKit/LayoutTests/http/tests/cachestorage/serviceworker/cache-storage-expected.txt
(right):

https://codereview.chromium.org/1485283002/diff/1/third_party/WebKit/LayoutTe...
third_party/WebKit/LayoutTests/http/tests/cachestorage/serviceworker/cache-storage-expected.txt:8:
FAIL CacheStorage.open with existing cache assert_not_equals: CacheStorage.open
should return a new Cache object each time its called. got disallowed value
object "[object Cache]"
https://crbug.com/429832

https://codereview.chromium.org/1485283002/diff/1/third_party/WebKit/LayoutTe...
File
third_party/WebKit/LayoutTests/http/tests/cachestorage/serviceworker/cache-storage-match-expected.txt
(right):

https://codereview.chromium.org/1485283002/diff/1/third_party/WebKit/LayoutTe...
third_party/WebKit/LayoutTests/http/tests/cachestorage/serviceworker/cache-storage-match-expected.txt:8:
FAIL CacheStorageMatch with no caches available but name provided assert_equals:
The match should reject with NotFoundError. expected "NotFoundError" but got
"Error"
https://crbug.com/482241

https://codereview.chromium.org/1485283002/diff/1/third_party/WebKit/LayoutTe...
File
third_party/WebKit/LayoutTests/http/tests/cachestorage/window/sandboxed-iframes-expected.txt
(right):

https://codereview.chromium.org/1485283002/diff/1/third_party/WebKit/LayoutTe...
third_party/WebKit/LayoutTests/http/tests/cachestorage/window/sandboxed-iframes-expected.txt:4:
FAIL Sandboxed iframe without allow-same-origin is denied access assert_equals:
Access should be denied if sandbox lacks allow-same-origin expected "denied" but
got "unexpecteddenied"
https://crbug.com/564245

[jsbell, 2015-12-01 20:53:35]

Description was changed from

==========
Sync Cache API tests w/ web-platform-tests

BUG=476814,482241,429832
==========

to

==========
Sync Cache API tests w/ web-platform-tests

Pull in some of the upstream changes made to the Service Worker Cache API tests.

BUG=476814,482241,429832,564245
==========

[yhirano, 2015-12-02 06:58:53]

https://codereview.chromium.org/1485283002/diff/1/third_party/WebKit/LayoutTe...
File
third_party/WebKit/LayoutTests/http/tests/cachestorage/serviceworker/cache-add-expected.txt
(right):

https://codereview.chromium.org/1485283002/diff/1/third_party/WebKit/LayoutTe...
third_party/WebKit/LayoutTests/http/tests/cachestorage/serviceworker/cache-add-expected.txt:18:
FAIL Cache.addAll called with the same Request object specified twice
promise_test: Unhandled rejection with value: "assert_promise_rejects:
Cache.addAll should throw InvalidStateError if the same request is added twice.
Promise did not reject."
On 2015/12/01 20:50:49, jsbell wrote:
> Test assumes that the body will be consumed. This looks wrong upstream if I'm
> reading the fetch spec correctly, since request's empty body won't be
disturbed.

A request with a null body should not be disturbed, but a request with an empty
body should be disturbed.

[jkarlin, 2015-12-02 16:42:45]

lgtm

https://codereview.chromium.org/1485283002/diff/1/third_party/WebKit/LayoutTe...
File
third_party/WebKit/LayoutTests/http/tests/cachestorage/serviceworker/cache-add-expected.txt
(right):

https://codereview.chromium.org/1485283002/diff/1/third_party/WebKit/LayoutTe...
third_party/WebKit/LayoutTests/http/tests/cachestorage/serviceworker/cache-add-expected.txt:18:
FAIL Cache.addAll called with the same Request object specified twice
promise_test: Unhandled rejection with value: "assert_promise_rejects:
Cache.addAll should throw InvalidStateError if the same request is added twice.
Promise did not reject."
On 2015/12/01 20:50:49, jsbell wrote:
> Test assumes that the body will be consumed. This looks wrong upstream if I'm
> reading the fetch spec correctly, since request's empty body won't be
disturbed.

Agreed. So should we incorporate the change and file a bug upstream?

[jsbell, 2015-12-02 17:17:46]

https://codereview.chromium.org/1485283002/diff/1/third_party/WebKit/LayoutTe...
File
third_party/WebKit/LayoutTests/http/tests/cachestorage/serviceworker/cache-add-expected.txt
(right):

https://codereview.chromium.org/1485283002/diff/1/third_party/WebKit/LayoutTe...
third_party/WebKit/LayoutTests/http/tests/cachestorage/serviceworker/cache-add-expected.txt:18:
FAIL Cache.addAll called with the same Request object specified twice
promise_test: Unhandled rejection with value: "assert_promise_rejects:
Cache.addAll should throw InvalidStateError if the same request is added twice.
Promise did not reject."
On 2015/12/02 16:42:45, jkarlin wrote:
> On 2015/12/01 20:50:49, jsbell wrote:
> > Test assumes that the body will be consumed. This looks wrong upstream if
I'm
> > reading the fetch spec correctly, since request's empty body won't be
> disturbed.
> 
> Agreed. So should we incorporate the change and file a bug upstream? 

+1, but I want to hear if yhirano agrees.

[yhirano, 2015-12-03 01:58:18]

lgtm

https://codereview.chromium.org/1485283002/diff/1/third_party/WebKit/LayoutTe...
File
third_party/WebKit/LayoutTests/http/tests/cachestorage/serviceworker/cache-add-expected.txt
(right):

https://codereview.chromium.org/1485283002/diff/1/third_party/WebKit/LayoutTe...
third_party/WebKit/LayoutTests/http/tests/cachestorage/serviceworker/cache-add-expected.txt:18:
FAIL Cache.addAll called with the same Request object specified twice
promise_test: Unhandled rejection with value: "assert_promise_rejects:
Cache.addAll should throw InvalidStateError if the same request is added twice.
Promise did not reject."
On 2015/12/02 17:17:46, jsbell wrote:
> On 2015/12/02 16:42:45, jkarlin wrote:
> > On 2015/12/01 20:50:49, jsbell wrote:
> > > Test assumes that the body will be consumed. This looks wrong upstream if
> I'm
> > > reading the fetch spec correctly, since request's empty body won't be
> > disturbed.
> > 
> > Agreed. So should we incorporate the change and file a bug upstream? 
> 
> +1, but I want to hear if yhirano agrees.

Oh, sorry, I misunderstood.
You're right, request's body is always null (not empty) because its method is
'GET', and hence the reuqest cannot be locked or disturbed.

[jsbell, 2015-12-03 17:02:24]

The CQ bit was checked by jsbell@chromium.org

[commit-bot: I haz the power, 2015-12-03 17:03:38]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1485283002/1
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1485283002/1

[commit-bot: I haz the power, 2015-12-03 18:28:45]

Description was changed from

==========
Sync Cache API tests w/ web-platform-tests

Pull in some of the upstream changes made to the Service Worker Cache API tests.

BUG=476814,482241,429832,564245
==========

to

==========
Sync Cache API tests w/ web-platform-tests

Pull in some of the upstream changes made to the Service Worker Cache API tests.

BUG=476814,482241,429832,564245
==========

[commit-bot: I haz the power, 2015-12-03 18:28:46]

Committed patchset #1 (id:1)

[commit-bot: I haz the power, 2015-12-03 18:29:38]

Description was changed from

==========
Sync Cache API tests w/ web-platform-tests

Pull in some of the upstream changes made to the Service Worker Cache API tests.

BUG=476814,482241,429832,564245
==========

to

==========
Sync Cache API tests w/ web-platform-tests

Pull in some of the upstream changes made to the Service Worker Cache API tests.

BUG=476814,482241,429832,564245

Committed: https://crrev.com/641689267fa6fea587b9fa5485ac42dc259fb14d
Cr-Commit-Position: refs/heads/master@{#363010}
==========

[commit-bot: I haz the power, 2015-12-03 18:29:38]

Patchset 1 (id:??) landed as
https://crrev.com/641689267fa6fea587b9fa5485ac42dc259fb14d
Cr-Commit-Position: refs/heads/master@{#363010}