Clean up URLFetcher unit tests, part 7.

Clean up URLFetcher unit tests, part 7.

Convert the BadHTTPS and URLFetcher reuse tests to use the shared test
fixture.  As a bonus, make sure the reuse test actually updates the
response body between requests.

BUG=471069
Committed: https://crrev.com/b948eae2b1c2733fefdcd064823656f26e1573d2
Cr-Commit-Position: refs/heads/master@{#325531}

[mmenke, 2015-04-16 19:11:39]

mmenke@chromium.org changed reviewers:
+ davidben@chromium.org

[mmenke, 2015-04-16 19:11:40]

At least this one's pretty trivial.

https://codereview.chromium.org/1091073003/diff/1/net/url_request/url_fetcher...
File net/url_request/url_fetcher_impl_unittest.cc (left):

https://codereview.chromium.org/1091073003/diff/1/net/url_request/url_fetcher...
net/url_request/url_fetcher_impl_unittest.cc:596: cert_dir_ =
cert_dir_.AppendASCII("certificates");
Didn't notice this wasn't even being used in my earlier CL that touched this.

[mmenke, 2015-04-16 19:33:28]

https://codereview.chromium.org/1091073003/diff/20001/net/url_request/url_fet...
File net/url_request/url_fetcher_impl_unittest.cc (right):

https://codereview.chromium.org/1091073003/diff/20001/net/url_request/url_fet...
net/url_request/url_fetcher_impl_unittest.cc:495: class URLFetcherBadHTTPSTest :
public URLFetcherTest {
Moved this test fixture (Which we have to keep) and its test to the bottom, so
order is consistent.  We'll be deleting the other fixtures, anyways, of course.

[davidben, 2015-04-16 21:48:39]

lgtm

https://codereview.chromium.org/1091073003/diff/20001/net/url_request/url_fet...
File net/url_request/url_fetcher_impl_unittest.cc (right):

https://codereview.chromium.org/1091073003/diff/20001/net/url_request/url_fet...
net/url_request/url_fetcher_impl_unittest.cc:1125: // request. Can't reuse the
old one because fetchers release it on completion.
Yuck. :-P I don't have a much better suggestion though.

https://codereview.chromium.org/1091073003/diff/20001/net/url_request/url_fet...
net/url_request/url_fetcher_impl_unittest.cc:1161: // cannot be. Can we get rid
of support for this?
Especially weird that you have to specify a new getter. What is this API...

The tests were added in http://codereview.chromium.org/5322015 which doesn't say
much about consumers who do this.

[mmenke, 2015-04-16 21:56:18]

The CQ bit was checked by mmenke@chromium.org

[mmenke, 2015-04-16 21:56:29]

Thanks!

https://codereview.chromium.org/1091073003/diff/20001/net/url_request/url_fet...
File net/url_request/url_fetcher_impl_unittest.cc (right):

https://codereview.chromium.org/1091073003/diff/20001/net/url_request/url_fet...
net/url_request/url_fetcher_impl_unittest.cc:1161: // cannot be. Can we get rid
of support for this?
On 2015/04/16 21:48:39, David Benjamin wrote:
> Especially weird that you have to specify a new getter. What is this API...

I think the intent is to prevent the fetcher from extending the life of the
getter.  Some getters own their request contexts, so that may have made sense,
if we didn't allow reuse.

> The tests were added in http://codereview.chromium.org/5322015 which doesn't
say
> much about consumers who do this.

[commit-bot: I haz the power, 2015-04-16 21:57:38]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1091073003/20001

[commit-bot: I haz the power, 2015-04-16 22:48:23]

Committed patchset #2 (id:20001)

[commit-bot: I haz the power, 2015-04-16 22:49:23]

Patchset 2 (id:??) landed as
https://crrev.com/b948eae2b1c2733fefdcd064823656f26e1573d2
Cr-Commit-Position: refs/heads/master@{#325531}