Remove URLRequestNonsecureConnection as it's unused

Remove URLRequestNonsecureConnection as it's unused

URLRequestNonsecureConnection is never instantiated, and can't ever be,
as it lacks a call to non-trivial constructor of its base.

Committed: https://crrev.com/eb6c9e4183ad218fc304c0b822b35befb3774186
Cr-Commit-Position: refs/heads/master@{#388087}

[caseq, 2016-04-18 21:59:09]

caseq@chromium.org changed reviewers:
+ estark@chromium.org, msw@chromium.org

[caseq, 2016-04-18 21:59:10]

Please take a look!

[msw, 2016-04-18 22:09:06]

lgtm with a nit

https://codereview.chromium.org/1899003002/diff/1/chrome/browser/ui/browser_b...
File chrome/browser/ui/browser_browsertest.cc (right):

https://codereview.chromium.org/1899003002/diff/1/chrome/browser/ui/browser_b...
chrome/browser/ui/browser_browsertest.cc:109: #include
"net/ssl/ssl_connection_status_flags.h"
nit: remove

[caseq, 2016-04-18 23:08:11]

The CQ bit was checked by caseq@chromium.org

[caseq, 2016-04-18 23:08:12]

The patchset sent to the CQ was uploaded after l-g-t-m from msw@chromium.org
Link to the patchset: https://codereview.chromium.org/1899003002/#ps20001
(title: "removed unused include as well")

[commit-bot: I haz the power, 2016-04-18 23:08:39]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1899003002/20001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1899003002/20001

[commit-bot: I haz the power, 2016-04-19 00:08:08]

Committed patchset #2 (id:20001)

[commit-bot: I haz the power, 2016-04-19 00:10:30]

Description was changed from

==========
Remove URLRequestNonsecureConnection as it's unused

URLRequestNonsecureConnection is never instantiated, and can't ever be,
as it lacks a call to non-trivial constructor of its base.
==========

to

==========
Remove URLRequestNonsecureConnection as it's unused

URLRequestNonsecureConnection is never instantiated, and can't ever be,
as it lacks a call to non-trivial constructor of its base.

Committed: https://crrev.com/eb6c9e4183ad218fc304c0b822b35befb3774186
Cr-Commit-Position: refs/heads/master@{#388087}
==========

[commit-bot: I haz the power, 2016-04-19 00:10:31]

Patchset 2 (id:??) landed as
https://crrev.com/eb6c9e4183ad218fc304c0b822b35befb3774186
Cr-Commit-Position: refs/heads/master@{#388087}

[estark, 2016-04-19 07:17:35]

I think this should be reverted. There's certainly a problem here -- the test
that was using this must be buggy. But with this change the test just doesn't
make any sense (it claims to be testing a connection with a nonsecure protocol).

I'll file a bug and investigate what the problem is with the test, but I think
this should be reverted in the meantime, as we *should* be using
URLRequestNonsecureConnection in the test.

[estark, 2016-04-19 07:21:13]

https://bugs.chromium.org/p/chromium/issues/detail?id=604629

[caseq, 2016-04-19 17:49:16]

On 2016/04/19 07:17:35, estark wrote:
> I think this should be reverted. There's certainly a problem here -- the test
> that was using this must be buggy. But with this change the test just doesn't
> make any sense (it claims to be testing a connection with a nonsecure
protocol).

This change makes sense because it removes dead code. Dead code is bad because
it reduces readability and may mask problems, as it happened in this particular
case.

> I'll file a bug and investigate what the problem is with the test, but I think
> this should be reverted in the meantime, as we *should* be using
> URLRequestNonsecureConnection in the test.

You can always revert it when you actually start using
URLRequestNonsecureConnection, there's no point in doing it before.

[estark, 2016-04-19 20:18:48]

On 2016/04/19 17:49:16, caseq wrote:
> On 2016/04/19 07:17:35, estark wrote:
> > I think this should be reverted. There's certainly a problem here -- the
test
> > that was using this must be buggy. But with this change the test just
doesn't
> > make any sense (it claims to be testing a connection with a nonsecure
> protocol).
> 
> This change makes sense because it removes dead code. Dead code is bad because
> it reduces readability and may mask problems, as it happened in this
particular
> case.

That's true but I think it also reduces readability to leave behind a
nonsensical test. I guess maybe not just deleting the dead code but also
disabling the test and leaving a TODO with a bug would be best. But anyway
hopefully I'll be able to fix the underlying problem soon so it doesn't really
matter.

> 
> > I'll file a bug and investigate what the problem is with the test, but I
think
> > this should be reverted in the meantime, as we *should* be using
> > URLRequestNonsecureConnection in the test.
> 
> You can always revert it when you actually start using
> URLRequestNonsecureConnection, there's no point in doing it before.