Remove ref-counting from StreamListenSocket

Remove ref-counting from StreamListenSocket

Changes APIs for StreamListenSocket, TCPListenSocket, UnixDomainSocket and derived test classes to use scoped_ptr instead of scoped_refptr.

BUG=263963
Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=220992

[szym, 2013-08-12 15:30:02]

ping

[joth, 2013-08-17 03:38:23]

On 2013/08/12 15:30:02, szym wrote:
> ping

android_webview/ lgtm rubberstamp

[pfeldman, 2013-08-30 09:18:25]

devtools lgtm

[szym, 2013-08-30 20:07:39]

Moving to akalin for net/ review.

[akalin, 2013-08-30 20:53:01]

ah, my favorite type of CL. net lgtm!

https://codereview.chromium.org/20142003/diff/51001/net/socket/tcp_listen_soc...
File net/socket/tcp_listen_socket.h (right):

https://codereview.chromium.org/20142003/diff/51001/net/socket/tcp_listen_soc...
net/socket/tcp_listen_socket.h:16: // Implements a TCP socket. Note that this is
ref counted.
update comment

https://codereview.chromium.org/20142003/diff/51001/net/socket/transport_clie...
File net/socket/transport_client_socket_unittest.cc (right):

https://codereview.chromium.org/20142003/diff/51001/net/socket/transport_clie...
net/socket/transport_client_socket_unittest.cc:53:
reinterpret_cast<TCPListenSocket*>(connection.release()));
I think you can just do connect_sock_ = connection.Pass(), right? Or at least
replace the (erroneous) reinterpret_cast with static_cast

[akalin, 2013-08-30 20:53:21]

Also mention in CL description that you're changing some APIs to work with
scoped_ptrs instead of raw pointers.

[szym, 2013-08-30 22:08:00]

tommi@chromium.org: Please review changes in:
  chrome_frame/*

jam@chromium.org: Please approve changes in:
  chrome/browser/android/*
  content/public/*

https://codereview.chromium.org/20142003/diff/51001/net/socket/tcp_listen_soc...
File net/socket/tcp_listen_socket.h (right):

https://codereview.chromium.org/20142003/diff/51001/net/socket/tcp_listen_soc...
net/socket/tcp_listen_socket.h:16: // Implements a TCP socket. Note that this is
ref counted.
On 2013/08/30 20:53:02, akalin wrote:
> update comment

Done.

https://codereview.chromium.org/20142003/diff/51001/net/socket/transport_clie...
File net/socket/transport_client_socket_unittest.cc (right):

https://codereview.chromium.org/20142003/diff/51001/net/socket/transport_clie...
net/socket/transport_client_socket_unittest.cc:53:
reinterpret_cast<TCPListenSocket*>(connection.release()));
On 2013/08/30 20:53:02, akalin wrote:
> I think you can just do connect_sock_ = connection.Pass(), right? Or at least
> replace the (erroneous) reinterpret_cast with static_cast

StreamListenSocket is the base class of TCPListenSocket so it needs a cast.
Replaced with static_cast.

[tommi (sloooow) - chröme, 2013-08-31 06:38:38]

lgtm for chrome_frame/*

[jam, 2013-09-03 16:38:43]

lgtm

[commit-bot: I haz the power, 2013-09-03 16:40:07]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/szym@chromium.org/20142003/71001

[commit-bot: I haz the power, 2013-09-03 17:15:47]

Retried try job too often on linux_chromeos for step(s) chromeos_unittests,
components_unittests, crypto_unittests
http://build.chromium.org/p/tryserver.chromium/buildstatus?builder=linux_chro...

[commit-bot: I haz the power, 2013-09-03 17:22:50]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/szym@chromium.org/20142003/71001

[commit-bot: I haz the power, 2013-09-03 19:03:09]

Change committed as 220992