Fix crash in client_socket_pool_base.cc.

Fix crash in client_socket_pool_base.cc.

The CHECK that was being hit showed a real problem: group's IsEmpty
should take into account the pending requests queue too.

I also made IsEmpty check for connecting requests queue emptiness,
so we can be sure that IsEmpty really means empty.

TEST=Added a regression test ClientSocketPoolBaseTest.GroupWithPendingRequestsIsNotEmpty to net_unittests.
http://crbug.com/17985

Committed: http://src.chromium.org/viewvc/chrome?view=rev&revision=21988

[Paweł Hajdan Jr., 2009-07-29 18:25:43]

Please review. It fixes a P1 bug.

[willchan no longer on Chromium, 2009-07-29 18:41:56]

lgtm

http://codereview.chromium.org/159597/diff/1/4
File net/socket/client_socket_pool_base_unittest.cc (right):

http://codereview.chromium.org/159597/diff/1/4#newcode978
Line 978: TEST_F(ClientSocketPoolBaseTest, GroupWithPendingRequestsIsNotEmpty) {
Please add a copy of this test for ClientSocketPoolBaseTest_LateBinding to make
sure it works with that.  I'll be deleting all these non-late binding tests
after late binding goes to 100%.

[Paweł Hajdan Jr., 2009-07-29 19:05:12]

Done.

eroman/wtc, could you take a look as well? Just to make sure the fix doesn't
introduce some other problems.

[eroman, 2009-07-29 20:12:53]

lgtm

[wtc, 2009-07-29 20:27:25]

LGTM.  Thanks!

http://codereview.chromium.org/159597/diff/1002/12
File net/socket/client_socket_pool_base_unittest.cc (right):

http://codereview.chromium.org/159597/diff/1002/12#newcode999
Line 999: // Release the first two sockets from "a", which will make place
Nit: "make place" => "make room"?  Or did you mean "make space"?

http://codereview.chromium.org/159597/diff/1002/12#newcode1437
Line 1437: // Release the first two sockets from "a", which will make place
Nit: "make place" => "make room"?