Adding an end to end test for QUIC

Adding an end to end test for QUIC, and fixing the bugs it turned up.

Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=191529

[Ryan Hamilton, 2013-03-29 14:22:40]

Other than the gyp file, this looks good.  I assume you are planning to push
some of these changes back to the internal version?  (Is that the CL you send to
Raman yesterday?)

https://codereview.chromium.org/13004022/diff/1/net/net.gyp
File net/net.gyp (right):

https://codereview.chromium.org/13004022/diff/1/net/net.gyp#newcode2544
net/net.gyp:2544: 'tools/quic/quic_server_session.cc',
It looks like some of these files are in both quic_unittests and quic_library. 
I think they should only be in one.

https://codereview.chromium.org/13004022/diff/1/net/tools/quic/quic_in_memory...
File net/tools/quic/quic_in_memory_cache.cc (right):

https://codereview.chromium.org/13004022/diff/1/net/tools/quic/quic_in_memory...
net/tools/quic/quic_in_memory_cache.cc:90: if
(responses_.find(GetKey(request_headers)) != responses_.end()) {
FWIW, we also have ContainsKey, which I think is in internal code as well.

[alyssar, 2013-03-29 15:31:35]

With the exception of the server/port work (which I'd be happy to back-port)
most of these were fixing porting errors or adding accessors which exist in
google3 but not chrome..

The cl I sent to Raman was cleanups from the original check-in.  Things like
fixing QuicClock (which isn't fixed in chrome yet - he'll probably do that
merge) and having small virtual functions in the .cc file.

https://codereview.chromium.org/13004022/diff/1/net/net.gyp
File net/net.gyp (right):

https://codereview.chromium.org/13004022/diff/1/net/net.gyp#newcode2544
net/net.gyp:2544: 'tools/quic/quic_server_session.cc',
On 2013/03/29 14:22:40, Ryan Hamilton wrote:
> It looks like some of these files are in both quic_unittests and quic_library.

> I think they should only be in one.

Which ones?  There's definitely overlap with quic_server and quic_client but
when I tried including them as deps it didn't work (maybe because they're
executables, maybe I did something wrong).  I don't see anything in quic_library
here.

I can create server_lib and client_lib if you'd prefer -- I thought you OKd
doing a copy/paste of src files was OK if the dependencies route didn't work.

https://codereview.chromium.org/13004022/diff/1/net/tools/quic/quic_in_memory...
File net/tools/quic/quic_in_memory_cache.cc (right):

https://codereview.chromium.org/13004022/diff/1/net/tools/quic/quic_in_memory...
net/tools/quic/quic_in_memory_cache.cc:90: if
(responses_.find(GetKey(request_headers)) != responses_.end()) {
On 2013/03/29 14:22:40, Ryan Hamilton wrote:
> FWIW, we also have ContainsKey, which I think is in internal code as well. 

in stl_util.h which I already had included even.  Thanks, fixed :-)

[Ryan Hamilton, 2013-03-29 15:50:55]

lgtm

https://codereview.chromium.org/13004022/diff/1/net/net.gyp
File net/net.gyp (right):

https://codereview.chromium.org/13004022/diff/1/net/net.gyp#newcode2544
net/net.gyp:2544: 'tools/quic/quic_server_session.cc',
On 2013/03/29 15:31:35, alyssar wrote:
> On 2013/03/29 14:22:40, Ryan Hamilton wrote:
> > It looks like some of these files are in both quic_unittests and
quic_library.
> 
> > I think they should only be in one.
> 
> Which ones?  There's definitely overlap with quic_server and quic_client but
> when I tried including them as deps it didn't work (maybe because they're
> executables, maybe I did something wrong).  I don't see anything in
quic_library
> here.
> 
> I can create server_lib and client_lib if you'd prefer -- I thought you OKd
> doing a copy/paste of src files was OK if the dependencies route didn't work.

Oh, you're right, I misread the target before this one.  As you say, the
duplication is in quic_client and quic_server.  I would recommend that the
duplication be eliminated either by simply moving the common files into
quic_library (leaving just _bin in quic_client and quic_server) or by creating
quic_client_lib and quic_server_lib.  Either approach is fine with me; a single
target might be easier.

I suspect that simply adding quic_client as a dependency of quic_unittests did
not work because quic_client is an executable.  I could imagine that not doing
the right thing, but gyp is a bit of a black box to me.  If we're unable to get
the dependency issues sorted out without duplication, I'll definitely OK a
commit with duplication, but I think we should be able to make this work.  

If this seem problematic and you'd really like to land as is and then clean up
in a follow up CL, that's totally fine with me too.  What do you think?

[alyssar, 2013-03-29 16:23:22]

https://codereview.chromium.org/13004022/diff/1/net/net.gyp
File net/net.gyp (right):

https://codereview.chromium.org/13004022/diff/1/net/net.gyp#newcode2544
net/net.gyp:2544: 'tools/quic/quic_server_session.cc',
I'm not a big fan of "everything in one big target" since it often leads to bad
file dependencies.  That said, we have the hierarchy I'd like on the google3
side and it seems to be the norm in this file, so I'll go with that.

[Ryan Hamilton, 2013-03-29 16:50:48]

LGTM

https://codereview.chromium.org/13004022/diff/1/net/net.gyp
File net/net.gyp (right):

https://codereview.chromium.org/13004022/diff/1/net/net.gyp#newcode2544
net/net.gyp:2544: 'tools/quic/quic_server_session.cc',
On 2013/03/29 16:23:22, alyssar wrote:
> I'm not a big fan of "everything in one big target" since it often leads to
bad
> file dependencies.  That said, we have the hierarchy I'd like on the google3
> side and it seems to be the norm in this file, so I'll go with that.

Yeah, I hear ya.  I'm not really sure why we don't split up net into smaller
pieces, but for whatever reason we tend not to.  *boggle*  On the other hand, it
is easier, so yay :>

[commit-bot: I haz the power, 2013-03-29 17:57:29]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/[email protected]/13004022/8001

[commit-bot: I haz the power, 2013-03-29 17:57:33]

Presubmit check for 13004022-8001 failed and returned exit status 1.

INFO:root:Found 13 file(s).

Running presubmit commit checks ...
Running /b/commit-queue/workdir/chromium/PRESUBMIT.py
Running /b/commit-queue/workdir/chromium/net/PRESUBMIT.py

** Presubmit Messages **
If this change has an associated bug, add BUG=[bug number].

** Presubmit Warnings **
License must match:
.*? Copyright (\(c\) )?(2013|2012|2011|2010|2009|2008|2007|2006|2006-2008) The
Chromium Authors\. All rights reserved\.\n.*? Use of this source code is
governed by a BSD-style license that can be\n.*? found in the LICENSE file\.(?:
\*/)?\n
Found a bad license header in these files:
  net/tools/quic/end_to_end_test.cc \
  net/tools/quic/test_tools/http_message_test_utils.cc \
  net/tools/quic/test_tools/http_message_test_utils.h \
  net/tools/quic/test_tools/quic_test_client.cc \
  net/tools/quic/test_tools/quic_test_client.h \
  net/tools/quic/test_tools/run_all_unittests.cc

Presubmit checks took 1.2s to calculate.

[commit-bot: I haz the power, 2013-03-29 19:39:59]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/[email protected]/13004022/14001

[commit-bot: I haz the power, 2013-03-30 17:00:12]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/[email protected]/13004022/14001

[commit-bot: I haz the power, 2013-03-30 17:50:13]

Change committed as 191529