Pepper: Add a test for PPB_TCPSocket_Private.

Pepper: Add a test for PPB_TCPSocket_Private.

Actually, the interface is currently known as PPB_Flash_TCPSocket.

BUG=103993
TEST=ui_tests OutOfProcessPPAPITest.TCPSocketPrivate
TBR=dmichael@chromium.org

Committed: http://src.chromium.org/viewvc/chrome?view=rev&revision=110026

[yzshen1, 2011-11-15 00:47:07]

Thanks!

http://codereview.chromium.org/8564034/diff/1/ppapi/tests/test_tcp_socket_pri...
File ppapi/tests/test_tcp_socket_private.cc (right):

http://codereview.chromium.org/8564034/diff/1/ppapi/tests/test_tcp_socket_pri...
ppapi/tests/test_tcp_socket_private.cc:11: //FIXME
Please use "// TODO(somebody): What to do."

http://codereview.chromium.org/8564034/diff/1/ppapi/tests/test_tcp_socket_pri...
ppapi/tests/test_tcp_socket_private.cc:18: #include "ppapi/c/pp_stdint.h"
This is not needed since it has been included in .h file. (And also line 9.)

http://codereview.chromium.org/8564034/diff/1/ppapi/tests/test_tcp_socket_pri...
ppapi/tests/test_tcp_socket_private.cc:53: // We need something to connect to,
so we connect to the HTTP server whence we
Haha!
So what is your plan to test the UDP API? :)

http://codereview.chromium.org/8564034/diff/1/ppapi/tests/test_tcp_socket_pri...
ppapi/tests/test_tcp_socket_private.cc:56: PP_URLComponents_Dev c;
This name is not very descriptive.

http://codereview.chromium.org/8564034/diff/1/ppapi/tests/test_tcp_socket_pri...
ppapi/tests/test_tcp_socket_private.cc:66: return false;
You may want to use EnsureRunningOverHTTP

http://codereview.chromium.org/8564034/diff/1/ppapi/tests/test_tcp_socket_pri...
ppapi/tests/test_tcp_socket_private.cc:190: s->push_back(buffer[i]);  //
Inefficiency FTW.
Maybe this is better: find out whether there is '\n' and where it is, and then
do an append().

[viettrungluu, 2011-11-15 01:08:26]

Thanks, new patch set up.

http://codereview.chromium.org/8564034/diff/1/ppapi/tests/test_tcp_socket_pri...
File ppapi/tests/test_tcp_socket_private.cc (right):

http://codereview.chromium.org/8564034/diff/1/ppapi/tests/test_tcp_socket_pri...
ppapi/tests/test_tcp_socket_private.cc:11: //FIXME
On 2011/11/15 00:47:07, yzshen1 wrote:
> Please use "// TODO(somebody): What to do."

Oops, leftover cruft.

http://codereview.chromium.org/8564034/diff/1/ppapi/tests/test_tcp_socket_pri...
ppapi/tests/test_tcp_socket_private.cc:18: #include "ppapi/c/pp_stdint.h"
On 2011/11/15 00:47:07, yzshen1 wrote:
> This is not needed since it has been included in .h file. (And also line 9.)

Fixed. (But, AFAIK, the C++ headers don't necessarily define uint16_t (or
int32_t, for that matter).)

http://codereview.chromium.org/8564034/diff/1/ppapi/tests/test_tcp_socket_pri...
ppapi/tests/test_tcp_socket_private.cc:53: // We need something to connect to,
so we connect to the HTTP server whence we
On 2011/11/15 00:47:07, yzshen1 wrote:
> Haha!
> So what is your plan to test the UDP API? :)

?!? Who knows. I don't even know how I'll properly test SSL.

http://codereview.chromium.org/8564034/diff/1/ppapi/tests/test_tcp_socket_pri...
ppapi/tests/test_tcp_socket_private.cc:56: PP_URLComponents_Dev c;
On 2011/11/15 00:47:07, yzshen1 wrote:
> This name is not very descriptive.

But it's wonderfully short? I changed it to "components".

http://codereview.chromium.org/8564034/diff/1/ppapi/tests/test_tcp_socket_pri...
ppapi/tests/test_tcp_socket_private.cc:66: return false;
On 2011/11/15 00:47:07, yzshen1 wrote:
> You may want to use EnsureRunningOverHTTP

I'll do that in addition to what I do here....

http://codereview.chromium.org/8564034/diff/1/ppapi/tests/test_tcp_socket_pri...
ppapi/tests/test_tcp_socket_private.cc:190: s->push_back(buffer[i]);  //
Inefficiency FTW.
On 2011/11/15 00:47:07, yzshen1 wrote:
> Maybe this is better: find out whether there is '\n' and where it is, and then
> do an append().

How about I just add a .reserve()?

[yzshen1, 2011-11-15 01:16:56]

lgtm

http://codereview.chromium.org/8564034/diff/1/ppapi/tests/test_tcp_socket_pri...
File ppapi/tests/test_tcp_socket_private.cc (right):

http://codereview.chromium.org/8564034/diff/1/ppapi/tests/test_tcp_socket_pri...
ppapi/tests/test_tcp_socket_private.cc:18: #include "ppapi/c/pp_stdint.h"
On 2011/11/15 01:08:26, viettrungluu wrote:
> On 2011/11/15 00:47:07, yzshen1 wrote:
> > This is not needed since it has been included in .h file. (And also line 9.)
> 
> Fixed. (But, AFAIK, the C++ headers don't necessarily define uint16_t (or
> int32_t, for that matter).)

What do you mean by "C++ headers"?

http://codereview.chromium.org/8564034/diff/1/ppapi/tests/test_tcp_socket_pri...
ppapi/tests/test_tcp_socket_private.cc:190: s->push_back(buffer[i]);  //
Inefficiency FTW.
On 2011/11/15 01:08:26, viettrungluu wrote:
> On 2011/11/15 00:47:07, yzshen1 wrote:
> > Maybe this is better: find out whether there is '\n' and where it is, and
then
> > do an append().
> 
> How about I just add a .reserve()?
Okay.

[viettrungluu, 2011-11-15 01:41:45]

Thanks again.

On 2011/11/15 01:16:56, yzshen1 wrote:
> lgtm
> 
>
http://codereview.chromium.org/8564034/diff/1/ppapi/tests/test_tcp_socket_pri...
> File ppapi/tests/test_tcp_socket_private.cc (right):
> 
>
http://codereview.chromium.org/8564034/diff/1/ppapi/tests/test_tcp_socket_pri...
> ppapi/tests/test_tcp_socket_private.cc:18: #include "ppapi/c/pp_stdint.h"
> On 2011/11/15 01:08:26, viettrungluu wrote:
> > On 2011/11/15 00:47:07, yzshen1 wrote:
> > > This is not needed since it has been included in .h file. (And also line
9.)
> > 
> > Fixed. (But, AFAIK, the C++ headers don't necessarily define uint16_t (or
> > int32_t, for that matter).)
> 
> What do you mean by "C++ headers"?

Oh, sorry, I misunderstood what you meant by your "line 9" comment. I guess I
should remove that include too. NM.

> 
>
http://codereview.chromium.org/8564034/diff/1/ppapi/tests/test_tcp_socket_pri...
> ppapi/tests/test_tcp_socket_private.cc:190: s->push_back(buffer[i]);  //
> Inefficiency FTW.
> On 2011/11/15 01:08:26, viettrungluu wrote:
> > On 2011/11/15 00:47:07, yzshen1 wrote:
> > > Maybe this is better: find out whether there is '\n' and where it is, and
> then
> > > do an append().
> > 
> > How about I just add a .reserve()?
> Okay.

[dmichael (off chromium), 2011-11-15 17:47:54]

lgtm