Add missing status codes to WebSocketError enum.

Add missing status codes to WebSocketError enum.

Add the close status codes that were missing from net::WebSocketError, and also
change the style of the enum from ALL_UPPER to kCamelCase.

Also add a constructor to WebSocketFrameHeader to minimise boilerplate when it
is used, and utility functions IsKnownDataOpCode() and IsKnownControlOpCode()
which will be used by WebSocketCommon. Made WebSocketFrameHeader uncopyable and
added a Clone() method for more style-guide compliance.

BUG=237496
TEST=net_unittests --gtest_filter='WebSocket*'

Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=198725

[tyoshino (SeeGerritForStatus), 2013-05-02 13:56:22]

https://codereview.chromium.org/14850012/diff/1/net/websockets/websocket_erro...
File net/websockets/websocket_errors.h (right):

https://codereview.chromium.org/14850012/diff/1/net/websockets/websocket_erro...
net/websockets/websocket_errors.h:36: // for dropping the control channel.
drop reason code of multiplexing spec are used only in multiplexing layer. can't
we separate them from the ones above that are used in the base WebSocket
protocol layer? even if we need to have some interface that can take/pass both,
we should group them into different enum, i think.

i'm sorry if i missed any point you told me before, but do you have any use case
of combined enum like this?

https://codereview.chromium.org/14850012/diff/1/net/websockets/websocket_erro...
net/websockets/websocket_errors.h:79: #endif  //
NET_WEBSOCKETS_WEBSOCKET_ERRORS_H_
good catch.

https://codereview.chromium.org/14850012/diff/1/net/websockets/websocket_erro...
File net/websockets/websocket_errors_unittest.cc (right):

https://codereview.chromium.org/14850012/diff/1/net/websockets/websocket_erro...
net/websockets/websocket_errors_unittest.cc:7: #include
"testing/gtest/include/gtest/gtest.h"
Please include "net/base/net_errors.h"

https://codereview.chromium.org/14850012/diff/1/net/websockets/websocket_fram...
File net/websockets/websocket_frame_unittest.cc (right):

https://codereview.chromium.org/14850012/diff/1/net/websockets/websocket_fram...
net/websockets/websocket_frame_unittest.cc:411: typedef WebSocketFrameHeader
Frame;
blank line please

https://codereview.chromium.org/14850012/diff/1/net/websockets/websocket_fram...
net/websockets/websocket_frame_unittest.cc:428:
EXPECT_FALSE(Frame::IsKnownDataOpCode(0xFF));
pick one value from 4-7 and another from 0xc-0xf, and test them.

https://codereview.chromium.org/14850012/diff/1/net/websockets/websocket_fram...
net/websockets/websocket_frame_unittest.cc:435: typedef WebSocketFrameHeader
Frame;
blank line please

https://codereview.chromium.org/14850012/diff/1/net/websockets/websocket_fram...
net/websockets/websocket_frame_unittest.cc:452:
EXPECT_FALSE(Frame::IsKnownControlOpCode(0xFF));
pick one value from 4-7 and another from 0xc-0xf, and test them.

[Adam Rice, 2013-05-04 09:51:53]

https://codereview.chromium.org/14850012/diff/1/net/websockets/websocket_erro...
File net/websockets/websocket_errors.h (right):

https://codereview.chromium.org/14850012/diff/1/net/websockets/websocket_erro...
net/websockets/websocket_errors.h:36: // for dropping the control channel.
On 2013/05/02 13:56:22, tyoshino wrote:
> drop reason code of multiplexing spec are used only in multiplexing layer.
can't
> we separate them from the ones above that are used in the base WebSocket
> protocol layer? even if we need to have some interface that can take/pass
both,
> we should group them into different enum, i think.
> 
> i'm sorry if i missed any point you told me before, but do you have any use
case
> of combined enum like this?

I thought it was necessary to include them to make the
WebSocketErrorToNetError() function useful. However, it made little difference
in practice. I have removed them.

https://codereview.chromium.org/14850012/diff/1/net/websockets/websocket_erro...
File net/websockets/websocket_errors_unittest.cc (right):

https://codereview.chromium.org/14850012/diff/1/net/websockets/websocket_erro...
net/websockets/websocket_errors_unittest.cc:7: #include
"testing/gtest/include/gtest/gtest.h"
On 2013/05/02 13:56:22, tyoshino wrote:
> Please include "net/base/net_errors.h"

Done.

https://codereview.chromium.org/14850012/diff/1/net/websockets/websocket_fram...
File net/websockets/websocket_frame_unittest.cc (right):

https://codereview.chromium.org/14850012/diff/1/net/websockets/websocket_fram...
net/websockets/websocket_frame_unittest.cc:411: typedef WebSocketFrameHeader
Frame;
On 2013/05/02 13:56:22, tyoshino wrote:
> blank line please

Done.

https://codereview.chromium.org/14850012/diff/1/net/websockets/websocket_fram...
net/websockets/websocket_frame_unittest.cc:428:
EXPECT_FALSE(Frame::IsKnownDataOpCode(0xFF));
On 2013/05/02 13:56:22, tyoshino wrote:
> pick one value from 4-7 and another from 0xc-0xf, and test them.

Done.

https://codereview.chromium.org/14850012/diff/1/net/websockets/websocket_fram...
net/websockets/websocket_frame_unittest.cc:435: typedef WebSocketFrameHeader
Frame;
On 2013/05/02 13:56:22, tyoshino wrote:
> blank line please

Done.

https://codereview.chromium.org/14850012/diff/1/net/websockets/websocket_fram...
net/websockets/websocket_frame_unittest.cc:452:
EXPECT_FALSE(Frame::IsKnownControlOpCode(0xFF));
On 2013/05/02 13:56:22, tyoshino wrote:
> pick one value from 4-7 and another from 0xc-0xf, and test them.

Done.

[tyoshino (SeeGerritForStatus), 2013-05-07 05:36:26]

https://codereview.chromium.org/14850012/diff/6001/net/websockets/websocket_e...
File net/websockets/websocket_errors.cc (left):

https://codereview.chromium.org/14850012/diff/6001/net/websockets/websocket_e...
net/websockets/websocket_errors.cc:20: NOTREACHED();
can we keep NOTREACHED()? will this function start receiving user input?

https://codereview.chromium.org/14850012/diff/6001/net/websockets/websocket_e...
File net/websockets/websocket_errors.h (right):

https://codereview.chromium.org/14850012/diff/6001/net/websockets/websocket_e...
net/websockets/websocket_errors.h:44: kWebSocketErrorBadFragmentation = 3009,
separate this block too. they're mux drop reason codes for logical channel
errors.

[Adam Rice, 2013-05-07 08:02:54]

https://codereview.chromium.org/14850012/diff/6001/net/websockets/websocket_e...
File net/websockets/websocket_errors.cc (left):

https://codereview.chromium.org/14850012/diff/6001/net/websockets/websocket_e...
net/websockets/websocket_errors.cc:20: NOTREACHED();
On 2013/05/07 05:36:26, tyoshino wrote:
> can we keep NOTREACHED()? will this function start receiving user input?

I am not sure, but it seems likely we will want to use this function on messages
received from the network. These may contain extension codes, or just invalid
data.

I am thinking of changing the function prototype from Error(WebSocketError) to
Error(unsigned short), but I want to wait and see how we use it.

To be honest, I am not sure how useful this function will really be. We might
end up not using it at all, or using different functions depending on what
extensions are in use.

https://codereview.chromium.org/14850012/diff/6001/net/websockets/websocket_e...
File net/websockets/websocket_errors.h (right):

https://codereview.chromium.org/14850012/diff/6001/net/websockets/websocket_e...
net/websockets/websocket_errors.h:44: kWebSocketErrorBadFragmentation = 3009,
On 2013/05/07 05:36:26, tyoshino wrote:
> separate this block too. they're mux drop reason codes for logical channel
> errors.

Sorry. I thought I had removed them, but I did not check closely enough.

[tyoshino (SeeGerritForStatus), 2013-05-07 08:09:35]

LGTM patch set 3

[commit-bot: I haz the power, 2013-05-07 08:09:43]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/ricea@chromium.org/14850012/14001

[commit-bot: I haz the power, 2013-05-07 14:59:21]

Change committed as 198725