On 28/12/2012 00:04, Konstantin Kolinko wrote:
> Looking at the current code, "new String()" call still remains in WsFrame.java
> Starting with Line 270:
>
> if (controlBufferBinary.remaining() > 0) {
> CoderResult cr = utf8DecoderControl.decode(
> controlBufferBinary, controlBufferText, true);
> if (cr.isError()) {
> controlBufferBinary.clear();
> controlBufferText.clear();
> throw new WsIOException(new CloseReason(
> CloseCodes.PROTOCOL_ERROR,
> sm.getString("wsFrame.invalidUtf8Close")));
> }
> reason = new String(controlBufferBinary.array(),
> controlBufferBinary.arrayOffset() +
> controlBufferBinary.position(),
> controlBufferBinary.remaining(), "UTF8");
> }
>
> Why use "new String(.., UTF8)" when there is
> "utf8DecoderControl.decode(..)" call several lines above?
>
> Isn't it supposed to use controlBufferText here ?
It is. The decoder catches the encoding problems which is why the
Autobahn tests pass but I should have removed the new String() at the
same time.
I've fixed this. Thanks yet again for the review.
Mark
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscr...@tomcat.apache.org
For additional commands, e-mail: dev-h...@tomcat.apache.org
