On 02/12/2014 09:24, Rémy Maucherat wrote: > 2014-12-02 10:12 GMT+01:00 Mark Thomas <ma...@apache.org>:My opinion is > that the default value currently being provided is just > >> plain wrong. The origin is meant to be "the script origin generating the >> WebSocket connection request" not the scheme, host and port of the >> WebSocket connection currently being made. >> > Well, I think we covered the topic then. I am not going to challenge the > TCK [there could be more problematic issues like batch mode flushing or > general output, so there's no way I'll do that for a trivial item]. If this > is important for you, you should veto this and I'll remove it.
I think there might a solution that works for everybody (thanks Konstantin). What about setting an origin header value of "null" in the client if none is set by the user? That is consistent with all the specifications and is arguably better than doing nothing which is what we are doing at the moment. Would that be sufficient to address the TCK issue as well? Mark --------------------------------------------------------------------- To unsubscribe, e-mail: dev-unsubscr...@tomcat.apache.org For additional commands, e-mail: dev-h...@tomcat.apache.org
