On 01/12/2014 21:09, Rémy Maucherat wrote: > 2014-12-01 21:44 GMT+01:00 Mark Thomas <ma...@apache.org>: > >> On 01/12/2014 17:56, r...@apache.org wrote: >>> Author: remm >>> Date: Mon Dec 1 17:56:10 2014 >>> New Revision: 1642722 >>> >>> URL: http://svn.apache.org/r1642722 >>> Log: >>> Don't call onMessage on whole empty messages. >> >> What is the justification for this? >> >> RFC 6455 permits zero length messages. >> >> There is nothing in the Java WebSocket spec I can recall that permits >> zero length messages to be skipped. >> >> Doing a close of a writer or OS following an exception causes an extra > empty message (so the client gets an empty message and a message created > from the error, and doesn't like the first one). A flush would do the same > too for no reason. So there's something that needs to be filtered out > there, that's probably not what was meant to happen.
OK. But this isn't the right fix for that problem. Do you have a recipe to reproduce this? My main question is what is triggering the exception? Mark --------------------------------------------------------------------- To unsubscribe, e-mail: dev-unsubscr...@tomcat.apache.org For additional commands, e-mail: dev-h...@tomcat.apache.org
