On Sat, Feb 03, 2007 at 12:00:24AM -0800, Sigbjorn Finne wrote: > I haven't fully trawled the mailing list archives to check if this has > been reported already, but neon-0.26.x has what looks like a > breaking change wrt. 0.25.x releases in lk_pre_send() -- it > now terminates the If: header thusly > > ne_buffer_czappend(req, "\n"); > > I believe this ought to be like it was previously, > > ne_buffer_zappend(req, EOL); > > That certainly fixes the issues I came across when trying to use > litmus-0.11.
Hi, thanks for reporting this - yes, that's a regression. (RFC 2616 actually recommends that a "tolerant" application accepts LF as the header line terminator, and most servers do, which is why this hasn't been spotted before; see section 19.3; but this is no excuse :) Regards, joe _______________________________________________ neon mailing list [email protected] http://mailman.webdav.org/mailman/listinfo/neon
