Sam Ruby wrote:
Julian Reschke wrote:
Tim Bray wrote:
Just to see if it uncovered any problems, I twiddled the ongoing software to generate a format-05 atom feed. It didn't uncover any problems that I could see. Norm's RNC schema was very valuable in debugging, not that there were many bugs. Check it out at http://www.tbray.org/ongoing/ongoing.atom
Same here (<http://greenbytes.de/tech/webdav/webdav.atom>).
I find your use of opaquelocktokens intriguing. My reading of the spec is that opaquelocktokens are sequences of 8, 4, 4, 12 hex digits, separated by dashes, optionally followed by a path.
In place of a path, I see what appears to be a timestamp. Can somebody cite a reference which permits such uris?
The timestamps I am using are legal path components, according to RFC2068, section 3.2.1 (which RCF2518's definition for the "opaquelocktoken" scheme refers to). Or am I missing something?
Best regards, Julian
-- <green/>bytes GmbH -- http://www.greenbytes.de -- tel:+492512807760
