> Is this expected behavior now for Twitter or is this a bug in the new > Twitter release (Not supporting the Expect: Continue header). I'm > writing a multi-service Windows Mobile application and I'd prefer not > changing global http connection settings if I don't have to.
Strictly speaking, according to the HTTP spec, it was a bug to *accept* Expect 100-continue and *not* return a 417 if you weren't prepared to handle it. See RFC 2616, 14.20. Thus, Twitter's Apache was buggy *before*, but not now. There are arguments for shooting first and asking questions later (i.e., send Expect: headers all the time even if the remote server doesn't handle it right), but clients that shoot first should be prepared to retry the request without Expect: because there are still many servers that don't know what to do with it. There are also ambiguous situations with servers that don't understand the header at all and just ignore it; see Section 8.2.3. -- ------------------------------------ personal: http://www.cameronkaiser.com/ -- Cameron Kaiser * Floodgap Systems * www.floodgap.com * [email protected] -- Ninety-nine percent of lawyers give the rest a bad name. -------------------
