So if not valid, what would be a better approach? Leaving out the content-length or setting it 0 would be valid I guess? What about the rest of the body. Each content-location could give back it's own header-informatie (like content-length), that would be good right? What about content-location, is it valid to put content-location inside the multiple parts of a multipart/parallel (or related) message? Trying to find ways within the specification. Roy T. Fielding wrote: > On Jun 11, 2009, at 12:14 PM, Robert de Wilde wrote: > >> Sorry, the last part was cut off! The 'complete' HTTP header: >> >> >> HTTP/1.1 302 Found >> Content-type: multipart/parallel; boundary="ping" >> Content-length: 2048; >> Accept-Ranges: bytes; > > No, the Content-Length determines the length of the message-body > including any multipart stuff within it. Multipart has no impact > on parsing the overall message. [And this is a bad use of 302.] > > ....Roy > >Received on Thursday, 11 June 2009 16:27:56 UTC
This archive was generated by hypermail 2.4.0 : Thursday, 2 February 2023 18:43:19 UTC