- From: Brian Smith <brian@briansmith.org>
- Date: Thu, 28 Feb 2008 16:14:06 -0800
- To: "'HTTP Working Group'" <ietf-http-wg@w3.org>
Frank Ellermann wrote: > Brian Smith wrote: > > > headers like Content-MD5 (recalculate MD5 if you > encode/decode using > > Content-Encoding) > [...] > > need special processing when you "regurgitate" them. > > What "special processing" apart from what is stated in RFC > 1864 do you have in mind ? > > | To generate the value of the Content-MD5 field, the MD5 > algorithm is > | computed on the canonical form of the MIME entity's object. In > | particular, this means that the sender applies the MD5 algorithm on > | the data immediately after conversion to canonical form, before > | applying any content-transfer-encoding, and that the receiver also > | applies the MD5 algorithm on the canonical form, after undoing any > | content-transfer-encoding. There is no content-transfer-encoding in HTTP, and Content-MD5/Content-Encoding processing works exactly the opposite way: if the request is compressed, you need to calculate the MD5 on the compressed request; if you later serve the uncompressed variant, then you need to recalculate the Content-MD5 or remove it. That is the "special processing" that I was referring to. - Brian
Received on Friday, 29 February 2008 00:25:52 UTC