- From: Julian Reschke <julian.reschke@gmx.de>
- Date: Tue, 16 Dec 2014 22:56:59 +0100
- To: "Roy T. Fielding" <fielding@gbiv.com>
- CC: Martin Thomson <martin.thomson@gmail.com>, HTTP Working Group <ietf-http-wg@w3.org>
On 2014-12-16 22:48, Roy T. Fielding wrote: > On Dec 16, 2014, at 1:36 PM, Julian Reschke wrote: > >> "OPTIONS uri" is fine, "OPTIONS *" is a problem... > > No, it isn't. I have implemented it. Yes, it is a special case, > but that's exactly why it was created -- to not look like a normal > request, not increment hit counts, not be blindly forwarded, and > not cause potentially expensive processing on the server. > > OPTIONS should be used whenever a NO-OP style of request is desired. > > ....Roy The problem with "OPTIONS *" is that id doesn't work in many frameworks. Thus it's a bad idea to *require* it's use anywhere. Is there a particular thing in this spec where "OPTIONS /" is worse than "OPTIONS *"? Best regards, Julian
Received on Tuesday, 16 December 2014 21:57:33 UTC