On Mon, Oct 19, 2009 at 12:28 PM, Tim Bray <[email protected]> wrote: > the notion that one might not have exact > control over the status code had never > occurred to me Even at the "Apache" level, getting the status code you want can be non-trivial... I'm not sure if Apache has ever fixed this bug, but for a long time, it was very difficult to issue a 226 status code.
bob wyman On Mon, Oct 19, 2009 at 12:28 PM, Tim Bray <[email protected]> wrote: > > On Mon, Oct 19, 2009 at 6:39 AM, Brett Slatkin <[email protected]> wrote: > > >> "The response from the subscriber's callback URL MUST be an HTTP > >> success (2xx) code." Why not just specify 202 Accepted here? What > >> other value might one expect to see? -T > > > > Practically speaking, it seems way more people have issues serving the > > correct HTTP status codes than one would expect. For instance, the > > default response code for more web frameworks is 200 if you do nothing > > explicit to modify it. So, if anything, that would advocate for the > > 200 response being the signal. However, is there a functional > > advantage for specifying 202 instead of 2xx? Seems like that would be > > more rigorous but also over-specified, since it does not help achieve > > a specific outcome? > > Good answer. I tend to work mostly down at the Rack or Apache levels > and thus the notion that one might not have exact control over the > status code had never occurred to me. Sigh. -T >
