On 13.07.2012 18:02, Jim Jagielski wrote:
If these can be added somewhat quickly, I'm willing to fast-track
them into 2.4.3.
I drafted a patch available at
http://people.apache.org/~rjung/patches/httpd-trunk-status-codes-iana.patch
Coments:
- I didn't fix the indentation in include/httpd.h in order
to keep the patch readable.
Some of the new codes have a short description which is a bit longer
than the longest one used up to now.
- I didn't "fix" the old define named "HTTP_REQUEST_URI_TOO_LARGE"
which should have been "HTTP_REQUEST_URI_TOO_LONG" since it
is defined in a public header file
- I included all changes proposed by Julian
- there is a big gap of unused numbers between 208 and 226 which
I filled with "unknown" as was done before due to the limitations
in ap_index_of_response() (focus on performance there)
- I added canned error strings for the new codes
- I did not yet define new error documents. The new pages could be
HTTP_PRECONDITION_REQUIRED 428
HTTP_TOO_MANY_REQUESTS 429
HTTP_REQUEST_HEADER_FIELDS_TOO_LARGE 431
HTTP_LOOP_DETECTED 508
HTTP_NETWORK_AUTHENTICATION_REQUIRED 511
Furthermore some other 4xx and 5xx codes already defined in httpd.h
also have no error page:
HTTP_PAYMENT_REQUIRED 402
HTTP_NOT_ACCEPTABLE 406
HTTP_PROXY_AUTHENTICATION_REQUIRED 407
HTTP_CONFLICT 409
HTTP_RANGE_NOT_SATISFIABLE 416
HTTP_EXPECTATION_FAILED 417
HTTP_UNPROCESSABLE_ENTITY 422
HTTP_LOCKED 423
HTTP_FAILED_DEPENDENCY 424
HTTP_UPGRADE_REQUIRED 426
HTTP_GATEWAY_TIME_OUT 504
HTTP_VERSION_NOT_SUPPORTED 505
HTTP_INSUFFICIENT_STORAGE 507
HTTP_NOT_EXTENDED 510
I guess that means defining ones for the new codes is not a must ...
- I did not check, which of the new codes actually should change
behaviour of the web server!!
Regards,
Rainer
On Jul 12, 2012, at 9:07 AM, Jim Jagielski wrote:
++1!
On Jul 12, 2012, at 2:34 AM, Julian Reschke wrote:
On 2012-07-11 19:15, Roy T. Fielding wrote:
I don't know of any issues with 308, and Julian generally knows what
he is doing with regard to HTTP. In general, we should consider
Thanks :-)
the IANA registry to be authoritative unless it is a known bug,
In which case we should fix the registry.
which means we should support everything in
http://www.iana.org/assignments/http-status-codes/http-status-codes.xml
Yes. If we want to get all of these in, I can open a separate ticket and
provide another patch.
Best regards, Julian
