www-apache-bugdb mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From John-Mark Gurney <gurne...@4j.lane.edu>
Subject Re: general/4033: HTTP/0.9 requests yield invalid response
Date Thu, 11 Mar 1999 21:40:03 GMT
The following reply was made to PR general/4033; it has been noted by GNATS.

From: John-Mark Gurney <gurney_j@4j.lane.edu>
To: Rodent of Unusual Size <Ken.Coar@Golux.Com>
Cc: Apache bug database <apbugs@Apache.Org>
Subject: Re: general/4033: HTTP/0.9 requests yield invalid response
Date: Thu, 11 Mar 1999 13:35:31 -0800

 Rodent of Unusual Size scribbled this message on Mar 11:
 > John-Mark Gurney wrote:
 > > 
 > >  oh well, you can close it, but it doesn't matter, the spec says that
 > >  a difference of major version numbers isn't transport compatible, and
 > >  apache is assuming that major version 0 of http is transport compatible
 > >  with major version 1...  it doesn't matter that it's a malformed HTTP/0.9
 > >  request, apache shouldn't return an HTTP/1.x response as it either should
 > >  return a simple request or return an invalid request...
 > 
 > You're missing the point, I think.  It *isn't* a 'malformed HTTP/0.9
 > request,' because HTTP/0.9 requests *never* have a version number.
 
 even though the specs say this (section 6):
 "A Simple-Response should only be sent in response [or case deleted] if
 the server only supports the more limited HTTP/0.9 protocol."
 
 > If anything, it's a malformed HTTP/1.0 request by virtue of having
 > included the version number at all, and the malformation is an
 
 yes if I generate a request of:
 GET /somefile HTTP/1293.398
 
 then you are ligitimate in returning a HTTP/1.x response if that is
 the "highest" major version of the speficiation...  but you server
 understands the HTTP/0.9 spec, so if you recieve a request that says
 it's HTTP/0.9, you have to respond with an HTTP/0.9 response...
 
 > unrecognised version string.  In which case an HTTP/1.0 response
 > is perfectly valid.  If a request doesn't include a version
 > identifier, it's HTTP/0.9 by definition.  If it does, it *has*
 > to be something other than HTTP/0.9.
 
 where in the specification does it say that HTTP/0.9 can not generate
 a full-request?
 
 oh well, until you can state in the spec that I am not correct, I'll
 still feel the way I do...  you might as well close the pr as it
 seems that we are not going to come to an agreement... lets agree
 to disagree...
 
 and thanks for all the work on Apache... it's a great product and will
 continue to use it...
 
 -- 
 John-Mark Gurney
 Eugene 4J School District

Mime
View raw message