Maybe a new component options would let the developer select the behaviour ?
Or maybe first try with on option and then automatically switch to the other
if it fails ?

--
francois.pie...@overbyte.be
The author of the freeware multi-tier middleware MidWare
The author of the freeware Internet Component Suite (ICS)
http://www.overbyte.be




-----Message d'origine-----
De : twsocket-boun...@elists.org [mailto:twsocket-boun...@elists.org] De la
part de Arno Garrels
Envoyé : dimanche 8 janvier 2012 19:30
À : ICS support mailing
Objet : Re: [twsocket] Digest authentication via THttpCli?

Arno Garrels wrote:
> Paul Read wrote:
>> THanks for that information I therefore tweaked 
>> 'AuthDigestParseChallenge' so that Info.Qop is set to 'auth' if no 
>> Qop value is given and now the right MD5 is calculated and the server 
>> accepts the data.
> 
> I'd say this is a server-side bug. 

Though it might be a ICS bug in the RFC 2069 implementation as well, but I
have no idea where. Digest calculation is simple in RFC 2069 and the same
calculation is also used as one part of the RFC 2617 calc. 

> It obviously understands a RFC 2617
> digest however sends an obsolete RFC 2069 WWW-Authenticate response 
> header.

Well, that seems OK as long as the server supports both RFC 2069 and RFC
2617 clients.  

> If I'm not totally misreading this sentence:
> "qop-options
>      This directive is optional, but is made so only for backward
>      compatibility with RFC 2069 [6];"
> 
> it means that if the qop directive is missing we have to assume RFC
> 2069 which calculates the digest differently. 

If not, ICS clients won't be able to authenticate with true old RFC 2069
servers...?     

--
Arno Garrels

--
To unsubscribe or change your settings for TWSocket mailing list please goto
http://lists.elists.org/cgi-bin/mailman/listinfo/twsocket
Visit our website at http://www.overbyte.be

--
To unsubscribe or change your settings for TWSocket mailing list
please goto http://lists.elists.org/cgi-bin/mailman/listinfo/twsocket
Visit our website at http://www.overbyte.be

Reply via email to