[jcifs] Re: SMB URL encoding/decoding

[Christopher R. Hertel]

James Nord wrote:
:
> >I'm not sure what you mean by "we are only allowed" but none of the
> >above is true (which may be your point). You cannot count certain
> >characters. The password field could have three ':' or fifteen '@'s in
> >it. It can have anything. *That's* the real problem.
> >
> No it can not.
> 
> again password != password field
> 
> the password field cannot contain any of,
> 
>       reserved      = ";" | "/" | "?" | ":" | "@" | "&" | "=" | "+" |
>                       "$" | ","
> 
> (note that is not an exhaustive list we also have öäåÖÄÅ....)

Much easier to denote what the field may contain.  :)

It may contain anything that <userinfo> may contain, including a colon.  The
*first* colon in <userinfo> is used as a delimiter.  Since the <password>
field is specifically not defined by the BNF grammar given in the RFC we
don't have anything that says it can't contain a colon.
...yes, I'm really annoying when it comes to fine details.

Since

        userinfo    = *( unreserved | escaped |
                         ";" | ":" | "&" | "=" | "+" | "$" | "," )

and we are allowing for

        userinfo    = user [ ":" password ]

we have

        password    = *( unreserved | escaped |
                         ";" | ":" | "&" | "=" | "+" | "$" | "," )

Chris -)-----

-- 
Samba Team -- http://www.samba.org/     -)-----   Christopher R. Hertel
jCIFS Team -- http://jcifs.samba.org/   -)-----   ubiqx development, uninq.
ubiqx Team -- http://www.ubiqx.org/     -)-----   crh at ubiqx.mn.org