[comp.protocols.iso.x400] Problem with mapping between RFC and X.400 adresses

cfe+@ANDREW.CMU.EDU ("Craig F. Everhart") (02/01/89)

Would it cause too many problems for the '.' character as used in X.400
attributes to be replaced by '-' when converted to domain-style notation?  As I
understand it, X.400/domain gateways need to know (at least) what the domain
names mean in X.400-speak, so they could easily know that the domain component
foo-bar had to be spelled out as the X.400 organization name Foo.Bar.

                Craig Everhart

P4403@com.qz.se (02/01/89)

>X-Bitnet-Sender: "Johan Lundberg TeleDelta (Swedish Telecom)" <P4403@COM.QZ.SE> 

A serius problem has been discovered with the current RFC-987
mapping
of
X.400
addresses to domain address that is used by the
academic comunity. The problem is the usage of the separator "."
(punctuation mark, full stop) between the "attributes" in the domain address.
     
In an X.400 address the "." is a leagal character in any of the attributes.
Many organizations make use of this character in their names
If such an organisationName or orgUnitName is used mapping from X.400 to RFC
will cause problem since the routing will probably fail.
And trying to map a sucsessfully transformed but not routable address
back again to X.400 will also fail.
     
One could perhaps circumvent this if one knows about those that uses "."
in thier names, but that will only work for the known ones and will make a
gateway more complex.
     
Has this been discussed and is there a solution to this problem?
     

pv@SUN.COM (Peter Vanderbilt) (02/02/89)

>  A serius problem has been discovered with the current RFC-987 mapping
>  of X.400 addresses to domain address that is used by the academic
>  comunity. The problem is the usage of the separator "." (punctuation
>  mark, full stop) between the "attributes" in the domain address.
>  
>  In an X.400 address the "." is a leagal character in any of the
>  attributes.

Attributes with a "." in them should not be mapped to the domain part
of the address but must be kept in the local part (using the
"/attr=value/" syntax).  The only attributes appearing in the domain
part must be in the domain syntax.

RFC 987 section 4.2.4, mapping B, point 2:

    Following the 4.3.1 hierarchy, if each successive
    component exists, and conforms to the syntax
    EBNF.domain-syntax (as defined in 4.3.1), allocate the
    next subdomain.

The reference to 4.3.1 should be 4.2.1:

     domain-syntax   = ALPHA [ *alphanumhyphen alphanum ]
     alphanum        = <ALPHA or DIGIT>
     alphanumhyphen  = <ALPHA or DIGIT or HYPHEN>

Pete

S.Kille@CS.UCL.AC.UK (Steve Kille) (02/02/89)

Phone: +44-1-380-7294

ermmmm - how about using the quoting mechanism specied in RFC 987??



Steve


 >From:  P4403@se.qz.com
 >To:    mhsnews <mhsnews@uninett.unit.runit.vax>
 >Subject: Problem with mapping between RFC and X.400 adresses
 >Date:  1 Feb 89 12:44 +0100

 >>X-Bitnet-Sender: "Johan Lundberg TeleDelta (Swedish Telecom)" <P4403@COM.QZ.SE> 

 >A serius problem has been discovered with the current RFC-987
 >mapping
 >of
 >X.400
 >addresses to domain address that is used by the
 >academic comunity. The problem is the usage of the separator "."
 >(punctuation mark, full stop) between the "attributes" in the domain address.
 >     
 >In an X.400 address the "." is a leagal character in any of the attributes.
 >Many organizations make use of this character in their names
 >If such an organisationName or orgUnitName is used mapping from X.400 to RFC
 >will cause problem since the routing will probably fail.
 >And trying to map a sucsessfully transformed but not routable address
 >back again to X.400 will also fail.
 >     
 >One could perhaps circumvent this if one knows about those that uses "."
 >in thier names, but that will only work for the known ones and will make a
 >gateway more complex.
 >     
 >Has this been discussed and is there a solution to this problem?
 >