[Fedora-directory-users] Schema Format

[Kevin Peterson]

When creating a custom schema file, whitespace appears to be
important. For example:

attributetypes: (1.3.6.1.4.1.2114.108.1.8.999.8
NAME ('inherent-3-D-shape-reverse')
SUP tsCaseIgnoreIA5String
SINGLE-VALUE)

attributetypes: (1.3.6.1.4.1.2114.108.1.8.999.9
NAME ('branch-continuity')
SUP tsCaseIgnoreIA5String
SINGLE-VALUE)

will not work, while

attributetypes: (1.3.6.1.4.1.2114.108.1.8.999.8 NAME
('inherent-3-D-shape-reverse') SUP tsCaseIgnoreIA5String SINGLE-VALUE)
attributetypes: (1.3.6.1.4.1.2114.108.1.8.999.9 NAME
('branch-continuity') SUP tsCaseIgnoreIA5String SINGLE-VALUE)

will work.

I did not see any mention of whitespace in RFC 2252 and I am wondering
1. Is this is expected behaviour, and 2. Is this documented anywhere,
because if it is perhaps there are more formatting rules that I am
missing.

[Nathan Kinder]

Hi Kevin,

There should be a single leading space to continue the line.  This is a 
standard descibed in RFC 2849.  Here's an example:

attributetypes: (1.3.6.1.4.1.2114.108.1.8.999.8
 NAME ('inherent-3-D-shape-reverse')
 SUP tsCaseIgnoreIA5String
 SINGLE-VALUE)

-NGK

Kevin Peterson wrote:

> When creating a custom schema file, whitespace appears to be
> important. For example:
>
> attributetypes: (1.3.6.1.4.1.2114.108.1.8.999.8
> NAME ('inherent-3-D-shape-reverse')
> SUP tsCaseIgnoreIA5String
> SINGLE-VALUE)
>
> attributetypes: (1.3.6.1.4.1.2114.108.1.8.999.9
> NAME ('branch-continuity')
> SUP tsCaseIgnoreIA5String
> SINGLE-VALUE)
>
> will not work, while
>
> attributetypes: (1.3.6.1.4.1.2114.108.1.8.999.8 NAME
> ('inherent-3-D-shape-reverse') SUP tsCaseIgnoreIA5String SINGLE-VALUE)
> attributetypes: (1.3.6.1.4.1.2114.108.1.8.999.9 NAME
> ('branch-continuity') SUP tsCaseIgnoreIA5String SINGLE-VALUE)
>
> will work.
>
> I did not see any mention of whitespace in RFC 2252 and I am wondering
> 1. Is this is expected behaviour, and 2. Is this documented anywhere,
> because if it is perhaps there are more formatting rules that I am
> missing.
>
> --
> Fedora-directory-users mailing list
> Fedora-directory-users(a)redhat.com
> https://www.redhat.com/mailman/listinfo/fedora-directory-users
>   
>

[Nathan Kinder]

Well, that didn't work.  My formatting was lost.  You need a space 
before the 2nd, 3rd, and 4th line.  Hopefully that's clear.

-NGK


Nathan Kinder wrote:

>  Hi Kevin,
>
>  There should be a single leading space to continue the line.  This is 
>  a standard descibed in RFC 2849.  Here's an example:
>
>  attributetypes: (1.3.6.1.4.1.2114.108.1.8.999.8
>  NAME ('inherent-3-D-shape-reverse')
>  SUP tsCaseIgnoreIA5String
>  SINGLE-VALUE)
>
>  -NGK
>
>  Kevin Peterson wrote:
>
> > When creating a custom schema file, whitespace appears to be
> > important. For example:
> >
> > attributetypes: (1.3.6.1.4.1.2114.108.1.8.999.8
> > NAME ('inherent-3-D-shape-reverse')
> > SUP tsCaseIgnoreIA5String
> > SINGLE-VALUE)
> >
> > attributetypes: (1.3.6.1.4.1.2114.108.1.8.999.9
> > NAME ('branch-continuity')
> > SUP tsCaseIgnoreIA5String
> > SINGLE-VALUE)
> >
> > will not work, while
> >
> > attributetypes: (1.3.6.1.4.1.2114.108.1.8.999.8 NAME
> > ('inherent-3-D-shape-reverse') SUP tsCaseIgnoreIA5String SINGLE-VALUE)
> > attributetypes: (1.3.6.1.4.1.2114.108.1.8.999.9 NAME
> > ('branch-continuity') SUP tsCaseIgnoreIA5String SINGLE-VALUE)
> >
> > will work.
> >
> > I did not see any mention of whitespace in RFC 2252 and I am wondering
> > 1. Is this is expected behaviour, and 2. Is this documented anywhere,
> > because if it is perhaps there are more formatting rules that I am
> > missing.
> >
> > -- 
> > Fedora-directory-users mailing list
> > Fedora-directory-users(a)redhat.com
> > https://www.redhat.com/mailman/listinfo/fedora-directory-users
> >  
> >
>
> ------------------------------------------------------------------------
>
> --
> Fedora-directory-users mailing list
> Fedora-directory-users(a)redhat.com
> https://www.redhat.com/mailman/listinfo/fedora-directory-users
>   
>

[Mike Jackson]

Kevin Peterson wrote:
>  When creating a custom schema file, whitespace appears to be
>  important. For example:
>  
>  attributetypes: (1.3.6.1.4.1.2114.108.1.8.999.8
>  NAME ('inherent-3-D-shape-reverse')
>  SUP tsCaseIgnoreIA5String
>  SINGLE-VALUE)
>  
>  attributetypes: (1.3.6.1.4.1.2114.108.1.8.999.9
>  NAME ('branch-continuity')
>  SUP tsCaseIgnoreIA5String
>  SINGLE-VALUE)
>  
>  will not work, while 
Of course it does not work ;-) See RFC 2849 for formatting rules. FDS 
uses an LDIF object for the schema files:

dn: cn=schema
objectClasses: ...
objectClasses: ...
attributeTypes: ...
attributeTypes: ...



I use this type of nice format; I even wrote a perl program to reformat 
long attribute values to the below format, for readability:

#
# foo schema
#
#
dn: cn=schema
#
################################
# attributes
################################
#
attributeTypes: (
   1.3.6.1.4.1.1.1.1.1.1.1
   NAME 'xxx'
   DESC 'xxx'
   SYNTAX 1.3.6.1.4.1.1466.115.121.1.26
   SINGLE-VALUE
   )
#
################################
#
attributeTypes: (
   1.3.6.1.4.1.1.1.1.1.1.2
   NAME 'xxy'
   DESC 'xxy'
   SYNTAX 1.3.6.1.4.1.1466.115.121.1.26
   SINGLE-VALUE
   )
#
################################
#
attributeTypes: (
   1.3.6.1.4.1.1.1.1.1.1.3
   NAME 'xxz'
   DESC 'xxz'
   SYNTAX 1.3.6.1.4.1.1466.115.121.1.26
   SINGLE-VALUE
   )
#
################################
# object classes
################################
#
objectClasses: (
   1.3.6.1.4.1.1.1.1.1.2.1
   NAME 'xyzclass'
   SUP top
   MUST ( cn $ xxx )
   MAY ( description $ xxy $ xxz )
   )
#
#
# EOF



Mike

-- 
LDAP Directory Consulting - http://www.netauth.com