Re: Cif dictionary version numbers

["Herbert J. Bernstein" <yaya@xxxxxxxxxxxxxxxxxxxxxxx>]

I would favor extending the definition to allow for all the
alternate dictionary version number schemes:

   h
   h.k
   h.k.l

Regards,
  Herbert

=====================================================
 Herbert J. Bernstein, Professor of Computer Science
   Dowling College, Kramer Science Center, KSC 121
        Idle Hour Blvd, Oakdale, NY, 11769

                 +1-631-244-3035
                 yaya@dowling.edu
=====================================================

On Wed, 20 Jul 2005, Brian McMahon wrote:

> > "Data value is not a correctly formatted number: _dictionary_version"
> >
> > According to ddl_core.dic the _type of _dictionary_version is numb.  1.0
> > and 1.1 are valid numbers but 1.0.1 and 2.3.1 are not, at least by my
> > understanding of the definition of a number in the CIF 1.1 specification.
> > This could be a serious issue if some computer program attempted to
> > compare dictionaries by converting the value of _dictionary_version to a
> > number.
>
> Oops. That's rather embarrassing.
>
> > So my question is, should the dictionary versions that do not meet the
> > dictionary specification be changed so that they do? i.e.
> >
> > 1.0.1 -> 1.1
> > 2.3.1 -> 2.4
>
> I think the three-level "numbering" scheme is useful in indicating the
> approximate level of change in an edition of the dictionaries - a change
> from 1.0.1 to 1.0.2 means that only some very minor changes have taken
> place, and one need not rush to upgrade. 1.1 to 1.2 probably means that new
> data items have been added, and perhaps some adjustment to some of the
> validation ranges. 1.1 to 2.0 probably means that there have been a lot of
> changes, perhaps even to the underlying data model - the documentation
> should be read very carefully. There's also benefit in allowing the
> _dictionary_version to have additional characters - "1.2beta" could be a
> useful way of flagging a preliminary release to developers.
>
> So I would favour leaving the version labelling as now, but changing the
> type of _dictionary_version in any new DDL1 release that results from this
> and other of the recent bug reports.
>
> What do others feel?
>
> Thanks
> Brian
> _______________________________________________
> cif-developers mailing list
> cif-developers@iucr.org
> http://scripts.iucr.org/mailman/listinfo/cif-developers
>
_______________________________________________
cif-developers mailing list
cif-developers@iucr.org
http://scripts.iucr.org/mailman/listinfo/cif-developers