Discussion List Archives

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Proposed DDLm change

Dear CIF developers,

The following message concerns software written to ingest CIF DDLm dictionaries, if your software does not do this you can ignore this message.

DDLm introduced a category called "dictionary_valid" (hereafter "dv" for brevity), which specifies which DDLm attributes are allowed in which scopes.  This category appears only in "ddl.dic", the file specifying the DDLm language.  The key for the category is a synthetic list constructed out of _dv.scope and _dv.option: dv.application = [_dv.scope , _dv.option].

I would like to propose replacing this synthetic key with the natural key consisting simply of the two data names _dv.scope and _dv.option.

The reason for doing this is simplicity: there is no need for the list to be constructed, no new information is provided by constructing the list compared to simply specifying that the two datanames form a key, and software is forced to take a further step to deconstruct the list in order to interpret the contents individually. In addition, as _dv.option has a default value of "Recommended", its explicit value should be included in the dv loop in any case to avoid the contradictions that would arise if it were absent.

Note that the only implications of this change would be for software that ingests ddl.dic in order to validate DDLm dictionaries. If you are an author of such software, please provide feedback on whether or not you are happy to support this change. Note that we have performed the same operation in many places for the core DDLm dictionary.

best wishes,
James.
--
T +61 (02) 9717 9907
F +61 (02) 9717 3145
M +61 (04) 0249 4148
_______________________________________________
cif-developers mailing list
cif-developers@iucr.org
http://mailman.iucr.org/cgi-bin/mailman/listinfo/cif-developers

Reply to: [list | sender only]