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

[ddlm-group] Feedback on draft CIF2 specification from JohnBollinger

Dear all,

If you haven't seen John Bollinger's useful feedback on our draft CIF2 specification, I recommend that you read it at

http://www.iucr.org/__data/iucr/lists/cif-developers/msg00269.html

My response is as follows:

Points 1-8,10-13, 16, 19,23-26: editorial points with which I agree
Point 9: if the CIF2 header characters are missing, I suggest the file is treated as a CIF1 file.  This should be clarified in the spec.
Point 14: matching with the global, save, data etc. keywords should be case-insensitive.  This should be clarified in the spec
Point 15: (disallowing delimiters in strings).  Despite John's objections, I think this is a useful simplification which helps newcomers.  I believe that the PDB has deliberately started 'normalising' their files to remove embedded delimiters, which indicates to me that embedded delimiters are not a useful CIF1 feature.
Point 17,20: Yes, whitespace is allowed between opening/closing brackets and the first/last element.  Should be clarified
Point 18,22: semicolon delimited strings should be allowed as list and table elements.  Clarify in spec.
Point 21: while triple-quote delimited strings as table labels should be OK, allowing non-delimited strings as labels would require further restricting the character set of non-delimited strings to avoid e.g. <semicolon> appearing and confusing the parser (as we have discussed previously).  Using quotes just makes those sort of problems go away

Do others have any further comments?

James.
--
T +61 (02) 9717 9907
F +61 (02) 9717 3145
M +61 (04) 0249 4148
_______________________________________________
ddlm-group mailing list
ddlm-group@iucr.org
http://scripts.iucr.org/mailman/listinfo/ddlm-group

Reply to: [list | sender only]