Ahhhhh…..I’d introduced a bug in the last build that primarily affected European diacritics when moving between MARC-8 to UTF-8. The issue at hand was how to deal with dangling or incorrectly coded diacritics. In coding a solution, I accidently removed some code that moved combining diacritics. In MARC-8, diacritic placement for most European diacritics should be: [diacritic][character being modified]. When moved into UTF-8, the characterset sequence changes to: [character-being modified][diacritic]. Well, in writing code to capture incomplete diacritics, I’d funzed up the logic that moves the characters (internally, I was clearing a tracking variable when I shouldn’t). Anyway, that should be corrected. The updated download can be found at: marcedit50_2005_07_25.zip.
With that said, there are two charactersets that need to be tested to make sure that they are working as documented. Both the Extended Arabic and Greek charactersets have combining characters. I’ve added the logic to the UTF-8 to MARC-8 converter to accommedate these combinations — but I need to find some sample records to make sure that this is working like I’m intending.
Z39.50 — just about finished. I just need to finish converting my Z39.50 database into XML and adding support for multiple connection formats in the metadata.
–Terry