1st commit done. It contains : * updater/updatedatabase : get rid with marc_* tables, but DON'T remove them. As a lot of things uses them, it would not be a good idea for instance to drop them. If you really want to play, you can rename them to test head without them but being still able to reintroduce them... * Biblio.pm : modify MARCgetbiblio to find the raw marc record in biblioitems.marc field, not from marc_subfield_table, modify MARCfindframeworkcode to find frameworkcode in biblio.frameworkcode, modify some other subs to use biblio.biblionumber & get rid of bibid. * other files : get rid of bibid and use biblionumber instead. What is broken : * does not do anything on zebra yet. * if you rename marc_subfield_table, you can't search anymore. * you can view a biblio & bibliodetails, go to MARC editor, but NOT save any modif. * don't try to add a biblio, it would add data poorly... (don't try to delete either, it may work, but that would be a surprise ;-) ) IMPORTANT NOTE : you need MARC::XML package (http://search.cpan.org/~esummers/MARC-XML-0.7/lib/MARC/File/XML.pm), that requires a recent version of MARC::Record Updatedatabase stores the iso2709 data in biblioitems.marc field & an xml version in biblioitems.marcxml Not sure we will keep it when releasing the stable version, but I think it's a good idea to have something readable in sql, at least for development stage. -- Paul POULAIN Consultant indépendant en logiciels libres responsable francophone de koha (SIGB libre http://www.koha-fr.org)