... but the 1st commits will arrive in a few hours, As I hack the code, I see that the field biblio.biblionumber can be called : * biblionumber * bib * oldbiblionumber * bibnum I plan to normalize & use only biblionumber that is more logic. Don't be surprised, if in the next days/weeks, the cvs head is badly broken. (that's why we have a 2.2 branch ;-) ) Every time I commit some code that breaks stability but is a step to 2.2, i'll throw a mail to koha-devel to explain what the commit does & breaks (at least what i'm sure it breaks ;-) ). My plan before the end of the week is (nothing commited yet) : 1/ modify SQL structure to get rid of marc_biblio / marc_subfield_table / marc_word => DONE 2/ modify Biblio.pm to handle biblios in the new structure => 10% done 3/ modify SearchMarc.pm to enable search on the new structure 4/ provide some zebra config files. of course, that will be highly unstable & only for true hackers, the one with a beard that love having bugs :o) -- Paul POULAIN Consultant indépendant en logiciels libres responsable francophone de koha (SIGB libre http://www.koha-fr.org)