Hello Petr, > Yes, we have actually changed the strip_accents function but the > result is that the tokenization has changed and the virtual global > index refuses to fully recognize this. [...]
> When looking directly into idxWORD01F. I see that there are words > created by the old tokenizer. When running bibindex -R -wglobal, the > task finishes almost immediately with "Selected indexes/recIDs are up > to date." Adding the --force parameter did not help either. Maybe it is similar to something that I experienced some years ago. Tibor ended up recomending me to directly truncate the index tables: http://cdsware.cern.ch/lists/project-cdsware-users/archive/msg01014.shtml > $ echo "TRUNCATE idxWORD55F" | /opt/cds-invenio/bin/dbexec > $ echo "TRUNCATE idxWORD55R" | /opt/cds-invenio/bin/dbexec > [...] In your case, change 55 for 01. Hope it helps, Ferran
