I figure something like this might be generally useful to anybody
using Solr, MARC and/or SolrMarc, so here it is:
I figure this script could also be a jumping off point for a script to
clean up bad MARC records that MARC4J rejects -- ruby-marc has a much
more lenient parser.
I don't actually have a setup where I can confirm if this script
actually works (!!) but if somebody wanted to try it out, there's
nothing here that could break anything (well, depending on how many
records are missing -- your system could slow down, I suppose).
Thanks,
-Ross.