Re: [OPEN-ILS-DEV] MARC Record Import from txt

[John Craig]

Elliot Voris wrote: Hi, Dev Mailing List   I’ve got a fresh installation of Evergreen 2.1.1, running on a fresh installation of Ubuntu 10.04.3.   I also have a very long list (23,000-something) of MARC records that I need to import into the system. I tested out doing this through the staff client. It worked in smaller batches, but would timeout when trying to do anything substantial. Right now; they’re in MarcXML format, in a .txt file. What’s the best way to import all my MARC records into the database? With some caveats, there's still some good info here: http://open-ils.org/dokuwiki/doku.php?id=evergreen-admin:importing:bibrecords&s[]=marc2bre&s[]=pl Under Evg 2.1.1, the direct_ingest.pl has been superceeded, so assuming your MARC XML records have no namespace identifier in front of the tags (search for "no namespace prefixes" in the link above to find the explanation and examples). You can skp the direct_ingest.pl step, and just do the pg_loader.pl step. The "Adding metarecords..." step is also obsolete. Look at the section "Example: Importing the Project Gutenberg records" to find the way to actually invoke pg_loader.pl and then run the script via psql. It'll still be somewhat slow, but it shouldn't time out on you. John   Thanks!   Elliot J Voris IT Director Saint Louis Christian College   evo...@slcconline.edu 314.837.6777 x 1250   __________ Information from ESET NOD32 Antivirus, version of virus signature database 6697 (20111209) __________ The message was checked by ESET NOD32 Antivirus. http://www.eset.com __________ Information from ESET NOD32 Antivirus, version of virus signature database 6697 (20111209) __________ The message was checked by ESET NOD32 Antivirus. http://www.eset.com