[Koha] Something new with bulkmarcimport.pl

[Andro Gagne]

Stephen,

         At 07:53 PM 1/5/06 -0500, you wrote:

>Andro, I've forgotten what Koha version you are using.

         I'm not at the museum as I write this.  If I remember correctly, 
I'm using ver. 2.2.2.

>   The Biblio.pm line
>numbers you list don't match my (2.2.4) Biblio.pm.
>
>In any case, I would guess that you have some MARC records in the file
>that have errors.

         That's what I was thinking at first, of course.  It looked like I 
was getting a steady stream of messages saying "DBD::mysql::db do failed: 
Access denied for user: 'myuserid at localhost' to
database 'libcat' at /usr/local/koha/intranet/modules/C4/Biblio.pm line 
331. [or 426.]" on the screen.  It looked like there were hundreds of lines 
of these messages--so many that I could not scroll back to the first 
one.  But when I searched for records, I found that there were quite a few 
records in the database.  And this was the first time I had attempted 
importing MARC records.  To confuse matters more, the last line of the 
script's output, "1572 MARC record done in 731.838590145111 seconds," made 
me wonder whether the script was telling me 1,572 records had been imported 
successfully or just telling me that 1,572 records had been processed, 
giving me no information about successful imports.  As this first import of 
records was done without mapping item information in the MARC records to 
the Koha database's item field, I could not get a report of the number of 
items in the OPAC.  I'm hoping I can use bulkmarcimport.pl to delete all of 
the current records in the OPAC database so I can properly map the item 
information tag(s) and re-import the records.  If I remember correctly, 
bulkmarcimport.pl has a switch that will allow me to delete the records in 
the OPAC.  If you have any advice for me on this point, I'd be very pleased 
to hear it.

>   There is another script in the intranet/scripts/misc
>directory called dumpmarc.pl that will simply list out your MARC file. But
>with 1500+ records, it wouldn't be much fun looking for stray errors!

         I am familiar with it, as I used it to look over a number of 
records to make sure they looked like they were in good shape before trying 
to import them.  However, as you can imagine, I did not examine all 1,572 
of them.  To resolve my issue with bulkmarcimport.pl, what I may need to do 
is to get a batch of only 10 records in a file, examine them in 
dumpmarc.pl, and then import them with bulkmarcimport.pl.  If I get the 
same kind of screen output from bulkmarcimport.pl with those records, I 
will have a much more manageable situation to deal with.

>You could also use this script to find the title of the last record in the
>file.  Do a search on that title and the Biblionumber on the screen should
>equal the number of records in your catalog -- assuming you haven't yet
>been doing any deleting of records.

         That's a very good suggestion, as I haven't deleted any of the 
imported records.  Thanks.

Best regards,

Andro P. Gagné
P.O. Box 715
Canandaigua, NY 14424