[Koha-devel] Cataloguing authorities 3.6/3.8

[Paul]

At 10:22 AM 10/1/2012 +1300, Robin Sheat wrote:
>Paul schreef op za 29-09-2012 om 17:54 [-0400]:
> > 17:32:00-29/09 zebraidx(2632) [warn] isamb: Inconsistent register (2)
> > zebraidx: isamb.c:1102: insert_leaf: Assertion `*lookahead_mode'
> > failed.
> > 17:32:00-29/09 zebraidx(2637) [warn] previous transaction didn't reach
> > commit
> >
> > that are "novel."
> >
> > Any suggestions, thoughts?
>
>Those errors are really annoying.
>
>Have a look to see if there's something funny about the MARC for biblio
>17765. Especially missing 999$c and $d (I think.)

I've been following up on on Robin's suggestion looking for "something 
funny" in our records that might explain current indexing (or at least 
reassure me that it's not a db rather than system error) and wrote a script 
to verify our inhouse barcodes:

mysql> select items.barcode, items.biblionumber, biblio.title, 
biblio.author, items.itemcallnumber from items join biblio on 
(items.biblionumber = biblio.biblionumber) where items.barcode NOT REGEXP 
'^[[:digit:]]{1}[[:alnum:]]{4}[[:digit:]]{8}$';
+------------------+--------------+----------------------+-------------------------+----------------+
| barcode          | biblionumber | title                | 
author                  | itemcallnumber |
+------------------+--------------+----------------------+-------------------------+----------------+
| 1AERO9107_221230 |        16603 | Heroes of the air,   | Fraser, Chelsea 
Curtis, | AERO-FRAS-1    |
| 1AERO111564      |        17408 | Fights and flights / | Samson, Charles 
Rumney, | AERO-SAMS-1    |
+------------------+--------------+----------------------+-------------------------+----------------+
2 rows in set (0.02 sec)

However, these are totally unfindable in Koha (by barcode, biblionumber, 
callnumber, etc) via any type of search (normal, advanced etc) [N.B. we 
have these two items under a different biblio, call, barcode, all valid and 
searchable.]

But, I can use 'Tools ==> Export bibliographic and holdings' and produce 
what appear to be perfectly valid Marc21 records

Could these 2 be (I'm wildly guessing) the reason for the "[warn] isamb: 
Inconsistent register (2)"?

Does anyone have a ready made method to delete a complete record in 
Mysql?  They're unwanted elements -- but how to remove them?

tnx and br -- paul