[Koha-devel] koha - length of records - separation of items from biblios

[LAURENT Henri-Damien]

Le 01/03/2011 23:04, Mike Hafen a écrit :
> I had an idea about this recently.  Perhaps it is possible to have in
> the items table a MARCXML column which stores the MARC just for the
> items, and have that be seperate from the biblio's MARCXML.  Then we the
> record is sent to the zebra server combine the two, so that zebra still
> has all the information we would like it to have.
There is an enhancement in bugzilla it is referenced as bug 5579 for
that and biblibre publicized a branch which is also pushed on
git.koha-community.org as new/bug_5579.

It is not exactly managing that removal the way you propose. In fact, it
builds on demand the marcfield for items.

But it is working pretty nicely, although you have to remove all the
items from marcxml (there is a script for that) before you use since on
MARCdetail.pl, it is taking items from items table + from marcxml (if
you havenot previously removed)

The improvement in circulation speed is really good.
But it would need thorough tests... If anyone is interested.
Be advised this branch should NOT be used in production at the moment.

-- 
Henri-Damein LAURENT

> 
> On Tue, Mar 1, 2011 at 10:31 AM, MJ Ray <mjr at phonecoop.coop
> <mailto:mjr at phonecoop.coop>> wrote:
> 
>     Linda Culberson asked:
>     > There is an "abandoned RFC" for "Stop copying item records to
>     bib's MARC
>     > and MARCXML RFC" .
>     >
>     > Could someone tell me why this was abandoned, because I think
>     having the
>     > item record data exists both in the items table and as MARC fields in
>     > the bib record is an a terrible shortcoming of Koha for academic and
>     > research libraries with large numbers of items for certain titles.
> 
>     My notes say that RFC was created by a LibLime worker, so a current or
>     former LibLime worker may have a definitive answer... BUT:
> 
>     I suspect it may have been abandoned because one of the MARC fields is
>     sent to the Zebra index server and there are benefits like faster
>     availability searches if the item data is indexed by Zebra.
> 
>     Hope that helps,
>     --
>     MJ Ray (slef), member of www.software.coop
>     <http://www.software.coop>, a for-more-than-profit co-op.
>     http://koha-community.org supporter, web and LMS developer,
>     statistician.
>     In My Opinion Only: see http://mjr.towers.org.uk/email.html
>     Available for hire for Koha work http://www.software.coop/products/koha