Re: metadata only collections and formatting strings

From Katherine Don
DateFri, 20 Dec 2002 11:03:01 +1300
Subject Re: metadata only collections and formatting strings
In-Reply-To (OF718AC095-54AE99ED-ON69256C94-00281735-69256C94-00281746-ntu-edu-au)
hi Stephen

You need to have Identifier metadata for each record - then it will work fine.
We normally generate a hash number for each one - you can use any string you like.
However, we use identifiers 1, 2, 3 etc for a separate purpose so dont just use numbers.
I used "record1", "record2" when I tried your collection. They need to be unique for
each record.
eg
<Metadata name="Identifier">record1</Metadata>

Then it should work fine.

By the way - have you thought about using mgpp instead of mg? You can index all your
metadata in one index, and combine searches on different fields.

change the collect.cfg file to say

buildtype mgpp
indexes metadata
defaultindex metadata

and comment out the old indexes lines.

to just do some specific indexes, do
indexes Title,Creator

cheers,
Katherine

Stephen.DeGabrielle@ntu.edu.au wrote:

> Thank you (_both_) for your help. I think I am having progress in skipping
> the import stage. (I am looking at splitplug and its 'children?' Referplug
> and the one for bibtextPlug - I am still learning perl; but was easier for
> me to write the convertor in AWK).
>
> This is what I have got to so far;
> I have created single doc.xml file (with a subsest of 5 records),
>
> It is only getting the last record; and complains
> "WARNING: AZList::classify called multiple times for NULL" at the end of
> the build, in addition to
>
> I think I am getting there- how do you induce the file to split?
>
> See
> http://aradadesktop/gsdl?site=localhost&a=p&p=about&c=intanmas&ct=0&l=en&w=iso-8859-1
>
> --
> Stephen De Gabrielle
> Digitisation Officer
> AraDA Project
> NTU Library
> +61 8 8946 7009
> http://www.ntu.edu.au/library