Z39.50 Batch Search and Queue
=============================

Staff Work Flow
---------------

 * Staff add records to bib record buckets
 * Staff select the new "Locate Z39.50 Matches" action for a selected bucket.
 * Staff choose which Z39.50 sources and indexes to search and the destination 
   queue.
 * Submitting the search fires a series of parallel Z39.50 searches across
   all selected Z39.50 sources.
 * Matches are added to the selected (Vandelay) queue.
 * Matched records may be manually or automatically overlaid to existing 
   catalog records using the existing Vandelay import/merge/overlay features.

Vandelay Limit to Bucket
------------------------

As a side effect of this feature, Vandelay now has a new option in the 
interface which allows staff to limit which catalog records to which an 
inbound record matches to bib records within a record bucket.  When a record
bucket and match set are chosen, only the records in the bucket can act as
merge/overlay targets for the inbound Vandelay records.

Configuration
-------------

Z39.50 Index Field Maps
~~~~~~~~~~~~~~~~~~~~~~~

These map bib record indexes (Metabib Fields and Record Attributes) to Z39.50
search attributes.  The purpose of the mapping is to allow the server to
determine which values to use for the automated Z39.50 searches.  For example,
if the Z39.50 "title" attribute is mapped to the "Uniform Title" Metabib Field,
the extracted value for "Uniform Title" for each record in the bucket will be
used as the "title" value in the Z39.50 search.

Mappings can be applied to specific Z39.50 attributes, which define an attribute
type and a Z39.50 source, or to generic attribute types (e.g. "title").  When
a specific attribute is used, the mapping will only be applied to searches 
directed at the Z39.50 source linked to the attribute.

The management interface can be found in the staff client under 

Admin => Server Administration => Z39.50 Index Field Maps

Metabib Field Additions
^^^^^^^^^^^^^^^^^^^^^^^

Stock config.metabib_field entries for author include additional author-
related data, like dates.  For example, a value for Personal Author might 
look like this:

'girdlestone cuthbert morton 1895 1975 creator'

In the context of searching Z39.50 servers, the extraneous data can be 
detrimental.  Creating a separate field definition without the extra data
is recommended.

[source,sql]
--------------------------------------------
INSERT INTO config.metabib_field 
    (field_class, name, label, format, xpath, search_field)
    VALUES (
        'author', 
        'personal - trimmed', 
        'Personal Author (trimmed)', 
        'mods32',
        '//mods32:mods/mods32:name[@type=''personal'' and mods32:role/mods32:roleTerm[text()=''creator'']]/mods32:namePart[not (@type)]',
        FALSE
    );

-- FULL BIB (OR INDEX-TARGETED) RE-INGEST REQUIRED
--------------------------------------------


Org Unit Settings
~~~~~~~~~~~~~~~~~

 * cat.z3950.batch.max_parallel 
  ** Maximum Parallel Z39.50 Batch Searches
  ** The maximum number of Z39.50 searches that can be in-flight at any given 
     time when performing batch Z39.50 searches
 * cat.z3950.batch.max_results
  ** Maximum Z39.50 Batch Search Results
  ** The maximum number of search results to retrieve and queue for each 
     record + Z39 source during batch Z39.50 searches