LP#1997485: Include authority headings in symspell dict
authorMike Rylander <mrylander@gmail.com>
Wed, 26 Apr 2023 17:53:45 +0000 (13:53 -0400)
committerGalen Charlton <gmc@equinoxOLI.org>
Thu, 27 Apr 2023 18:00:50 +0000 (18:00 +0000)
This patch adjusts the generation of search suggestions
based on authority headings by:

* Including all indexed authority headings (1XX, 4XX, and 5XX) from
  authority records for author, title, and subject headings in the
  search suggestion dictionary.
* As a consequence, suggestions can be based entirely on headings
  found in authority 4XX and 5XX fields without those headings
  also needing to be directly in at least one bibliographic record.

Signed-off-by: Mike Rylander <mrylander@gmail.com>
Signed-off-by: Galen Charlton <gmc@equinoxOLI.org>

Open-ILS/src/sql/Pg/upgrade/YYYY.schema.DYM-authority-data-dictionary.sql [new file with mode: 0644]

diff --git a/Open-ILS/src/sql/Pg/upgrade/YYYY.schema.DYM-authority-data-dictionary.sql b/Open-ILS/src/sql/Pg/upgrade/YYYY.schema.DYM-authority-data-dictionary.sql
new file mode 100644 (file)
index 0000000..55fd419
--- /dev/null
@@ -0,0 +1,94 @@
+BEGIN;
+
+-- Bring authorized headings into the symspell dictionary. Side
+-- loader should be used for Real Sites.  See below the COMMIT.
+/*
+SELECT  search.symspell_build_and_merge_entries(h.value, m.field_class, NULL)
+  FROM  authority.simple_heading h
+        JOIN authority.control_set_auth_field_metabib_field_map_refs a ON (a.authority_field = a.atag)
+        JOIN config.metabib_field m ON (a.metabib_field=m.id);
+*/
+
+CREATE OR REPLACE FUNCTION search.symspell_maintain_entries () RETURNS TRIGGER AS $f$
+DECLARE
+    search_class    TEXT;
+    new_value       TEXT := NULL;
+    old_value       TEXT := NULL;
+BEGIN
+
+    IF TG_TABLE_SCHEMA = 'authority' THEN
+        SELECT  m.field_class INTO search_class
+          FROM  authority.control_set_auth_field_metabib_field_map_refs a
+                JOIN config.metabib_field m ON (a.metabib_field=m.id)
+          WHERE a.authority_field = NEW.atag;
+
+        IF NOT FOUND THEN
+            RETURN NULL;
+        END IF;
+    ELSE
+        search_class := COALESCE(TG_ARGV[0], SPLIT_PART(TG_TABLE_NAME,'_',1));
+    END IF;
+
+    IF TG_OP IN ('INSERT', 'UPDATE') THEN
+        new_value := NEW.value;
+    END IF;
+
+    IF TG_OP IN ('DELETE', 'UPDATE') THEN
+        old_value := OLD.value;
+    END IF;
+
+    PERFORM * FROM search.symspell_build_and_merge_entries(new_value, search_class, old_value);
+
+    RETURN NULL; -- always fired AFTER
+END;
+$f$ LANGUAGE PLPGSQL;
+
+CREATE TRIGGER maintain_symspell_entries_tgr
+    AFTER INSERT OR UPDATE OR DELETE ON authority.simple_heading
+    FOR EACH ROW EXECUTE PROCEDURE search.symspell_maintain_entries();
+
+COMMIT;
+
+-- Generate symspell sideloader data with authority headings included.
+
+/*
+
+\a
+\t
+
+\o title
+select value from metabib.title_field_entry;
+select  h.value
+  from  authority.simple_heading h
+        join authority.control_set_auth_field_metabib_field_map_refs a on (a.authority_field = a.atag)
+        join config.metabib_field m on (a.metabib_field=m.id and m.field_class='title');
+
+\o author
+select value from metabib.author_field_entry;
+select  h.value
+  from  authority.simple_heading h
+        join authority.control_set_auth_field_metabib_field_map_refs a on (a.authority_field = a.atag)
+        join config.metabib_field m on (a.metabib_field=m.id and m.field_class='author');
+
+\o subject
+select value from metabib.subject_field_entry;
+select  h.value
+  from  authority.simple_heading h
+        join authority.control_set_auth_field_metabib_field_map_refs a on (a.authority_field = a.atag)
+        join config.metabib_field m on (a.metabib_field=m.id and m.field_class='subject');
+
+\o series
+select value from metabib.series_field_entry;
+
+\o identifier
+select value from metabib.identifier_field_entry;
+
+\o keyword
+select value from metabib.keyword_field_entry;
+
+\o
+\a
+\t
+
+*/
+