X-Git-Url: http://git.equinoxoli.org/?p=migration-tools.git;a=blobdiff_plain;f=sql%2Fbase%2Fbase.sql;h=0e6c24afffaf40f591a932c9d4e148b932602491;hp=30951e0a3b1d13d33f7aa2fe9813dd67fa06e0c9;hb=2463e7151e631cb22c8032053ba246bcb7da59e7;hpb=4622e6339ce2a078cdb7ea128cc2a41fe803e50e diff --git a/sql/base/base.sql b/sql/base/base.sql index 30951e0..0e6c24a 100644 --- a/sql/base/base.sql +++ b/sql/base/base.sql @@ -244,7 +244,7 @@ CREATE OR REPLACE FUNCTION migration_tools.build_specific_base_staging_table (TE BEGIN base_staging_table = REPLACE( production_table, '.', '_' ); --RAISE INFO 'In migration_tools.build_specific_base_staging_table(%,%) -> %', migration_schema, production_table, base_staging_table; - PERFORM migration_tools.exec( $1, 'CREATE UNLOGGED TABLE ' || migration_schema || '.' || base_staging_table || ' ( LIKE ' || production_table || ' INCLUDING DEFAULTS EXCLUDING CONSTRAINTS );' ); + PERFORM migration_tools.exec( $1, 'CREATE TABLE ' || migration_schema || '.' || base_staging_table || ' ( LIKE ' || production_table || ' INCLUDING DEFAULTS EXCLUDING CONSTRAINTS );' ); PERFORM migration_tools.exec( $1, ' INSERT INTO ' || migration_schema || '.fields_requiring_mapping SELECT table_schema, table_name, column_name, data_type @@ -272,7 +272,7 @@ CREATE OR REPLACE FUNCTION migration_tools.create_linked_legacy_table_from (TEXT column_list TEXT := ''; column_count INTEGER := 0; BEGIN - create_sql := 'CREATE UNLOGGED TABLE ' || migration_schema || '.' || parent_table || '_legacy ( '; + create_sql := 'CREATE TABLE ' || migration_schema || '.' || parent_table || '_legacy ( '; FOR columns IN SELECT table_schema, table_name, column_name, data_type FROM information_schema.columns @@ -573,6 +573,98 @@ CREATE OR REPLACE FUNCTION migration_tools.name_parse_out_fuller_last_first_midd END; $$ LANGUAGE PLPGSQL STRICT IMMUTABLE; +CREATE OR REPLACE FUNCTION migration_tools.name_parse_out_fuller_last_first_middle_and_random_affix2 (TEXT) RETURNS TEXT[] AS $$ + DECLARE + full_name TEXT := $1; + temp TEXT; + family_name TEXT := ''; + first_given_name TEXT := ''; + second_given_name TEXT := ''; + suffix TEXT := ''; + prefix TEXT := ''; + BEGIN + temp := BTRIM(full_name); + -- Use values, not structure, for prefix/suffix, unless we come up with a better idea + --IF temp ~ '^\S{2,}\.' THEN + -- prefix := REGEXP_REPLACE(temp, '^(\S{2,}\.).*$','\1'); + -- temp := BTRIM(REGEXP_REPLACE(temp, '^\S{2,}\.(.*)$','\1')); + --END IF; + --IF temp ~ '\S{2,}\.$' THEN + -- suffix := REGEXP_REPLACE(temp, '^.*(\S{2,}\.)$','\1'); + -- temp := REGEXP_REPLACE(temp, '^(.*)\S{2,}\.$','\1'); + --END IF; + IF temp ilike '%MR.%' THEN + prefix := 'Mr.'; + temp := BTRIM(REGEXP_REPLACE( temp, E'MR\.\\s*', '', 'i' )); + END IF; + IF temp ilike '%MRS.%' THEN + prefix := 'Mrs.'; + temp := BTRIM(REGEXP_REPLACE( temp, E'MRS\.\\s*', '', 'i' )); + END IF; + IF temp ilike '%MS.%' THEN + prefix := 'Ms.'; + temp := BTRIM(REGEXP_REPLACE( temp, E'MS\.\\s*', '', 'i' )); + END IF; + IF temp ilike '%DR.%' THEN + prefix := 'Dr.'; + temp := BTRIM(REGEXP_REPLACE( temp, E'DR\.\\s*', '', 'i' )); + END IF; + IF temp ilike '%JR.%' THEN + suffix := 'Jr.'; + temp := BTRIM(REGEXP_REPLACE( temp, E'JR\.\\s*', '', 'i' )); + END IF; + IF temp ilike '%JR,%' THEN + suffix := 'Jr.'; + temp := BTRIM(REGEXP_REPLACE( temp, E'JR,\\s*', ',', 'i' )); + END IF; + IF temp ilike '%SR.%' THEN + suffix := 'Sr.'; + temp := BTRIM(REGEXP_REPLACE( temp, E'SR\.\\s*', '', 'i' )); + END IF; + IF temp ilike '%SR,%' THEN + suffix := 'Sr.'; + temp := BTRIM(REGEXP_REPLACE( temp, E'SR,\\s*', ',', 'i' )); + END IF; + IF temp like '%III%' THEN + suffix := 'III'; + temp := BTRIM(REGEXP_REPLACE( temp, E'III', '' )); + END IF; + IF temp like '%II%' THEN + suffix := 'II'; + temp := BTRIM(REGEXP_REPLACE( temp, E'II', '' )); + END IF; + + IF temp ~ ',' THEN + family_name = BTRIM(REGEXP_REPLACE(temp,'^(.*?,).*$','\1')); + temp := BTRIM(REPLACE( temp, family_name, '' )); + family_name := REPLACE( family_name, ',', '' ); + IF temp ~ ' ' THEN + first_given_name := BTRIM( REGEXP_REPLACE(temp,'^(.+)\s(.+)$','\1') ); + second_given_name := BTRIM( REGEXP_REPLACE(temp,'^(.+)\s(.+)$','\2') ); + ELSE + first_given_name := temp; + second_given_name := ''; + END IF; + ELSE + IF temp ~ '^\S+\s+\S+\s+\S+$' THEN + first_given_name := BTRIM( REGEXP_REPLACE(temp,'^(\S+)\s*(\S+)\s*(\S+)$','\1') ); + second_given_name := BTRIM( REGEXP_REPLACE(temp,'^(\S+)\s*(\S+)\s*(\S+)$','\2') ); + family_name := BTRIM( REGEXP_REPLACE(temp,'^(\S+)\s*(\S+)\s*(\S+)$','\3') ); + ELSE + first_given_name := BTRIM( REGEXP_REPLACE(temp,'^(\S+)\s*(\S+)$','\1') ); + second_given_name := temp; + family_name := BTRIM( REGEXP_REPLACE(temp,'^(\S+)\s*(\S+)$','\2') ); + END IF; + END IF; + + family_name := BTRIM(REPLACE(REPLACE(family_name,',',''),'"','')); + first_given_name := BTRIM(REPLACE(REPLACE(first_given_name,',',''),'"','')); + second_given_name := BTRIM(REPLACE(REPLACE(second_given_name,',',''),'"','')); + + RETURN ARRAY[ family_name, prefix, first_given_name, second_given_name, suffix ]; + END; +$$ LANGUAGE PLPGSQL STRICT IMMUTABLE; + CREATE OR REPLACE FUNCTION migration_tools.address_parse_out_citystatezip (TEXT) RETURNS TEXT[] AS $$ DECLARE city_state_zip TEXT := $1; @@ -688,7 +780,7 @@ CREATE OR REPLACE FUNCTION migration_tools.parse_out_address2 (TEXT) RETURNS TEX $$ LANGUAGE PLPERLU STABLE; DROP TABLE IF EXISTS migration_tools.usps_suffixes; -CREATE UNLOGGED TABLE migration_tools.usps_suffixes ( suffix_from TEXT, suffix_to TEXT ); +CREATE TABLE migration_tools.usps_suffixes ( suffix_from TEXT, suffix_to TEXT ); INSERT INTO migration_tools.usps_suffixes VALUES ('ALLEE','ALY'), ('ALLEY','ALY'), @@ -3451,6 +3543,27 @@ CREATE OR REPLACE FUNCTION migration_tools.get_marc_tags (TEXT, TEXT, TEXT, TEXT return \@texts; $$ LANGUAGE PLPERLU STABLE; +CREATE OR REPLACE FUNCTION migration_tools.get_marc_tags_filtered (TEXT, TEXT, TEXT, TEXT, TEXT) RETURNS TEXT[] AS $$ + my ($marcxml, $tag, $subfield, $delimiter, $match) = @_; + + use MARC::Record; + use MARC::File::XML; + use MARC::Field; + + my @fields; + eval { + my $marc = MARC::Record->new_from_xml($marcxml, 'UTF-8'); + @fields = $marc->field($tag); + }; + my @texts; + foreach my $field (@fields) { + if ($field->as_string() =~ qr/$match/) { + push @texts, $field->as_string($subfield,$delimiter); + } + } + return \@texts; +$$ LANGUAGE PLPERLU STABLE; + CREATE OR REPLACE FUNCTION migration_tools.find_hold_matrix_matchpoint (INTEGER) RETURNS INTEGER AS $$ SELECT action.find_hold_matrix_matchpoint( (SELECT pickup_lib FROM action.hold_request WHERE id = $1), @@ -4248,6 +4361,135 @@ return $marc_xml->as_xml_record(); $function$; +DROP FUNCTION IF EXISTS migration_tools.owner_change_sf9_substring_match(TEXT,TEXT,TEXT,TEXT); +CREATE OR REPLACE FUNCTION migration_tools.owner_change_sf9_substring_match (marc_xml TEXT, substring_old_value TEXT, new_value TEXT, fix_indicators TEXT) + RETURNS TEXT + LANGUAGE plperlu +AS $function$ +use strict; +use warnings; + +use MARC::Record; +use MARC::File::XML (BinaryEncoding => 'utf8'); + +binmode(STDERR, ':bytes'); +binmode(STDOUT, ':utf8'); +binmode(STDERR, ':utf8'); + +my $marc_xml = shift; +my $substring_old_value = shift; +my $new_value = shift; +my $fix_indicators = shift; + +$marc_xml =~ s/(.........)./${1}a/; + +eval { + $marc_xml = MARC::Record->new_from_xml($marc_xml); +}; +if ($@) { + #elog("could not parse $bibid: $@\n"); + import MARC::File::XML (BinaryEncoding => 'utf8'); + return $marc_xml; +} + +my @uris = $marc_xml->field('856'); +return $marc_xml->as_xml_record() unless @uris; + +foreach my $field (@uris) { + my $ind1 = $field->indicator('1'); + if (defined $ind1) { + if ($ind1 ne '1' && $ind1 ne '4' && $fix_indicators eq 'true') { + $field->set_indicator(1,'4'); + } + } + my $ind2 = $field->indicator('2'); + if (defined $ind2) { + if ($ind2 ne '0' && $ind2 ne '1' && $fix_indicators eq 'true') { + $field->set_indicator(2,'0'); + } + } + if ($field->as_string('9') =~ qr/$substring_old_value/) { + $field->delete_subfield('9'); + $field->add_subfields( '9' => $new_value ); + } + $marc_xml->delete_field($field); # -- we're going to dedup and add them back +} + +my %hash = (map { ($_->as_usmarc => $_) } @uris); # -- courtesy of an old Mike Rylander post :-) +$marc_xml->insert_fields_ordered( values( %hash ) ); + +return $marc_xml->as_xml_record(); + +$function$; + +DROP FUNCTION IF EXISTS migration_tools.owner_change_sf9_substring_match2(TEXT,TEXT,TEXT,TEXT,TEXT); +CREATE OR REPLACE FUNCTION migration_tools.owner_change_sf9_substring_match2 (marc_xml TEXT, qualifying_match TEXT, substring_old_value TEXT, new_value TEXT, fix_indicators TEXT) + RETURNS TEXT + LANGUAGE plperlu +AS $function$ +use strict; +use warnings; + +use MARC::Record; +use MARC::File::XML (BinaryEncoding => 'utf8'); + +binmode(STDERR, ':bytes'); +binmode(STDOUT, ':utf8'); +binmode(STDERR, ':utf8'); + +my $marc_xml = shift; +my $qualifying_match = shift; +my $substring_old_value = shift; +my $new_value = shift; +my $fix_indicators = shift; + +$marc_xml =~ s/(.........)./${1}a/; + +eval { + $marc_xml = MARC::Record->new_from_xml($marc_xml); +}; +if ($@) { + #elog("could not parse $bibid: $@\n"); + import MARC::File::XML (BinaryEncoding => 'utf8'); + return $marc_xml; +} + +my @unqualified_uris = $marc_xml->field('856'); +my @uris = (); +foreach my $field (@unqualified_uris) { + if ($field->as_string() =~ qr/$qualifying_match/) { + push @uris, $field; + } +} +return $marc_xml->as_xml_record() unless @uris; + +foreach my $field (@uris) { + my $ind1 = $field->indicator('1'); + if (defined $ind1) { + if ($ind1 ne '1' && $ind1 ne '4' && $fix_indicators eq 'true') { + $field->set_indicator(1,'4'); + } + } + my $ind2 = $field->indicator('2'); + if (defined $ind2) { + if ($ind2 ne '0' && $ind2 ne '1' && $fix_indicators eq 'true') { + $field->set_indicator(2,'0'); + } + } + if ($field->as_string('9') =~ qr/$substring_old_value/) { + $field->delete_subfield('9'); + $field->add_subfields( '9' => $new_value ); + } + $marc_xml->delete_field($field); # -- we're going to dedup and add them back +} + +my %hash = (map { ($_->as_usmarc => $_) } @uris); # -- courtesy of an old Mike Rylander post :-) +$marc_xml->insert_fields_ordered( values( %hash ) ); + +return $marc_xml->as_xml_record(); + +$function$; + -- strip marc tag DROP FUNCTION IF EXISTS migration_tools.strip_tag(TEXT,TEXT); CREATE OR REPLACE FUNCTION migration_tools.strip_tag(marc TEXT, tag TEXT) @@ -5068,3 +5310,86 @@ BEGIN RETURN TRUE; END $function$; + + +-- convenience function for handling item barcode collisions in asset_copy_legacy + +CREATE OR REPLACE FUNCTION migration_tools.handle_asset_barcode_collisions(migration_schema TEXT) RETURNS VOID AS $function$ +DECLARE + x_barcode TEXT; + x_id BIGINT; + row_count NUMERIC; + internal_collision_count NUMERIC := 0; + incumbent_collision_count NUMERIC := 0; +BEGIN + FOR x_barcode IN SELECT barcode FROM asset_copy_legacy WHERE x_migrate GROUP BY 1 HAVING COUNT(*) > 1 + LOOP + FOR x_id IN SELECT id FROM asset_copy WHERE barcode = x_barcode + LOOP + UPDATE asset_copy SET barcode = migration_schema || '_internal_collision_' || id || '_' || barcode WHERE id = x_id; + GET DIAGNOSTICS row_count = ROW_COUNT; + internal_collision_count := internal_collision_count + row_count; + END LOOP; + END LOOP; + RAISE INFO '% internal collisions', internal_collision_count; + FOR x_barcode IN SELECT a.barcode FROM asset.copy a, asset_copy_legacy b WHERE x_migrate AND a.deleted IS FALSE AND a.barcode = b.barcode + LOOP + FOR x_id IN SELECT id FROM asset_copy_legacy WHERE barcode = x_barcode + LOOP + UPDATE asset_copy_legacy SET barcode = migration_schema || '_incumbent_collision_' || id || '_' || barcode WHERE id = x_id; + GET DIAGNOSTICS row_count = ROW_COUNT; + incumbent_collision_count := incumbent_collision_count + row_count; + END LOOP; + END LOOP; + RAISE INFO '% incumbent collisions', incumbent_collision_count; +END +$function$ LANGUAGE plpgsql; + +-- convenience function for handling patron barcode/usrname collisions in actor_usr_legacy +-- this should be ran prior to populating actor_card + +CREATE OR REPLACE FUNCTION migration_tools.handle_actor_barcode_collisions(migration_schema TEXT) RETURNS VOID AS $function$ +DECLARE + x_barcode TEXT; + x_id BIGINT; + row_count NUMERIC; + internal_collision_count NUMERIC := 0; + incumbent_barcode_collision_count NUMERIC := 0; + incumbent_usrname_collision_count NUMERIC := 0; +BEGIN + FOR x_barcode IN SELECT usrname FROM actor_usr_legacy WHERE x_migrate GROUP BY 1 HAVING COUNT(*) > 1 + LOOP + FOR x_id IN SELECT id FROM actor_usr_legacy WHERE x_migrate AND usrname = x_barcode + LOOP + UPDATE actor_usr_legacy SET usrname = migration_schema || '_internal_collision_' || id || '_' || usrname WHERE id = x_id; + GET DIAGNOSTICS row_count = ROW_COUNT; + internal_collision_count := internal_collision_count + row_count; + END LOOP; + END LOOP; + RAISE INFO '% internal usrname/barcode collisions', internal_collision_count; + + FOR x_barcode IN + SELECT a.barcode FROM actor.card a, actor_usr_legacy b WHERE x_migrate AND a.barcode = b.usrname + LOOP + FOR x_id IN SELECT DISTINCT id FROM actor_usr_legacy WHERE x_migrate AND usrname = x_barcode + LOOP + UPDATE actor_usr_legacy SET usrname = migration_schema || '_incumbent_barcode_collision_' || id || '_' || usrname WHERE id = x_id; + GET DIAGNOSTICS row_count = ROW_COUNT; + incumbent_barcode_collision_count := incumbent_barcode_collision_count + row_count; + END LOOP; + END LOOP; + RAISE INFO '% incumbent barcode collisions', incumbent_barcode_collision_count; + + FOR x_barcode IN + SELECT a.usrname FROM actor.usr a, actor_usr_legacy b WHERE x_migrate AND a.deleted IS FALSE AND a.usrname = b.usrname + LOOP + FOR x_id IN SELECT DISTINCT id FROM actor_usr_legacy WHERE x_migrate AND usrname = x_barcode + LOOP + UPDATE actor_usr_legacy SET usrname = migration_schema || '_incumbent_usrname_collision_' || id || '_' || usrname WHERE id = x_id; + GET DIAGNOSTICS row_count = ROW_COUNT; + incumbent_usrname_collision_count := incumbent_usrname_collision_count + row_count; + END LOOP; + END LOOP; + RAISE INFO '% incumbent usrname collisions (post barcode collision munging)', incumbent_usrname_collision_count; +END +$function$ LANGUAGE plpgsql;