release 5.0.0post8

[nominatim.git] / lib-sql / tokenizer / icu_tokenizer.sql

diff --git a/lib-sql/tokenizer/icu_tokenizer.sql b/lib-sql/tokenizer/icu_tokenizer.sql
index 6092319a0578d338915b6890902d71df7ec90b1f..f0c30f1b786471d210cd96b28fd49fd149487ea2 100644 (file)
--- a/lib-sql/tokenizer/icu_tokenizer.sql
+++ b/lib-sql/tokenizer/icu_tokenizer.sql
@@ -1,3 +1,10 @@
+-- SPDX-License-Identifier: GPL-2.0-only
+--
+-- This file is part of Nominatim. (https://nominatim.org)
+--
+-- Copyright (C) 2022 by the Nominatim developer community.
+-- For a full list of authors see the git log.
+
 -- Get tokens used for searching the given place.
 --
 -- These are the tokens that will be saved in the search_name table.
@@ -5,7 +12,7 @@ CREATE OR REPLACE FUNCTION token_get_name_search_tokens(info JSONB)
   RETURNS INTEGER[]
 AS $$
   SELECT (info->>'names')::INTEGER[]
-$$ LANGUAGE SQL IMMUTABLE STRICT;
+$$ LANGUAGE SQL IMMUTABLE STRICT PARALLEL SAFE;
 
 
 -- Get tokens for matching the place name against others.
@@ -15,7 +22,7 @@ CREATE OR REPLACE FUNCTION token_get_name_match_tokens(info JSONB)
   RETURNS INTEGER[]
 AS $$
   SELECT (info->>'names')::INTEGER[]
-$$ LANGUAGE SQL IMMUTABLE STRICT;
+$$ LANGUAGE SQL IMMUTABLE STRICT PARALLEL SAFE;
 
 
 -- Return the housenumber tokens applicable for the place.
@@ -23,7 +30,7 @@ CREATE OR REPLACE FUNCTION token_get_housenumber_search_tokens(info JSONB)
   RETURNS INTEGER[]
 AS $$
   SELECT (info->>'hnr_tokens')::INTEGER[]
-$$ LANGUAGE SQL IMMUTABLE STRICT;
+$$ LANGUAGE SQL IMMUTABLE STRICT PARALLEL SAFE;
 
 
 -- Return the housenumber in the form that it can be matched during search.
@@ -31,70 +38,77 @@ CREATE OR REPLACE FUNCTION token_normalized_housenumber(info JSONB)
   RETURNS TEXT
 AS $$
   SELECT info->>'hnr';
-$$ LANGUAGE SQL IMMUTABLE STRICT;
+$$ LANGUAGE SQL IMMUTABLE STRICT PARALLEL SAFE;
+
+
+CREATE OR REPLACE FUNCTION token_is_street_address(info JSONB)
+  RETURNS BOOLEAN
+AS $$
+  SELECT info->>'street' is not null or info->>'place' is null;
+$$ LANGUAGE SQL IMMUTABLE PARALLEL SAFE;
 
 
 CREATE OR REPLACE FUNCTION token_has_addr_street(info JSONB)
   RETURNS BOOLEAN
 AS $$
-  SELECT info->>'street' is not null;
-$$ LANGUAGE SQL IMMUTABLE;
+  SELECT info->>'street' is not null and info->>'street' != '{}';
+$$ LANGUAGE SQL IMMUTABLE PARALLEL SAFE;
 
 
 CREATE OR REPLACE FUNCTION token_has_addr_place(info JSONB)
   RETURNS BOOLEAN
 AS $$
   SELECT info->>'place' is not null;
-$$ LANGUAGE SQL IMMUTABLE;
+$$ LANGUAGE SQL IMMUTABLE PARALLEL SAFE;
 
 
 CREATE OR REPLACE FUNCTION token_matches_street(info JSONB, street_tokens INTEGER[])
   RETURNS BOOLEAN
 AS $$
-  SELECT (info->>'street')::INTEGER[] <@ street_tokens
-$$ LANGUAGE SQL IMMUTABLE STRICT;
+  SELECT (info->>'street')::INTEGER[] && street_tokens
+$$ LANGUAGE SQL IMMUTABLE STRICT PARALLEL SAFE;
 
 
 CREATE OR REPLACE FUNCTION token_matches_place(info JSONB, place_tokens INTEGER[])
   RETURNS BOOLEAN
 AS $$
   SELECT (info->>'place')::INTEGER[] <@ place_tokens
-$$ LANGUAGE SQL IMMUTABLE STRICT;
+$$ LANGUAGE SQL IMMUTABLE STRICT PARALLEL SAFE;
 
 
 CREATE OR REPLACE FUNCTION token_addr_place_search_tokens(info JSONB)
   RETURNS INTEGER[]
 AS $$
   SELECT (info->>'place')::INTEGER[]
-$$ LANGUAGE SQL IMMUTABLE STRICT;
+$$ LANGUAGE SQL IMMUTABLE STRICT PARALLEL SAFE;
 
 
 CREATE OR REPLACE FUNCTION token_get_address_keys(info JSONB)
   RETURNS SETOF TEXT
 AS $$
   SELECT * FROM jsonb_object_keys(info->'addr');
-$$ LANGUAGE SQL IMMUTABLE STRICT;
+$$ LANGUAGE SQL IMMUTABLE STRICT PARALLEL SAFE;
 
 
 CREATE OR REPLACE FUNCTION token_get_address_search_tokens(info JSONB, key TEXT)
   RETURNS INTEGER[]
 AS $$
   SELECT (info->'addr'->>key)::INTEGER[];
-$$ LANGUAGE SQL IMMUTABLE STRICT;
+$$ LANGUAGE SQL IMMUTABLE STRICT PARALLEL SAFE;
 
 
 CREATE OR REPLACE FUNCTION token_matches_address(info JSONB, key TEXT, tokens INTEGER[])
   RETURNS BOOLEAN
 AS $$
   SELECT (info->'addr'->>key)::INTEGER[] <@ tokens;
-$$ LANGUAGE SQL IMMUTABLE STRICT;
+$$ LANGUAGE SQL IMMUTABLE STRICT PARALLEL SAFE;
 
 
-CREATE OR REPLACE FUNCTION token_normalized_postcode(postcode TEXT)
+CREATE OR REPLACE FUNCTION token_get_postcode(info JSONB)
   RETURNS TEXT
 AS $$
-  SELECT CASE WHEN postcode SIMILAR TO '%(,|;)%' THEN NULL ELSE upper(trim(postcode))END;
-$$ LANGUAGE SQL IMMUTABLE STRICT;
+  SELECT info->>'postcode';
+$$ LANGUAGE SQL IMMUTABLE STRICT PARALLEL SAFE;
 
 
 -- Return token info that should be saved permanently in the database.
@@ -102,7 +116,7 @@ CREATE OR REPLACE FUNCTION token_strip_info(info JSONB)
   RETURNS JSONB
 AS $$
   SELECT NULL::JSONB;
-$$ LANGUAGE SQL IMMUTABLE STRICT;
+$$ LANGUAGE SQL IMMUTABLE STRICT PARALLEL SAFE;
 
 --------------- private functions ----------------------------------------------
 
@@ -193,3 +207,49 @@ BEGIN
 END;
 $$
 LANGUAGE plpgsql;
+
+
+CREATE OR REPLACE FUNCTION create_analyzed_hnr_id(norm_term TEXT, lookup_terms TEXT[])
+  RETURNS INTEGER
+  AS $$
+DECLARE
+  return_id INTEGER;
+BEGIN
+  SELECT min(word_id) INTO return_id
+    FROM word WHERE word = norm_term and type = 'H';
+
+  IF return_id IS NULL THEN
+    return_id := nextval('seq_word');
+    INSERT INTO word (word_id, word_token, type, word, info)
+      SELECT return_id, lookup_term, 'H', norm_term,
+             json_build_object('lookup', lookup_terms[1])
+        FROM unnest(lookup_terms) as lookup_term;
+  END IF;
+
+  RETURN return_id;
+END;
+$$
+LANGUAGE plpgsql;
+
+CREATE OR REPLACE FUNCTION create_postcode_word(postcode TEXT, lookup_terms TEXT[])
+  RETURNS BOOLEAN
+  AS $$
+DECLARE
+  existing INTEGER;
+BEGIN
+  SELECT count(*) INTO existing
+    FROM word WHERE word = postcode and type = 'P';
+
+  IF existing > 0 THEN
+    RETURN TRUE;
+  END IF;
+
+  -- postcodes don't need word ids
+  INSERT INTO word (word_token, type, word)
+    SELECT lookup_term, 'P', postcode FROM unnest(lookup_terms) as lookup_term;
+
+  RETURN FALSE;
+END;
+$$
+LANGUAGE plpgsql;
+