Merge pull request #3099 from lonvia/determine-place-address-from-tokenizer

[nominatim.git] / lib-sql / functions / placex_triggers.sql

diff --git a/lib-sql/functions/placex_triggers.sql b/lib-sql/functions/placex_triggers.sql
index 37b57b6467ac01f17bf4014b161cec5490cd4831..a252e9aacd1d4d9de035b4884b935ce92b850fe7 100644 (file)
--- a/lib-sql/functions/placex_triggers.sql
+++ b/lib-sql/functions/placex_triggers.sql
@@ -16,7 +16,9 @@ CREATE TYPE prepare_update_info AS (
   country_code TEXT,
   class TEXT,
   type TEXT,
   country_code TEXT,
   class TEXT,
   type TEXT,

-  linked_place_id BIGINT
+  linked_place_id BIGINT,
+  centroid_x float,
+  centroid_y float

 );
 
 -- Retrieve the data needed by the indexer for updating the place.
 );
 
 -- Retrieve the data needed by the indexer for updating the place.


@@ -47,7 +49,7 @@ BEGIN
                and rank_search = 30 AND ST_GeometryType(geometry) in ('ST_Polygon','ST_MultiPolygon')
          LIMIT 1;
     ELSE
                and rank_search = 30 AND ST_GeometryType(geometry) in ('ST_Polygon','ST_MultiPolygon')
          LIMIT 1;
     ELSE

-      -- See if we can inherit addtional address tags from an interpolation.
+      -- See if we can inherit additional address tags from an interpolation.

       -- These will become permanent.
       FOR location IN
         SELECT (address - 'interpolation'::text - 'housenumber'::text) as address
       -- These will become permanent.
       FOR location IN
         SELECT (address - 'interpolation'::text - 'housenumber'::text) as address


@@ -71,6 +73,8 @@ BEGIN
   result.type := p.type;
   result.country_code := p.country_code;
   result.rank_address := p.rank_address;
   result.type := p.type;
   result.country_code := p.country_code;
   result.rank_address := p.rank_address;

+  result.centroid_x := ST_X(p.centroid);
+  result.centroid_y := ST_Y(p.centroid);

 
   -- Names of linked places need to be merged in, so search for a linkable
   -- place already here.
 
   -- Names of linked places need to be merged in, so search for a linkable
   -- place already here.


@@ -103,12 +107,17 @@ LANGUAGE plpgsql STABLE;
 
 
 CREATE OR REPLACE FUNCTION find_associated_street(poi_osm_type CHAR(1),
 
 
 CREATE OR REPLACE FUNCTION find_associated_street(poi_osm_type CHAR(1),

-                                                  poi_osm_id BIGINT)
+                                                  poi_osm_id BIGINT,
+                                                  bbox GEOMETRY)

   RETURNS BIGINT
   AS $$
 DECLARE
   location RECORD;
   parent RECORD;
   RETURNS BIGINT
   AS $$
 DECLARE
   location RECORD;
   parent RECORD;

+  result BIGINT;
+  distance FLOAT;
+  new_distance FLOAT;
+  waygeom GEOMETRY;

 BEGIN
   FOR location IN
     SELECT members FROM planet_osm_rels
 BEGIN
   FOR location IN
     SELECT members FROM planet_osm_rels


@@ -119,19 +128,34 @@ BEGIN
     FOR i IN 1..array_upper(location.members, 1) BY 2 LOOP
       IF location.members[i+1] = 'street' THEN
         FOR parent IN
     FOR i IN 1..array_upper(location.members, 1) BY 2 LOOP
       IF location.members[i+1] = 'street' THEN
         FOR parent IN

-          SELECT place_id from placex
+          SELECT place_id, geometry
+           FROM placex

            WHERE osm_type = upper(substring(location.members[i], 1, 1))::char(1)
                  and osm_id = substring(location.members[i], 2)::bigint
                  and name is not null
                  and rank_search between 26 and 27
         LOOP
            WHERE osm_type = upper(substring(location.members[i], 1, 1))::char(1)
                  and osm_id = substring(location.members[i], 2)::bigint
                  and name is not null
                  and rank_search between 26 and 27
         LOOP

-          RETURN parent.place_id;
+          -- Find the closest 'street' member.
+          -- Avoid distance computation for the frequent case where there is
+          -- only one street member.
+          IF waygeom is null THEN
+            result := parent.place_id;
+            waygeom := parent.geometry;
+          ELSE
+            distance := coalesce(distance, ST_Distance(waygeom, bbox));
+            new_distance := ST_Distance(parent.geometry, bbox);
+            IF new_distance < distance THEN
+              distance := new_distance;
+              result := parent.place_id;
+              waygeom := parent.geometry;
+            END IF;
+          END IF;

         END LOOP;
       END IF;
     END LOOP;
   END LOOP;
 
         END LOOP;
       END IF;
     END LOOP;
   END LOOP;
 

-  RETURN NULL;
+  RETURN result;

 END;
 $$
 LANGUAGE plpgsql STABLE;
 END;
 $$
 LANGUAGE plpgsql STABLE;


@@ -158,7 +182,7 @@ BEGIN
   {% if debug %}RAISE WARNING 'finding street for % %', poi_osm_type, poi_osm_id;{% endif %}
 
   -- Is this object part of an associatedStreet relation?
   {% if debug %}RAISE WARNING 'finding street for % %', poi_osm_type, poi_osm_id;{% endif %}
 
   -- Is this object part of an associatedStreet relation?

-  parent_place_id := find_associated_street(poi_osm_type, poi_osm_id);
+  parent_place_id := find_associated_street(poi_osm_type, poi_osm_id, bbox);

 
   IF parent_place_id is null THEN
     parent_place_id := find_parent_for_address(token_info, poi_partition, bbox);
 
   IF parent_place_id is null THEN
     parent_place_id := find_parent_for_address(token_info, poi_partition, bbox);


@@ -181,7 +205,7 @@ BEGIN
         RETURN location.place_id;
       END IF;
 
         RETURN location.place_id;
       END IF;
 

-      parent_place_id := find_associated_street('W', location.osm_id);
+      parent_place_id := find_associated_street('W', location.osm_id, bbox);

     END LOOP;
   END IF;
 
     END LOOP;
   END IF;
 


@@ -193,6 +217,7 @@ BEGIN
         SELECT place_id FROM placex
          WHERE bbox && geometry AND _ST_Covers(geometry, ST_Centroid(bbox))
                AND rank_address between 5 and 25
         SELECT place_id FROM placex
          WHERE bbox && geometry AND _ST_Covers(geometry, ST_Centroid(bbox))
                AND rank_address between 5 and 25

+               AND ST_GeometryType(geometry) in ('ST_Polygon','ST_MultiPolygon')

          ORDER BY rank_address desc
       LOOP
         RETURN location.place_id;
          ORDER BY rank_address desc
       LOOP
         RETURN location.place_id;


@@ -208,6 +233,7 @@ BEGIN
         SELECT place_id FROM placex
          WHERE bbox && geometry AND _ST_Covers(geometry, ST_Centroid(bbox))
                AND rank_address between 5 and 25
         SELECT place_id FROM placex
          WHERE bbox && geometry AND _ST_Covers(geometry, ST_Centroid(bbox))
                AND rank_address between 5 and 25

+               AND ST_GeometryType(geometry) in ('ST_Polygon','ST_MultiPolygon')

         ORDER BY rank_address desc
       LOOP
         RETURN location.place_id;
         ORDER BY rank_address desc
       LOOP
         RETURN location.place_id;


@@ -271,7 +297,9 @@ BEGIN
 
   -- If extratags has a place tag, look for linked nodes by their place type.
   -- Area and node still have to have the same name.
 
   -- If extratags has a place tag, look for linked nodes by their place type.
   -- Area and node still have to have the same name.

-  IF bnd.extratags ? 'place' and bnd_name is not null THEN
+  IF bnd.extratags ? 'place' and bnd.extratags->'place' != 'postcode'
+     and bnd_name is not null
+  THEN

     FOR linked_placex IN
       SELECT * FROM placex
       WHERE (position(lower(name->'name') in bnd_name) > 0
     FOR linked_placex IN
       SELECT * FROM placex
       WHERE (position(lower(name->'name') in bnd_name) > 0


@@ -280,7 +308,6 @@ BEGIN
         AND placex.osm_type = 'N'
         AND (placex.linked_place_id is null or placex.linked_place_id = bnd.place_id)
         AND placex.rank_search < 26 -- needed to select the right index
         AND placex.osm_type = 'N'
         AND (placex.linked_place_id is null or placex.linked_place_id = bnd.place_id)
         AND placex.rank_search < 26 -- needed to select the right index

-        AND placex.type != 'postcode'

         AND ST_Covers(bnd.geometry, placex.geometry)
     LOOP
       {% if debug %}RAISE WARNING 'Found type-matching place node %', linked_placex.osm_id;{% endif %}
         AND ST_Covers(bnd.geometry, placex.geometry)
     LOOP
       {% if debug %}RAISE WARNING 'Found type-matching place node %', linked_placex.osm_id;{% endif %}


@@ -842,7 +869,8 @@ BEGIN
       FROM placex
       WHERE osm_type = 'R' and class = 'boundary' and type = 'administrative'
             and admin_level < NEW.admin_level and admin_level > 3
       FROM placex
       WHERE osm_type = 'R' and class = 'boundary' and type = 'administrative'
             and admin_level < NEW.admin_level and admin_level > 3

-            and rank_address > 0
+            and rank_address between 1 and 25 -- for index selection
+            and ST_GeometryType(geometry) in ('ST_Polygon','ST_MultiPolygon') -- for index selection

             and geometry && NEW.centroid and _ST_Covers(geometry, NEW.centroid)
       ORDER BY admin_level desc LIMIT 1
     LOOP
             and geometry && NEW.centroid and _ST_Covers(geometry, NEW.centroid)
       ORDER BY admin_level desc LIMIT 1
     LOOP


@@ -864,29 +892,61 @@ BEGIN
 
     IF NEW.rank_address > 9 THEN
         -- Second check that the boundary is not completely contained in a
 
     IF NEW.rank_address > 9 THEN
         -- Second check that the boundary is not completely contained in a

-        -- place area with a higher address rank
+        -- place area with a equal or higher address rank.

         FOR location IN
         FOR location IN

-          SELECT rank_address FROM placex
-          WHERE class = 'place' and rank_address < 24
-                and rank_address > NEW.rank_address
+          SELECT rank_address
+          FROM placex,
+               LATERAL compute_place_rank(country_code, 'A', class, type,
+                                          admin_level, False, null) prank
+          WHERE class = 'place' and rank_address between 1 and 23
+                and prank.address_rank >= NEW.rank_address
+                and ST_GeometryType(geometry) in ('ST_Polygon','ST_MultiPolygon') -- select right index

                 and geometry && NEW.geometry
                 and geometry ~ NEW.geometry -- needed because ST_Relate does not do bbox cover test
                 and ST_Relate(geometry, NEW.geometry, 'T*T***FF*') -- contains but not equal
                 and geometry && NEW.geometry
                 and geometry ~ NEW.geometry -- needed because ST_Relate does not do bbox cover test
                 and ST_Relate(geometry, NEW.geometry, 'T*T***FF*') -- contains but not equal

-          ORDER BY rank_address desc LIMIT 1
+          ORDER BY prank.address_rank desc LIMIT 1

         LOOP
           NEW.rank_address := location.rank_address + 2;
         END LOOP;
     END IF;
         LOOP
           NEW.rank_address := location.rank_address + 2;
         END LOOP;
     END IF;

+  ELSEIF NEW.class = 'place'
+         and ST_GeometryType(NEW.geometry) in ('ST_Polygon', 'ST_MultiPolygon')
+         and NEW.rank_address between 16 and 23
+  THEN
+    -- For place areas make sure they are not completely contained in an area
+    -- with a equal or higher address rank.
+    FOR location IN
+          SELECT rank_address
+          FROM placex,
+               LATERAL compute_place_rank(country_code, 'A', class, type,
+                                          admin_level, False, null) prank
+          WHERE prank.address_rank < 24
+                and rank_address between 1 and 25 -- select right index
+                and ST_GeometryType(geometry) in ('ST_Polygon','ST_MultiPolygon') -- select right index
+                and prank.address_rank >= NEW.rank_address
+                and geometry && NEW.geometry
+                and geometry ~ NEW.geometry -- needed because ST_Relate does not do bbox cover test
+                and ST_Relate(geometry, NEW.geometry, 'T*T***FF*') -- contains but not equal
+          ORDER BY prank.address_rank desc LIMIT 1
+        LOOP
+          NEW.rank_address := location.rank_address + 2;
+        END LOOP;

   ELSEIF NEW.class = 'place' and NEW.osm_type = 'N'
   ELSEIF NEW.class = 'place' and NEW.osm_type = 'N'

-     and NEW.rank_address between 16 and 23
+         and NEW.rank_address between 16 and 23

   THEN
   THEN

-    -- If a place node is contained in a admin boundary with the same address level
-    -- and has not been linked, then make the node a subpart by increasing the
-    -- address rank (city level and above).
+    -- If a place node is contained in an admin or place boundary with the same
+    -- address level and has not been linked, then make the node a subpart
+    -- by increasing the address rank (city level and above).

     FOR location IN
     FOR location IN

-        SELECT rank_address FROM placex
-        WHERE osm_type = 'R' and class = 'boundary' and type = 'administrative'
-              and rank_address = NEW.rank_address
+        SELECT rank_address
+        FROM placex,
+             LATERAL compute_place_rank(country_code, 'A', class, type,
+                                        admin_level, False, null) prank
+        WHERE osm_type = 'R'
+              and rank_address between 1 and 25 -- select right index
+              and ST_GeometryType(geometry) in ('ST_Polygon','ST_MultiPolygon') -- select right index
+              and ((class = 'place' and prank.address_rank = NEW.rank_address)
+                   or (class = 'boundary' and rank_address = NEW.rank_address))

               and geometry && NEW.centroid and _ST_Covers(geometry, NEW.centroid)
         LIMIT 1
     LOOP
               and geometry && NEW.centroid and _ST_Covers(geometry, NEW.centroid)
         LIMIT 1
     LOOP


@@ -925,7 +985,7 @@ BEGIN
 
   NEW.importance := null;
   SELECT wikipedia, importance
 
   NEW.importance := null;
   SELECT wikipedia, importance

-    FROM compute_importance(NEW.extratags, NEW.country_code, NEW.osm_type, NEW.osm_id)
+    FROM compute_importance(NEW.extratags, NEW.country_code, NEW.rank_search, NEW.centroid)

     INTO NEW.wikipedia,NEW.importance;
 
 {% if debug %}RAISE WARNING 'Importance computed from wikipedia: %', NEW.importance;{% endif %}
     INTO NEW.wikipedia,NEW.importance;
 
 {% if debug %}RAISE WARNING 'Importance computed from wikipedia: %', NEW.importance;{% endif %}


@@ -936,7 +996,7 @@ BEGIN
 
     {% if debug %}RAISE WARNING 'finding street for % %', NEW.osm_type, NEW.osm_id;{% endif %}
     NEW.parent_place_id := null;
 
     {% if debug %}RAISE WARNING 'finding street for % %', NEW.osm_type, NEW.osm_id;{% endif %}
     NEW.parent_place_id := null;

-    is_place_address := coalesce(not NEW.address ? 'street' and NEW.address ? 'place', FALSE);
+    is_place_address := not token_is_street_address(NEW.token_info);

 
     -- We have to find our parent road.
     NEW.parent_place_id := find_parent_for_poi(NEW.osm_type, NEW.osm_id,
 
     -- We have to find our parent road.
     NEW.parent_place_id := find_parent_for_poi(NEW.osm_type, NEW.osm_id,


@@ -953,7 +1013,7 @@ BEGIN
       SELECT p.country_code, p.postcode, p.name FROM placex p
        WHERE p.place_id = NEW.parent_place_id INTO location;
 
       SELECT p.country_code, p.postcode, p.name FROM placex p
        WHERE p.place_id = NEW.parent_place_id INTO location;
 

-      IF is_place_address THEN
+      IF is_place_address and NEW.address ? 'place' THEN

         -- Check if the addr:place tag is part of the parent name
         SELECT count(*) INTO i
           FROM svals(location.name) AS pname WHERE pname = NEW.address->'place';
         -- Check if the addr:place tag is part of the parent name
         SELECT count(*) INTO i
           FROM svals(location.name) AS pname WHERE pname = NEW.address->'place';


@@ -966,7 +1026,7 @@ BEGIN
       {% if debug %}RAISE WARNING 'Got parent details from search name';{% endif %}
 
       -- determine postcode
       {% if debug %}RAISE WARNING 'Got parent details from search name';{% endif %}
 
       -- determine postcode

-      NEW.postcode := coalesce(token_normalized_postcode(NEW.address->'postcode'),
+      NEW.postcode := coalesce(token_get_postcode(NEW.token_info),

                                location.postcode,
                                get_nearest_postcode(NEW.country_code, NEW.centroid));
 
                                location.postcode,
                                get_nearest_postcode(NEW.country_code, NEW.centroid));
 


@@ -1006,8 +1066,8 @@ BEGIN
   {% if debug %}RAISE WARNING 'Using full index mode for % %', NEW.osm_type, NEW.osm_id;{% endif %}
   IF linked_place is not null THEN
     -- Recompute the ranks here as the ones from the linked place might
   {% if debug %}RAISE WARNING 'Using full index mode for % %', NEW.osm_type, NEW.osm_id;{% endif %}
   IF linked_place is not null THEN
     -- Recompute the ranks here as the ones from the linked place might

-    -- have been shifted to accomodate surrounding boundaries.
-    SELECT place_id, osm_id, class, type, extratags,
+    -- have been shifted to accommodate surrounding boundaries.
+    SELECT place_id, osm_id, class, type, extratags, rank_search,

            centroid, geometry,
            (compute_place_rank(country_code, osm_type, class, type, admin_level,
                               (extratags->'capital') = 'yes', null)).*
            centroid, geometry,
            (compute_place_rank(country_code, osm_type, class, type, admin_level,
                               (extratags->'capital') = 'yes', null)).*


@@ -1036,7 +1096,9 @@ BEGIN
                      || coalesce(NEW.extratags, ''::hstore);
 
     -- mark the linked place (excludes from search results)
                      || coalesce(NEW.extratags, ''::hstore);
 
     -- mark the linked place (excludes from search results)

-    UPDATE placex set linked_place_id = NEW.place_id
+    -- Force reindexing to remove any traces from the search indexes and
+    -- reset the address rank if necessary.
+    UPDATE placex set linked_place_id = NEW.place_id, indexed_status = 2

       WHERE place_id = location.place_id;
     -- ensure that those places are not found anymore
     {% if 'search_name' in db.tables %}
       WHERE place_id = location.place_id;
     -- ensure that those places are not found anymore
     {% if 'search_name' in db.tables %}


@@ -1046,7 +1108,7 @@ BEGIN
 
     SELECT wikipedia, importance
       FROM compute_importance(location.extratags, NEW.country_code,
 
     SELECT wikipedia, importance
       FROM compute_importance(location.extratags, NEW.country_code,

-                              'N', location.osm_id)
+                              location.rank_search, NEW.centroid)

       INTO linked_wikipedia,linked_importance;
 
     -- Use the maximum importance if one could be computed from the linked object.
       INTO linked_wikipedia,linked_importance;
 
     -- Use the maximum importance if one could be computed from the linked object.


@@ -1058,7 +1120,7 @@ BEGIN
   ELSE
     -- No linked place? As a last resort check if the boundary is tagged with
     -- a place type and adapt the rank address.
   ELSE
     -- No linked place? As a last resort check if the boundary is tagged with
     -- a place type and adapt the rank address.

-    IF NEW.rank_address > 0 and NEW.extratags ? 'place' THEN
+    IF NEW.rank_address between 4 and 25 and NEW.extratags ? 'place' THEN

       SELECT address_rank INTO place_address_level
         FROM compute_place_rank(NEW.country_code, 'A', 'place',
                                 NEW.extratags->'place', 0::SMALLINT, False, null);
       SELECT address_rank INTO place_address_level
         FROM compute_place_rank(NEW.country_code, 'A', 'place',
                                 NEW.extratags->'place', 0::SMALLINT, False, null);


@@ -1069,13 +1131,22 @@ BEGIN
     END IF;
   END IF;
 
     END IF;
   END IF;
 

+  {% if not disable_diff_updates %}
+  IF OLD.rank_address != NEW.rank_address THEN
+    -- After a rank shift all addresses containing us must be updated.
+    UPDATE placex p SET indexed_status = 2 FROM place_addressline pa
+      WHERE pa.address_place_id = NEW.place_id and p.place_id = pa.place_id
+            and p.indexed_status = 0 and p.rank_address between 4 and 25;
+  END IF;
+  {% endif %}
+

   IF NEW.admin_level = 2
      AND NEW.class = 'boundary' AND NEW.type = 'administrative'
      AND NEW.country_code IS NOT NULL AND NEW.osm_type = 'R'
   THEN
     -- Update the list of country names.
     -- Only take the name from the largest area for the given country code
   IF NEW.admin_level = 2
      AND NEW.class = 'boundary' AND NEW.type = 'administrative'
      AND NEW.country_code IS NOT NULL AND NEW.osm_type = 'R'
   THEN
     -- Update the list of country names.
     -- Only take the name from the largest area for the given country code

-    -- in the hope that this is the authoritive one.
+    -- in the hope that this is the authoritative one.

     -- Also replace any old names so that all mapping mistakes can
     -- be fixed through regular OSM updates.
     FOR location IN
     -- Also replace any old names so that all mapping mistakes can
     -- be fixed through regular OSM updates.
     FOR location IN


@@ -1122,8 +1193,7 @@ BEGIN
 
   {% if debug %}RAISE WARNING 'RETURN insert_addresslines: %, %, %', NEW.parent_place_id, NEW.postcode, nameaddress_vector;{% endif %}
 
 
   {% if debug %}RAISE WARNING 'RETURN insert_addresslines: %, %, %', NEW.parent_place_id, NEW.postcode, nameaddress_vector;{% endif %}
 

-  NEW.postcode := coalesce(token_normalized_postcode(NEW.address->'postcode'),
-                           NEW.postcode);
+  NEW.postcode := coalesce(token_get_postcode(NEW.token_info), NEW.postcode);

 
   -- if we have a name add this to the name search table
   IF NEW.name IS NOT NULL THEN
 
   -- if we have a name add this to the name search table
   IF NEW.name IS NOT NULL THEN


@@ -1160,11 +1230,15 @@ BEGIN
     {% endif %}
   END IF;
 
     {% endif %}
   END IF;
 

-  IF NEW.postcode is null AND NEW.rank_search > 8 THEN
+  IF NEW.postcode is null AND NEW.rank_search > 8
+     AND (NEW.rank_address > 0
+          OR ST_GeometryType(NEW.geometry) not in ('ST_LineString','ST_MultiLineString')
+          OR ST_Length(NEW.geometry) < 0.02)
+  THEN

     NEW.postcode := get_nearest_postcode(NEW.country_code, NEW.geometry);
   END IF;
 
     NEW.postcode := get_nearest_postcode(NEW.country_code, NEW.geometry);
   END IF;
 

-  {% if debug %}RAISE WARNING 'place update % % finsihed.', NEW.osm_type, NEW.osm_id;{% endif %}
+  {% if debug %}RAISE WARNING 'place update % % finished.', NEW.osm_type, NEW.osm_id;{% endif %}

 
   NEW.token_info := token_strip_info(NEW.token_info);
   RETURN NEW;
 
   NEW.token_info := token_strip_info(NEW.token_info);
   RETURN NEW;