remove postcodes entirely from indexing

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

diff --git a/sql/functions/placex_triggers.sql b/sql/functions/placex_triggers.sql
index 5c04ad065a9c58211e0bbb3912ac457f4adc76fa..0102d2c819d83c59f8a6047e732102e67d2341bb 100644 (file)
--- a/sql/functions/placex_triggers.sql
+++ b/sql/functions/placex_triggers.sql
@@ -1,45 +1,5 @@
 -- Trigger functions for the placex table.
 
 -- Trigger functions for the placex table.
 

-CREATE OR REPLACE FUNCTION get_rel_node_members(members TEXT[], memberLabels TEXT[])
-  RETURNS SETOF BIGINT
-  AS $$
-DECLARE
-  i INTEGER;
-BEGIN
-  FOR i IN 1..ARRAY_UPPER(members,1) BY 2 LOOP
-    IF members[i+1] = ANY(memberLabels)
-       AND upper(substring(members[i], 1, 1))::char(1) = 'N'
-    THEN
-      RETURN NEXT substring(members[i], 2)::bigint;
-    END IF;
-  END LOOP;
-
-  RETURN;
-END;
-$$
-LANGUAGE plpgsql IMMUTABLE;
-
--- copy 'name' to or from the default language (if there is a default language)
-CREATE OR REPLACE FUNCTION add_default_place_name(country_code VARCHAR(2),
-                                                  INOUT name HSTORE)
-  AS $$
-DECLARE
-  default_language VARCHAR(10);
-BEGIN
-  IF name is not null AND array_upper(akeys(name),1) > 1 THEN
-    default_language := get_country_language_code(country_code);
-    IF default_language IS NOT NULL THEN
-      IF name ? 'name' AND NOT name ? ('name:'||default_language) THEN
-        name := name || hstore(('name:'||default_language), (name -> 'name'));
-      ELSEIF name ? ('name:'||default_language) AND NOT name ? 'name' THEN
-        name := name || hstore('name', (name -> ('name:'||default_language)));
-      END IF;
-    END IF;
-  END IF;
-END;
-$$
-LANGUAGE plpgsql IMMUTABLE;
-

 -- Find the parent road of a POI.
 --
 -- \returns Place ID of parent object or NULL if none
 -- Find the parent road of a POI.
 --
 -- \returns Place ID of parent object or NULL if none


@@ -71,7 +31,6 @@ BEGIN
     LOOP
       FOR i IN 1..array_upper(location.members, 1) BY 2 LOOP
         IF location.members[i+1] = 'street' THEN
     LOOP
       FOR i IN 1..array_upper(location.members, 1) BY 2 LOOP
         IF location.members[i+1] = 'street' THEN

-          --DEBUG: RAISE WARNING 'node in relation %',relation;

           FOR parent IN
             SELECT place_id from placex
              WHERE osm_type = 'W' and osm_id = substring(location.members[i],2)::bigint
           FOR parent IN
             SELECT place_id from placex
              WHERE osm_type = 'W' and osm_id = substring(location.members[i],2)::bigint


@@ -133,18 +92,18 @@ BEGIN
     END IF;
 
     IF fallback THEN
     END IF;
 
     IF fallback THEN

-      IF ST_Area(bbox) < 0.01 THEN
+      IF ST_Area(bbox) < 0.005 THEN

         -- for smaller features get the nearest road
         SELECT getNearestRoadPlaceId(poi_partition, bbox) INTO parent_place_id;
         --DEBUG: RAISE WARNING 'Checked for nearest way (%)', parent_place_id;
       ELSE
         -- for larger features simply find the area with the largest rank that
         -- for smaller features get the nearest road
         SELECT getNearestRoadPlaceId(poi_partition, bbox) INTO parent_place_id;
         --DEBUG: RAISE WARNING 'Checked for nearest way (%)', parent_place_id;
       ELSE
         -- for larger features simply find the area with the largest rank that

-        -- contains the bbox
+        -- contains the bbox, only use addressable features

         FOR location IN
           SELECT place_id FROM placex
             WHERE bbox @ geometry AND _ST_Covers(geometry, ST_Centroid(bbox))
         FOR location IN
           SELECT place_id FROM placex
             WHERE bbox @ geometry AND _ST_Covers(geometry, ST_Centroid(bbox))

-                  AND rank_search between 5 and 25
-            ORDER BY rank_search desc
+                  AND rank_address between 5 and 25
+            ORDER BY rank_address desc

         LOOP
             RETURN location.place_id;
         END LOOP;
         LOOP
             RETURN location.place_id;
         END LOOP;


@@ -432,7 +391,6 @@ DECLARE
   country_code VARCHAR(2);
   diameter FLOAT;
   classtable TEXT;
   country_code VARCHAR(2);
   diameter FLOAT;
   classtable TEXT;

-  classtype TEXT;

 BEGIN
   --DEBUG: RAISE WARNING '% % % %',NEW.osm_type,NEW.osm_id,NEW.class,NEW.type;
 
 BEGIN
   --DEBUG: RAISE WARNING '% % % %',NEW.osm_type,NEW.osm_id,NEW.class,NEW.type;
 


@@ -450,8 +408,8 @@ BEGIN
     is_area := ST_GeometryType(NEW.geometry) IN ('ST_Polygon','ST_MultiPolygon');
 
     IF NEW.class in ('place','boundary')
     is_area := ST_GeometryType(NEW.geometry) IN ('ST_Polygon','ST_MultiPolygon');
 
     IF NEW.class in ('place','boundary')

-       AND NEW.type in ('postcode','postal_code') THEN
-
+       AND NEW.type in ('postcode','postal_code')
+    THEN

       IF NEW.address IS NULL OR NOT NEW.address ? 'postcode' THEN
           -- most likely just a part of a multipolygon postcode boundary, throw it away
           RETURN NULL;
       IF NEW.address IS NULL OR NOT NEW.address ? 'postcode' THEN
           -- most likely just a part of a multipolygon postcode boundary, throw it away
           RETURN NULL;


@@ -459,63 +417,28 @@ BEGIN
 
       NEW.name := hstore('ref', NEW.address->'postcode');
 
 
       NEW.name := hstore('ref', NEW.address->'postcode');
 

-      SELECT * FROM get_postcode_rank(NEW.country_code, NEW.address->'postcode')
-        INTO NEW.rank_search, NEW.rank_address;
-
-      IF NOT is_area THEN
-          NEW.rank_address := 0;
-      END IF;

     ELSEIF NEW.class = 'boundary' AND NOT is_area THEN
     ELSEIF NEW.class = 'boundary' AND NOT is_area THEN

-        return NULL;
+        RETURN NULL;

     ELSEIF NEW.class = 'boundary' AND NEW.type = 'administrative'
     ELSEIF NEW.class = 'boundary' AND NEW.type = 'administrative'

-           AND NEW.admin_level <= 4 AND NEW.osm_type = 'W' THEN
-        return NULL;
-    ELSEIF NEW.osm_type = 'N' AND NEW.class = 'highway' THEN
-        NEW.rank_search = 30;
-        NEW.rank_address = 0;
-    ELSEIF NEW.class = 'landuse' AND NOT is_area THEN
-        NEW.rank_search = 30;
-        NEW.rank_address = 0;
-    ELSE
-      -- do table lookup stuff
-      IF NEW.class = 'boundary' and NEW.type = 'administrative' THEN
-        classtype = NEW.type || NEW.admin_level::TEXT;
-      ELSE
-        classtype = NEW.type;
-      END IF;
-      SELECT l.rank_search, l.rank_address FROM address_levels l
-       WHERE (l.country_code = NEW.country_code or l.country_code is NULL)
-             AND l.class = NEW.class AND (l.type = classtype or l.type is NULL)
-       ORDER BY l.country_code, l.class, l.type LIMIT 1
-        INTO NEW.rank_search, NEW.rank_address;
-
-      IF NEW.rank_search is NULL THEN
-        NEW.rank_search := 30;
-      END IF;
-
-      IF NEW.rank_address is NULL THEN
-        NEW.rank_address := 30;
-      END IF;
+           AND NEW.admin_level <= 4 AND NEW.osm_type = 'W'
+    THEN
+        RETURN NULL;

     END IF;
 
     END IF;
 

-    -- some postcorrections
-    IF NEW.class = 'waterway' AND NEW.osm_type = 'R' THEN
-        -- Slightly promote waterway relations so that they are processed
-        -- before their members.
-        NEW.rank_search := NEW.rank_search - 1;
-    END IF;
+    SELECT * INTO NEW.rank_search, NEW.rank_address
+      FROM compute_place_rank(NEW.country_code,
+                              CASE WHEN is_area THEN 'A' ELSE NEW.osm_type END,
+                              NEW.class, NEW.type, NEW.admin_level,
+                              (NEW.extratags->'capital') = 'yes',
+                              NEW.address->'postcode');

 
 

-    IF (NEW.extratags -> 'capital') = 'yes' THEN
-      NEW.rank_search := NEW.rank_search - 1;
+    -- a country code make no sense below rank 4 (country)
+    IF NEW.rank_search < 4 THEN
+      NEW.country_code := NULL;

     END IF;
 
   END IF;
 
     END IF;
 
   END IF;
 

-  -- a country code make no sense below rank 4 (country)
-  IF NEW.rank_search < 4 THEN
-    NEW.country_code := NULL;
-  END IF;
-

   --DEBUG: RAISE WARNING 'placex_insert:END: % % % %',NEW.osm_type,NEW.osm_id,NEW.class,NEW.type;
 
   RETURN NEW; -- %DIFFUPDATES% The following is not needed until doing diff updates, and slows the main index process down
   --DEBUG: RAISE WARNING 'placex_insert:END: % % % %',NEW.osm_type,NEW.osm_id,NEW.class,NEW.type;
 
   RETURN NEW; -- %DIFFUPDATES% The following is not needed until doing diff updates, and slows the main index process down


@@ -538,25 +461,7 @@ BEGIN
       END IF;
     ELSE
       -- mark nearby items for re-indexing, where 'nearby' depends on the features rank_search and is a complete guess :(
       END IF;
     ELSE
       -- mark nearby items for re-indexing, where 'nearby' depends on the features rank_search and is a complete guess :(

-      diameter := 0;
-      -- 16 = city, anything higher than city is effectively ignored (polygon required!)
-      IF NEW.type='postcode' THEN
-        diameter := 0.05;
-      ELSEIF NEW.rank_search < 16 THEN
-        diameter := 0;
-      ELSEIF NEW.rank_search < 18 THEN
-        diameter := 0.1;
-      ELSEIF NEW.rank_search < 20 THEN
-        diameter := 0.05;
-      ELSEIF NEW.rank_search = 21 THEN
-        diameter := 0.001;
-      ELSEIF NEW.rank_search < 24 THEN
-        diameter := 0.02;
-      ELSEIF NEW.rank_search < 26 THEN
-        diameter := 0.002; -- 100 to 200 meters
-      ELSEIF NEW.rank_search < 28 THEN
-        diameter := 0.001; -- 50 to 100 meters
-      END IF;
+      diameter := update_place_diameter(NEW.rank_search);

       IF diameter > 0 THEN
   --      RAISE WARNING 'placex point insert: % % % % %',NEW.osm_type,NEW.osm_id,NEW.class,NEW.type,diameter;
         IF NEW.rank_search >= 26 THEN
       IF diameter > 0 THEN
   --      RAISE WARNING 'placex point insert: % % % % %',NEW.osm_type,NEW.osm_id,NEW.class,NEW.type,diameter;
         IF NEW.rank_search >= 26 THEN


@@ -592,6 +497,32 @@ END;
 $$
 LANGUAGE plpgsql;
 
 $$
 LANGUAGE plpgsql;
 

+CREATE OR REPLACE FUNCTION get_parent_address_level(geom GEOMETRY, in_level SMALLINT)
+  RETURNS SMALLINT
+  AS $$
+DECLARE
+  address_rank SMALLINT;
+BEGIN
+  IF in_level <= 3 or in_level > 15 THEN
+    address_rank := 3;
+  ELSE
+    SELECT rank_address INTO address_rank
+      FROM placex
+      WHERE osm_type = 'R' and class = 'boundary' and type = 'administrative'
+            and admin_level < in_level
+            and geometry && geom and ST_Covers(geometry, geom)
+      ORDER BY admin_level desc LIMIT 1;
+  END IF;
+
+  IF address_rank is NULL or address_rank <= 3 THEN
+    RETURN 3;
+  END IF;
+
+  RETURN address_rank;
+END;
+$$
+LANGUAGE plpgsql;
+

 
 CREATE OR REPLACE FUNCTION placex_update()
   RETURNS TRIGGER
 
 CREATE OR REPLACE FUNCTION placex_update()
   RETURNS TRIGGER


@@ -602,6 +533,8 @@ DECLARE
   relation_members TEXT[];
 
   centroid GEOMETRY;
   relation_members TEXT[];
 
   centroid GEOMETRY;

+  parent_address_level SMALLINT;
+  place_address_level SMALLINT;

 
   addr_street TEXT;
   addr_place TEXT;
 
   addr_street TEXT;
   addr_place TEXT;


@@ -641,11 +574,52 @@ BEGIN
          where linked_place_id = NEW.place_id;
   -- update not necessary for osmline, cause linked_place_id does not exist
 
          where linked_place_id = NEW.place_id;
   -- update not necessary for osmline, cause linked_place_id does not exist
 

+  NEW.extratags := NEW.extratags - 'linked_place'::TEXT;
+

   IF NEW.linked_place_id is not null THEN
     --DEBUG: RAISE WARNING 'place already linked to %', NEW.linked_place_id;
     RETURN NEW;
   END IF;
 
   IF NEW.linked_place_id is not null THEN
     --DEBUG: RAISE WARNING 'place already linked to %', NEW.linked_place_id;
     RETURN NEW;
   END IF;
 

+  -- Postcodes are just here to compute the centroids. They are not searchable
+  -- unless they are a boundary=postal_code.
+  -- There was an error in the style so that boundary=postal_code used to be
+  -- imported as place=postcode. That's why relations are allowed to pass here.
+  -- This can go away in a couple of versions.
+  IF NEW.class = 'place'  and NEW.type = 'postcode' and NEW.osm_type != 'R' THEN
+    RETURN NEW;
+  END IF;
+
+  -- Speed up searches - just use the centroid of the feature
+  -- cheaper but less acurate
+  NEW.centroid := ST_PointOnSurface(NEW.geometry);
+  --DEBUG: RAISE WARNING 'Computing preliminary centroid at %',ST_AsText(NEW.centroid);
+
+  -- recompute the ranks, they might change when linking changes
+  SELECT * INTO NEW.rank_search, NEW.rank_address
+    FROM compute_place_rank(NEW.country_code,
+                            CASE WHEN ST_GeometryType(NEW.geometry)
+                                        IN ('ST_Polygon','ST_MultiPolygon')
+                            THEN 'A' ELSE NEW.osm_type END,
+                            NEW.class, NEW.type, NEW.admin_level,
+                            (NEW.extratags->'capital') = 'yes',
+                            NEW.address->'postcode');
+  -- We must always increase the address level relative to the admin boundary.
+  IF NEW.class = 'boundary' and NEW.type = 'administrative'
+     and NEW.osm_type = 'R' and NEW.rank_address > 0
+  THEN
+    parent_address_level := get_parent_address_level(NEW.centroid, NEW.admin_level);
+    IF parent_address_level >= NEW.rank_address THEN
+      IF parent_address_level >= 24 THEN
+        NEW.rank_address := 25;
+      ELSE
+        NEW.rank_address := parent_address_level + 2;
+      END IF;
+    END IF;
+  ELSE
+    parent_address_level := 3;
+  END IF;
+

   --DEBUG: RAISE WARNING 'Copy over address tags';
   -- housenumber is a computed field, so start with an empty value
   NEW.housenumber := NULL;
   --DEBUG: RAISE WARNING 'Copy over address tags';
   -- housenumber is a computed field, so start with an empty value
   NEW.housenumber := NULL;


@@ -674,11 +648,6 @@ BEGIN
       END IF;
   END IF;
 
       END IF;
   END IF;
 

-  -- Speed up searches - just use the centroid of the feature
-  -- cheaper but less acurate
-  NEW.centroid := ST_PointOnSurface(NEW.geometry);
-  --DEBUG: RAISE WARNING 'Computing preliminary centroid at %',ST_AsText(NEW.centroid);
-

   NEW.postcode := null;
 
   -- recalculate country and partition
   NEW.postcode := null;
 
   -- recalculate country and partition


@@ -715,7 +684,9 @@ BEGIN
                   and ( relation_members[i+1] != 'side_stream' or NEW.name->'name' = name->'name')
                 LOOP
                   UPDATE placex SET linked_place_id = NEW.place_id WHERE place_id = linked_node_id;
                   and ( relation_members[i+1] != 'side_stream' or NEW.name->'name' = name->'name')
                 LOOP
                   UPDATE placex SET linked_place_id = NEW.place_id WHERE place_id = linked_node_id;

-                  DELETE FROM search_name WHERE place_id = linked_node_id;
+                  IF NOT %REVERSE-ONLY% THEN
+                    DELETE FROM search_name WHERE place_id = linked_node_id;
+                  END IF;

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


@@ -753,7 +724,6 @@ BEGIN
         NEW.housenumber := location.address->'housenumber';
         addr_street := location.address->'street';
         addr_place := location.address->'place';
         NEW.housenumber := location.address->'housenumber';
         addr_street := location.address->'street';
         addr_place := location.address->'place';

-        --DEBUG: RAISE WARNING 'Found surrounding building % %', location.osm_type, location.osm_id;

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


@@ -839,8 +809,10 @@ BEGIN
         NEW.centroid := centroid;
     END IF;
 
         NEW.centroid := centroid;
     END IF;
 

-    -- Use the address rank of the linked place, if it has one
-    IF location.rank_address between 5 and 25 THEN
+    --DEBUG: RAISE WARNING 'parent address: % rank address: %', parent_address_level, location.rank_address;
+    IF location.rank_address > parent_address_level
+       and location.rank_address < 26
+    THEN

       NEW.rank_address := location.rank_address;
     END IF;
 
       NEW.rank_address := location.rank_address;
     END IF;
 


@@ -858,7 +830,10 @@ BEGIN
     UPDATE placex set linked_place_id = NEW.place_id
       WHERE place_id = location.place_id;
     -- ensure that those places are not found anymore
     UPDATE placex set linked_place_id = NEW.place_id
       WHERE place_id = location.place_id;
     -- ensure that those places are not found anymore

-    DELETE FROM search_name WHERE place_id = location.place_id;
+    IF NOT %REVERSE-ONLY% THEN
+      DELETE FROM search_name WHERE place_id = location.place_id;
+    END IF;
+    PERFORM deleteLocationArea(NEW.partition, location.place_id, NEW.rank_search);

 
     SELECT wikipedia, importance
       FROM compute_importance(location.extratags, NEW.country_code,
 
     SELECT wikipedia, importance
       FROM compute_importance(location.extratags, NEW.country_code,


@@ -871,6 +846,18 @@ BEGIN
     THEN
       NEW.importance = linked_importance;
     END IF;
     THEN
       NEW.importance = linked_importance;
     END IF;

+  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
+      SELECT address_rank INTO place_address_level
+        FROM compute_place_rank(NEW.country_code, 'A', 'place',
+                                NEW.extratags->'place', 0::SMALLINT, False, null);
+      IF place_address_level > parent_address_level and
+         place_address_level < 26 THEN
+        NEW.rank_address := place_address_level;
+      END IF;
+    END IF;

   END IF;
 
   -- Initialise the name vector using our name
   END IF;
 
   -- Initialise the name vector using our name


@@ -887,7 +874,9 @@ BEGIN
   END IF;
 
   SELECT * FROM insert_addresslines(NEW.place_id, NEW.partition,
   END IF;
 
   SELECT * FROM insert_addresslines(NEW.place_id, NEW.partition,

-                                    NEW.rank_search, NEW.address,
+                                    CASE WHEN NEW.rank_address = 0
+                                      THEN NEW.rank_search ELSE NEW.rank_address END,
+                                    NEW.address,

                                     CASE WHEN NEW.rank_search >= 26
                                              AND NEW.rank_search < 30
                                       THEN NEW.geometry ELSE NEW.centroid END)
                                     CASE WHEN NEW.rank_search >= 26
                                              AND NEW.rank_search < 30
                                       THEN NEW.geometry ELSE NEW.centroid END)


@@ -949,10 +938,14 @@ DECLARE
 BEGIN
   -- RAISE WARNING 'placex_delete % %',OLD.osm_type,OLD.osm_id;
 
 BEGIN
   -- RAISE WARNING 'placex_delete % %',OLD.osm_type,OLD.osm_id;
 

-  update placex set linked_place_id = null, indexed_status = 2 where linked_place_id = OLD.place_id and indexed_status = 0;
-  --DEBUG: RAISE WARNING 'placex_delete:01 % %',OLD.osm_type,OLD.osm_id;
-  update placex set linked_place_id = null where linked_place_id = OLD.place_id;
-  --DEBUG: RAISE WARNING 'placex_delete:02 % %',OLD.osm_type,OLD.osm_id;
+  IF OLD.linked_place_id is null THEN
+    update placex set linked_place_id = null, indexed_status = 2 where linked_place_id = OLD.place_id and indexed_status = 0;
+    --DEBUG: RAISE WARNING 'placex_delete:01 % %',OLD.osm_type,OLD.osm_id;
+    update placex set linked_place_id = null where linked_place_id = OLD.place_id;
+    --DEBUG: RAISE WARNING 'placex_delete:02 % %',OLD.osm_type,OLD.osm_id;
+  ELSE
+    update placex set indexed_status = 2 where place_id = OLD.linked_place_id and indexed_status = 0;
+  END IF;

 
   IF OLD.rank_address < 30 THEN
 
 
   IF OLD.rank_address < 30 THEN