Documnt place

Makefile

@@ -1,8 +1,7 @@
 all: build/openmaptiles.tm2source/data.yml build/mapping.yaml build/tileset.sql
 
 .PHONY: docs
-docs: layers/railway/README.md layers/boundary/README.md layers/water/README.md layers/building/README.md layers/highway/README.md layers/highway_name/README.md layers/poi/README.md
+docs: layers/railway/README.md layers/boundary/README.md layers/water/README.md layers/building/README.md layers/highway/README.md layers/highway_name/README.md layers/poi/README.md layers/place/README.md
-
 
 build/openmaptiles.tm2source/data.yml:
 	mkdir -p build/openmaptiles.tm2source && generate-tm2source openmaptiles.yaml --host="postgres" --port=5432 --database="openmaptiles" --user="openmaptiles" --password="openmaptiles" > build/openmaptiles.tm2source/data.yml
@@ -34,5 +33,8 @@ layers/water/README.md:
 layers/building/README.md:
 	generate-doc layers/building/building.yaml > layers/building/README.md
 
+layers/place/README.md:
+	generate-doc layers/place/place.yaml --diagram layers/place/mapping > layers/place/README.md
+
 clean:
 	rm -f build/openmaptiles.tm2source/data.yml && rm -f build/mapping.yaml && rm -f build/tileset.sql && rm -f layers/**/README.md&& rm -f layers/**/*.png

layers/place/README.md

@@ -0,0 +1,25 @@
+# place
+
+The place layer consists out of [countries](http://wiki.openstreetmap.org/wiki/Tag:place%3Dcountry),
+[states](http://wiki.openstreetmap.org/wiki/Tag:place%3Dstate) and [cities](http://wiki.openstreetmap.org/wiki/Key:place).
+Apart from the roads this is also one of the more important layers to create a beautiful map.
+We suggest you use different font styles and sizes to create a text hierarchy.
+
+## Fields
+
+- **name_en**: The english `name:en` value if available.
+- **name**: The OSM [`name`](http://wiki.openstreetmap.org/wiki/Key:name) value of the POI.
+- **rank**: Countries, states and the most important cities all have a `rank` field ranging from `1` to `6` which
+marks the importance of the feature. Less important places do not have a `rank`.
+Use this to build a text hierarchy. The rank value originates from Natural Earth data and is either the
+original `scalerank` for cities or the original `labelrank` for countries and states.
+- **class**: Distinguish between `country`, `state` and other city classes like
+`city`, `town`, `village`, `hamlet`, `suburb`, `neighbourhood` or `isolated_dwelling`.
+Use this to separately style the different places according to their importance (usually country and state different
+than cities).
+
+## Mapping
+
+![](mapping.png)
+
+

layers/place/mapping.yaml

@@ -53,7 +53,7 @@ tables:
       place:
       - state
   city_point:
-    type: geometry
+    type: point
     fields:
     - name: osm_id
       type: id
@@ -71,13 +71,12 @@ tables:
     filters:
       exclude_tags:
       - [ "name", "__nil__" ]
-    type_mappings:
+    mapping:
-      points:
+      place:
-        place:
+      - city
-        - city
+      - town
-        - town
+      - village
-        - village
+      - hamlet
-        - hamlet
+      - suburb
-        - suburb
+      - neighbourhood
-        - neighbourhood
+      - isolated_dwelling
-        - isolated_dwelling

layers/place/place.yaml

@@ -3,16 +3,25 @@ layer:
   description: |
       The place layer consists out of [countries](http://wiki.openstreetmap.org/wiki/Tag:place%3Dcountry),
       [states](http://wiki.openstreetmap.org/wiki/Tag:place%3Dstate) and [cities](http://wiki.openstreetmap.org/wiki/Key:place).
-      For countries, states and places at low zoom levels (`z0` to `z7`) the
+      Apart from the roads this is also one of the more important layers to create a beautiful map.
-      [Populated Places](http://www.naturalearthdata.com/downloads/10m-cultural-vectors/10m-populated-places/) dataset from NaturalEarth is merged
+      We suggest you use different font styles and sizes to create a text hierarchy.
-      with OSM data to only show the most relevant places for the given zoom level. All features derived from NaturalEarth have a `scalerank` field.
+  fields:
-      States and countries additionally contain `postal` and `abbreviation` fields for styles that want to fit a lot of data with small labels on the map. 
+    name: The OSM [`name`](http://wiki.openstreetmap.org/wiki/Key:name) value of the POI.
-      The `place` layer only contains more important states (or similar entities) from the US, Russia, China and Brazil.
+    name_en: The english `name:en` value if available.
-
+    class: |
+        Distinguish between `country`, `state` and other city classes like
+        `city`, `town`, `village`, `hamlet`, `suburb`, `neighbourhood` or `isolated_dwelling`.
+        Use this to separately style the different places according to their importance (usually country and state different
+        than cities).
+    rank: |
+        Countries, states and the most important cities all have a `rank` field ranging from `1` to `6` which
+        marks the importance of the feature. Less important places do not have a `rank`.
+        Use this to build a text hierarchy. The rank value originates from Natural Earth data and is either the
+        original `scalerank` for cities or the original `labelrank` for countries and states.
   buffer_size: 128
   datasource:
     geometry_field: geometry
-    query: (SELECT geometry, name, name_en, class, abbrev scalerank FROM layer_place(!bbox!, z(!scale_denominator!), !pixel_width!)) AS t
+    query: (SELECT geometry, name, name_en, class, rank FROM layer_place(!bbox!, z(!scale_denominator!), !pixel_width!)) AS t
 schema:
   - ./types.sql
   - ./city.sql