move PlaceInfo into data submodule

[nominatim.git] / nominatim / tools / country_info.py

diff --git a/nominatim/tools/country_info.py b/nominatim/tools/country_info.py
index 1f7b799610743926f513f67fe3e40b20e76e6397..d754b4ddb029365b22d2cc7a77ccaeefc49a2719 100644 (file)
--- a/nominatim/tools/country_info.py
+++ b/nominatim/tools/country_info.py
@@ -1,11 +1,41 @@
+# 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.

 """
 Functions for importing and managing static country information.
 """
 import psycopg2.extras
 """
 Functions for importing and managing static country information.
 """
 import psycopg2.extras

-import yaml

 
 from nominatim.db import utils as db_utils
 from nominatim.db.connection import connect
 
 from nominatim.db import utils as db_utils
 from nominatim.db.connection import connect

+from nominatim.errors import UsageError
+
+def _flatten_name_list(names):
+    if names is None:
+        return {}
+
+    if not isinstance(names, dict):
+        raise UsageError("Expected key-value list for names in country_settings.py")
+
+    flat = {}
+    for prefix, remain in names.items():
+        if isinstance(remain, str):
+            flat[prefix] = remain
+        elif not isinstance(remain, dict):
+            raise UsageError("Entries in names must be key-value lists.")
+        else:
+            for suffix, name in remain.items():
+                if suffix == 'default':
+                    flat[prefix] = name
+                else:
+                    flat[f'{prefix}:{suffix}'] = name
+
+    return flat
+
+

 
 class _CountryInfo:
     """ Caches country-specific properties from the configuration file.
 
 class _CountryInfo:
     """ Caches country-specific properties from the configuration file.


@@ -14,30 +44,66 @@ class _CountryInfo:
     def __init__(self):
         self._info = {}
 
     def __init__(self):
         self._info = {}
 

-    def load(self, configfile):
+
+    def load(self, config):
+        """ Load the country properties from the configuration files,
+            if they are not loaded yet.
+        """

         if not self._info:
         if not self._info:

-            self._info = yaml.safe_load(configfile.read_text())
+            self._info = config.load_sub_configuration('country_settings.yaml')
+            for prop in self._info.values():
+                # Convert languages into a list for simpler handling.
+                if 'languages' not in prop:
+                    prop['languages'] = []
+                elif not isinstance(prop['languages'], list):
+                    prop['languages'] = [x.strip()
+                                         for x in prop['languages'].split(',')]
+                prop['names'] = _flatten_name_list(prop.get('names'))
+

 
     def items(self):
 
     def items(self):

+        """ Return tuples of (country_code, property dict) as iterable.
+        """

         return self._info.items()
 
         return self._info.items()
 

+    def get(self, country_code):
+        """ Get country information for the country with the given country code.
+        """
+        return self._info.get(country_code, {})
+
+

 
 _COUNTRY_INFO = _CountryInfo()
 
 
 _COUNTRY_INFO = _CountryInfo()
 

-def setup_country_config(configfile):
+
+def setup_country_config(config):

     """ Load country properties from the configuration file.
         Needs to be called before using any other functions in this
         file.
     """
     """ Load country properties from the configuration file.
         Needs to be called before using any other functions in this
         file.
     """

-    _COUNTRY_INFO.load(configfile)
-    print(_COUNTRY_INFO._info)
+    _COUNTRY_INFO.load(config)
+
+
+def iterate(prop=None):
+    """ Iterate over country code and properties.
+
+        When `prop` is None, all countries are returned with their complete
+        set of properties.
+
+        If `prop` is given, then only countries are returned where the
+        given property is set. The second item of the tuple contains only
+        the content of the given property.
+    """
+    if prop is None:
+        return _COUNTRY_INFO.items()
+
+    return ((c, p[prop]) for c, p in _COUNTRY_INFO.items() if prop in p)

 
 
 def setup_country_tables(dsn, sql_dir, ignore_partitions=False):
     """ Create and populate the tables with basic static data that provides
         the background for geocoding. Data is assumed to not yet exist.
     """
 
 
 def setup_country_tables(dsn, sql_dir, ignore_partitions=False):
     """ Create and populate the tables with basic static data that provides
         the background for geocoding. Data is assumed to not yet exist.
     """

-    db_utils.execute_file(dsn, sql_dir / 'country_name.sql')

     db_utils.execute_file(dsn, sql_dir / 'country_osm_grid.sql.gz')
 
     params = []
     db_utils.execute_file(dsn, sql_dir / 'country_osm_grid.sql.gz')
 
     params = []


@@ -47,19 +113,25 @@ def setup_country_tables(dsn, sql_dir, ignore_partitions=False):
                 partition = 0
             else:
                 partition = props.get('partition')
                 partition = 0
             else:
                 partition = props.get('partition')

-            if ',' in (props.get('languages', ',') or ','):
-                lang = None
-            else:
-                lang = props['languages']
-            params.append((ccode, partition, lang))
+            lang = props['languages'][0] if len(
+                props['languages']) == 1 else None

 
 

+            params.append((ccode, props['names'], lang, partition))

     with connect(dsn) as conn:
         with conn.cursor() as cur:
     with connect(dsn) as conn:
         with conn.cursor() as cur:

+            psycopg2.extras.register_hstore(cur)
+            cur.execute(
+                """ CREATE TABLE public.country_name (
+                        country_code character varying(2),
+                        name public.hstore,
+                        derived_name public.hstore,
+                        country_default_language_code text,
+                        partition integer
+                    ); """)

             cur.execute_values(
             cur.execute_values(

-                """ UPDATE country_name
-                    SET partition = part, country_default_language_code = lang
-                    FROM (VALUES %s) AS v (cc, part, lang)
-                    WHERE country_code = v.cc""", params)
+                """ INSERT INTO public.country_name
+                    (country_code, name, country_default_language_code, partition) VALUES %s
+                """, params)

         conn.commit()
 
 
         conn.commit()
 
 


@@ -69,12 +141,9 @@ def create_country_names(conn, tokenizer, languages=None):
         empty then only name translations for the given languages are added
         to the index.
     """
         empty then only name translations for the given languages are added
         to the index.
     """

-    if languages:
-        languages = languages.split(',')
-

     def _include_key(key):
     def _include_key(key):

-        return key == 'name' or \
-               (key.startswith('name:') and (not languages or key[5:] in languages))
+        return ':' not in key or not languages or \
+               key[key.index(':') + 1:] in languages

 
     with conn.cursor() as cur:
         psycopg2.extras.register_hstore(cur)
 
     with conn.cursor() as cur:
         psycopg2.extras.register_hstore(cur)


@@ -84,14 +153,10 @@ def create_country_names(conn, tokenizer, languages=None):
         with tokenizer.name_analyzer() as analyzer:
             for code, name in cur:
                 names = {'countrycode': code}
         with tokenizer.name_analyzer() as analyzer:
             for code, name in cur:
                 names = {'countrycode': code}

-                if code == 'gb':
-                    names['short_name'] = 'UK'
-                if code == 'us':
-                    names['short_name'] = 'United States'

 
                 # country names (only in languages as provided)
                 if name:
 
                 # country names (only in languages as provided)
                 if name:

-                    names.update(((k, v) for k, v in name.items() if _include_key(k)))
+                    names.update({k : v for k, v in name.items() if _include_key(k)})

 
                 analyzer.add_country_names(code, names)
 
 
                 analyzer.add_country_names(code, names)