sp_wiki_loader: add more default languages

[nominatim.git] / nominatim / tools / special_phrases / sp_wiki_loader.py

diff --git a/nominatim/tools/special_phrases/sp_wiki_loader.py b/nominatim/tools/special_phrases/sp_wiki_loader.py
index 11e59b45a33ea88eb6098ba7b74d01bd8dd1dd93..cbeaeabfdfae752b11825b33c544e4d6476f3b8f 100644 (file)
--- a/nominatim/tools/special_phrases/sp_wiki_loader.py
+++ b/nominatim/tools/special_phrases/sp_wiki_loader.py
@@ -1,71 +1,68 @@
+# 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.
 """
     Module containing the SPWikiLoader class.
 """
-import logging
+from typing import Iterable
 import re
+import logging
+
+from nominatim.config import Configuration
 from nominatim.tools.special_phrases.special_phrase import SpecialPhrase
-from nominatim.tools.special_phrases.sp_loader import SPLoader
 from nominatim.tools.exec_utils import get_url
 
 LOG = logging.getLogger()
-class SPWikiLoader(SPLoader):
+
+def _get_wiki_content(lang: str) -> str:
+    """
+        Request and return the wiki page's content
+        corresponding to special phrases for a given lang.
+        Requested URL Example :
+            https://wiki.openstreetmap.org/wiki/Special:Export/Nominatim/Special_Phrases/EN
+    """
+    url = 'https://wiki.openstreetmap.org/wiki/Special:Export/Nominatim/Special_Phrases/' \
+          + lang.upper()
+    return get_url(url)
+
+
+class SPWikiLoader:
     """
         Handles loading of special phrases from the wiki.
     """
-    def __init__(self, config, languages=None):
-        if languages is not None and not isinstance(languages, list):
-            raise TypeError('The \'languages\' parameter should be of type list.')
-        super().__init__()
+    def __init__(self, config: Configuration) -> None:
         self.config = config
-        #Compile the regex here to increase performances.
+        # Compile the regex here to increase performances.
         self.occurence_pattern = re.compile(
             r'\| *([^\|]+) *\|\| *([^\|]+) *\|\| *([^\|]+) *\|\| *([^\|]+) *\|\| *([\-YN])'
         )
-        self.languages = self._load_languages() if not languages else languages
+        # Hack around a bug where building=yes was imported with quotes into the wiki
+        self.type_fix_pattern = re.compile(r'\"|"')
 
-    def __next__(self):
-        if not self.languages:
-            raise StopIteration
+        self.languages = self.config.get_str_list('LANGUAGES') or \
+                         ['af', 'ar', 'br', 'ca', 'cs', 'de', 'en', 'es',
+                          'et', 'eu', 'fa', 'fi', 'fr', 'gl', 'hr', 'hu',
+                          'ia', 'is', 'it', 'ja', 'mk', 'nl', 'no', 'pl',
+                          'ps', 'pt', 'ru', 'sk', 'sl', 'sv', 'uk', 'vi',
+                          'lv', 'tr']
 
-        lang = self.languages.pop(0)
-        loaded_xml = SPWikiLoader._get_wiki_content(lang)
-        LOG.warning('Importing phrases for lang: %s...', lang)
-        return self.parse_xml(loaded_xml)
 
-    def parse_xml(self, xml):
-        """
-            Parses XML content and extracts special phrases from it.
-            Return a list of SpecialPhrase.
+    def generate_phrases(self) -> Iterable[SpecialPhrase]:
+        """ Download the wiki pages for the configured languages
+            and extract the phrases from the page.
         """
-        #One match will be of format [label, class, type, operator, plural]
-        matches = self.occurence_pattern.findall(xml)
-        returned_phrases = set()
-        for match in matches:
-            returned_phrases.add(
-                SpecialPhrase(match[0], match[1], match[2], match[3])
-            )
-        return returned_phrases
+        for lang in self.languages:
+            LOG.warning('Importing phrases for lang: %s...', lang)
+            loaded_xml = _get_wiki_content(lang)
 
-    def _load_languages(self):
-        """
-            Get list of all languages from env config file
-            or default if there is no languages configured.
-            The system will extract special phrases only from all specified languages.
-        """
-        default_languages = [
-            'af', 'ar', 'br', 'ca', 'cs', 'de', 'en', 'es',
-            'et', 'eu', 'fa', 'fi', 'fr', 'gl', 'hr', 'hu',
-            'ia', 'is', 'it', 'ja', 'mk', 'nl', 'no', 'pl',
-            'ps', 'pt', 'ru', 'sk', 'sl', 'sv', 'uk', 'vi']
-        return self.config.LANGUAGES.split(',') if self.config.LANGUAGES else default_languages
+            # One match will be of format [label, class, type, operator, plural]
+            matches = self.occurence_pattern.findall(loaded_xml)
 
-    @staticmethod
-    def _get_wiki_content(lang):
-        """
-            Request and return the wiki page's content
-            corresponding to special phrases for a given lang.
-            Requested URL Example :
-                https://wiki.openstreetmap.org/wiki/Special:Export/Nominatim/Special_Phrases/EN
-        """
-        url = 'https://wiki.openstreetmap.org/wiki/Special:Export/Nominatim/Special_Phrases/' + lang.upper() # pylint: disable=line-too-long
-        return get_url(url)
+            for match in matches:
+                yield SpecialPhrase(match[0],
+                                    match[1],
+                                    self.type_fix_pattern.sub('', match[2]),
+                                    match[3])