convert special phrase loaders to generators

Generators simplify the code quite a bit compared to the previous
Iterator approach.

nominatim/tools/special_phrases/sp_csv_loader.py

@@ -11,43 +11,31 @@
 """
 import csv
 import os
-from collections.abc import Iterator
 from nominatim.tools.special_phrases.special_phrase import SpecialPhrase
 from nominatim.errors import UsageError
 
-class SPCsvLoader(Iterator):
+class SPCsvLoader:
     """
         Handles loading of special phrases from external csv file.
     """
     def __init__(self, csv_path):
         super().__init__()
         self.csv_path = csv_path
-        self.has_been_read = False
 
-    def __next__(self):
-        if self.has_been_read:
-            raise StopIteration()
 
-        self.has_been_read = True
-        self.check_csv_validity()
-        return self.parse_csv()
-
-    def parse_csv(self):
-        """
-            Open and parse the given csv file.
+    def generate_phrases(self):
+        """ Open and parse the given csv file.
             Create the corresponding SpecialPhrases.
         """
-        phrases = set()
+        self._check_csv_validity()
 
         with open(self.csv_path, encoding='utf-8') as fd:
             reader = csv.DictReader(fd, delimiter=',')
             for row in reader:
-                phrases.add(
-                    SpecialPhrase(row['phrase'], row['class'], row['type'], row['operator'])
-                )
-        return phrases
+                yield SpecialPhrase(row['phrase'], row['class'], row['type'], row['operator'])
 
-    def check_csv_validity(self):
+
+    def _check_csv_validity(self):
         """
             Check that the csv file has the right extension.
         """

nominatim/tools/special_phrases/sp_importer.py

@@ -62,11 +62,10 @@ class SPImporter():
         # Store pairs of class/type for further processing
         class_type_pairs = set()
 
-        for loaded_phrases in self.sp_loader:
-            for phrase in loaded_phrases:
-                result = self._process_phrase(phrase)
-                if result:
-                    class_type_pairs.add(result)
+        for phrase in self.sp_loader.generate_phrases():
+            result = self._process_phrase(phrase)
+            if result:
+                class_type_pairs.add(result)
 
         self._create_place_classtype_table_and_indexes(class_type_pairs)
         if should_replace:

nominatim/tools/special_phrases/sp_wiki_loader.py

@@ -9,12 +9,24 @@
 """
 import re
 import logging
-from collections.abc import Iterator
 from nominatim.tools.special_phrases.special_phrase import SpecialPhrase
 from nominatim.tools.exec_utils import get_url
 
 LOG = logging.getLogger()
-class SPWikiLoader(Iterator):
+
+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()
+    return get_url(url)
+
+
+class SPWikiLoader:
     """
         Handles loading of special phrases from the wiki.
     """
@@ -27,28 +39,21 @@ class SPWikiLoader(Iterator):
         )
         self._load_languages()
 
-    def __next__(self):
-        if not self.languages:
-            raise StopIteration
 
-        lang = self.languages.pop(0)
-        loaded_xml = self._get_wiki_content(lang)
-        LOG.warning('Importing phrases for lang: %s...', lang)
-        return self.parse_xml(loaded_xml)
+    def generate_phrases(self):
+        """ Download the wiki pages for the configured languages
+            and extract the phrases from the page.
+        """
+        for lang in self.languages:
+            LOG.warning('Importing phrases for lang: %s...', lang)
+            loaded_xml = _get_wiki_content(lang)
+
+            # One match will be of format [label, class, type, operator, plural]
+            matches = self.occurence_pattern.findall(loaded_xml)
+
+            for match in matches:
+                yield SpecialPhrase(match[0], match[1], match[2], match[3])
 
-    def parse_xml(self, xml):
-        """
-            Parses XML content and extracts special phrases from it.
-            Return a list of SpecialPhrase.
-        """
-        # 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
 
     def _load_languages(self):
         """
@@ -64,15 +69,3 @@ class SPWikiLoader(Iterator):
             'et', 'eu', 'fa', 'fi', 'fr', 'gl', 'hr', 'hu',
             'ia', 'is', 'it', 'ja', 'mk', 'nl', 'no', 'pl',
             'ps', 'pt', 'ru', 'sk', 'sl', 'sv', 'uk', 'vi']
-
-    @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()
-        return get_url(url)