Merge pull request #2786 from lonvia/export-centroid-for-tokenizer

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

diff --git a/nominatim/tools/special_phrases/sp_csv_loader.py b/nominatim/tools/special_phrases/sp_csv_loader.py
index a32a4b39e5a74387c144fdc5f41c4b7ecd638f94..400f9fa91aa3efec500a8e40b3e7f1df08e609bf 100644 (file)
--- a/nominatim/tools/special_phrases/sp_csv_loader.py
+++ b/nominatim/tools/special_phrases/sp_csv_loader.py
@@ -9,49 +9,37 @@
 
     The class allows to load phrases from a csv file.
 """
 
     The class allows to load phrases from a csv file.
 """

+from typing import Iterable

 import csv
 import os
 import csv
 import os

-from collections.abc import Iterator

 from nominatim.tools.special_phrases.special_phrase import SpecialPhrase
 from nominatim.errors import UsageError
 
 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.
     """
     """
         Handles loading of special phrases from external csv file.
     """

-    def __init__(self, csv_path):
-        super().__init__()
+    def __init__(self, csv_path: str) -> None:

         self.csv_path = csv_path
         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) -> Iterable[SpecialPhrase]:
+        """ Open and parse the given csv file.

             Create the corresponding SpecialPhrases.
         """
             Create the corresponding SpecialPhrases.
         """

-        phrases = set()
+        self._check_csv_validity()

 
 

-        with open(self.csv_path) as file:
-            reader = csv.DictReader(file, delimiter=',')
+        with open(self.csv_path, encoding='utf-8') as fd:
+            reader = csv.DictReader(fd, delimiter=',')

             for row in reader:
             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) -> None:

         """
             Check that the csv file has the right extension.
         """
         _, extension = os.path.splitext(self.csv_path)
 
         if extension != '.csv':
         """
             Check that the csv file has the right extension.
         """
         _, extension = os.path.splitext(self.csv_path)
 
         if extension != '.csv':

-            raise UsageError('The file {} is not a csv file.'.format(self.csv_path))
+            raise UsageError(f'The file {self.csv_path} is not a csv file.')