Moved KANJI_MAP to global variable

diff --git a/nominatim/tokenizer/sanitizers/kanji_utils.py b/nominatim/tokenizer/sanitizers/kanji_utils.py
deleted file mode 100644 (file)
index 6956152..0000000
--- a/nominatim/tokenizer/sanitizers/kanji_utils.py
+++ /dev/null
@@ -1,36 +0,0 @@
-# SPDX-License-Identifier: GPL-3.0-or-later
-#
-# 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.
-"""
-This is a file for a function that converts Kanji (Japanese) numerals to Arabic numerals.
-"""
-
-def convert_kanji_sequence_to_number(sequence: str) -> str:
-    """Converts Kanji numbers to Arabic numbers
-    """
-    kanji_map = {
-      '零': '0',
-      '一': '1',
-      '二': '2',
-      '三': '3',
-      '四': '4',
-      '五': '5',
-      '六': '6',
-      '七': '7',
-      '八': '8',
-      '九': '9'
-    }
-    converted = ''
-    current_number = ''
-    for char in sequence:
-        if char in kanji_map:
-            current_number += kanji_map[char]
-        else:
-            converted += current_number
-            current_number = ''
-            converted += char
-    converted += current_number
-    return converted

diff --git a/nominatim/tokenizer/sanitizers/tag_japanese.py b/nominatim/tokenizer/sanitizers/tag_japanese.py
index dffd95595d3f7e532e2c402428c39ed7229c58d4..723a6dfaa27bca7b53854ec01504ff5d23d5a39b 100644 (file)
--- a/nominatim/tokenizer/sanitizers/tag_japanese.py
+++ b/nominatim/tokenizer/sanitizers/tag_japanese.py
@@ -17,7 +17,34 @@ from typing import List, Optional
 from nominatim.tokenizer.sanitizers.base import ProcessInfo
 from nominatim.tokenizer.sanitizers.config import SanitizerConfig
 from nominatim.data.place_name import PlaceName
 from nominatim.tokenizer.sanitizers.base import ProcessInfo
 from nominatim.tokenizer.sanitizers.config import SanitizerConfig
 from nominatim.data.place_name import PlaceName

-from nominatim.tokenizer.sanitizers.kanji_utils import convert_kanji_sequence_to_number
+
+KANJI_MAP = {
+      '零': '0',
+      '一': '1',
+      '二': '2',
+      '三': '3',
+      '四': '4',
+      '五': '5',
+      '六': '6',
+      '七': '7',
+      '八': '8',
+      '九': '9'
+    }
+
+def convert_kanji_sequence_to_number(sequence: str) -> str:
+    """Converts Kanji numbers to Arabic numbers
+    """
+    converted = ''
+    current_number = ''
+    for char in sequence:
+        if char in KANJI_MAP:
+            current_number += KANJI_MAP[char]
+        else:
+            converted += current_number
+            current_number = ''
+            converted += char
+    converted += current_number
+    return converted

 
 def create(_: SanitizerConfig) -> Callable[[ProcessInfo], None]:
 #def create(config: SanitizerConfig) -> Callable[[ProcessInfo],None]:
 
 def create(_: SanitizerConfig) -> Callable[[ProcessInfo], None]:
 #def create(config: SanitizerConfig) -> Callable[[ProcessInfo],None]: