Merge pull request #3110 from lonvia/sql-lambda-queries

[nominatim.git] / nominatim / tokenizer / sanitizers / delete_tags.py

diff --git a/nominatim/tokenizer/sanitizers/delete_tags.py b/nominatim/tokenizer/sanitizers/delete_tags.py
index fd35de488c617f0624271323fe525e23f5ff8b2f..f18a894b07e480fdab61251dba246a4f65c925ad 100644 (file)
--- a/nominatim/tokenizer/sanitizers/delete_tags.py
+++ b/nominatim/tokenizer/sanitizers/delete_tags.py
@@ -54,8 +54,7 @@ Arguments:
 \r
 \r
 """\r
-from typing import Callable, List, Optional, Pattern, Tuple, Sequence\r
-import re\r
+from typing import Callable, List, Tuple, Sequence\r
 \r
 from nominatim.tokenizer.sanitizers.base import ProcessInfo\r
 from nominatim.data.place_name import PlaceName\r
@@ -65,37 +64,33 @@ class _TagSanitizer:
 \r
     def __init__(self, config: SanitizerConfig) -> None:\r
         self.type = config.get('type', 'name')\r
-        self.filter_kind = config.get_filter_kind()\r
+        self.filter_kind = config.get_filter('filter-kind')\r
         self.country_codes = config.get_string_list('country_code', [])\r
-        self.allowed_ranks = self._set_allowed_ranks( \\r
-                                            config.get_string_list('rank_address', ['0-30']))\r
+        self.filter_suffix = config.get_filter('suffix')\r
+        self.filter_name = config.get_filter('name')\r
+        self.allowed_ranks = self._set_allowed_ranks(\r
+            config.get_string_list("rank_address", ["0-30"])\r
+        )\r
 \r
         self.has_country_code = config.get('country_code', None) is not None\r
 \r
-        suffixregexps = config.get_string_list('suffix', [r'[\s\S]*'])\r
-        self.suffix_regexp = [re.compile(r) for r in suffixregexps]\r
-\r
-        nameregexps = config.get_string_list('name', [r'[\s\S]*'])\r
-        self.name_regexp = [re.compile(r) for r in nameregexps]\r
-\r
-\r
 \r
     def __call__(self, obj: ProcessInfo) -> None:\r
         tags = obj.names if self.type == 'name' else obj.address\r
 \r
-        if (not tags or\r
-             self.has_country_code and\r
-              obj.place.country_code not in self.country_codes or\r
-               not self.allowed_ranks[obj.place.rank_address]):\r
+        if not tags \\r
+           or not self.allowed_ranks[obj.place.rank_address] \\r
+           or self.has_country_code \\r
+           and obj.place.country_code not in self.country_codes:\r
             return\r
 \r
         filtered_tags: List[PlaceName] = []\r
 \r
         for tag in tags:\r
 \r
-            if (not self.filter_kind(tag.kind) or\r
-                  not self._matches(tag.suffix, self.suffix_regexp) or\r
-                    not self._matches(tag.name, self.name_regexp)):\r
+            if not self.filter_kind(tag.kind) \\r
+               or not self.filter_suffix(tag.suffix or '') \\r
+               or not self.filter_name(tag.name):\r
                 filtered_tags.append(tag)\r
 \r
 \r
@@ -117,7 +112,7 @@ class _TagSanitizer:
         for rank in ranks:\r
             intvl = [int(x) for x in rank.split('-')]\r
 \r
-            start, end = (intvl[0], intvl[0]) if len(intvl) == 1 else (intvl[0], intvl[1])\r
+            start, end = intvl[0], intvl[0] if len(intvl) == 1 else intvl[1]\r
 \r
             for i in range(start, end + 1):\r
                 allowed_ranks[i] = True\r
@@ -126,17 +121,6 @@ class _TagSanitizer:
         return tuple(allowed_ranks)\r
 \r
 \r
-    def _matches(self, value: Optional[str], patterns: List[Pattern[str]]) -> bool:\r
-        """ Returns True if the given value fully matches any of the regular\r
-            expression pattern in the list. Otherwise, returns False.\r
-\r
-            Note that if the value is None, it is taken as an empty string.\r
-        """\r
-        target = '' if value is None else value\r
-        return any(r.fullmatch(target) is not None for r in patterns)\r
-\r
-\r
-\r
 def create(config: SanitizerConfig) -> Callable[[ProcessInfo], None]:\r
     """ Create a function to process removal of certain tags.\r
     """\r