Merge pull request #2294 from lonvia/update-actions

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

diff --git a/nominatim/tools/special_phrases.py b/nominatim/tools/special_phrases.py
index 0c1258fefb0b9cece4c4e7ca88869b39fd97ab0e..28a077012024ac3bc49d60f18be87c382e05a0b9 100644 (file)
--- a/nominatim/tools/special_phrases.py
+++ b/nominatim/tools/special_phrases.py
@@ -27,7 +27,7 @@ class SpecialPhrasesImporter():
         self.black_list, self.white_list = self._load_white_and_black_lists()
         #Compile the regex here to increase performances.
         self.occurence_pattern = re.compile(
         self.black_list, self.white_list = self._load_white_and_black_lists()
         #Compile the regex here to increase performances.
         self.occurence_pattern = re.compile(

-            r'\| ([^\|]+) \|\| ([^\|]+) \|\| ([^\|]+) \|\| ([^\|]+) \|\| ([\-YN])'
+            r'\| *([^\|]+) *\|\| *([^\|]+) *\|\| *([^\|]+) *\|\| *([^\|]+) *\|\| *([\-YN])'

         )
         self.sanity_check_pattern = re.compile(r'^\w+$')
         self.transliterator = Transliterator.createFromRules("special-phrases normalizer",
         )
         self.sanity_check_pattern = re.compile(r'^\w+$')
         self.transliterator = Transliterator.createFromRules("special-phrases normalizer",


@@ -116,7 +116,7 @@ class SpecialPhrasesImporter():
         if self.config.PHRASE_CONFIG:
             settings_path = self._convert_php_settings_if_needed(self.config.PHRASE_CONFIG)
 
         if self.config.PHRASE_CONFIG:
             settings_path = self._convert_php_settings_if_needed(self.config.PHRASE_CONFIG)
 

-        with open(settings_path, "r") as json_settings:
+        with settings_path.open("r") as json_settings:

             settings = json.load(json_settings)
         return settings['blackList'], settings['whiteList']
 
             settings = json.load(json_settings)
         return settings['blackList'], settings['whiteList']
 


@@ -153,8 +153,10 @@ class SpecialPhrasesImporter():
         class_matchs = self.sanity_check_pattern.findall(phrase_class)
 
         if len(class_matchs) < 1 or len(type_matchs) < 1:
         class_matchs = self.sanity_check_pattern.findall(phrase_class)
 
         if len(class_matchs) < 1 or len(type_matchs) < 1:

-            raise UsageError("Bad class/type for language {}: {}={}".format(
-                lang, phrase_class, phrase_type))
+            LOG.warning("Bad class/type for language %s: %s=%s. It will not be imported",
+                        lang, phrase_class, phrase_type)
+            return False
+        return True

 
     def _process_xml_content(self, xml_content, lang):
         """
 
     def _process_xml_content(self, xml_content, lang):
         """


@@ -205,7 +207,8 @@ class SpecialPhrasesImporter():
                 continue
 
             #sanity check, in case somebody added garbage in the wiki
                 continue
 
             #sanity check, in case somebody added garbage in the wiki

-            self._check_sanity(lang, phrase_class, phrase_type)
+            if not self._check_sanity(lang, phrase_class, phrase_type):
+                continue

 
             class_type_pairs.add((phrase_class, phrase_type))
 
 
             class_type_pairs.add((phrase_class, phrase_type))