X-Git-Url: http://git.kaliko.me/?a=blobdiff_plain;f=sima%2Flib%2Fsimastr.py;h=56edbb4d7e68b82a816ad2d30cb6a99698d78864;hb=e0c9aed7aff3a338a023eb4dd9d2732546387ba0;hp=56cd2423a0474b540df2c4f010ee89f017585b0e;hpb=1d41464ccb6ff66441947eef0305518e3ce79a77;p=mpd-sima.git

diff --git a/sima/lib/simastr.py b/sima/lib/simastr.py
index 56cd242..56edbb4 100644
--- a/sima/lib/simastr.py
+++ b/sima/lib/simastr.py
@@ -1,5 +1,4 @@
 # -*- coding: utf-8 -*-
-
 #
 # Copyright (c) 2009, 2010, 2013 Jack Kaliko <kaliko@azylum.org>
 #
@@ -18,7 +17,7 @@
 #  If not, see <http://www.gnu.org/licenses/>.
 #
 
-"""
+r"""
 SimaStr
 
 Special unicode() subclass to perform fuzzy match on specific strings with
@@ -70,7 +69,7 @@ __version__ = '0.4'
 
 # IMPORTS
 import unicodedata
-from re import (compile, U, I)
+from re import compile as re_compile, U, I
 
 from ..utils.leven import levenshtein_ratio
 
@@ -80,6 +79,8 @@ class SimaStr(str):
     Specific string object for artist names and song titles.
     Here follows some class variables for regex to run on strings.
     """
+    diafilter = True
+    leven_ratio = 0.82
     regexp_dict = dict()
 
     # Leading patterns: The Le Les
@@ -92,21 +93,24 @@ class SimaStr(str):
     # Trailing patterns: ! ? live
     # TODO: add "concert" key word
     #       add "Live at <somewhere>"
-    regexp_dict.update({'trail': '([- !?\.]|\(? ?[Ll]ive ?\)?)'})
+    regexp_dict.update({'trail': r'([- !?\.]|\(? ?[Ll]ive ?\)?)'})
 
-    reg_lead = compile('^(?P<lead>%(lead)s )(?P<root0>.*)$' % regexp_dict, I | U)
-    reg_midl = compile('^(?P<root0>.*)(?P<mid> %(mid)s )(?P<root1>.*)' % regexp_dict, U)
-    reg_trail = compile('^(?P<root0>.*?)(?P<trail>%(trail)s+$)' % regexp_dict, U)
+    reg_lead = re_compile('^(?P<lead>%(lead)s )(?P<root0>.*)$' % regexp_dict, I | U)
+    reg_midl = re_compile('^(?P<root0>.*)(?P<mid> %(mid)s )(?P<root1>.*)' % regexp_dict, U)
+    reg_trail = re_compile('^(?P<root0>.*?)(?P<trail>%(trail)s+$)' % regexp_dict, U)
 
     def __init__(self, fuzzstr):
         """
         """
-        super().__init__(fuzzstr)
         self.orig = str(fuzzstr)
         self.stripped = str(fuzzstr.strip())
         # fuzzy computation
         self._get_root()
-        self.remove_diacritics()
+        if self.__class__.diafilter:
+            self.remove_diacritics()
+
+    def __new__(cls, fuzzstr):
+        return super(SimaStr, cls).__new__(cls, fuzzstr)
 
     def _get_root(self):
         """
@@ -129,6 +133,7 @@ class SimaStr(str):
             self.stripped = sea.group('root0')
 
     def remove_diacritics(self):
+        """converting diacritics"""
         self.stripped = ''.join(x for x in
                                 unicodedata.normalize('NFKD', self.stripped)
                                 if unicodedata.category(x) != 'Mn')
@@ -143,7 +148,7 @@ class SimaStr(str):
                                    other.stripped.lower())
         if hash(self) == hash(other):
             return True
-        return levenr >= 0.82
+        return levenr >= self.__class__.leven_ratio
 
     def __ne__(self, other):
         if not isinstance(other, SimaStr):