support ignoring characters

Author: mollerhoj

tests/test_unidecode.py

@@ -68,6 +68,22 @@ def test_ascii(self):
 
         wlog.stop()
 
+
+    def test_ignore(self):
+
+        wlog = WarningLogger()
+        wlog.start("should be ignored")
+
+        t = 'æøå'
+        r = self.unidecode(t, ignore='æøå')
+        self.assertEqual(r, t)
+        self.assertEqual(type(r), str)
+
+        # unicode objects shouldn't raise warnings
+        self.assertEqual(0, len(wlog.log))
+
+        wlog.stop()
+
     def test_bmp(self):
         for n in range(0,0x10000):
             # skip over surrogate pairs, which throw a warning

unidecode/__init__.py

@@ -28,7 +28,7 @@ def _warn_if_not_unicode(string):
                         RuntimeWarning, 2)
 
 
-def unidecode_expect_ascii(string):
+def unidecode_expect_ascii(string, ignore=''):
     """Transliterate an Unicode object into an ASCII string
 
     >>> unidecode(u"\u5317\u4EB0")
@@ -47,30 +47,34 @@ def unidecode_expect_ascii(string):
     try:
         bytestring = string.encode('ASCII')
     except UnicodeEncodeError:
-        return _unidecode(string)
+        return _unidecode(string, ignore)
     if version_info[0] >= 3:
         return string
     else:
         return bytestring
 
-def unidecode_expect_nonascii(string):
+def unidecode_expect_nonascii(string, ignore=''):
     """Transliterate an Unicode object into an ASCII string
 
     >>> unidecode(u"\u5317\u4EB0")
     "Bei Jing "
     """
 
     _warn_if_not_unicode(string)
-    return _unidecode(string)
+    return _unidecode(string, ignore)
 
 unidecode = unidecode_expect_ascii
 
-def _unidecode(string):
+def _unidecode(string, ignore=''):
     retval = []
 
     for char in string:
         codepoint = ord(char)
 
+        if char in ignore:
+            retval.append(str(char))
+            continue
+
         if codepoint < 0x80: # Basic ASCII
             retval.append(str(char))
             continue