MD improvement on trailing data and long foreign (non-pure latin) data (

#124)

* ❇️ Always feed a final non printable character to detect trailing mess

some MD plugins require a "separator" character before assessing the current buffer.

* ❇️ Never ignore non-pure latin (foreign) word that are too long

charset_normalizer/md.py

@@ -252,6 +252,8 @@ class SuperWeirdWordPlugin(MessDetectorPlugin):
     def __init__(self) -> None:
         self._word_count = 0  # type: int
         self._bad_word_count = 0  # type: int
+        self._foreign_long_count = 0  # type: int
+
         self._is_current_word_bad = False  # type: bool
         self._foreign_long_watch = False  # type: bool
 
@@ -271,7 +273,7 @@ def feed(self, character: str) -> None:
                 self._buffer_accent_count += 1
             if (
                 self._foreign_long_watch is False
-                and is_latin(character) is False
+                and (is_latin(character) is False or is_accentuated(character))
                 and is_cjk(character) is False
                 and is_hangul(character) is False
                 and is_katakana(character) is False
@@ -293,6 +295,7 @@ def feed(self, character: str) -> None:
             if buffer_length >= 4 and self._buffer_accent_count / buffer_length > 0.34:
                 self._is_current_word_bad = True
             if buffer_length >= 24 and self._foreign_long_watch:
+                self._foreign_long_count += 1
                 self._is_current_word_bad = True
 
             if self._is_current_word_bad:
@@ -319,10 +322,11 @@ def reset(self) -> None:  # pragma: no cover
         self._word_count = 0
         self._character_count = 0
         self._bad_character_count = 0
+        self._foreign_long_count = 0
 
     @property
     def ratio(self) -> float:
-        if self._word_count <= 10:
+        if self._word_count <= 10 and self._foreign_long_count == 0:
             return 0.0
 
         return self._bad_character_count / self._character_count
@@ -515,7 +519,7 @@ def mess_ratio(
         md_class() for md_class in MessDetectorPlugin.__subclasses__()
     ]  # type: List[MessDetectorPlugin]
 
-    length = len(decoded_sequence)  # type: int
+    length = len(decoded_sequence) + 1  # type: int
 
     mean_mess_ratio = 0.0  # type: float
 
@@ -526,7 +530,7 @@ def mess_ratio(
     else:
         intermediary_mean_mess_ratio_calc = 128
 
-    for character, index in zip(decoded_sequence, range(0, length)):
+    for character, index in zip(decoded_sequence + "\n", range(0, length)):
         for detector in detectors:
             if detector.eligible(character):
                 detector.feed(character)