Improvement patch on jp (#30)

* More punctuation sign hardcoded. * Do not trigger suspicious range on hiragana x katakana * bump version minor
jawah · Dec 16, 2019 · 48c2e6b · 48c2e6b
1 parent b0e4e94
commit 48c2e6b
Show file tree

Hide file tree

Showing 2 changed files with 5 additions and 2 deletions.
diff --git a/charset_normalizer/unicode.py b/charset_normalizer/unicode.py
@@ -69,7 +69,7 @@ def is_punc(letter):
         return r_name is not None and \
                ("Punctuation" in r_name or
                'Forms' in r_name or
-               letter in 'º¯—–‒‐⁃«‹?!;.:^$*»£¹¿~ª؟©±¡{}[]|¼½¾⅕⅙⅛™℠‼⁇❝❞¶⁋√↑↓�¤`')
+               letter in set('º¯—–‒‐⁃«‹?!;.:^$¥*»£¹¿~ª؟©±¡{}[]|½⅓⅔¼¾⅕⅖⅗⅘⅙⅚⅐⅛⅜⅝⅞⅑⅒™℠¬‼⁇❝❞¶⁋√↑↓�¤`¨'))
 
     @staticmethod
     @lru_cache(maxsize=8192)
@@ -141,6 +141,9 @@ def is_suspiciously_successive_range(range_name_a, range_name_b):
         if 'CJK' in range_name_a and range_name_b in ['Katakana', 'Hiragana']:
             return False
 
+        if range_name_a in ['Katakana', 'Hiragana'] and range_name_b in ['Katakana', 'Hiragana']:
+            return False
+
         return True
 
     @staticmethod

diff --git a/charset_normalizer/version.py b/charset_normalizer/version.py
@@ -2,5 +2,5 @@
 Expose version
 """
 
-__version__ = "1.3.2"
+__version__ = "1.3.3"
 VERSION = __version__.split('.')