Add get_emoji_regex function

Author: alexmick

emoji_data_python/__init__.py

@@ -4,24 +4,25 @@
 
 from .conversion import unified_to_char
 from .emoji_char import EmojiChar
-from .replacement import replace_colons
+from .replacement import replace_colons, get_emoji_regex
 from .search import all_doublebyte, find_by_shortname, find_by_name
 
 # Read json data on module load to be cached
 with open(path.join(path.dirname(__file__), 'data/emoji.json'), 'r') as full_data:
     # Load and parse emoji data from json into EmojiChar objects
     emoji_data = [EmojiChar(data_blob) for data_blob in json.loads(full_data.read())]  # type: List[EmojiChar]
 
-    # Build a cached dictionary of short names for quicker access, short code keys are normalized with underscores
-    emoji_short_names = {
-        emoji.short_name.replace('-', '_'): emoji for emoji in emoji_data
-    }  # type: Dict[str, EmojiChar]
+# Build a cached dictionary of short names for quicker access, short code keys are normalized with underscores
+emoji_short_names = {
+    emoji.short_name.replace('-', '_'): emoji for emoji in emoji_data
+}  # type: Dict[str, EmojiChar]
 
-    # Add other short names if they are not already used as a primary short name for an other emoji
-    for emoji in emoji_data:
-        for short_name in emoji.short_names:
-            if short_name not in emoji_short_names:
-                emoji_short_names[short_name] = emoji
+# Add other short names if they are not already used as a primary short name for an other emoji
+for emoji in emoji_data:
+    for short_name in emoji.short_names:
+        if short_name not in emoji_short_names:
+            emoji_short_names[short_name] = emoji
 
 
-__all__ = ['unified_to_char', 'EmojiChar', 'replace_colons', 'all_doublebyte', 'find_by_shortname', 'find_by_name', 'emoji_data', 'emoji_short_names']
+__all__ = ['unified_to_char', 'EmojiChar', 'replace_colons', 'get_emoji_regex',
+           'all_doublebyte', 'find_by_shortname', 'find_by_name', 'emoji_data', 'emoji_short_names']

emoji_data_python/replacement.py

@@ -1,12 +1,12 @@
-from re import sub
+import re
 
 
 def replace_colons(text: str, strip: bool=False) -> str:
     """Parses a string with colon encoded emoji and renders found emoji.
     Unknown emoji are left as is unless `strip` is set to `True`
 
     :param text: String of text to parse and replace
-    :param strip: Whether to strip unknown codes or to leave them as `:unkown:`
+    :param strip: Whether to strip unknown codes or to leave them as `:unknown:`
 
     >>> emoji_data_python.replace_colons('Hello world ! :wave::skin-tone-3: :earth_africa: :exclamation:')
     'Hello world ! 👋🏼 🌍 ❗'
@@ -27,4 +27,19 @@ def emoji_repl(matchobj) -> str:
 
         return res
 
-    return sub(r'\:[a-zA-Z0-9-_+]+\:(\:skin-tone-[2-6]\:)?', emoji_repl, text)
+    return re.sub(r'\:[a-zA-Z0-9-_+]+\:(\:skin-tone-[2-6]\:)?', emoji_repl, text)
+
+
+def get_emoji_regex():
+    """Returns a regex to match any emoji
+
+    >>> emoji_data_python.get_emoji_regex().findall('Hello world ! 👋🏼 🌍 ❗')
+    ['👋', '🏼', '🌍', '❗']
+    """
+    from emoji_data_python import emoji_data
+    # Sort emojis by length to make sure mulit-character emojis are
+    # matched first
+
+    emojis = sorted([emoji.char for emoji in emoji_data], key=len, reverse=True)
+    pattern = u'(' + u'|'.join(re.escape(u) for u in emojis) + u')'
+    return re.compile(pattern)

tests/test_replacement.py

@@ -1,6 +1,6 @@
 import unittest
 
-from emoji_data_python import replace_colons
+from emoji_data_python import replace_colons, get_emoji_regex
 
 
 class ReplaceColonsTestCase(unittest.TestCase):
@@ -40,3 +40,10 @@ def test_multiline_sentence(self):
 How are you :question:""")
         )
 
+    def test_emoji_regex(self):
+        regex = get_emoji_regex()
+        self.assertRegex('😄', regex)
+        self.assertRegex('👪', regex)
+        self.assertNotRegex('hello :wave: l → ▶', regex)
+        res = regex.findall('💩💩 🏼 bla 👋🏼')
+        self.assertEqual(len(res), 5)  # Wave + skin tone counts as two