Merge pull request #291 from lovetox/typing

Tests: Add type hints

.github/workflows/pythonTests.yml

@@ -42,7 +42,7 @@ jobs:
 
       - name: Install Dependencies
         run: |
-          python -m pip install .
+          python -m pip install .[dev]
 
       - name: Pyright
         uses: jakebailey/pyright-action@v2

pyproject.toml

@@ -67,8 +67,5 @@ exclude = [
   ".git",
   ".venv",
   "build",
-  "docs",
-  "example",
-  "tests",
   "utils",
 ]

tests/test_analyze.py

@@ -15,44 +15,57 @@ def test_analyze():
 
     result = list(emoji.analyze('abc\U0001F472'))
     assert len(result) == 1
+    assert not isinstance(result[0].value, str)
     assert result[0].value.emoji == '\U0001F472'
 
     result = list(emoji.analyze('abc\U0001F472', non_emoji=True))
     assert result[0].value == 'a'
+    assert not isinstance(result[3].value, str)
     assert result[3].value.emoji == '\U0001F472'
 
     result = list(emoji.analyze('\U0001F477\U0001F3FB\U0000200D\U00002640'))
     assert len(result) == 1
+    assert not isinstance(result[0].value, str)
     assert result[0].value.emoji == '\U0001F477\U0001F3FB\U0000200D\U00002640'
 
     result = list(emoji.analyze('\U0001F477\U0001F3FC\U0001F477\U0001F3FB\U0000200D\U00002640'))
     assert len(result) == 2
+    assert not isinstance(result[0].value, str)
     assert result[0].value.emoji == '\U0001F477\U0001F3FC'
+    assert not isinstance(result[1].value, str)
     assert result[1].value.emoji == '\U0001F477\U0001F3FB\U0000200D\U00002640'
 
 
 def test_analyze_non_rgi_zwj():
     result = list(emoji.analyze('\U0001F468\U0001F3FF\U0000200D\U0001F469\U0001F3FB\U0000200D\U0001F467\U0001F3FD'))
     assert len(result) == 1
+    assert not isinstance(result[0].value, str)
     assert result[0].value.emoji == '\U0001F468\U0001F3FF\U0000200D\U0001F469\U0001F3FB\U0000200D\U0001F467\U0001F3FD'
 
     result = list(emoji.analyze('\U0001F468\U0001F3FF\U0000200D\U0001F469\U0001F3FB\U0000200D\U0001F467\U0001F3FD', join_emoji=False))
     assert len(result) == 3
+    assert not isinstance(result[0].value, str)
     assert result[0].value.emoji == '\U0001F468\U0001F3FF'
+    assert not isinstance(result[1].value, str)
     assert result[1].value.emoji == '\U0001F469\U0001F3FB'
+    assert not isinstance(result[2].value, str)
     assert result[2].value.emoji == '\U0001F467\U0001F3FD'
 
     result = list(emoji.analyze('\U0001F468\U0001F3FF\U0000200D\U0001F469\U0001F3FB\U0000200D\U0001F467\U0001F3FDx', join_emoji=False, non_emoji=True))
     assert len(result) == 6
+    assert not isinstance(result[0].value, str)
     assert result[0].value.emoji == '\U0001F468\U0001F3FF'
     assert result[1].value == '\U0000200D'
+    assert not isinstance(result[2].value, str)
     assert result[2].value.emoji == '\U0001F469\U0001F3FB'
     assert result[3].value == '\U0000200D'
+    assert not isinstance(result[4].value, str)
     assert result[4].value.emoji == '\U0001F467\U0001F3FD'
     assert result[5].value == 'x'
 
     result = list(emoji.analyze('\U0001F468\U0001F3FF\U0000200D\U0001F469\U0001F3FB\U0000200D\U0001F467\U0001F3FDx', join_emoji=True, non_emoji=True))
     assert len(result) == 2
+    assert not isinstance(result[0].value, str)
     assert result[0].value.emoji == '\U0001F468\U0001F3FF\U0000200D\U0001F469\U0001F3FB\U0000200D\U0001F467\U0001F3FD'
     assert result[1].value == 'x'
 
@@ -72,23 +85,6 @@ def test_analyze_non_rgi_zwj():
     assert isinstance(result[3].value, emoji.EmojiMatch)
 
 
-def test_emoji_match():
-    s = 'a\U0001F309b'
-    token = next(emoji.analyze(s))
-    assert isinstance(token, emoji.Token)
-
-    assert token.chars == s[1:-1]
-
-    match = token.value
-
-    assert isinstance(match, emoji.EmojiMatch)
-    assert match.emoji == s[1:-1]
-    assert match.start == 1
-    assert match.end == 2
-    assert match.is_zwj() == False
-    assert str(match).startswith('EmojiMatch(')
-
-
 def test_emoji_match():
     s = 'a\U0001F309b'
     token = next(emoji.analyze(s))

tests/test_core.py

@@ -2,17 +2,21 @@
 
 import random
 import re
-import emoji
+from typing import Any, Callable, Dict, List, Tuple, Union
+from typing_extensions import Literal
+import emoji.unicode_codes
 import pytest
 import unicodedata
 
+_NormalizationForm = Literal['NFC', 'NFD', 'NFKC', 'NFKD']
+
 # Build all language packs (i.e. fill the cache):
 emoji.emojize("", language="alias")
 for lang_code in emoji.LANGUAGES:
     emoji.emojize("", language=lang_code)
 
 
-def ascii(s):
+def ascii(s: str) -> str:
     # return escaped Code points \U000AB123
     return s.encode("unicode-escape").decode()
 
@@ -24,13 +28,13 @@ def all_language_and_alias_packs():
         yield (lang_code, emoji.unicode_codes.get_emoji_unicode_dict(lang_code))
 
 
-def normalize(form, s):
+def normalize(form: _NormalizationForm, s: str) -> str:
     return unicodedata.normalize(form, s)
 
 
 def test_emojize_name_only():
     # Check that the regular expression emoji.core._EMOJI_NAME_PATTERN contains all the necesseary characters
-    from emoji.core import _EMOJI_NAME_PATTERN
+    from emoji.core import _EMOJI_NAME_PATTERN  # pyright: ignore [reportPrivateUsage]
 
     pattern = re.compile('[^%s]' % (_EMOJI_NAME_PATTERN, ))
 
@@ -59,7 +63,7 @@ def test_emojize_name_only():
 
 def test_regular_expression_minimal():
     # Check that the regular expression emoji.core._EMOJI_NAME_PATTERN only contains the necesseary characters
-    from emoji.core import _EMOJI_NAME_PATTERN
+    from emoji.core import _EMOJI_NAME_PATTERN  # pyright: ignore [reportPrivateUsage]
 
     pattern_str = '[^%s]' % (_EMOJI_NAME_PATTERN, )
     i = 2
@@ -70,7 +74,7 @@ def test_regular_expression_minimal():
             continue
         pattern = re.compile(pattern_str.replace(c, ''))
         failed = False
-        for lang_code, emoji_pack in all_language_and_alias_packs():
+        for _, emoji_pack in all_language_and_alias_packs():
             for name_in_db in emoji_pack.keys():
                 name_in_db = name_in_db[1:-1]
                 names = [
@@ -112,45 +116,46 @@ def test_emojize_complicated_string():
 
 
 def test_emojize_languages():
-    for lang_code, emoji_pack in emoji.unicode_codes._EMOJI_UNICODE.items():
+    for lang_code, emoji_pack in emoji.unicode_codes._EMOJI_UNICODE.items():  # pyright: ignore [reportPrivateUsage]
         for name, emj in emoji_pack.items():
             assert emoji.emojize(name, language=lang_code) == emj
 
 
 def test_demojize_languages():
-    for lang_code, emoji_pack in emoji.unicode_codes._EMOJI_UNICODE.items():
+    for lang_code, emoji_pack in emoji.unicode_codes._EMOJI_UNICODE.items():  # pyright: ignore [reportPrivateUsage]
         for name, emj in emoji_pack.items():
             assert emoji.demojize(emj, language=lang_code) == name
 
 
 def test_emojize_variant():
-    def remove_variant(s): return re.sub('[\ufe0e\ufe0f]$', '', s)
+    def remove_variant(s: str) -> str:
+        return re.sub('[\ufe0e\ufe0f]$', '', s)
 
     assert emoji.emojize(
-        ':Taurus:', variant=None) == emoji.unicode_codes._EMOJI_UNICODE['en'][':Taurus:']
+        ':Taurus:', variant=None) == emoji.unicode_codes._EMOJI_UNICODE['en'][':Taurus:']  # pyright: ignore [reportPrivateUsage]
     assert emoji.emojize(':Taurus:', variant=None) == emoji.emojize(':Taurus:')
     assert emoji.emojize(':Taurus:', variant='text_type') == remove_variant(
-        emoji.unicode_codes._EMOJI_UNICODE['en'][':Taurus:']) + '\ufe0e'
+        emoji.unicode_codes._EMOJI_UNICODE['en'][':Taurus:']) + '\ufe0e'  # pyright: ignore [reportPrivateUsage]
     assert emoji.emojize(':Taurus:', variant='emoji_type') == remove_variant(
-        emoji.unicode_codes._EMOJI_UNICODE['en'][':Taurus:']) + '\ufe0f'
+        emoji.unicode_codes._EMOJI_UNICODE['en'][':Taurus:']) + '\ufe0f'  # pyright: ignore [reportPrivateUsage]
 
     assert emoji.emojize(
-        ':admission_tickets:', variant=None) == emoji.unicode_codes._EMOJI_UNICODE['en'][':admission_tickets:']
+        ':admission_tickets:', variant=None) == emoji.unicode_codes._EMOJI_UNICODE['en'][':admission_tickets:']  # pyright: ignore [reportPrivateUsage]
     assert emoji.emojize(':admission_tickets:', variant=None) == emoji.emojize(
         ':admission_tickets:')
     assert emoji.emojize(':admission_tickets:', variant='text_type') == remove_variant(
-        emoji.unicode_codes._EMOJI_UNICODE['en'][':admission_tickets:']) + '\ufe0e'
+        emoji.unicode_codes._EMOJI_UNICODE['en'][':admission_tickets:']) + '\ufe0e'  # pyright: ignore [reportPrivateUsage]
     assert emoji.emojize(':admission_tickets:', variant='emoji_type') == remove_variant(
-        emoji.unicode_codes._EMOJI_UNICODE['en'][':admission_tickets:']) + '\ufe0f'
+        emoji.unicode_codes._EMOJI_UNICODE['en'][':admission_tickets:']) + '\ufe0f'  # pyright: ignore [reportPrivateUsage]
 
     with pytest.raises(ValueError):
-        emoji.emojize(':admission_tickets:', variant=False)
+        emoji.emojize(':admission_tickets:', variant=False)  # pyright: ignore [reportArgumentType]
 
     with pytest.raises(ValueError):
-        emoji.emojize(':admission_tickets:', variant=True)
+        emoji.emojize(':admission_tickets:', variant=True)  # pyright: ignore [reportArgumentType]
 
     with pytest.raises(ValueError):
-        emoji.emojize(':admission_tickets:', variant='wrong')
+        emoji.emojize(':admission_tickets:', variant='wrong')  # pyright: ignore [reportArgumentType]
 
     assert emoji.emojize(":football:") == ':football:'
     assert emoji.emojize(":football:", variant="text_type") == ':football:'
@@ -198,11 +203,11 @@ def test_emojize_version():
     assert emoji.emojize("Biking :man_biking: is in 4.0", version=3.0, handle_version=lambda e, data: '<emoji>') == "Biking <emoji> is in 4.0"
     assert emoji.emojize("Biking :man_biking: is in 4.0", version=3.0, handle_version=lambda e, data: data["fr"]) == "Biking :cycliste_homme: is in 4.0"
 
-    def f(emj, data):
+    def f(emj: str, data: Dict[str, str]) -> str:
         assert data['E'] == 5
+        return ''
 
-    assert emoji.emojize(':bowl_with_spoon:', version=-
-                         1, handle_version=f) == ''
+    assert emoji.emojize(':bowl_with_spoon:', version=-1, handle_version=f) == ''
     assert emoji.emojize(':bowl_with_spoon:') == '\U0001F963'
     assert emoji.emojize(':bowl_with_spoon:', version=4) == ''
     assert emoji.emojize(':bowl_with_spoon:', version=4.9) == ''
@@ -348,7 +353,7 @@ def test_replace_emoji():
     assert emoji.replace_emoji('Hello 🇫🇷👌') == 'Hello '
     assert emoji.replace_emoji('Hello 🇫🇷👌', 'x') == 'Hello xx'
 
-    def replace(emj, data):
+    def replace(emj: str, data: Dict[str, str]) -> str:
         assert emj in ["🇫🇷", "👌"]
         return 'x'
     assert emoji.replace_emoji('Hello 🇫🇷👌', replace) == 'Hello xx'
@@ -373,7 +378,7 @@ def test_long_emoji():
 
 
 def test_untranslated():
-    for emj, item in emoji.EMOJI_DATA.items():
+    for item in emoji.EMOJI_DATA.values():
         if item['status'] != emoji.STATUS['fully_qualified']:
             continue
         if 'es' not in item:
@@ -408,9 +413,16 @@ def test_text():
 Excepteur sint occaecat cupidatat non proident, sunt in culpa qui officia deserunt mollit anim id est laborum.
 """
 
-    def add_random_emoji(text, lst, select=lambda emj_data: emj_data['en']):
+    def default_select(emj_data: Dict[str, Any]) -> str:
+        return emj_data['en']
+
+    def add_random_emoji(
+        text: str,
+        lst: List[Tuple[str, Dict[str, Any]]],
+        select: Callable[[Dict[str, Any]], Union[str, Literal[False]]] = default_select
+    ) -> Tuple[str, str, List[str]]:
 
-        emoji_list = []
+        emoji_list: List[str] = []
         text_with_unicode = ""
         text_with_placeholder = ""
         for i in range(0, len(text), 10):
@@ -439,7 +451,7 @@ def add_random_emoji(text, lst, select=lambda emj_data: emj_data['en']):
 
         return text_with_unicode, text_with_placeholder, emoji_list
 
-    def clean(s):
+    def clean(s: str) -> str:
         return s.replace('\u200d', '').replace('\ufe0f', '')
 
     all_emoji_list = list(emoji.EMOJI_DATA.items())
@@ -456,8 +468,10 @@ def clean(s):
         assert lis['emoji'] == emoji_list[i]
 
     # qualified emoji from "es"
-    selector = lambda emoji_data: emoji_data["es"] if "es" in emoji_data else False
-    text_with_unicode, text_with_placeholder, emoji_list = add_random_emoji(text, qualified_emoji_list, selector)
+    def select_es(emj_data: Dict[str, Any]) -> Union[str, Literal[False]]:
+        return emj_data["es"] if "es" in emj_data else False
+
+    text_with_unicode, text_with_placeholder, emoji_list = add_random_emoji(text, qualified_emoji_list, select=select_es)
     assert emoji.demojize(text_with_unicode, language="es") == text_with_placeholder
     assert emoji.emojize(text_with_placeholder, language="es") == text_with_unicode
     if not UCS2:
@@ -467,8 +481,10 @@ def clean(s):
         assert lis['emoji'] == emoji_list[i]
 
     # qualified emoji from "alias"
-    selector = lambda emoji_data: emoji_data["alias"][0] if "alias" in emoji_data else False
-    text_with_unicode, text_with_placeholder, emoji_list = add_random_emoji(text, qualified_emoji_list, selector)
+    def select_alias(emj_data: Dict[str, Any]) -> Union[str, Literal[False]]:
+        return emj_data["alias"][0] if "alias" in emj_data else False
+
+    text_with_unicode, text_with_placeholder, emoji_list = add_random_emoji(text, qualified_emoji_list, select=select_alias)
     assert emoji.demojize(text_with_unicode, language="alias") == text_with_placeholder
     assert emoji.emojize(text_with_placeholder, language="alias") == text_with_unicode
     if not UCS2:
@@ -490,7 +506,7 @@ def clean(s):
 
 def test_text_multiple_times():
     # Run test_text() multiple times because it relies on a random text
-    for i in range(100):
+    for _ in range(100):
         test_text()
 
 

tests/test_dict.py

@@ -8,7 +8,7 @@
 def test_all_languages_list():
     """Compare all language keys in EMOJI_DATA with the emoji.LANGUAGES list"""
 
-    langs = set()
+    langs: set[str] = set()
     for item in emoji.EMOJI_DATA.values():
         langs.update(item.keys())
     all_languages = {lang for lang in langs if len(lang) == 2 and lang.lower() == lang}
@@ -25,10 +25,10 @@ def test_emoji_versions():
         assert v >= 0.6
 
 
-def check_duplicate_names(lang):
+def check_duplicate_names(lang: str):
     """Check that there are no duplicate names in the fully_qualified except for different variants"""
     seen = {}
-    for emj, item in emoji.EMOJI_DATA.items():
+    for item in emoji.EMOJI_DATA.values():
         if item["status"] > emoji.STATUS["fully_qualified"]:
             continue
 

tests/test_nfkc.py

@@ -1,11 +1,14 @@
 """Unittests for canonically equivalent Unicode sequences"""
 
 import sys
-import emoji
 import unicodedata
+import emoji
+from typing_extensions import Literal
+
 
+_NormalizationForm = Literal['NFC', 'NFD', 'NFKC', 'NFKD']
 
-def is_normalized(form, s):
+def is_normalized(form: _NormalizationForm, s: str) -> bool:
     if sys.version_info >= (3, 8):
         return unicodedata.is_normalized(form, s)
     else:

tests/test_unicode_codes.py

@@ -1,8 +1,6 @@
 """Unittests for emoji.unicode_codes."""
 
-
-import emoji
-
+import emoji.unicode_codes
 
 # Build all language packs (i.e. fill the cache):
 emoji.emojize("", language="alias")
@@ -13,8 +11,8 @@
 def test_emoji_english_names():
 
     for language, group in (
-            ('en', emoji.unicode_codes._EMOJI_UNICODE['en']),
-            ('alias', emoji.unicode_codes._ALIASES_UNICODE)
+            ('en', emoji.unicode_codes._EMOJI_UNICODE['en']),  # pyright: ignore [reportPrivateUsage]
+            ('alias', emoji.unicode_codes._ALIASES_UNICODE)  # pyright: ignore [reportPrivateUsage]
     ):
         for name, ucode in group.items():
             assert name.startswith(':') and name.endswith(':') and len(name) >= 3
@@ -26,14 +24,14 @@ def test_compare_normal_and_aliases():
     # There should always be more aliases than normal codes
     # since the aliases contain the normal codes
 
-    assert len(emoji.unicode_codes._EMOJI_UNICODE['en']) < len(
-        emoji.unicode_codes._ALIASES_UNICODE)
+    assert len(emoji.unicode_codes._EMOJI_UNICODE['en']) < len(  # pyright: ignore [reportPrivateUsage]
+        emoji.unicode_codes._ALIASES_UNICODE)  # pyright: ignore [reportPrivateUsage]
 
 
 def test_no_alias_duplicates():
     # There should not be two emoji with the same alias
     # (aliases still can be the same as another 'en'-name)
-    all_aliases = set()
+    all_aliases: set[str] = set()
     for data in emoji.EMOJI_DATA.values():
         if data['status'] <= emoji.STATUS['fully_qualified'] and 'alias' in data:
             for alias in data['alias']: