Merge pull request #1362 from bact/fix-type-tests

Add type ignore comments for type tests

Author: bact

pythainlp/util/digitconv.py

@@ -89,8 +89,10 @@ def thai_digit_to_arabic_digit(text: str) -> str:
         thai_digit_to_arabic_digit(text)
         # output: เป็นจำนวน 123,400.25 บาท
     """
-    if not text or not isinstance(text, str):
+    if not isinstance(text, str):
         raise TypeError("The text must be str type.")
+    if not text:
+        return ""
 
     return text.translate(_thai_arabic_translate_table)
 
@@ -114,8 +116,10 @@ def arabic_digit_to_thai_digit(text: str) -> str:
         arabic_digit_to_thai_digit(text)
         # output: เป็นจำนวน ๑๒๓,๔๐๐.๒๕ บาท
     """
-    if not text or not isinstance(text, str):
+    if not isinstance(text, str):
         raise TypeError("The text must be str type.")
+    if not text:
+        return ""
 
     # Convert Arabic to Thai numerals
     return text.translate(_arabic_thai_translate_table)
@@ -139,8 +143,10 @@ def digit_to_text(text: str) -> str:
         digit_to_text("๕๖๗")
         # output: 'ห้าหกเจ็ด'
     """
-    if not text or not isinstance(text, str):
+    if not isinstance(text, str):
         raise TypeError("The text must be str type.")
+    if not text:
+        return ""
 
     # Convert Thai numerals to Arabic ones
     text = text.translate(_thai_arabic_translate_table)
@@ -180,7 +186,7 @@ def text_to_arabic_digit(text: str) -> str:
     """
     if not isinstance(text, str):
         raise TypeError("The text must be str type.")
-    elif not text or text not in _spell_digit:
+    if not text or text not in _spell_digit:
         return ""
 
     return _spell_digit[text]
@@ -217,7 +223,7 @@ def text_to_thai_digit(text: str) -> str:
     """
     if not isinstance(text, str):
         raise TypeError("The text must be str type.")
-    elif not text:
+    if not text:
         return ""
 
     return arabic_digit_to_thai_digit(text_to_arabic_digit(text))

tests/compact/testc_tools.py

@@ -38,7 +38,7 @@ def test_misspell_edge_cases(self):
         self.assertEqual(len(result), 1)
         # Edge case: None raises TypeError
         with self.assertRaises(TypeError):
-            misspell(None)
+            misspell(None)  # type: ignore[arg-type]
 
     def test_misspell_naive(self):
         for text in self.texts:
@@ -95,12 +95,14 @@ def test_search_location_of_character(self):
         # Test Thai characters
         loc = search_location_of_character("ก")
         self.assertIsNotNone(loc)
-        self.assertEqual(len(loc), 4)  # (language_ix, is_shift, row, pos)
+        # loc shape is (language_ix, is_shift, row, pos)
+        self.assertEqual(len(loc), 4)  # type: ignore[arg-type]
 
         # Test English characters
         loc = search_location_of_character("a")
         self.assertIsNotNone(loc)
-        self.assertEqual(len(loc), 4)
+        # loc shape is (language_ix, is_shift, row, pos)
+        self.assertEqual(len(loc), 4)  # type: ignore[arg-type]
 
         # Test shifted characters
         loc = search_location_of_character("A")

tests/core/test_cli.py

@@ -4,7 +4,7 @@
 
 import io
 import unittest
-from argparse import ArgumentError
+from argparse import ArgumentError, ArgumentParser
 from contextlib import redirect_stderr, redirect_stdout
 from unittest.mock import MagicMock, patch
 
@@ -18,8 +18,9 @@
 
 class CliTestCase(unittest.TestCase):
     def test_cli(self):
+        parser = ArgumentParser()
         with self.assertRaises((ArgumentError, SystemExit)):
-            cli.exit_if_empty("", None)
+            cli.exit_if_empty("", parser)
 
     def test_cli_main(self):
         # Suppress output to keep test log clean

tests/core/test_khavee.py

@@ -177,7 +177,7 @@ def setUp(self):
 
     def test_non_string_raises_type_error(self):
         with self.assertRaises(TypeError):
-            self.kv.check_aek_too(123)
+            self.kv.check_aek_too(123)  # type: ignore[arg-type]
 
     def test_dead_syllable_as_aek_flag(self):
         self.assertEqual(self.kv.check_aek_too("บท", dead_syllable_as_aek=True), "aek")
@@ -187,7 +187,7 @@ def test_dead_syllable_without_flag_returns_false(self):
 
     def test_list_with_non_string_element_raises(self):
         with self.assertRaises(TypeError):
-            self.kv.check_aek_too(["ไก่", 42])
+            self.kv.check_aek_too(["ไก่", 42])  # type: ignore[list-item]
 
     def test_both_tone_marks_returns_false(self):
         # word with both ่ and ้ should return False

tests/core/test_morpheme.py

@@ -21,7 +21,7 @@ def test_nighit(self):
             nighit("สํ", "มาร")  # consonant ม is not in any supported group
 
     def test_is_native_thai(self):
-        self.assertFalse(is_native_thai(None), False)
+        self.assertFalse(is_native_thai(None), False)  # type: ignore[arg-type]
         self.assertFalse(is_native_thai(""), False)
         self.assertFalse(is_native_thai("116"), False)
         self.assertFalse(is_native_thai("abc"), False)

tests/core/test_soundex.py

@@ -27,7 +27,7 @@ def test_soundex(self):
         self.assertIsNotNone(soundex("a", engine="complete_soundex"))
         self.assertIsNotNone(soundex("a", engine="XXX"))
 
-        self.assertEqual(lk82(None), "")
+        self.assertEqual(lk82(None), "")  # type: ignore
         self.assertEqual(lk82(""), "")
         self.assertEqual(lk82("เหตุ"), lk82("เหด"))
         self.assertEqual(lk82("รถ"), "ร3000")
@@ -43,13 +43,13 @@ def test_soundex(self):
         self.assertIsNotNone(lk82("หืออือ"))
         self.assertEqual(lk82("น์"), "")
 
-        self.assertEqual(udom83(None), "")
+        self.assertEqual(udom83(None), "")  # type: ignore
         self.assertEqual(udom83(""), "")
         self.assertEqual(udom83("เหตุ"), udom83("เหด"))
         self.assertEqual(udom83("รถ"), "ร800000")
         self.assertEqual(udom83("น์"), "")
 
-        self.assertEqual(metasound(None), "")
+        self.assertEqual(metasound(None), "")  # type: ignore
         self.assertEqual(metasound(""), "")
         self.assertEqual(metasound("เหตุ"), metasound("เหด"))
         self.assertEqual(metasound("รักษ์"), metasound("รัก"))
@@ -70,7 +70,7 @@ def test_soundex(self):
         self.assertIsNotNone(metasound("สุวรรณา"))
         self.assertIsNotNone(metasound("ดอยบอย"))
 
-        self.assertEqual(prayut_and_somchaip(None), "")
+        self.assertEqual(prayut_and_somchaip(None), "")  # type: ignore
         self.assertEqual(prayut_and_somchaip(""), "")
         self.assertEqual(prayut_and_somchaip("vp"), "11")
         self.assertIsNotNone(prayut_and_somchaip("บา"))
@@ -86,7 +86,7 @@ def test_soundex(self):
         self.assertIsNotNone(prayut_and_somchaip("ว้าว"))
 
         # Test complete_soundex
-        self.assertEqual(complete_soundex(None), "")
+        self.assertEqual(complete_soundex(None), "")  # type: ignore
         self.assertEqual(complete_soundex(""), "")
 
         # Single syllable test cases from the paper

tests/core/test_spell.py

@@ -17,7 +17,7 @@
 
 class SpellTestCase(unittest.TestCase):
     def test_spell(self):
-        self.assertEqual(spell(None), [""])
+        self.assertEqual(spell(None), [""])  # type: ignore
         self.assertEqual(spell(""), [""])
 
         result = spell("เน้ร")
@@ -29,7 +29,7 @@ def test_spell(self):
         self.assertGreater(len(result), 0)
 
     def test_word_correct(self):
-        self.assertEqual(correct(None), "")
+        self.assertEqual(correct(None), "")  # type: ignore
         self.assertEqual(correct(""), "")
         self.assertEqual(correct("1"), "1")
         self.assertEqual(correct("05"), "05")
@@ -53,7 +53,7 @@ def test_norvig_spell_checker(self):
         self.assertGreater(dict_size, 30000)  # Should have substantial words
         self.assertLess(dict_size, len(orst) + 1000)  # Should not exceed ORST by much
 
-        user_dict = [
+        user_list_tuple = [
             ("การงาน", 31),  # longer than max_len
             ("กาม", 1),  # fewer than min_freq
             ("กาล0", 64),  # has digit
@@ -63,20 +63,20 @@ def test_norvig_spell_checker(self):
             ("การ", 42),  # OK
         ]
         checker = NorvigSpellChecker(
-            custom_dict=user_dict, min_freq=2, max_len=5
+            custom_dict=user_list_tuple, min_freq=2, max_len=5
         )
         self.assertEqual(len(checker.dictionary()), 1)
 
-        user_dict = [
+        user_list_str = [
             "เอกราช",
             "ปลอดภัย",
             "เศรษฐกิจ",
             "เสมอภาค",
             "เสรีภาพ",
             "การศึกษา",
         ]
-        checker = NorvigSpellChecker(custom_dict=user_dict)
-        self.assertEqual(len(checker.dictionary()), len(user_dict))
+        checker = NorvigSpellChecker(custom_dict=user_list_str)
+        self.assertEqual(len(checker.dictionary()), len(user_list_str))
 
         user_dict = {
             "พหลโยธิน": 1,
@@ -92,9 +92,9 @@ def test_norvig_spell_checker(self):
         # as it has frequency less than default min_freq (2)
         self.assertEqual(len(checker.dictionary()), len(user_dict) - 1)
 
-        user_dict = [24, 6, 2475]
+        user_list_int = [24, 6, 2475]
         with self.assertRaises(TypeError):
-            _ = NorvigSpellChecker(custom_dict=user_dict)
+            _ = NorvigSpellChecker(custom_dict=user_list_int)  # type: ignore[arg-type]
 
     def test_issue_680_orst_filtering(self):
         """Test for issue #680: Spell checker uses only ORST words.

tests/core/test_tag.py

@@ -28,7 +28,7 @@ def setUpClass(cls) -> None:
         download("blackboard_unigram_tagger")
 
     def test_pos_tag(self):
-        self.assertEqual(pos_tag(None), [])
+        self.assertEqual(pos_tag(None), [])  # type: ignore[arg-type]
         self.assertEqual(pos_tag([]), [])
         self.assertEqual(
             pos_tag(["นักเรียน", "ถาม", "ครู"]),
@@ -38,13 +38,13 @@ def test_pos_tag(self):
             len(pos_tag(["การ", "เดินทาง", "มี", "ความ", "ท้าทาย"])), 5
         )
 
-        self.assertEqual(unigram.tag(None, corpus="pud"), [])
+        self.assertEqual(unigram.tag(None, corpus="pud"), [])  # type: ignore[arg-type]
         self.assertEqual(unigram.tag([], corpus="pud"), [])
-        self.assertEqual(unigram.tag(None, corpus="orchid"), [])
+        self.assertEqual(unigram.tag(None, corpus="orchid"), [])  # type: ignore[arg-type]
         self.assertEqual(unigram.tag([], corpus="orchid"), [])
-        self.assertEqual(unigram.tag(None, corpus="blackboard"), [])
+        self.assertEqual(unigram.tag(None, corpus="blackboard"), [])  # type: ignore[arg-type]
         self.assertEqual(unigram.tag([], corpus="blackboard"), [])
-        self.assertEqual(unigram.tag(None, corpus="tud"), [])
+        self.assertEqual(unigram.tag(None, corpus="tud"), [])  # type: ignore[arg-type]
         self.assertEqual(unigram.tag([], corpus="tud"), [])
         self.assertIsNotNone(
             pos_tag(TEST_TOKENS, engine="unigram", corpus="orchid")
@@ -85,7 +85,7 @@ def test_pos_tag(self):
             pos_tag(["ความ", "พอเพียง"], corpus="orchid_ud")[0][1], "NOUN"
         )
 
-        self.assertEqual(pos_tag_sents(None), [])
+        self.assertEqual(pos_tag_sents(None), [])  # type: ignore[arg-type]
         self.assertEqual(pos_tag_sents([]), [])
         self.assertEqual(
             pos_tag_sents([["ผม", "กิน", "ข้าว"], ["แมว", "วิ่ง"]]),
@@ -123,15 +123,15 @@ def setUpClass(cls) -> None:
         download("blackboard_pt_tagger")
 
     def test_perceptron_tagger(self):
-        self.assertEqual(perceptron.tag(None, corpus="orchid"), [])
+        self.assertEqual(perceptron.tag(None, corpus="orchid"), [])  # type: ignore[arg-type]
         self.assertEqual(perceptron.tag([], corpus="orchid"), [])
-        self.assertEqual(perceptron.tag(None, corpus="orchid_ud"), [])
+        self.assertEqual(perceptron.tag(None, corpus="orchid_ud"), [])  # type: ignore[arg-type]
         self.assertEqual(perceptron.tag([], corpus="orchid_ud"), [])
-        self.assertEqual(perceptron.tag(None, corpus="pud"), [])
+        self.assertEqual(perceptron.tag(None, corpus="pud"), [])  # type: ignore[arg-type]
         self.assertEqual(perceptron.tag([], corpus="pud"), [])
-        self.assertEqual(perceptron.tag(None, corpus="blackboard"), [])
+        self.assertEqual(perceptron.tag(None, corpus="blackboard"), [])  # type: ignore[arg-type]
         self.assertEqual(perceptron.tag([], corpus="blackboard"), [])
-        self.assertEqual(perceptron.tag(None, corpus="tud"), [])
+        self.assertEqual(perceptron.tag(None, corpus="tud"), [])  # type: ignore[arg-type]
         self.assertEqual(perceptron.tag([], corpus="tud"), [])
 
         self.assertIsNotNone(

tests/core/test_tokenize.py

@@ -296,7 +296,7 @@ def test_Tokenizer(self):
         _tokenizer = Tokenizer(word_dict_trie())
         self.assertEqual(_tokenizer.word_tokenize(""), [])
         _tokenizer.set_tokenize_engine("longest")
-        self.assertEqual(_tokenizer.word_tokenize(None), [])
+        self.assertEqual(_tokenizer.word_tokenize(None), [])  # type: ignore[arg-type]
 
         _tokenizer = Tokenizer()
         self.assertEqual(_tokenizer.word_tokenize("ก"), ["ก"])
@@ -335,7 +335,7 @@ def test_sent_tokenize(self):
             sent_tokenize("ฉันไป กิน", engine="XX")  # engine does not exist
 
     def test_subword_tokenize(self):
-        self.assertEqual(subword_tokenize(None), [])
+        self.assertEqual(subword_tokenize(None), [])  # type: ignore[arg-type]
         self.assertEqual(subword_tokenize(""), [])
         self.assertIsInstance(
             subword_tokenize("สวัสดีดาวอังคาร", engine="tcc"), list
@@ -345,7 +345,7 @@ def test_subword_tokenize(self):
             subword_tokenize("สวัสดีดาวอังคาร", engine="tcc_p"), list
         )
         self.assertNotIn("า", subword_tokenize("สวัสดีดาวอังคาร", engine="tcc_p"))
-        self.assertEqual(subword_tokenize(None, engine="etcc"), [])
+        self.assertEqual(subword_tokenize(None, engine="etcc"), [])  # type: ignore[arg-type]
         self.assertEqual(subword_tokenize("", engine="etcc"), [])
         self.assertIsInstance(
             subword_tokenize("สวัสดิีดาวอังคาร", engine="etcc"), list
@@ -495,7 +495,7 @@ def test_mm(self):
         self.assertIsNotNone(
             multi_cut.find_all_segment("รถไฟฟ้ากรุงเทพมหานครBTS")
         )
-        self.assertEqual(multi_cut.find_all_segment(None), [])
+        self.assertEqual(multi_cut.find_all_segment(None), [])  # type: ignore[arg-type]
 
     def test_newmm(self):
         assert_segment_handles_none_and_empty(self, newmm.segment)
@@ -686,8 +686,8 @@ def test_tcc_p(self):
         self.assertEqual(tcc_p.tcc_pos(""), set())
         # tcc_pos_array: edge cases
         self.assertIsInstance(tcc_p.tcc_pos_array(""), bytearray)
-        self.assertIsInstance(tcc_p.tcc_pos_array(None), bytearray)
-        self.assertIsInstance(tcc_p.tcc_pos_array(42), bytearray)
+        self.assertIsInstance(tcc_p.tcc_pos_array(None), bytearray)  # type: ignore[arg-type]
+        self.assertIsInstance(tcc_p.tcc_pos_array(42), bytearray)    # type: ignore[arg-type]
         # valid text: array length must equal len(text)+1 and mark boundaries
         arr = tcc_p.tcc_pos_array("ประเทศ")
         self.assertEqual(len(arr), len("ประเทศ") + 1)

tests/core/test_transliterate.py

@@ -52,13 +52,13 @@
 
 class TransliterateTestCase(unittest.TestCase):
     def test_romanize(self):
-        self.assertEqual(romanize(None), "")
+        self.assertEqual(romanize(None), "")  # type: ignore[arg-type]
         self.assertEqual(romanize(""), "")
         self.assertEqual(romanize("แมว"), "maeo")
 
     def test_romanize_royin_basic(self):
         for word, expect in BASIC_TESTS.items():
-            self.assertEqual(romanize(word, engine="royin"), expect)
+            self.assertEqual(romanize(word, engine="royin"), expect)  # type: ignore[arg-type]
 
     def test_romanize_royin_consistency(self):
         for word, part1, part2 in CONSISTENCY_TESTS: