Merge pull request #1360 from bact/fix-mypy-test

Fix mypy issues in tests.extra

Author: bact

tests/extra/testx_translate.py

@@ -15,7 +15,8 @@
 class TranslateTestCaseX(unittest.TestCase):
     def test_translate(self):
         # remove("scb_1m_th-en_spm")
-        self.assertIsNone(download_model_all())
+        result = download_model_all()
+        self.assertIsNone(result)
         self.th_en_translator = ThEnTranslator()
         self.assertIsNotNone(
             self.th_en_translator.translate(

tests/extra/testx_translate_helpers.py

@@ -57,7 +57,7 @@ def test_prepare_text_with_exclusions_none(self):
     def test_prepare_text_with_exclusions_empty_list(self):
         """Test with empty exclusion list"""
         text = "Hello world"
-        exclude_words = []
+        exclude_words: list[str] = []
         prepared, mapping = _prepare_text_with_exclusions(
             text, exclude_words
         )
@@ -93,7 +93,7 @@ def test_restore_excluded_words_multiple(self):
     def test_restore_excluded_words_empty_map(self):
         """Test restoring with empty mapping"""
         translated = "Hello world"
-        mapping = {}
+        mapping: dict[str, str] = {}
         restored = _restore_excluded_words(translated, mapping)
 
         # Text should be unchanged

tests/extra/testx_ulmfit.py

@@ -4,6 +4,7 @@
 
 import pickle
 import unittest
+from typing import cast
 
 import fastai
 import pandas as pd
@@ -19,7 +20,7 @@
     untar_data,
 )
 
-from pythainlp.tokenize import THAI2FIT_TOKENIZER
+from pythainlp.tokenize import thai2fit_tokenizer
 from pythainlp.ulmfit import (
     THWIKI_LSTM,
     ThaiTokenizer,
@@ -53,12 +54,14 @@ class UlmfitTestCaseX(unittest.TestCase):
     def test_ThaiTokenizer(self):
         self.thai = ThaiTokenizer()
         self.assertIsNotNone(self.thai.tokenizer("ทดสอบการตัดคำ"))
-        self.assertIsNone(self.thai.add_special_cases(["แมว"]))
+        result = self.thai.add_special_cases(["แมว"])
+        self.assertIsNone(result)
 
     def test_BaseTokenizer(self):
         self.base = base_tokenizer(lang="th")
         self.assertIsNotNone(self.base.tokenizer("ทดสอบ การ ตัด คำ"))
-        self.assertIsNone(self.base.add_special_cases(["แมว"]))
+        result = self.base.add_special_cases(["แมว"])
+        self.assertIsNone(result)
 
     def test_load_pretrained(self):
         self.assertIsNotNone(THWIKI_LSTM)
@@ -179,7 +182,7 @@ def test_process_thai_dense(self):
             text,
             pre_rules=pre_rules_th,
             post_rules=post_rules_th,
-            tok_func=THAI2FIT_TOKENIZER.word_tokenize,
+            tok_func=thai2fit_tokenizer().word_tokenize,
         )
 
         # after pre_rules_th
@@ -225,7 +228,7 @@ def test_document_vector(self):
         thwiki = THWIKI_LSTM
         # Security note: pickle.load() executes arbitrary code if file is malicious.
         # These corpus files come from a trusted source with MD5 verification.
-        with open(thwiki["itos_fname"], "rb") as f:
+        with open(cast(str, thwiki["itos_fname"]), "rb") as f:
             thwiki_itos = pickle.load(f)
         thwiki_vocab = fastai.text.transform.Vocab(thwiki_itos)
         tt = Tokenizer(