Formatting

Author: bact

pyproject.toml

@@ -50,6 +50,7 @@ classifiers = [
     "Programming Language :: Python :: 3.11",
     "Programming Language :: Python :: 3.12",
     "Programming Language :: Python :: 3.13",
+    "Programming Language :: Python :: 3.14",
     "Intended Audience :: Developers",
     "Natural Language :: Thai",
     "Topic :: Scientific/Engineering :: Artificial Intelligence",
@@ -185,6 +186,7 @@ extra = [
     "pandas>=0.24",
     "ssg>=0.0.8",
     "symspellpy>=6.7.6",
+    "tltk>=1.10",
 ]
 
 # Full dependencies - pinned where available

tests/compact/testc_util.py

@@ -2,8 +2,7 @@
 # SPDX-FileType: SOURCE
 # SPDX-License-Identifier: Apache-2.0
 
-"""Unit tests for pythainlp.util module.
-"""
+"""Unit tests for pythainlp.util module."""
 
 import unittest
 

tests/core/test_robustness.py

@@ -253,4 +253,3 @@ def test_word_tokenize_with_very_long_strings(self):
                         f"word_tokenize (engine={engine}) failed with "
                         f"very long string (index={i}): {e}"
                     )
-

tests/core/test_tag.py

@@ -94,6 +94,7 @@ def test_NER_error_handling(self):
         with self.assertRaises(ValueError):
             NER(engine="thainer", corpus="cat")
 
+
 class PerceptronTaggerTestCase(unittest.TestCase):
     """Test pythainlp.tag.PerceptronTagger
 
@@ -202,6 +203,7 @@ def test_perceptron_tagger_custom(self):
         with self.assertRaises(IOError):
             tagger.load("ptagger_notexistX4AcOcX.pkl")  # file does not exist
 
+
 class TagLocationsTestCase(unittest.TestCase):
     """Test pythainlp.tag.locations"""
 

tests/extra/testx_augment.py

@@ -22,13 +22,13 @@ def setUp(self):
         self.text2 = "เราอยู่ที่มหาวิทยาลัยขอนแก่น"
 
     def test_WordNetAug(self):
-        nltk.download('omw-1.4', force=True)  # load wordnet
+        nltk.download("omw-1.4", force=True)  # load wordnet
         wordnetaug = WordNetAug()
         self.assertIsNotNone(wordnetaug.augment(self.text))
         self.assertIsNotNone(wordnetaug.find_synonyms("ผม", pos=None))
         self.assertIsNotNone(wordnetaug.augment(self.text, postag=False))
-        self.assertIsNone(postype2wordnet('n', 'abc'))
-        self.assertIsNotNone(postype2wordnet('NOUN', 'orchid'))
+        self.assertIsNone(postype2wordnet("n", "abc"))
+        self.assertIsNotNone(postype2wordnet("NOUN", "orchid"))
 
     # def test_Thai2fitAug(self):
     #     _aug = Thai2fitAug()

tests/extra/testx_spell.py

@@ -3,7 +3,6 @@
 # SPDX-License-Identifier: Apache-2.0
 
 # Tests for spell functions that need extra dependencies
-# Note: Tests requiring phunspell/torch/HuggingFace Hub have been moved to tests.noauto
 
 import unittest
 
@@ -20,7 +19,6 @@
 
 class SpellTestCaseX(unittest.TestCase):
     def test_spell(self):
-        # Tests for symspellpy only (phunspell moved to noauto)
         result = spell("เน้ร", engine="symspellpy")
         self.assertIsInstance(result, list)
         self.assertGreater(len(result), 0)
@@ -30,17 +28,14 @@ def test_spell(self):
         self.assertGreater(len(result), 0)
 
     def test_word_correct(self):
-        # Tests for symspellpy only (phunspell and wanchanberta moved to noauto)
         result = correct("ทดสอง", engine="symspellpy")
         self.assertIsInstance(result, str)
         self.assertNotEqual(result, "")
 
     def test_spell_sent(self):
-        # Tests for symspellpy only (phunspell moved to noauto)
         self.assertIsNotNone(spell_sent(SENT_TOKS, engine="symspellpy"))
 
     def test_correct_sent(self):
-        # Tests for symspellpy only (phunspell and wanchanberta moved to noauto)
         self.assertIsNotNone(correct_sent(SENT_TOKS, engine="symspellpy"))
         self.assertIsNotNone(symspellpy.correct_sent(SENT_TOKS))
 
@@ -56,4 +51,3 @@ def test_spell_tltk(self):
         result = spell("เดก", engine="tltk")
         self.assertIsInstance(result, list)
         self.assertGreater(len(result), 0)
-

tests/extra/testx_tag.py

@@ -3,7 +3,6 @@
 # SPDX-License-Identifier: Apache-2.0
 
 # Tests for tag functions that need extra dependencies
-# Note: Tests requiring transformers have been moved to tests.noauto
 
 import unittest
 
@@ -12,7 +11,6 @@
 
 
 class TagTestCaseX(unittest.TestCase):
-    # Tests for ThaiNameTagger (doesn't require transformers)
 
     def test_thai_name_tagger_1_5(self):
         ner = ThaiNameTagger(version="1.5")
@@ -153,4 +151,3 @@ def test_tltk_ner(self):
                 tag=True,
             )
         )
-

tests/extra/testx_tokenize.py

@@ -3,7 +3,6 @@
 # SPDX-License-Identifier: Apache-2.0
 
 # Tests for tokenize functions that need extra dependencies
-# Note: Tests requiring TensorFlow/Keras/torch/transformers have been moved to tests.noauto
 
 import unittest
 
@@ -145,5 +144,3 @@ class WordTokenizeTLTKTestCaseX(unittest.TestCase):
 
     def test_word_tokenize_tltk(self):
         self.assertIsNotNone(word_tokenize(TEXT_1, engine="tltk"))
-
-

tests/extra/testx_translate.py

@@ -71,7 +71,7 @@ def test_translate(self):
         #     )
         # )
         with self.assertRaises(ValueError):
-            self.th_cat_translator = Translate('th', 'cat', engine="fkfj")
+            self.th_cat_translator = Translate("th", "cat", engine="fkfj")
 
     def test_word_translate(self):
         self.assertIsNone(word_translate("cat", src="en", target="th"))

tests/extra/testx_util.py

@@ -2,8 +2,7 @@
 # SPDX-FileType: SOURCE
 # SPDX-License-Identifier: Apache-2.0
 
-"""Unit tests for pythainlp.util module.
-"""
+"""Unit tests for pythainlp.util module."""
 
 import unittest