Merge pull request #1256 from PyThaiNLP/copilot/increase-test-coverage

Increase test coverage for core utilities and misspell modules

Author: bact

pyproject.toml

@@ -226,7 +226,7 @@ full = [
     "sefr_cut>=1.1",
     "symspellpy==6.9.0",
     "thai-nner==0.3",
-    "tltk>=1.6.8,<2",
+    "tltk>=1.10,<2",
     "torch>=1.13.1,<3",
     "transformers==4.57.6",
     "ufal.chu-liu-edmonds==1.0.3",

tests/README.md

@@ -22,7 +22,7 @@ supported Python versions and operating systems:
 
 The CI/CD test workflow is at
 <https://github.com/PyThaiNLP/pythainlp/blob/dev/.github/workflows/unittest.yml>.
-        
+
 ## Core tests (test_*.py)
 
 - Run `unittest tests.core`
@@ -65,7 +65,7 @@ The CI/CD test workflow is at
 A comprehensive test suite within core tests that tests edge cases important
 for real-world usage:
 
-- Empty strings and various whitespace handling (spaces, tabs, unicode spaces)
+- Empty strings and various whitespace handling (spaces, tabs, Unicode spaces)
 - Special characters from encoding issues, BOM, terminal copy/paste
 - Truncated/malformed Unicode and surrogate pairs
 - Emoji and modern Unicode sequences (ZWJ, modifiers, flags)

tests/compact/testc_tools.py

@@ -6,7 +6,11 @@
 
 import numpy as np
 
-from pythainlp.tools.misspell import misspell
+from pythainlp.tools.misspell import (
+    find_misspell_candidates,
+    misspell,
+    search_location_of_character,
+)
 
 
 def _count_difference(st1: str, st2: str) -> int:
@@ -85,3 +89,58 @@ def test_misspell_with_ratio_100_percent(self):
                 2,
                 f"expect len(text)-2 misspells with ratio=1.5. (Δ={diff})",
             )
+
+    def test_search_location_of_character(self):
+        """Test search_location_of_character function."""
+        # Test Thai characters
+        loc = search_location_of_character("ก")
+        self.assertIsNotNone(loc)
+        self.assertEqual(len(loc), 4)  # (language_ix, is_shift, row, pos)
+
+        # Test English characters
+        loc = search_location_of_character("a")
+        self.assertIsNotNone(loc)
+        self.assertEqual(len(loc), 4)
+
+        # Test shifted characters
+        loc = search_location_of_character("A")
+        self.assertIsNotNone(loc)
+
+        # Test numbers
+        loc = search_location_of_character("1")
+        self.assertIsNotNone(loc)
+
+        # Test character not in keyboard
+        loc = search_location_of_character("€")
+        self.assertIsNone(loc)
+
+        # Test empty string
+        # Note: Empty string returns a location because Python's "in" operator
+        # matches empty string at the beginning of any string
+        loc = search_location_of_character("")
+        self.assertIsNotNone(loc)
+
+    def test_find_misspell_candidates(self):
+        """Test find_misspell_candidates function."""
+        # Test Thai character
+        candidates = find_misspell_candidates("ก")
+        self.assertIsNotNone(candidates)
+        self.assertIsInstance(candidates, list)
+        self.assertGreater(len(candidates), 0)
+
+        # Test English character
+        candidates = find_misspell_candidates("a")
+        self.assertIsNotNone(candidates)
+        self.assertIsInstance(candidates, list)
+        self.assertGreater(len(candidates), 0)
+
+        # Test character not in keyboard
+        candidates = find_misspell_candidates("€")
+        self.assertIsNone(candidates)
+
+        # Test that candidates are different from input
+        candidates = find_misspell_candidates("ด")
+        if candidates:
+            for candidate in candidates:
+                # Candidates should be strings
+                self.assertIsInstance(candidate, str)

tests/core/test_ancient.py

@@ -23,6 +23,11 @@ def test_aksonhan_to_current(self):
         self.assertEqual(aksonhan_to_current("บงงคบบ"), "บังคับ")
         self.assertEqual(aksonhan_to_current("สรรเพชญ"), "สรรเพชญ")
 
+        # Edge cases
+        self.assertEqual(aksonhan_to_current(""), "")  # empty string
+        self.assertEqual(aksonhan_to_current("ก"), "ก")  # single char
+        self.assertEqual(aksonhan_to_current("กา"), "กา")  # two chars
+
     def test_convert_currency(self):
         self.assertEqual(
             convert_currency(80, "บาท")["ตำลึง"],
@@ -44,3 +49,24 @@ def test_convert_currency(self):
             convert_currency(1,"ชั่ง")["ชั่ง"],
             1.0
         )
+
+        # Test all supported units
+        result = convert_currency(1, "บาท")
+        self.assertIn("เบี้ย", result)
+        self.assertIn("อัฐ", result)
+        self.assertIn("ไพ", result)
+        self.assertIn("เฟื้อง", result)
+        self.assertIn("สลึง", result)
+        self.assertIn("ตำลึง", result)
+
+        # Test with zero value
+        result = convert_currency(0, "บาท")
+        self.assertEqual(result["บาท"], 0.0)
+
+        # Test with fractional value
+        result = convert_currency(0.5, "บาท")
+        self.assertEqual(result["บาท"], 0.5)
+
+        # Test invalid unit
+        with self.assertRaises(NotImplementedError):
+            convert_currency(1, "invalid_unit")

tests/core/test_tools.py

@@ -5,12 +5,14 @@
 import os
 import tempfile
 import unittest
+import warnings
 
 from pythainlp.tools import (
     get_full_data_path,
     get_pythainlp_data_path,
     get_pythainlp_path,
 )
+from pythainlp.tools.core import safe_print, warn_deprecation
 
 
 class ToolsTestCase(unittest.TestCase):
@@ -57,3 +59,69 @@ def test_custom_data_dir(self):
                     os.environ["PYTHAINLP_DATA_DIR"] = original_value
                 elif "PYTHAINLP_DATA_DIR" in os.environ:
                     del os.environ["PYTHAINLP_DATA_DIR"]
+
+    def test_warn_deprecation(self):
+        """Test deprecation warning function."""
+        # Test basic deprecation warning
+        with warnings.catch_warnings(record=True) as w:
+            warnings.simplefilter("always")
+            warn_deprecation("old_func")
+            self.assertEqual(len(w), 1)
+            self.assertTrue(issubclass(w[0].category, DeprecationWarning))
+            self.assertIn("old_func", str(w[0].message))
+            self.assertIn("deprecated", str(w[0].message))
+
+        # Test with replacement function
+        with warnings.catch_warnings(record=True) as w:
+            warnings.simplefilter("always")
+            warn_deprecation("old_func", replacing_func="new_func")
+            self.assertEqual(len(w), 1)
+            self.assertIn("old_func", str(w[0].message))
+            self.assertIn("new_func", str(w[0].message))
+
+        # Test with version information
+        with warnings.catch_warnings(record=True) as w:
+            warnings.simplefilter("always")
+            warn_deprecation(
+                "old_func",
+                deprecated_version="1.0",
+                removal_version="2.0"
+            )
+            self.assertEqual(len(w), 1)
+            self.assertIn("1.0", str(w[0].message))
+            self.assertIn("2.0", str(w[0].message))
+
+        # Test with all parameters
+        with warnings.catch_warnings(record=True) as w:
+            warnings.simplefilter("always")
+            warn_deprecation(
+                "old_func",
+                replacing_func="new_func",
+                deprecated_version="1.0",
+                removal_version="2.0"
+            )
+            self.assertEqual(len(w), 1)
+            message = str(w[0].message)
+            self.assertIn("old_func", message)
+            self.assertIn("new_func", message)
+            self.assertIn("1.0", message)
+            self.assertIn("2.0", message)
+
+    def test_safe_print(self):
+        """Test safe_print function."""
+        # Test normal printing
+        safe_print("Hello, World!")
+        safe_print("สวัสดีครับ")
+
+        # Test with Unicode characters
+        safe_print("Hello 👋 World")
+        safe_print("ภาษาไทย")
+
+        # Test with empty string
+        safe_print("")
+
+        # Test with special characters
+        safe_print("\n\t")
+
+        # Note: Testing actual UnicodeEncodeError is environment-dependent
+        # and would require mocking sys.stdout.encoding

tests/core/test_util.py

@@ -97,6 +97,14 @@ def test_collate(self):
             collate(["ไก่", "เป็ด", "หมู", "วัว"], reverse=True),
             ["หมู", "วัว", "เป็ด", "ไก่"],
         )
+        # Edge cases: mixed Thai and numbers
+        self.assertEqual(collate(["ก", "1", "ข"]), ["1", "ก", "ข"])
+        # Edge cases: with spaces (spaces sort before letters)
+        result = collate([" ก", "ก", "  ก"])
+        self.assertEqual(len(result), 3)
+        self.assertIn(" ก", result)
+        self.assertIn("ก", result)
+        self.assertIn("  ก", result)
 
     # ### pythainlp.util.numtoword
 
@@ -720,6 +728,11 @@ def test_emoji_to_thai(self):
             ":ธง_ไทย: นี่คือธงประเทศไทย",
         )
 
+        # Edge cases
+        self.assertEqual(emoji_to_thai(""), "")  # empty string
+        self.assertEqual(emoji_to_thai("no emoji"), "no emoji")  # no emoji
+        self.assertEqual(emoji_to_thai("ไม่มีอีโมจิ"), "ไม่มีอีโมจิ")  # Thai no emoji
+
     def test_sound_syllable(self):
         test = [
             ("มา", "live"),
@@ -863,6 +876,10 @@ def test_syllable_open_close_detector(self):
 
     def test_to_idna(self):
         self.assertEqual(to_idna("คนละครึ่ง.com"), "xn--42caj4e6bk1f5b1j.com")
+        # Additional test cases for IDNA encoding
+        self.assertEqual(to_idna("ไทย.com"), "xn--o3cw4h.com")
+        self.assertEqual(to_idna("example.com"), "example.com")  # ASCII unchanged
+        self.assertEqual(to_idna("ภาษาไทย.th"), "xn--o3crh0a8bb0k.th")
 
     def test_thai_strptime(self):
         self.assertIsNotNone(
@@ -923,6 +940,11 @@ def test_tis620_to_utf8(self):
         self.assertEqual(
             tis620_to_utf8("¡ÃзÃǧÍصÊÒË¡ÃÃÁ"), "กระทรวงอุตสาหกรรม"
         )
+        # Additional test cases
+        self.assertEqual(tis620_to_utf8("»ÃÐà·Èä·Â"), "ประเทศไทย")
+        self.assertEqual(tis620_to_utf8("ÀÒÉÒä·Â"), "ภาษาไทย")
+        # Empty string
+        self.assertEqual(tis620_to_utf8(""), "")
 
     def test_remove_repeat_consonants(self):
         # update of pythainlp.copus.thai_words() able to break this
@@ -988,11 +1010,20 @@ def test_longest_common_subsequence(self):
         self.assertEqual(longest_common_subsequence("ABCBDAB", "BDCAB"), "BDAB")
         self.assertEqual(longest_common_subsequence("AGGTAB", "GXTXAYB"), "GTAB")
         self.assertEqual(longest_common_subsequence("ABCDGH", "AEDFHR"), "ADH")
+
+        # Edge cases
+        self.assertEqual(longest_common_subsequence("", ""), "")  # empty strings
+        self.assertEqual(longest_common_subsequence("ABC", ""), "")  # one empty
+        self.assertEqual(longest_common_subsequence("", "ABC"), "")  # other empty
+        self.assertEqual(longest_common_subsequence("A", "A"), "A")  # single char match
+        self.assertEqual(longest_common_subsequence("A", "B"), "")  # single char no match
+        self.assertEqual(longest_common_subsequence("ABC", "ABC"), "ABC")  # identical
+        self.assertEqual(longest_common_subsequence("ABC", "XYZ"), "")  # no common chars
         self.assertEqual(longest_common_subsequence("ABC", "AC"), "AC")
-        self.assertEqual(longest_common_subsequence("ABC", "DEF"), "")
-        self.assertEqual(longest_common_subsequence("", "ABC"), "")
-        self.assertEqual(longest_common_subsequence("ABC", ""), "")
-        self.assertEqual(longest_common_subsequence("", ""), "")
+
+        # Thai text
+        self.assertEqual(longest_common_subsequence("ไทย", "ไทย"), "ไทย")
+        self.assertEqual(longest_common_subsequence("ภาษาไทย", "ไทย"), "ไทย")
 
     def test_analyze_thai_text(self):
         self.assertEqual(