Test on windows

Author: dflook

test/test_utf8_encoding.py

@@ -0,0 +1,97 @@
+import pytest
+import python_minifier
+import tempfile
+import os
+
+
+def test_minify_utf8_file():
+    """Test minifying a Python file with UTF-8 characters not in Windows default encoding."""
+
+    # Create Python source with UTF-8 characters that are not in Windows-1252
+    # Using Greek letters, Cyrillic, and mathematical symbols
+    source_code = '''# -*- coding: utf-8 -*-
+"""
+This module contains UTF-8 characters that are not in Windows-1252 encoding:
+- Greek: α β γ δ ε ζ η θ ι κ λ μ ν ξ ο π ρ σ τ υ φ χ ψ ω
+- Cyrillic: а б в г д е ё ж з и й к л м н о п р с т у ф х ц ч ш щ ъ ы ь э ю я
+- Mathematical: ∀ ∃ ∈ ∉ ∅ ∞ ∑ ∏ √ ∫ ∇ ∂ ≠ ≤ ≥ ≈ ≡ ⊂ ⊃ ⊆ ⊇
+- Arrows: ← → ↑ ↓ ↔ ↕ ↖ ↗ ↘ ↙
+"""
+
+def greet_in_greek():
+    return "Γεια σας κόσμος"  # "Hello world" in Greek
+
+def mathematical_formula():
+    # Using mathematical symbols in comments
+    # ∀x ∈ ℝ: x² ≥ 0
+    return "∑ from i=1 to ∞ of 1/i² = π²/6"
+
+def arrow_symbols():
+    directions = {
+        "left": "←",
+        "right": "→", 
+        "up": "↑",
+        "down": "↓"
+    }
+    return directions
+
+if __name__ == "__main__":
+    print(greet_in_greek())
+    print(greet_in_russian())
+    print(mathematical_formula())
+    print(arrow_symbols())
+'''
+
+    # Write to temporary file with UTF-8 encoding
+    with tempfile.NamedTemporaryFile(mode='w', suffix='.py', delete=False, encoding='utf-8') as f:
+        f.write(source_code)
+        temp_file = f.name
+
+    try:
+        # Read the file and minify it
+        with open(temp_file, 'r', encoding='utf-8') as f:
+            original_content = f.read()
+
+        # This should work - minify the UTF-8 content
+        minified = python_minifier.minify(original_content)
+
+        # Verify the minified code still contains the UTF-8 characters
+        assert "Γεια σας κόσμος" in minified
+        assert "∑ from i=1 to ∞" in minified
+        assert "←" in minified and "→" in minified
+
+    finally:
+        # Clean up
+        os.unlink(temp_file)
+
+
+def test_minify_utf8_file_direct():
+    """Test minifying a file directly with UTF-8 characters."""
+
+    # Create Python source with UTF-8 characters
+    source_code = '''# UTF-8 test file
+def emoji_function():
+    """Function with emoji and special characters: 🐍 ∆ ∑ ∫ ∞"""
+    return "Python is 🐍 awesome! Math symbols: ∆x ≈ 0, ∑∞ = ∞"
+
+class UnicodeClass:
+    """Class with unicode: ñ ü ö ä ë ï ÿ"""
+    def __init__(self):
+        self.message = "Héllö Wörld with àccénts!"
+        
+    def get_symbols(self):
+        return "Symbols: ™ © ® ° ± × ÷ ≠ ≤ ≥"
+'''
+
+    # Test direct minification
+    minified = python_minifier.minify(source_code)
+
+    # Verify UTF-8 characters are preserved
+    assert "🐍" in minified
+    assert "∆" in minified
+    assert "∑" in minified
+    assert "∞" in minified
+    assert "Héllö" in minified
+    assert "àccénts" in minified
+    assert "™" in minified
+    assert "©" in minified