Test on windows

Author: dflook

test/test_windows_encoding_regression.py

@@ -0,0 +1,156 @@
+import pytest
+import tempfile
+import os
+import subprocess
+import sys
+import locale
+
+
+def test_cli_output_flag_with_unicode():
+    """Regression test for GitHub issues #2, #57, #59, #68, #113, #123, #129.
+
+    Tests that the CLI tool can write Unicode characters to files using --output flag.
+    This should fail on Windows without proper encoding handling.
+    """
+    # Minimal source with all problematic Unicode characters from reported issues
+    source_code = 'print("❌ ✓ 🐍 Привет © ∀")'
+
+    # Create temporary source file
+    with tempfile.NamedTemporaryFile(mode='w', suffix='.py', delete=False, encoding='utf-8') as source_file:
+        source_file.write(source_code)
+        source_path = source_file.name
+
+    # Create temporary output file path
+    with tempfile.NamedTemporaryFile(mode='w', suffix='.min.py', delete=False) as output_file:
+        output_path = output_file.name
+
+    try:
+        # Remove output file so pyminify can create it
+        os.unlink(output_path)
+
+        # Run pyminify CLI with --output flag (this should reproduce Windows encoding errors)
+        result = subprocess.run([
+            sys.executable, '-m', 'python_minifier',
+            source_path, '--output', output_path
+        ], capture_output=True, text=True, timeout=30)
+
+        if result.returncode != 0:
+            # This is expected on Windows with encoding issues
+            print(f"CLI failed as expected on Windows: {result.stderr}")
+            assert "UnicodeEncodeError" in result.stderr or "charmap" in result.stderr or "gbk" in result.stderr
+        else:
+            # If it succeeds, verify the output file contains Unicode characters
+            with open(output_path, 'r', encoding='utf-8') as f:
+                minified_content = f.read()
+
+            # Verify problematic Unicode characters are preserved
+            assert "❌" in minified_content  # Issue #113
+            assert "✓" in minified_content   # Issue #129
+            assert "🐍" in minified_content  # General emoji
+            assert "Привет мир" in minified_content  # Issue #123
+            assert "©" in minified_content   # Issue #59
+            assert "∀" in minified_content   # Mathematical symbols
+
+    finally:
+        # Cleanup
+        if os.path.exists(source_path):
+            os.unlink(source_path)
+        if os.path.exists(output_path):
+            os.unlink(output_path)
+
+
+def test_cli_in_place_with_unicode():
+    """Regression test for --in-place flag with Unicode characters.
+
+    Tests GitHub issues #57, #68 where --in-place fails on Windows.
+    """
+    source_code = 'print("❌ ✓ 🐍 Привет © ∀")'
+
+    # Create temporary file
+    with tempfile.NamedTemporaryFile(mode='w', suffix='.py', delete=False, encoding='utf-8') as temp_file:
+        temp_file.write(source_code)
+        temp_path = temp_file.name
+
+    try:
+        # Run pyminify with --in-place flag
+        result = subprocess.run([
+            sys.executable, '-m', 'python_minifier',
+            temp_path, '--in-place'
+        ], capture_output=True, text=True, timeout=30)
+
+        if result.returncode != 0:
+            # Expected failure on Windows
+            print(f"--in-place failed as expected on Windows: {result.stderr}")
+            assert "UnicodeEncodeError" in result.stderr or "charmap" in result.stderr or "gbk" in result.stderr
+        else:
+            # If successful, verify Unicode characters are preserved
+            with open(temp_path, 'r', encoding='utf-8') as f:
+                content = f.read()
+
+            assert "✓" in content
+            assert "❌" in content
+            assert "🐍" in content
+            assert "Привет" in content
+            assert "©" in content
+
+    finally:
+        if os.path.exists(temp_path):
+            os.unlink(temp_path)
+
+
+def test_cli_stdout_works_with_unicode():
+    """Verify that stdout output works fine (as reported in issues).
+
+    All GitHub issues mention that stdout output works, only file output fails.
+    """
+    source_code = 'print("❌ ✓ 🐍 Привет © ∀")'
+
+    with tempfile.NamedTemporaryFile(mode='w', suffix='.py', delete=False, encoding='utf-8') as temp_file:
+        temp_file.write(source_code)
+        temp_path = temp_file.name
+
+    try:
+        # Run without --output or --in-place (should output to stdout)
+        result = subprocess.run([
+            sys.executable, '-m', 'python_minifier', temp_path
+        ], capture_output=True, text=True, timeout=30)
+
+        # This should always succeed (as confirmed by GitHub issues)
+        assert result.returncode == 0, f"Stdout output failed: {result.stderr}"
+
+        # Verify Unicode characters in stdout
+        assert "❌" in result.stdout
+        assert "✓" in result.stdout
+        assert "🐍" in result.stdout
+        assert "Привет" in result.stdout
+        assert "©" in result.stdout
+
+    finally:
+        os.unlink(temp_path)
+
+
+@pytest.mark.skipif(os.name != 'nt', reason="Windows-specific encoding test")
+def test_windows_default_encoding_detection():
+    """Test to detect Windows default encoding that causes issues."""
+
+    # Check what encoding Python would use on Windows for file operations
+    default_encoding = locale.getpreferredencoding()
+
+    # On problematic Windows systems, this is often cp1252, gbk, or similar
+    print(f"Windows default encoding: {default_encoding}")
+
+    # This test documents the encoding environment for debugging
+    assert default_encoding is not None
+
+
+def test_system_encoding_info():
+    """Diagnostic test to understand system encoding setup."""
+
+    print(f"System default encoding: {sys.getdefaultencoding()}")
+    print(f"Filesystem encoding: {sys.getfilesystemencoding()}")
+    print(f"Preferred encoding: {locale.getpreferredencoding()}")
+    print(f"Platform: {sys.platform}")
+
+    # This test always passes but provides diagnostic information
+    assert True
+