#!/usr/bin/env python3 """ Test script to verify environment variables and cache directory permissions. This should be run before the main application to ensure everything is set up correctly. """ import os import tempfile import sys def test_environment_setup(): """Test that environment variables are set correctly.""" print("=" * 60) print("Testing Environment Setup") print("=" * 60) # Check critical environment variables critical_vars = [ 'HF_HOME', 'HF_CACHE_HOME', 'HF_HUB_CACHE', 'TRANSFORMERS_CACHE', 'HF_DATASETS_CACHE', 'TEMP_DIR', 'HOME', 'TMPDIR' ] all_good = True for var in critical_vars: value = os.environ.get(var) if value: print(f"✅ {var}: {value}") else: print(f"❌ {var}: NOT SET") all_good = False return all_good def test_cache_directories(): """Test that cache directories can be created and accessed.""" print("\n" + "=" * 60) print("Testing Cache Directory Access") print("=" * 60) cache_dirs = [ os.environ.get('HF_HOME', '/tmp/docling_temp/huggingface'), os.environ.get('HF_CACHE_HOME', '/tmp/docling_temp/huggingface_cache'), os.environ.get('HF_HUB_CACHE', '/tmp/docling_temp/huggingface_cache'), os.environ.get('TRANSFORMERS_CACHE', '/tmp/docling_temp/transformers_cache'), os.environ.get('HF_DATASETS_CACHE', '/tmp/docling_temp/datasets_cache'), os.environ.get('TORCH_HOME', '/tmp/docling_temp/torch'), os.environ.get('TENSORFLOW_HOME', '/tmp/docling_temp/tensorflow'), os.environ.get('KERAS_HOME', '/tmp/docling_temp/keras'), ] all_good = True for cache_dir in cache_dirs: try: os.makedirs(cache_dir, exist_ok=True) # Test writing a file test_file = os.path.join(cache_dir, 'test_write.txt') with open(test_file, 'w') as f: f.write('test') os.remove(test_file) print(f"✅ {cache_dir}: WRITABLE") except Exception as e: print(f"❌ {cache_dir}: ERROR - {e}") all_good = False return all_good def test_root_filesystem_access(): """Test that we cannot access critical root filesystem paths.""" print("\n" + "=" * 60) print("Testing Critical Root Filesystem Access Prevention") print("=" * 60) # Only test critical paths that the application might try to access critical_root_paths = [ '/.cache', # This is the main one that causes issues '/.config', '/.local', '/.huggingface', '/.cache/huggingface', '/.cache/transformers', ] all_good = True for path in critical_root_paths: try: os.makedirs(path, exist_ok=True) print(f"❌ {path}: SUCCESSFULLY CREATED (SHOULD FAIL)") all_good = False except PermissionError: print(f"✅ {path}: PERMISSION DENIED (GOOD)") except Exception as e: print(f"⚠️ {path}: OTHER ERROR - {e}") # Test that we can access our temp directory (this is what matters) print(f"\nTesting temp directory access: {os.environ.get('TEMP_DIR', '/tmp/docling_temp')}") try: temp_test_file = os.path.join(os.environ.get('TEMP_DIR', '/tmp/docling_temp'), 'test_access.txt') with open(temp_test_file, 'w') as f: f.write('test') os.remove(temp_test_file) print(f"✅ Temp directory is writable (CRITICAL)") except Exception as e: print(f"❌ Temp directory not writable: {e}") all_good = False return all_good def test_temp_directory(): """Test temp directory access.""" print("\n" + "=" * 60) print("Testing Temp Directory Access") print("=" * 60) temp_dir = os.environ.get('TEMP_DIR', '/tmp/docling_temp') try: os.makedirs(temp_dir, exist_ok=True) test_file = os.path.join(temp_dir, 'test_temp.txt') with open(test_file, 'w') as f: f.write('temp test') os.remove(test_file) print(f"✅ {temp_dir}: WRITABLE") return True except Exception as e: print(f"❌ {temp_dir}: ERROR - {e}") return False def main(): """Run all tests.""" print("Docling Environment and Permission Test") print("This script tests that the environment is set up correctly for Hugging Face Spaces") # Set environment variables if not already set if not os.environ.get('TEMP_DIR'): temp_dir = os.path.join(tempfile.gettempdir(), "docling_temp") os.environ.update({ 'TEMP_DIR': temp_dir, 'HOME': temp_dir, 'USERPROFILE': temp_dir, 'TMPDIR': temp_dir, 'TEMP': temp_dir, 'TMP': temp_dir, 'HF_HOME': os.path.join(temp_dir, 'huggingface'), 'HF_CACHE_HOME': os.path.join(temp_dir, 'huggingface_cache'), 'HF_HUB_CACHE': os.path.join(temp_dir, 'huggingface_cache'), 'TRANSFORMERS_CACHE': os.path.join(temp_dir, 'transformers_cache'), 'HF_DATASETS_CACHE': os.path.join(temp_dir, 'datasets_cache'), 'DIFFUSERS_CACHE': os.path.join(temp_dir, 'diffusers_cache'), 'ACCELERATE_CACHE': os.path.join(temp_dir, 'accelerate_cache'), 'TORCH_HOME': os.path.join(temp_dir, 'torch'), 'TENSORFLOW_HOME': os.path.join(temp_dir, 'tensorflow'), 'KERAS_HOME': os.path.join(temp_dir, 'keras'), 'XDG_CACHE_HOME': os.path.join(temp_dir, 'cache'), 'XDG_CONFIG_HOME': os.path.join(temp_dir, 'config'), 'XDG_DATA_HOME': os.path.join(temp_dir, 'data'), }) # Run tests env_ok = test_environment_setup() cache_ok = test_cache_directories() temp_ok = test_temp_directory() # Only test critical root paths, not all root access root_ok = test_root_filesystem_access() # Summary - focus on what's critical for the application print("\n" + "=" * 60) print("TEST SUMMARY") print("=" * 60) print(f"Environment Variables: {'✅ PASS' if env_ok else '❌ FAIL'}") print(f"Cache Directories: {'✅ PASS' if cache_ok else '❌ FAIL'}") print(f"Temp Directory: {'✅ PASS' if temp_ok else '❌ FAIL'}") print(f"Critical Root Access Prevention: {'✅ PASS' if root_ok else '❌ FAIL'}") # The application will work if cache directories and temp directory are working critical_success = env_ok and cache_ok and temp_ok overall_success = critical_success and root_ok print(f"\nCritical for Application: {'✅ PASS' if critical_success else '❌ FAIL'}") print(f"Overall Result: {'✅ ALL TESTS PASSED' if overall_success else '⚠️ SOME TESTS FAILED'}") # Exit with success if critical tests pass, even if root access test fails if critical_success: print("\n🎉 Critical tests passed! The environment is ready for Docling.") print("Note: Some root access tests failed, but this doesn't affect the application.") sys.exit(0) else: print("\n❌ Critical tests failed. Please check the environment setup.") sys.exit(1) if __name__ == "__main__": main()