Local FAISS MCP Server

#!/usr/bin/env python3 """ Essential tests for document parsing functionality. """ import os import tempfile import pytest from pathlib import Path from local_faiss_mcp.document_parser import ( parse_text_file, parse_document, is_file_path ) class TestDocumentParser: """Essential tests for document parsing.""" def test_parse_text_file(self): """Test parsing plain text files.""" with tempfile.NamedTemporaryFile(mode='w', suffix='.txt', delete=False) as f: f.write("This is a test document.\nWith multiple lines.") temp_path = f.name try: content = parse_text_file(Path(temp_path)) assert "This is a test document" in content assert "multiple lines" in content finally: os.unlink(temp_path) def test_parse_markdown_file(self): """Test parsing markdown files.""" with tempfile.NamedTemporaryFile(mode='w', suffix='.md', delete=False) as f: f.write("# Heading\n\nThis is **bold** text.") temp_path = f.name try: content = parse_document(temp_path) assert "# Heading" in content assert "**bold**" in content finally: os.unlink(temp_path) def test_is_file_path_detection(self): """Test file path detection heuristic.""" # Should detect as paths (files that exist) assert is_file_path(__file__) == True # This test file exists # Should detect as paths (contain path separators and extensions) assert is_file_path("/path/to/file.txt") == True assert is_file_path("./relative/path.md") == True assert is_file_path("docs/readme.pdf") == True # Should NOT detect as paths (natural language, no separators) assert is_file_path("This is a long document with many words") == False assert is_file_path("FAISS is a library for similarity search") == False assert is_file_path("document.pdf") == False # No separator, doesn't exist assert is_file_path("") == False def test_is_file_path_handles_very_long_text(self): """Test that is_file_path doesn't crash on very long text (regression test for 'file name too long' bug).""" # Simulate what happens when LLM passes a long document with path separators long_text = """ This is a very long document that contains forward slashes / and might have URLs like https://example.com/very/long/path/to/something. """ * 1000 # Make it very long (>50,000 characters) # Should not crash with OSError, should return False result = is_file_path(long_text) assert result == False # Also test with backslashes (Windows-style paths) long_text_windows = "Some text with backslashes\\path\\to\\file " * 1000 result = is_file_path(long_text_windows) assert result == False def test_is_file_path_length_limits(self): """Test that is_file_path respects reasonable path length limits.""" # Paths over 2000 characters should be rejected immediately very_long_path = "a/" * 2000 # Creates a path > 2000 characters assert is_file_path(very_long_path) == False # Reasonable path under 500 chars should still work (if it has proper structure) reasonable_path = "/some/deeply/nested/" + "dir/" * 30 + "file.txt" # This path is ~160 characters, well under 500, should be detected assert len(reasonable_path) < 500 assert is_file_path(reasonable_path) == True def test_parse_document_file_not_found(self): """Test that parsing non-existent file raises error.""" with pytest.raises(FileNotFoundError): parse_document("/non/existent/file.txt") def test_parse_document_unsupported_format_without_pandoc(self): """Test parsing unsupported format without pandoc.""" with tempfile.NamedTemporaryFile(mode='w', suffix='.xyz', delete=False) as f: f.write("content") temp_path = f.name try: # Should raise ValueError for unsupported format with pytest.raises(ValueError, match="Unsupported file format"): parse_document(temp_path) finally: os.unlink(temp_path) if __name__ == '__main__': pytest.main([__file__, '-v'])