from __future__ import annotations try: from docx import Document as _Doc Document = _Doc except Exception: Document = None def parse_docx(file_path: str) -> str: """Parse a DOCX file and return its text content. The function preserves paragraph breaks by inserting a newline between paragraphs. Empty documents yield an empty string. Raises: ValueError: If the file is not a valid DOCX document or cannot be read. """ if Document is None: raise ValueError("DOCX library is not installed") try: doc = Document(file_path) except Exception as exc: # pragma: no cover - surface invalid DOCX raise ValueError(f"Invalid DOCX file: {exc}") from exc paragraphs = [para.text for para in doc.paragraphs if para.text is not None] # Join with newline to preserve paragraph breaks return "\n".join(paragraphs).strip()