ingestion/docx_reader.py

from docx import Document

def extract_text_from_docx(path):
    doc = Document(path)
    return "\n".join(p.text for p in doc.paragraphs)