ingestion.py

from pypdf import PdfReader
import docx2txt

def read_file(file):
    if file.name.endswith(".pdf"):
        reader = PdfReader(file)
        return " ".join(p.extract_text() or "" for p in reader.pages)
    if file.name.endswith(".docx"):
        return docx2txt.process(file)
    return file.read().decode("utf-8")