Add validation of processed lists in the document

    "update_notes": True,

    "update_references": True,

    "update_lists": True,

    "repair_invalid_numbering_references": True,

    "update_body_text_style": True,

    "add_no_break_hyphens": True,

    "update_references_style": True,

        if os.path.exists(tmp_path):

            os.remove(tmp_path)

def repair_invalid_numbering_references(docx_input, docx_output):

    """

    Repair invalid paragraph numbering references that can make DOCX files

    unreadable in Word.

    A paragraph numbering reference is considered invalid when:

    - w:numPr is missing w:numId or w:ilvl

    - w:numId points to a non-existing w:num in numbering.xml

    - w:ilvl is not defined for the referenced abstract numbering definition

    """

    ns = {"w": "http://schemas.openxmlformats.org/wordprocessingml/2006/main"}

    with zipfile.ZipFile(docx_input, 'r') as zin:

        xml_data = zin.read("word/document.xml")

        try:

            numbering_data = zin.read("word/numbering.xml")

            numbering_root = etree.fromstring(numbering_data)

        except KeyError:

            numbering_root = None

    # Nothing to validate if there is no numbering part.

    if numbering_root is None:

        return

    root = etree.fromstring(xml_data)

    num_to_abstract = {}

    abstract_levels = {}

    for abstract in numbering_root.xpath('.//w:abstractNum', namespaces=ns):

        abstract_id = abstract.get(f"{{{ns['w']}}}abstractNumId")

        levels = set()

        for lvl in abstract.xpath('./w:lvl', namespaces=ns):

            ilvl = lvl.get(f"{{{ns['w']}}}ilvl")

            if ilvl is not None:

                levels.add(ilvl)

        abstract_levels[abstract_id] = levels

    for num in numbering_root.xpath('.//w:num', namespaces=ns):

        num_id = num.get(f"{{{ns['w']}}}numId")

        abstract_id_elem = num.xpath('./w:abstractNumId', namespaces=ns)

        abstract_id = abstract_id_elem[0].get(f"{{{ns['w']}}}val") if abstract_id_elem else None

        num_to_abstract[num_id] = abstract_id

    repaired = 0

    for para in root.xpath('.//w:p', namespaces=ns):

        ppr = para.xpath('./w:pPr', namespaces=ns)

        if not ppr:

            continue

        ppr = ppr[0]

        numpr = ppr.xpath('./w:numPr', namespaces=ns)

        if not numpr:

            continue

        numpr = numpr[0]

        num_id_elem = numpr.xpath('./w:numId', namespaces=ns)

        ilvl_elem = numpr.xpath('./w:ilvl', namespaces=ns)

        invalid_ref = False

        if not num_id_elem or not ilvl_elem:

            invalid_ref = True

        else:

            num_id = num_id_elem[0].get(f"{{{ns['w']}}}val")

            ilvl = ilvl_elem[0].get(f"{{{ns['w']}}}val")

            abstract_id = num_to_abstract.get(num_id)

            valid_levels = abstract_levels.get(abstract_id, set())

            if abstract_id is None or ilvl not in valid_levels:

                invalid_ref = True

        if invalid_ref:

            ppr.remove(numpr)

            repaired += 1

    if repaired == 0:

        print("List numbering validation: no invalid numbering references found")

        return

    print(f"List numbering validation: repaired {repaired} invalid numbering reference(s)")

    xml_data = etree.tostring(root, xml_declaration=True, encoding="UTF-8", standalone="yes")

    tmp_fd, tmp_path = tempfile.mkstemp(suffix=".docx")

    os.close(tmp_fd)

    try:

        with zipfile.ZipFile(docx_input, 'r') as zin, zipfile.ZipFile(tmp_path, 'w', zipfile.ZIP_DEFLATED) as zout:

            for item in zin.infolist():

                if item.filename != "word/document.xml":

                    data = zin.read(item.filename)

                    zout.writestr(item.filename, data)

            zout.writestr("word/document.xml", xml_data)

        shutil.move(tmp_path, docx_output)

        os.chmod(docx_output, 0o644)

    finally:

        if os.path.exists(tmp_path):

            os.remove(tmp_path)

def update_table_captions(docx_input, docx_output):

    """

    Updates table caption styles from 'TableCaption' to 'TF' in a DOCX file.

        ("update_notes", update_notes),

        ("update_references", update_references),

        ("update_lists", update_lists),

        ("repair_invalid_numbering_references", repair_invalid_numbering_references),

        ("update_body_text_style", update_body_text_style),

        ("add_no_break_hyphens", add_no_break_hyphens),

        ("update_references_style", update_references_style),