Skip to content

trading_agent.attachments.word_text_reader

Reads the text of a Word document (.docx) with python-docx: its paragraphs, then its tables.

The older .doc format is a different, binary format that python-docx cannot open, so it is not accepted.

Typical usage example:

extracted = WordTextReader().read(Path('plan.docx'))

WordTextReader

Bases: TextReader

Reads a .docx file.

read 

read(path)

Reads the paragraphs, and each table row with its cells separated by " | ".

Parameters:

Name Type Description Default
path Path

The document.

 required

Returns:

Type Description
ExtractedText

Its text.

Raises:

Type Description
TextReadError

The file is damaged or is not a .docx file.
Source code in src/trading_agent/attachments/word_text_reader.py 
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
def read(self, path: Path) -> text_reader.ExtractedText:
    """Reads the paragraphs, and each table row with its cells separated by " | ".

    Args:
        path (Path): The document.

    Returns:
        text_reader.ExtractedText: Its text.

    Raises:
        TextReadError: The file is damaged or is not a .docx file.
    """
    try:
        document = docx.Document(str(path))
    except (
        docx.opc.exceptions.PackageNotFoundError,
        zipfile.BadZipFile,
        KeyError,
        ValueError,
    ) as error:
        raise text_reader.TextReadError(
            f'the Word document cannot be read ({type(error).__name__}: {error})'
        ) from error
    lines = []
    for paragraph in document.paragraphs:
        lines.append(paragraph.text)
    for number, table in enumerate(document.tables, start=1):
        lines.append('')
        lines.append(f'--- Table {number} ---')
        for row in table.rows:
            cells = []
            for cell in row.cells:
                cells.append(cell.text.strip())
            lines.append(' | '.join(cells))
    return text_reader.ExtractedText(text='\n'.join(lines).strip())