Text and files¶

The API supplies document text in one of two ways:

• Inline text. The text is in the metadata. Use doc.content.
• A file. The text is in a PDF or other file. Download the file.

Inline text¶

These document types contain the full text inline:

doc.content is a dictionary of field name to text:

doc = next(drs.iter_documents(DocType.FAR, limit=1))
print(doc.content["drs:farSectionRule"])

For other document types, doc.content is empty. Use catalog.get_doctype(code).has_inline_content to check a type.

Download the main file¶

Download into memory:

data = drs.download(doc)

Download to a directory. The client uses the file name from the document. If the path ends with /, the client makes the directory if it does not exist:

path = drs.download_to(doc, "downloads/")

Download to a file path:

path = drs.download_to(doc, "downloads/ac-39-8a.pdf")

The client writes to a temporary file, then renames it. If the download fails, no partial file stays at the destination.

Some documents have no file. For these, doc.download_url is None and download raises DRSError.

Download attachments¶

Some documents have more files. doc.has_attachments is True for these documents.

for attachment in drs.get_attachments(doc):
    drs.download_to(attachment, "downloads/")

If has_attachments is False, get_attachments returns an empty list and does not send a request.

Extract text from files¶

This package does not extract text from PDF files. Use a PDF library for that step.