Download shared directory from a seafile server

download_seafile.py

#!/usr/bin/env python
'''
Download files from a password protected seafile share.
Converted to use python from https://gist.github.com/vmx/ed6f91b93bc3c4f7e152dcd5f377ecef
'''
import requests
from argparse import ArgumentParser
import logging
from urllib.parse import urlparse
from pathlib import Path
from tqdm import tqdm
from getpass import getpass
import sys

log = logging.getLogger('seafile')

parser = ArgumentParser()
parser.add_argument('url')
parser.add_argument('directory')
parser.add_argument('outputdir', type=Path)
parser.add_argument('--password')


def download_file(session, url, path, chunk_size=10240, progress=False, **kwargs):
    log.info(f"Downloading {url} to {path}")
    name = urlparse(url).path.split("/")[-1]
    path = Path(path)

    with session.get(url, stream=True, timeout=5, **kwargs) as r:
        # make sure the request is successful
        r.raise_for_status()

        total = float(r.headers.get("Content-Length", float("inf")))
        pbar = tqdm(
            total=total,
            disable=not progress,
            unit="B",
            unit_scale=True,
            desc=f"Downloading {name}",
        )

        # open a .part file to avoid creating
        # a broken file at the intended location
        part_file = path.with_suffix(path.suffix + ".part")

        try:

            part_file.parent.mkdir(parents=True, exist_ok=True)
            with part_file.open("wb") as f:
                for chunk in r.iter_content(chunk_size=chunk_size):
                    f.write(chunk)
                    pbar.update(len(chunk))
        except:  # we really want to catch everythin here
            # cleanup part file if something goes wrong
            if part_file.is_file():
                part_file.unlink()
            raise

    # when successful, move to intended location
    part_file.rename(path)


def main():
    args = parser.parse_args()
    password = args.password or getpass()
    logging.basicConfig(level=logging.INFO)

    parsed_url = urlparse(args.url.rstrip('/'))
    token = parsed_url.path.split('/')[-1]

    # make sure we have a single '/' at the end
    url = args.url.rstrip('/') + '/'

    session = requests.session()
    session.headers['referer'] = url

    # get the csrf token cookie
    session.get(url)

    # login
    session.post(
        # make sure there is / at the end
        url,
        data={
            'csrfmiddlewaretoken': session.cookies['sfcsrftoken'],
            'token': token,
            'password': password,
        }
    )
    if "sessionid" not in session.cookies:
        log.critical('Login failed')
        sys.exit(1)

    # get file listing
    api_url = f'{parsed_url.scheme}://{parsed_url.hostname}/api/v2.1/share-links/{token}/dirents/'
    ret = session.get(api_url, params=dict(path=args.directory))
    if ret.status_code == 404:
        logging.critical(f"Did not find path {args.directory} for given URL")
        sys.exit(1)
        
    paths = [Path(p['file_path']) for p in ret.json()['dirent_list']]

    args.outputdir.mkdir(parents=True, exist_ok=True)

    for path in paths:
        download_file(
            session,
            f'{url}files/',
            args.outputdir / path.name,
            params=dict(p=str(path), dl=1),
            progress=True,
        )


if __name__ == '__main__':
    main()