Interface with http://docverter.com

docvert.py

import re
import requests
from os.path import basename
from urllib.parse import urljoin


def generate_files(filenames):
    return [
        (
            'input_files[]',
            (
                basename(filename),
                file_data,
                'application/octet-stream'
            )
        )
        for filename, file_data in filenames
    ]


def get_args():
    import argparse

    parser = argparse.ArgumentParser()
    parser.add_argument('input_files', nargs='+')
    parser.add_argument('output_filename')
    parser.add_argument(
        '--mend-relative-links',
        '-m'
    )

    return parser.parse_args()

LINK_RE = re.compile(r'\[([^\]]*)\]\(([^\)]*)\)')


def mend_relative_links(data, url):
    return LINK_RE.sub(
        lambda match: '[{}]({})'.format(
            match.groups()[0],
            urljoin(url, match.groups()[1])
        ),
        data
    )


def get_pdf(files):
    r = requests.post(
        'http://c.docverter.com/convert',
        files=files,
        data={
            'from': 'markdown',
            'to': 'pdf'
        }
    )

    if r.headers['content-type'] != 'application/pdf':
        raise Exception(r.json()['error'])

    return r.content


def read(path):
    if path.startswith('http'):
        return requests.get(path).text
    else:
        with open(path) as fh:
            return fh.read()


def main():
    args = get_args()

    print(args.input_files)

    input_files = [
        (filename, read(filename))
        for filename in args.input_files
    ]

    if args.mend_relative_links is not None:
        input_files = [
            (
                filename,
                mend_relative_links(file_data, args.mend_relative_links)
            )
            for filename, file_data in input_files
        ]

    files = generate_files(input_files)

    pdf = get_pdf(files)

    with open(args.output_filename, 'wb') as fh:
        fh.write(pdf)

if __name__ == '__main__':
    main()