patent2table.py

"""
Script that takes patent url from Google Patents and creates nice little HTML table with patent metadata, then opens it
with your default browser. From there, you can copy the table to the clipboard and paste it in spreadsheet, nicely
formatted and parsed as a proper table. Apart from spreadsheet, any program that can process HTML tables from clipboard
will also work.

URL formats:
- https://patents.google.com/patent/[PATENT ID]
- https://patentimages.storage.googleapis.com/.../[PATENT ID].pdf

Examples:
- https://patents.google.com/patent/US3798032
- https://patentimages.storage.googleapis.com/21/6e/08/1fc5322ac7d593/US3798032.pdf

"""

import os
import logging
import tempfile
from pathlib import Path
import webbrowser

import requests
import bs4

logging.basicConfig(level=logging.INFO)


HTML_TPL = '''
<table border="1" style="font-family: arial">
<tr>
<td>!</td>
<td><a href="{url}">{title}</a></td>
<td>patent</td>
<td>{patent_num}</td>
<td>{date_issued}</td>
<td>{date_submitted}</td>
<td>{assignee}</td>
</tr>
</table>
'''

s = requests.Session()


def get_patent_data(url):
    def get_meta(filter):
        el = soup.select_one(f'meta{filter}')
        if el is None:
            return ''
        return el.attrs['content']

    if not url.startswith('https://patents.google.com/patent/'):
        basename = os.path.basename(url)
        patent_num = os.path.splitext(basename)[0]

        patent_url = f'https://patents.google.com/patent/{patent_num}'
    else:
        patent_url = url

    logging.info('Final url: %s', patent_url)

    html = s.get(patent_url).content
    soup = bs4.BeautifulSoup(html, 'html.parser')

    title = get_meta('[name="DC.title"]')
    date_issued = get_meta('[name="DC.date"][scheme="issue"]').split('-')[0]
    date_submitted = get_meta('[name="DC.date"][scheme="dateSubmitted"]').split('-')[0]
    assignee = get_meta('[name="DC.contributor"][scheme="assignee"]')
    patent_num = get_meta('[name="citation_patent_number"]').replace(':', '')

    patent_data = dict(
        title=title,
        date_issued=date_issued,
        date_submitted=date_submitted,
        assignee=assignee,
        url=patent_url,
        patent_num=patent_num
    )

    return patent_data


if __name__ == '__main__':
    url = 'https://patents.google.com/patent/US3798032'

    patent_data = get_patent_data(url)

    tmpdir = tempfile.gettempdir()
    html_path = Path(tmpdir) / 'patent.html'

    with open(html_path, 'w') as tmpf:
        html_out = HTML_TPL.format(**patent_data)
        tmpf.write(html_out)

    logging.info('Output path: %s', html_path)

    webbrowser.open(html_path)