This uses the Mercury parser (https://mercury.postlight.com/web-parser/) and other tools to convert an article from URL to an HTML file. Image URLs are converted into inline Base-64 encoded images. You can run the command with the following code: ./mercury_parse.py --url=<URL> --htmlfile=<htmlfile>. <url> has to exist, and <htmlfile> has to end …

mercury_parse.py

#!/usr/bin/env python

import os, sys, requests, base64, urllib.request, codecs
from bs4 import BeautifulSoup
from optparse import OptionParser

_apiKey = 'nGc0ya2J7z2aalFrGa8Gx3Q1o8grGFsn3cz58EJy'

def get_content_to_file( url, htmlfilename ):
    with requests.Session( ) as s:
        s.headers =  { 'Content-Type' : 'application/json',
                       'x-api-key' : _apiKey }
        response = s.get( 'https://mercury.postlight.com/parser',
                          params = { 'url' : url })
        if response.status_code != 200:
            return 'Error, no data from %s' % url
        data = response.json( )
        content = data['content']
        title = data['title']
        date_publish_string = data['date_published']
        excerpt = data['excerpt']
        url = data['url']
        html = BeautifulSoup( content, 'lxml' )
        #
        ## now all png objects to inline
        for img in html.find_all('img'):
            imgURL = img['src']
            if imgURL.lower( ).endswith('.png'):                
                img_64 = "data:image/png;base64," + str(base64.b64encode( urllib.request.urlopen( imgURL ).read( ) ),'utf8')
            elif imgURL.lower( ).endswith( '.jpg' ):
                img_64 = "data:image/jpg;base64," + str(base64.b64encode( urllib.request.urlopen( imgURL ).read( ) ),'utf8')
            else:
                img_64 = None
            #
            if img_64 is not None:
                img['src'] = img_64

        htag = html.new_tag( 'head' )
        mtag = html.new_tag( 'meta' )
        mtag['charset'] = 'utf-8'
        htag.append( mtag )
        html.insert(0, htag )
        with codecs.open( htmlfilename, 'w', 'utf-8') as openfile:
            openfile.write('%s\n' % html.prettify( ) )

if __name__=='__main__':
    parser = OptionParser( )
    parser.add_option('--url', dest='url', type=str, action='store',
                      help = "Name of the URL to output into an HTML file.")
    parser.add_option('--htmlfile', dest='htmlfile', type=str, action='store',
                      help = 'Name of the HTML file to store the underlying data.')
    opts, args = parser.parse_args( )
    assert(all(map(lambda tok: tok is not None, ( opts.url, opts.htmlfile ) ) ) )
    assert( os.path.basename( opts.htmlfile ).endswith( '.html' ) )
    get_content_to_file( opts.url, opts.htmlfile )