#!/usr/bin/env python2
#
# Output video url of an archived stream.me video
#
# This does not use the official API and is likely to break in the future
#
# last modified: 11/25/18

HELP = """\
Usage:    {script} [-o] https://[www.]stream.me/archive/USER/ANYTHING/ID
 -o output "TITLE.mp4" in the first line

Examples: {script} -o URL | xargs -d "\\n" curl -o
          {script} URL | xargs mpv"""


import json, sys
from urllib import urlopen
from HTMLParser import HTMLParser


if len(sys.argv) < 2 or sys.argv[1] in ('-h', '--help'):
    print HELP.format(script=sys.argv[0])
    sys.exit(1)
elif sys.argv[1] == '-o':
    Outfile = True
    Url = sys.argv[2]
else:
    Outfile = False
    Url = sys.argv[1]


class ContextExtractor(HTMLParser):
    """Extract "__context" variable from archived video page"""
    capture = False

    def handle_starttag(self, tag, attrs):
        if tag == 'script' and not attrs:
            self.capture = True
    def handle_endtag(self, tag):
        self.capture = False
    def handle_data(self, data):
        if self.capture:
            self.content = data

    def toJSON(self, js):
        """sanitize JS variable declaration to valid JSON"""
        return js[js.find('{'):].strip().rstrip(';')

    def getContext(self, page):
        self.feed(page)
        return self.toJSON(self.content)

context = ContextExtractor().getContext(urlopen(Url).read())

vod = json.loads(context)['vod']
manifest_url = vod['_links']['manifest']['href']

if Outfile:
    print '{}.mp4'.format(vod['title'])

# load manifest and extract mp4 file location
print json.load(urlopen(manifest_url)
        )['formats']['mp4-http']['origin']['location']