svtplay-dl/lib/svtplay_dl/service/justin.py

# ex:ts=4:sw=4:sts=4:et
# -*- tab-width: 4; c-basic-offset: 4; indent-tabs-mode: nil -*-

# pylint has issues with urlparse: "some types could not be inferred"
# pylint: disable=E1103

from __future__ import absolute_import
import sys
import re
import xml.etree.ElementTree as ET

from svtplay_dl.utils.urllib import urlparse
from svtplay_dl.service import Service
from svtplay_dl.utils import get_http_data, select_quality, check_redirect, is_py2_old
from svtplay_dl.log import log
from svtplay_dl.fetcher.rtmp import download_rtmp
from svtplay_dl.fetcher.http import download_http

class Justin(Service):
    # Justin and Twitch uses language subdomains, e.g. en.www.twitch.tv. They
    # are usually two characters, but may have a country suffix as well (e.g.
    # zh-tw, zh-cn and pt-br.
    supported_domains_re = [
        r'^(?:(?:[a-z]{2}-)?[a-z]{2}\.)?(www\.)?twitch\.tv$',
        r'^(?:(?:[a-z]{2}-)?[a-z]{2}\.)?(www\.)?justin\.tv$']

    def get(self, options):
        parse = urlparse(self.url)
        match = re.search(r"/[-a-zA-Z0-9_]+/c/(\d+)", parse.path)
        if match:
            url = "http://api.justin.tv/api/broadcast/by_chapter/%s.xml?onsite=true" % match.group(1)
            data = get_http_data(url)
            xml = ET.XML(data)
            url = xml.find("archive").find("video_file_url").text

            download_http(options, url)
        else:
            match = re.search(r"/(.*)", parse.path)
            if match:
                user = match.group(1)
                data = get_http_data(url)
                match = re.search(r"embedSWF\(\"(.*)\", \"live", data)
                if not match:
                    log.error("Can't find swf file.")
                    sys.exit(2)
                options.other = check_redirect(match.group(1))
                url = "http://usher.justin.tv/find/%s.xml?type=any&p=2321" % user
                options.live = True
                data = get_http_data(url)
                data = re.sub(r"<(\d+)", r"<_\g<1>", data)
                data = re.sub(r"</(\d+)", r"</_\g<1>", data)
                xml = ET.XML(data)
                if is_py2_old:
                    sa = list(xml)
                else:
                    sa = list(xml)
                streams = {}
                for i in sa:
                    try:
                        stream = {}
                        stream["token"] = i.find("token").text
                        stream["url"] = "%s/%s" % (i.find("connect").text, i.find("play").text)
                        streams[int(i.find("video_height").text)] = stream
                    except AttributeError:
                        pass
                if len(streams) > 0:
                    test = select_quality(options, streams)
                    options.other = "-j '%s' -W %s" % (test["token"], options.other)
                    options.resume = False
                    download_rtmp(options, test["url"])
                else:
                    log.error("Can't find any streams")
                    sys.exit(2)