youtube-dl/youtube_dl/extractor/moviefap.py

from __future__ import unicode_literals

import re

from .common import InfoExtractor
from ..utils import str_to_int


class MovieFapIE(InfoExtractor):
    _VALID_URL = r'https?://(?:www\.)?moviefap\.com/videos/(?P<id>[0-9a-f]+)/(?P<name>[a-z-_]+)'
    _TESTS = [{
        'url': 'http://www.moviefap.com/videos/e5da0d3edce5404418f5/jeune-couple-russe.html',
        'md5': 'fa56683e291fc80635907168a743c9ad',
        'info_dict': {
            'id': 'e5da0d3edce5404418f5',
            'ext': 'flv',
            'title': 'Jeune Couple Russe',
            'description': 'Amateur',
            'thumbnail': 'http://pic.moviefap.com/thumbs/e5/949-18l.jpg',
            'uploader_id': 'whiskeyjar',
            'display_id': 'jeune-couple-russe'
        }
    }, {
        'url': 'http://www.moviefap.com/videos/3080837f6712355015c2/busty-british-blonde-takes-backdoor-in-fake-taxi.html',
        'md5': 'bedef72cb23d27a20755fc430a6d7a0e',
        'info_dict': {
            'id': '3080837f6712355015c2',
            'ext': 'mp4',
            'title': 'Busty British blonde takes backdoor in fake taxi',
            'description': 'Big boobs British blonde flashing in fake taxi then giving titsjob and rimjob in the back seat before getting big cock up her tight ass',
            'thumbnail': 'http://img.moviefap.com/a16:9w990r/thumbs/30/322021-18l.jpg',
            'uploader_id': 'momcikoper',
            'display_id': 'busty-british-blonde-takes-backdoor-in-fake-taxi'
        }
    }]

    @staticmethod
    def __get_thumbnail_data(xml):

        """
        Constructs a list of video thumbnails from timeline preview images.
        :param xml: the information XML document to parse
        """

        timeline = xml.find('timeline')
        if timeline is None:
            # not all videos have the data - ah well
            return []

        # get the required information from the XML
        width = str_to_int(timeline.find('imageWidth').text)
        height = str_to_int(timeline.find('imageHeight').text)
        first = str_to_int(timeline.find('imageFirst').text)
        last = str_to_int(timeline.find('imageLast').text)
        pattern = timeline.find('imagePattern').text

        # generate the list of thumbnail information dicts
        thumbnails = []
        for i in range(first, last + 1):
            thumbnails.append({
                'url': pattern.replace('#', str(i)),
                'width': width,
                'height': height
            })
        return thumbnails

    def _real_extract(self, url):

        # find the video ID
        video_id = self._match_id(url)

        # retrieve the page HTML
        webpage = self._download_webpage(url, video_id)

        # find the URL of the XML document detailing video download URLs
        info_url = self._html_search_regex(r'flashvars\.config = escape\("(.+?)"', webpage, 'player parameters')

        # download that XML
        xml = self._download_xml(info_url, video_id)

        # create dictionary of properties we know so far, or can find easily
        info = {
            'id': video_id,
            'title': self._html_search_regex(r'<div id="view_title"><h1>(.*?)</h1>', webpage, 'title'),
            'display_id': re.compile(self._VALID_URL).match(url).group('name'),
            'thumbnails': self.__get_thumbnail_data(xml),
            'thumbnail': xml.find('startThumb').text,
            'description': self._html_search_regex(r'name="description" value="(.*?)"', webpage, 'description'),
            'uploader_id': self._html_search_regex(r'name="username" value="(.*?)"', webpage, 'uploader_id'),
            'view_count': str_to_int(self._html_search_regex(r'<br>Views <strong>([0-9]+)</strong>', webpage, 'view_count')),
            'average_rating': float(self._html_search_regex(r'Current Rating<br> <strong>(.*?)</strong>', webpage, 'average_rating')),
            'comment_count': str_to_int(self._html_search_regex(r'<span id="comCount">([0-9]+)</span>', webpage, 'comment_count')),
            'age_limit': 18,
            'webpage_url': self._html_search_regex(r'name="link" value="(.*?)"', webpage, 'webpage_url'),
            'categories': self._html_search_regex(r'</div>\s*(.*?)\s*<br>', webpage, 'categories').split(', ')
        }

        # find and add the format
        if xml.find('videoConfig') is not None:
            info['ext'] = xml.find('videoConfig').find('type').text
        else:
            info['ext'] = 'flv'  # guess...

        # work out the video URL(s)
        if xml.find('videoLink') is not None:
            # single format available
            info['url'] = xml.find('videoLink').text
        else:
            # multiple formats available
            info['formats'] = []

            # N.B. formats are already in ascending order of quality
            for item in xml.find('quality').findall('item'):
                info['formats'].append({
                    'url': item.find('videoLink').text,
                    'resolution': item.find('res').text  # 480p etc.
                })

        return info