From 8652770bd23ff5f46c5687d94f71cec08d2c5886 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Sergey=20M=E2=80=A4?= Date: Wed, 17 Aug 2016 05:44:46 +0700 Subject: [PATCH] [keezmovies] Improve and modernize --- youtube_dl/extractor/keezmovies.py | 138 +++++++++++++++++++++-------- 1 file changed, 99 insertions(+), 39 deletions(-) diff --git a/youtube_dl/extractor/keezmovies.py b/youtube_dl/extractor/keezmovies.py index 126ca13df..ad2f8a8c8 100644 --- a/youtube_dl/extractor/keezmovies.py +++ b/youtube_dl/extractor/keezmovies.py @@ -3,64 +3,124 @@ from __future__ import unicode_literals import re from .common import InfoExtractor +from ..aes import aes_decrypt_text +from ..compat import ( + compat_str, + compat_urllib_parse_unquote, +) from ..utils import ( - sanitized_Request, - url_basename, + determine_ext, + ExtractorError, + int_or_none, + str_to_int, + strip_or_none, ) class KeezMoviesIE(InfoExtractor): - _VALID_URL = r'https?://(?:www\.)?keezmovies\.com/video/.+?(?P[0-9]+)(?:[/?&]|$)' - _TEST = { + _VALID_URL = r'https?://(?:www\.)?keezmovies\.com/video/(?:(?P[^/]+)-)?(?P\d+)' + _TESTS = [{ 'url': 'http://www.keezmovies.com/video/petite-asian-lady-mai-playing-in-bathtub-1214711', 'md5': '1c1e75d22ffa53320f45eeb07bc4cdc0', 'info_dict': { 'id': '1214711', + 'display_id': 'petite-asian-lady-mai-playing-in-bathtub', 'ext': 'mp4', 'title': 'Petite Asian Lady Mai Playing In Bathtub', - 'age_limit': 18, 'thumbnail': 're:^https?://.*\.jpg$', + 'view_count': int, + 'age_limit': 18, } - } + }, { + 'url': 'http://www.keezmovies.com/video/1214711', + 'only_matching': True, + }] - def _real_extract(self, url): - video_id = self._match_id(url) + def _extract_info(self, url): + mobj = re.match(self._VALID_URL, url) + video_id = mobj.group('id') + display_id = mobj.group('display_id') or video_id - req = sanitized_Request(url) - req.add_header('Cookie', 'age_verified=1') - webpage = self._download_webpage(req, video_id) - - # embedded video - mobj = re.search(r'href="([^"]+)">', webpage) - if mobj: - embedded_url = mobj.group(1) - return self.url_result(embedded_url) - - video_title = self._html_search_regex( - r'

]*>([^<]+)', webpage, 'title') - flashvars = self._parse_json(self._search_regex( - r'var\s+flashvars\s*=\s*([^;]+);', webpage, 'flashvars'), video_id) + webpage = self._download_webpage( + url, display_id, headers={'Cookie': 'age_verified=1'}) formats = [] - for height in (180, 240, 480): - if flashvars.get('quality_%dp' % height): - video_url = flashvars['quality_%dp' % height] - a_format = { - 'url': video_url, - 'height': height, - 'format_id': '%dp' % height, - } - filename_parts = url_basename(video_url).split('_') - if len(filename_parts) >= 2 and re.match(r'\d+[Kk]', filename_parts[1]): - a_format['tbr'] = int(filename_parts[1][:-1]) - formats.append(a_format) + format_urls = set() - age_limit = self._rta_search(webpage) + title = None + thumbnail = None + duration = None + encrypted = False - return { + def extract_format(format_url, height=None): + if not isinstance(format_url, compat_str) or not format_url.startswith('http'): + return + if format_url in format_urls: + return + format_urls.add(format_url) + tbr = int_or_none(self._search_regex( + r'[/_](\d+)[kK][/_]', format_url, 'tbr', default=None)) + if not height: + height = int_or_none(self._search_regex( + r'[/_](\d+)[pP][/_]', format_url, 'height', default=None)) + if encrypted: + format_url = aes_decrypt_text( + video_url, title, 32).decode('utf-8') + formats.append({ + 'url': format_url, + 'format_id': '%dp' % height if height else None, + 'height': height, + 'tbr': tbr, + }) + + flashvars = self._parse_json( + self._search_regex( + r'flashvars\s*=\s*({.+?});', webpage, + 'flashvars', default='{}'), + display_id, fatal=False) + + if flashvars: + title = flashvars.get('video_title') + thumbnail = flashvars.get('image_url') + duration = int_or_none(flashvars.get('video_duration')) + encrypted = flashvars.get('encrypted') is True + for key, value in flashvars.items(): + mobj = re.search(r'quality_(\d+)[pP]', key) + if mobj: + extract_format(value, int(mobj.group(1))) + video_url = flashvars.get('video_url') + if video_url and determine_ext(video_url, None): + extract_format(video_url) + + video_url = self._html_search_regex( + r'flashvars\.video_url\s*=\s*(["\'])(?Phttp.+?)\1', + webpage, 'video url', default=None, group='url') + if video_url: + extract_format(compat_urllib_parse_unquote(video_url)) + + if not formats: + if 'title="This video is no longer available"' in webpage: + raise ExtractorError( + 'Video %s is no longer available' % video_id, expected=True) + + self._sort_formats(formats) + + if not title: + title = self._html_search_regex( + r']*>([^<]+)', webpage, 'title') + + return webpage, { 'id': video_id, - 'title': video_title, + 'display_id': display_id, + 'title': strip_or_none(title), + 'thumbnail': thumbnail, + 'duration': duration, + 'age_limit': 18, 'formats': formats, - 'age_limit': age_limit, - 'thumbnail': flashvars.get('image_url') } + + def _real_extract(self, url): + webpage, info = self._extract_info(url) + info['view_count'] = str_to_int(self._search_regex( + r'([\d,.]+) Views?', webpage, 'view count', fatal=False)) + return info