Merge pull request #3 from unit193/master

[youtubedl] / youtube_dl / extractor / youporn.py

diff --git a/youtube_dl/extractor/youporn.py b/youtube_dl/extractor/youporn.py
index c85fd4b5af0ccdd3f259bd403ddd4311f2de5fdb..d4eccb4b2a48efafec0232a451b3ee617e6bc859 100644 (file)
--- a/youtube_dl/extractor/youporn.py
+++ b/youtube_dl/extractor/youporn.py
@@ -1,138 +1,192 @@
-import json
-import os
+from __future__ import unicode_literals
+

 import re
 import re

-import sys

 
 from .common import InfoExtractor
 from ..utils import (
 
 from .common import InfoExtractor
 from ..utils import (

-    compat_urllib_parse_urlparse,
-    compat_urllib_request,
-
-    ExtractorError,
+    int_or_none,
+    sanitized_Request,
+    str_to_int,

     unescapeHTML,
     unified_strdate,
     unescapeHTML,
     unified_strdate,

+    url_or_none,

 )
 )

-from ..aes import (
-    aes_decrypt_text
-)
+from ..aes import aes_decrypt_text
+

 
 class YouPornIE(InfoExtractor):
 
 class YouPornIE(InfoExtractor):

-    _VALID_URL = r'^(?:https?://)?(?:\w+\.)?youporn\.com/watch/(?P<videoid>[0-9]+)/(?P<title>[^/]+)'
-    _TEST = {
-        u'url': u'http://www.youporn.com/watch/505835/sex-ed-is-it-safe-to-masturbate-daily/',
-        u'file': u'505835.mp4',
-        u'md5': u'71ec5fcfddacf80f495efa8b6a8d9a89',
-        u'info_dict': {
-            u"upload_date": u"20101221", 
-            u"description": u"Love & Sex Answers: http://bit.ly/DanAndJenn -- Is It Unhealthy To Masturbate Daily?", 
-            u"uploader": u"Ask Dan And Jennifer", 
-            u"title": u"Sex Ed: Is It Safe To Masturbate Daily?"
-        }
-    }
-
-    def _print_formats(self, formats):
-        """Print all available formats"""
-        print(u'Available formats:')
-        print(u'ext\t\tformat')
-        print(u'---------------------------------')
-        for format in formats:
-            print(u'%s\t\t%s'  % (format['ext'], format['format']))
-
-    def _specific(self, req_format, formats):
-        for x in formats:
-            if x["format"] == req_format:
-                return x
-        return None
+    _VALID_URL = r'https?://(?:www\.)?youporn\.com/watch/(?P<id>\d+)/(?P<display_id>[^/?#&]+)'
+    _TESTS = [{
+        'url': 'http://www.youporn.com/watch/505835/sex-ed-is-it-safe-to-masturbate-daily/',
+        'md5': '3744d24c50438cf5b6f6d59feb5055c2',
+        'info_dict': {
+            'id': '505835',
+            'display_id': 'sex-ed-is-it-safe-to-masturbate-daily',
+            'ext': 'mp4',
+            'title': 'Sex Ed: Is It Safe To Masturbate Daily?',
+            'description': 'Love & Sex Answers: http://bit.ly/DanAndJenn -- Is It Unhealthy To Masturbate Daily?',
+            'thumbnail': r're:^https?://.*\.jpg$',
+            'uploader': 'Ask Dan And Jennifer',
+            'upload_date': '20101217',
+            'average_rating': int,
+            'view_count': int,
+            'comment_count': int,
+            'categories': list,
+            'tags': list,
+            'age_limit': 18,
+        },
+    }, {
+        # Unknown uploader
+        'url': 'http://www.youporn.com/watch/561726/big-tits-awesome-brunette-on-amazing-webcam-show/?from=related3&al=2&from_id=561726&pos=4',
+        'info_dict': {
+            'id': '561726',
+            'display_id': 'big-tits-awesome-brunette-on-amazing-webcam-show',
+            'ext': 'mp4',
+            'title': 'Big Tits Awesome Brunette On amazing webcam show',
+            'description': 'http://sweetlivegirls.com Big Tits Awesome Brunette On amazing webcam show.mp4',
+            'thumbnail': r're:^https?://.*\.jpg$',
+            'uploader': 'Unknown',
+            'upload_date': '20110418',
+            'average_rating': int,
+            'view_count': int,
+            'comment_count': int,
+            'categories': list,
+            'tags': list,
+            'age_limit': 18,
+        },
+        'params': {
+            'skip_download': True,
+        },
+    }]

 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)

-        video_id = mobj.group('videoid')
-
-        req = compat_urllib_request.Request(url)
-        req.add_header('Cookie', 'age_verified=1')
-        webpage = self._download_webpage(req, video_id)
-
-        # Get JSON parameters
-        json_params = self._search_regex(r'var currentVideo = new Video\((.*)\);', webpage, u'JSON parameters')
-        try:
-            params = json.loads(json_params)
-        except:
-            raise ExtractorError(u'Invalid JSON')
-
-        self.report_extraction(video_id)
-        try:
-            video_title = params['title']
-            upload_date = unified_strdate(params['release_date_f'])
-            video_description = params['description']
-            video_uploader = params['submitted_by']
-            thumbnail = params['thumbnails'][0]['image']
-        except KeyError:
-            raise ExtractorError('Missing JSON parameter: ' + sys.exc_info()[1])
-
-        # Get all of the formats available
-        DOWNLOAD_LIST_RE = r'(?s)<ul class="downloadList">(?P<download_list>.*?)</ul>'
-        download_list_html = self._search_regex(DOWNLOAD_LIST_RE,
-            webpage, u'download list').strip()
-
-        # Get all of the links from the page
-        LINK_RE = r'(?s)<a href="(?P<url>[^"]+)">'
-        links = re.findall(LINK_RE, download_list_html)
-        
-        # Get link of hd video if available
-        mobj = re.search(r'var encryptedQuality720URL = \'(?P<encrypted_video_url>[a-zA-Z0-9+/]+={0,2})\';', webpage)
-        if mobj != None:
-            encrypted_video_url = mobj.group(u'encrypted_video_url')
-            video_url = aes_decrypt_text(encrypted_video_url, video_title, 32).decode('utf-8')
-            links = [video_url] + links
-        
-        if not links:
-            raise ExtractorError(u'ERROR: no known formats available for video')
-
-        self.to_screen(u'Links found: %d' % len(links))
+        video_id = mobj.group('id')
+        display_id = mobj.group('display_id')
+
+        request = sanitized_Request(url)
+        request.add_header('Cookie', 'age_verified=1')
+        webpage = self._download_webpage(request, display_id)
+
+        title = self._html_search_regex(
+            r'(?s)<div[^>]+class=["\']watchVideoTitle[^>]+>(.+?)</div>',
+            webpage, 'title', default=None) or self._og_search_title(
+            webpage, default=None) or self._html_search_meta(
+            'title', webpage, fatal=True)
+
+        links = []
+
+        # Main source
+        definitions = self._parse_json(
+            self._search_regex(
+                r'mediaDefinition\s*=\s*(\[.+?\]);', webpage,
+                'media definitions', default='[]'),
+            video_id, fatal=False)
+        if definitions:
+            for definition in definitions:
+                if not isinstance(definition, dict):
+                    continue
+                video_url = url_or_none(definition.get('videoUrl'))
+                if video_url:
+                    links.append(video_url)
+
+        # Fallback #1, this also contains extra low quality 180p format
+        for _, link in re.findall(r'<a[^>]+href=(["\'])(http.+?)\1[^>]+title=["\']Download [Vv]ideo', webpage):
+            links.append(link)
+
+        # Fallback #2 (unavailable as at 22.06.2017)
+        sources = self._search_regex(
+            r'(?s)sources\s*:\s*({.+?})', webpage, 'sources', default=None)
+        if sources:
+            for _, link in re.findall(r'[^:]+\s*:\s*(["\'])(http.+?)\1', sources):
+                links.append(link)
+
+        # Fallback #3 (unavailable as at 22.06.2017)
+        for _, link in re.findall(
+                r'(?:videoSrc|videoIpadUrl|html5PlayerSrc)\s*[:=]\s*(["\'])(http.+?)\1', webpage):
+            links.append(link)
+
+        # Fallback #4, encrypted links (unavailable as at 22.06.2017)
+        for _, encrypted_link in re.findall(
+                r'encryptedQuality\d{3,4}URL\s*=\s*(["\'])([\da-zA-Z+/=]+)\1', webpage):
+            links.append(aes_decrypt_text(encrypted_link, title, 32).decode('utf-8'))

 
         formats = []
 
         formats = []

-        for link in links:
-
-            # A link looks like this:
-            # http://cdn1.download.youporn.phncdn.com/201210/31/8004515/480p_370k_8004515/YouPorn%20-%20Nubile%20Films%20The%20Pillow%20Fight.mp4?nvb=20121113051249&nva=20121114051249&ir=1200&sr=1200&hash=014b882080310e95fb6a0
-            # A path looks like this:
-            # /201210/31/8004515/480p_370k_8004515/YouPorn%20-%20Nubile%20Films%20The%20Pillow%20Fight.mp4
-            video_url = unescapeHTML( link )
-            path = compat_urllib_parse_urlparse( video_url ).path
-            extension = os.path.splitext( path )[1][1:]
-            format = path.split('/')[4].split('_')[:2]
-            # size = format[0]
-            # bitrate = format[1]
-            format = "-".join( format )
-            # title = u'%s-%s-%s' % (video_title, size, bitrate)
-
-            formats.append({
-                'id': video_id,
+        for video_url in set(unescapeHTML(link) for link in links):
+            f = {

                 'url': video_url,
                 'url': video_url,

-                'uploader': video_uploader,
-                'upload_date': upload_date,
-                'title': video_title,
-                'ext': extension,
-                'format': format,
-                'thumbnail': thumbnail,
-                'description': video_description
-            })
-
-        if self._downloader.params.get('listformats', None):
-            self._print_formats(formats)
-            return
-
-        req_format = self._downloader.params.get('format', 'best')
-        self.to_screen(u'Format: %s' % req_format)
-
-        if req_format is None or req_format == 'best':
-            return [formats[0]]
-        elif req_format == 'worst':
-            return [formats[-1]]
-        elif req_format in ('-1', 'all'):
-            return formats
-        else:
-            format = self._specific( req_format, formats )
-            if format is None:
-                raise ExtractorError(u'Requested format not available')
-            return [format]
+            }
+            # Video URL's path looks like this:
+            #  /201012/17/505835/720p_1500k_505835/YouPorn%20-%20Sex%20Ed%20Is%20It%20Safe%20To%20Masturbate%20Daily.mp4
+            #  /201012/17/505835/vl_240p_240k_505835/YouPorn%20-%20Sex%20Ed%20Is%20It%20Safe%20To%20Masturbate%20Daily.mp4
+            # We will benefit from it by extracting some metadata
+            mobj = re.search(r'(?P<height>\d{3,4})[pP]_(?P<bitrate>\d+)[kK]_\d+/', video_url)
+            if mobj:
+                height = int(mobj.group('height'))
+                bitrate = int(mobj.group('bitrate'))
+                f.update({
+                    'format_id': '%dp-%dk' % (height, bitrate),
+                    'height': height,
+                    'tbr': bitrate,
+                })
+            formats.append(f)
+        self._sort_formats(formats)
+
+        description = self._html_search_regex(
+            r'(?s)<div[^>]+\bid=["\']description["\'][^>]*>(.+?)</div>',
+            webpage, 'description',
+            default=None) or self._og_search_description(
+            webpage, default=None)
+        thumbnail = self._search_regex(
+            r'(?:imageurl\s*=|poster\s*:)\s*(["\'])(?P<thumbnail>.+?)\1',
+            webpage, 'thumbnail', fatal=False, group='thumbnail')
+
+        uploader = self._html_search_regex(
+            r'(?s)<div[^>]+class=["\']submitByLink["\'][^>]*>(.+?)</div>',
+            webpage, 'uploader', fatal=False)
+        upload_date = unified_strdate(self._html_search_regex(
+            [r'Date\s+[Aa]dded:\s*<span>([^<]+)',
+             r'(?s)<div[^>]+class=["\']videoInfo(?:Date|Time)["\'][^>]*>(.+?)</div>'],
+            webpage, 'upload date', fatal=False))
+
+        age_limit = self._rta_search(webpage)
+
+        average_rating = int_or_none(self._search_regex(
+            r'<div[^>]+class=["\']videoRatingPercentage["\'][^>]*>(\d+)%</div>',
+            webpage, 'average rating', fatal=False))
+
+        view_count = str_to_int(self._search_regex(
+            r'(?s)<div[^>]+class=(["\']).*?\bvideoInfoViews\b.*?\1[^>]*>.*?(?P<count>[\d,.]+)<',
+            webpage, 'view count', fatal=False, group='count'))
+        comment_count = str_to_int(self._search_regex(
+            r'>All [Cc]omments? \(([\d,.]+)\)',
+            webpage, 'comment count', fatal=False))
+
+        def extract_tag_box(regex, title):
+            tag_box = self._search_regex(regex, webpage, title, default=None)
+            if not tag_box:
+                return []
+            return re.findall(r'<a[^>]+href=[^>]+>([^<]+)', tag_box)
+
+        categories = extract_tag_box(
+            r'(?s)Categories:.*?</[^>]+>(.+?)</div>', 'categories')
+        tags = extract_tag_box(
+            r'(?s)Tags:.*?</div>\s*<div[^>]+class=["\']tagBoxContent["\'][^>]*>(.+?)</div>',
+            'tags')
+
+        return {
+            'id': video_id,
+            'display_id': display_id,
+            'title': title,
+            'description': description,
+            'thumbnail': thumbnail,
+            'uploader': uploader,
+            'upload_date': upload_date,
+            'average_rating': average_rating,
+            'view_count': view_count,
+            'comment_count': comment_count,
+            'categories': categories,
+            'tags': tags,
+            'age_limit': age_limit,
+            'formats': formats,
+        }