[youtube] Fix extraction.

[youtube-dl] / youtube_dl / extractor / eporner.py

diff --git a/youtube_dl/extractor/eporner.py b/youtube_dl/extractor/eporner.py
index 3071da6dbb9c3a4289102dd470785f45cd385d4e..fe42821c731c711e8f0974fd4ce48f5c9aee8e8f 100644 (file)
--- a/youtube_dl/extractor/eporner.py
+++ b/youtube_dl/extractor/eporner.py
@@ -4,42 +4,126 @@ from __future__ import unicode_literals
 import re
 
 from .common import InfoExtractor
-from ..utils import int_or_none
+from ..utils import (
+    encode_base_n,
+    ExtractorError,
+    int_or_none,
+    merge_dicts,
+    parse_duration,
+    str_to_int,
+    url_or_none,
+)
+
 
 class EpornerIE(InfoExtractor):
-    _VALID_URL = r'http?://(?:www\.)?eporner\.com/hd-porn/(?P<id>\d+)/(?P<title_dash>[\w-]+)/?'
-    _TEST = {
+    _VALID_URL = r'https?://(?:www\.)?eporner\.com/(?:hd-porn|embed)/(?P<id>\w+)(?:/(?P<display_id>[\w-]+))?'
+    _TESTS = [{
         'url': 'http://www.eporner.com/hd-porn/95008/Infamous-Tiffany-Teen-Strip-Tease-Video/',
-        'md5': '3b427ae4b9d60619106de3185c2987cd',
+        'md5': '39d486f046212d8e1b911c52ab4691f8',
         'info_dict': {
-            'id': '95008',
-            'ext': 'flv',
+            'id': 'qlDUmNsj6VS',
+            'display_id': 'Infamous-Tiffany-Teen-Strip-Tease-Video',
+            'ext': 'mp4',
             'title': 'Infamous Tiffany Teen Strip Tease Video',
-            'duration': 194
+            'description': 'md5:764f39abf932daafa37485eb46efa152',
+            'timestamp': 1232520922,
+            'upload_date': '20090121',
+            'duration': 1838,
+            'view_count': int,
+            'age_limit': 18,
+        },
+        'params': {
+            'proxy': '127.0.0.1:8118'
         }
-    }
+    }, {
+        # New (May 2016) URL layout
+        'url': 'http://www.eporner.com/hd-porn/3YRUtzMcWn0/Star-Wars-XXX-Parody/',
+        'only_matching': True,
+    }, {
+        'url': 'http://www.eporner.com/hd-porn/3YRUtzMcWn0',
+        'only_matching': True,
+    }, {
+        'url': 'http://www.eporner.com/hd-porn/3YRUtzMcWn0',
+        'only_matching': True,
+    }]
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
         video_id = mobj.group('id')
-        webpage = self._download_webpage(url, video_id)
-        title = self._html_search_regex(r'<title>(.*?) - EPORNER', webpage, 'title')
+        display_id = mobj.group('display_id') or video_id
+
+        webpage, urlh = self._download_webpage_handle(url, display_id)
+
+        video_id = self._match_id(urlh.geturl())
+
+        hash = self._search_regex(
+            r'hash\s*:\s*["\']([\da-f]{32})', webpage, 'hash')
+
+        title = self._og_search_title(webpage, default=None) or self._html_search_regex(
+            r'<title>(.+?) - EPORNER', webpage, 'title')
+
+        # Reverse engineered from vjs.js
+        def calc_hash(s):
+            return ''.join((encode_base_n(int(s[lb:lb + 8], 16), 36) for lb in range(0, 32, 8)))
 
-        redirect_code = self._html_search_regex(r'<script type="text/javascript" src="/config5/'+str(video_id)+'/([a-f\d]+)/">', webpage, 'redirect_code')
-        redirect_url = 'http://www.eporner.com/config5/' + str(video_id) +'/'+ redirect_code
-        webpage2 = self._download_webpage(redirect_url, video_id)
-        video_url = self._html_search_regex(r'file: "(.*?)",', webpage2, 'video_url')
+        video = self._download_json(
+            'http://www.eporner.com/xhr/video/%s' % video_id,
+            display_id, note='Downloading video JSON',
+            query={
+                'hash': calc_hash(hash),
+                'device': 'generic',
+                'domain': 'www.eporner.com',
+                'fallback': 'false',
+            })
 
-        mobj = re.search(r'class="mbtim">(?P<minutes>\d+):(?P<seconds>\d+)</div>', webpage)
-        duration = int(mobj.group('minutes')) * 60 + int(mobj.group('seconds')) if mobj else None
+        if video.get('available') is False:
+            raise ExtractorError(
+                '%s said: %s' % (self.IE_NAME, video['message']), expected=True)
 
-        mobj = re.search(r'id="cinemaviews">((?P<thousands>\d+),)?(?P<units>\d+)<small>views', webpage)
-        view_count = int(mobj.group('thousands')) * 1000 + int(mobj.group('units')) if mobj else None
+        sources = video['sources']
 
-        return {
+        formats = []
+        for kind, formats_dict in sources.items():
+            if not isinstance(formats_dict, dict):
+                continue
+            for format_id, format_dict in formats_dict.items():
+                if not isinstance(format_dict, dict):
+                    continue
+                src = url_or_none(format_dict.get('src'))
+                if not src or not src.startswith('http'):
+                    continue
+                if kind == 'hls':
+                    formats.extend(self._extract_m3u8_formats(
+                        src, display_id, 'mp4', entry_protocol='m3u8_native',
+                        m3u8_id=kind, fatal=False))
+                else:
+                    height = int_or_none(self._search_regex(
+                        r'(\d+)[pP]', format_id, 'height', default=None))
+                    fps = int_or_none(self._search_regex(
+                        r'(\d+)fps', format_id, 'fps', default=None))
+
+                    formats.append({
+                        'url': src,
+                        'format_id': format_id,
+                        'height': height,
+                        'fps': fps,
+                    })
+        self._sort_formats(formats)
+
+        json_ld = self._search_json_ld(webpage, display_id, default={})
+
+        duration = parse_duration(self._html_search_meta(
+            'duration', webpage, default=None))
+        view_count = str_to_int(self._search_regex(
+            r'id="cinemaviews">\s*([0-9,]+)\s*<small>views',
+            webpage, 'view count', fatal=False))
+
+        return merge_dicts(json_ld, {
             'id': video_id,
-            'url': video_url,
+            'display_id': display_id,
             'title': title,
-            'duration': int_or_none(duration),
-            'view_count': int_or_none(view_count),
-        }
+            'duration': duration,
+            'view_count': view_count,
+            'formats': formats,
+            'age_limit': 18,
+        })