]> Raphaƫl G. Git Repositories - youtubedl/blobdiff - youtube_dl/extractor/pornhd.py
Merge tag 'upstream/2016.06.25'
[youtubedl] / youtube_dl / extractor / pornhd.py
index bac484c67dbb01bbafa319c117b7c6d152b7dd5d..8df12eec0d44c371d99b536b55694cfd2211f9d0 100644 (file)
@@ -1,20 +1,32 @@
 from __future__ import unicode_literals
 
 import re
-import json
 
 from .common import InfoExtractor
 from ..utils import (
+    ExtractorError,
     int_or_none,
     js_to_json,
-    qualities,
-    determine_ext,
 )
 
 
 class PornHdIE(InfoExtractor):
-    _VALID_URL = r'http://(?:www\.)?pornhd\.com/(?:[a-z]{2,4}/)?videos/(?P<id>\d+)(?:/(?P<display_id>.+))?'
-    _TEST = {
+    _VALID_URL = r'https?://(?:www\.)?pornhd\.com/(?:[a-z]{2,4}/)?videos/(?P<id>\d+)(?:/(?P<display_id>.+))?'
+    _TESTS = [{
+        'url': 'http://www.pornhd.com/videos/9864/selfie-restroom-masturbation-fun-with-chubby-cutie-hd-porn-video',
+        'md5': 'c8b964b1f0a4b5f7f28ae3a5c9f86ad5',
+        'info_dict': {
+            'id': '9864',
+            'display_id': 'selfie-restroom-masturbation-fun-with-chubby-cutie-hd-porn-video',
+            'ext': 'mp4',
+            'title': 'Restroom selfie masturbation',
+            'description': 'md5:3748420395e03e31ac96857a8f125b2b',
+            'thumbnail': 're:^https?://.*\.jpg',
+            'view_count': int,
+            'age_limit': 18,
+        }
+    }, {
+        # removed video
         'url': 'http://www.pornhd.com/videos/1962/sierra-day-gets-his-cum-all-over-herself-hd-porn-video',
         'md5': '956b8ca569f7f4d8ec563e2c41598441',
         'info_dict': {
@@ -26,8 +38,9 @@ class PornHdIE(InfoExtractor):
             'thumbnail': 're:^https?://.*\.jpg',
             'view_count': int,
             'age_limit': 18,
-        }
-    }
+        },
+        'skip': 'Not available anymore',
+    }]
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
@@ -37,23 +50,40 @@ class PornHdIE(InfoExtractor):
         webpage = self._download_webpage(url, display_id or video_id)
 
         title = self._html_search_regex(
-            r'<title>(.+) porn HD.+?</title>', webpage, 'title')
+            [r'<span[^>]+class=["\']video-name["\'][^>]*>([^<]+)',
+             r'<title>(.+?) - .*?[Pp]ornHD.*?</title>'], webpage, 'title')
+
+        sources = self._parse_json(js_to_json(self._search_regex(
+            r"(?s)'sources'\s*:\s*(\{.+?\})\s*\}[;,)]",
+            webpage, 'sources', default='{}')), video_id)
+
+        if not sources:
+            message = self._html_search_regex(
+                r'(?s)<(div|p)[^>]+class="no-video"[^>]*>(?P<value>.+?)</\1',
+                webpage, 'error message', group='value')
+            raise ExtractorError('%s said: %s' % (self.IE_NAME, message), expected=True)
+
+        formats = []
+        for format_id, video_url in sources.items():
+            if not video_url:
+                continue
+            height = int_or_none(self._search_regex(
+                r'^(\d+)[pP]', format_id, 'height', default=None))
+            formats.append({
+                'url': video_url,
+                'format_id': format_id,
+                'height': height,
+            })
+        self._sort_formats(formats)
+
         description = self._html_search_regex(
-            r'<div class="description">([^<]+)</div>', webpage, 'description', fatal=False)
+            r'<(div|p)[^>]+class="description"[^>]*>(?P<value>[^<]+)</\1',
+            webpage, 'description', fatal=False, group='value')
         view_count = int_or_none(self._html_search_regex(
-            r'(\d+) views\s*</span>', webpage, 'view count', fatal=False))
+            r'(\d+) views\s*<', webpage, 'view count', fatal=False))
         thumbnail = self._search_regex(
             r"'poster'\s*:\s*'([^']+)'", webpage, 'thumbnail', fatal=False)
 
-        quality = qualities(['SD', 'HD'])
-        formats = [{
-            'url': source['file'],
-            'format_id': '%s-%s' % (source['label'], determine_ext(source['file'])),
-            'quality': quality(source['label']),
-        } for source in json.loads(js_to_json(self._search_regex(
-            r"(?s)'sources'\s*:\s*(\[.+?\])", webpage, 'sources')))]
-        self._sort_formats(formats)
-
         return {
             'id': video_id,
             'display_id': display_id,