debian/changelog: Annotate the log with bugs to close.

[youtubedl] / youtube_dl / extractor / pornhd.py

diff --git a/youtube_dl/extractor/pornhd.py b/youtube_dl/extractor/pornhd.py
index 718fe9aba5fc710ee5efc47c2bbae2b02fc3c117..b52879c7a444fef98b2354660025958a7c5ce180 100644 (file)
--- a/youtube_dl/extractor/pornhd.py
+++ b/youtube_dl/extractor/pornhd.py
@@ -1,77 +1,93 @@
 from __future__ import unicode_literals
 
 import re
 from __future__ import unicode_literals
 
 import re

-import json

 
 from .common import InfoExtractor
 
 from .common import InfoExtractor

-from ..utils import int_or_none
+from ..utils import (
+    ExtractorError,
+    int_or_none,
+    js_to_json,
+)

 
 
 class PornHdIE(InfoExtractor):
 
 
 class PornHdIE(InfoExtractor):

-    _VALID_URL = r'http://(?:www\.)?pornhd\.com/(?:[a-z]{2,4}/)?videos/(?P<id>\d+)'
-    _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': r'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': {
             'id': '1962',
         'url': 'http://www.pornhd.com/videos/1962/sierra-day-gets-his-cum-all-over-herself-hd-porn-video',
         'md5': '956b8ca569f7f4d8ec563e2c41598441',
         'info_dict': {
             'id': '1962',

+            'display_id': 'sierra-day-gets-his-cum-all-over-herself-hd-porn-video',

             'ext': 'mp4',
             'title': 'Sierra loves doing laundry',
             'description': 'md5:8ff0523848ac2b8f9b065ba781ccf294',
             'ext': 'mp4',
             'title': 'Sierra loves doing laundry',
             'description': 'md5:8ff0523848ac2b8f9b065ba781ccf294',

+            'thumbnail': r're:^https?://.*\.jpg',
+            'view_count': int,

             'age_limit': 18,
             'age_limit': 18,

-        }
-    }
+        },
+        'skip': 'Not available anymore',
+    }]

 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
         video_id = mobj.group('id')
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
         video_id = mobj.group('id')

+        display_id = mobj.group('display_id')

 
 

-        webpage = self._download_webpage(url, video_id)
+        webpage = self._download_webpage(url, display_id or video_id)

 
 

-        title = self._og_search_title(webpage)
-        TITLE_SUFFIX = ' porn HD Video | PornHD.com '
-        if title.endswith(TITLE_SUFFIX):
-            title = title[:-len(TITLE_SUFFIX)]
+        title = self._html_search_regex(
+            [r'<span[^>]+class=["\']video-name["\'][^>]*>([^<]+)',
+             r'<title>(.+?) - .*?[Pp]ornHD.*?</title>'], webpage, 'title')

 
 

-        description = self._html_search_regex(
-            r'<div class="description">([^<]+)</div>', webpage, 'description', fatal=False)
-        view_count = int_or_none(self._html_search_regex(
-            r'(\d+) views      </span>', webpage, 'view count', fatal=False))
-
-        formats = [
-            {
-                'url': format_url,
-                'ext': format.lower(),
-                'format_id': '%s-%s' % (format.lower(), quality.lower()),
-                'quality': 1 if quality.lower() == 'high' else 0,
-            } for format, quality, format_url in re.findall(
-                r'var __video([\da-zA-Z]+?)(Low|High)StreamUrl = \'(http://.+?)\?noProxy=1\'', webpage)
-        ]
+        sources = self._parse_json(js_to_json(self._search_regex(
+            r"(?s)sources'?\s*[:=]\s*(\{.+?\})",
+            webpage, 'sources', default='{}')), video_id)

 
 

-        mobj = re.search(r'flashVars = (?P<flashvars>{.+?});', webpage)
-        if mobj:
-            flashvars = json.loads(mobj.group('flashvars'))
-            formats.extend([
-                {
-                    'url': flashvars['hashlink'].replace('?noProxy=1', ''),
-                    'ext': 'flv',
-                    'format_id': 'flv-low',
-                    'quality': 0,
-                },
-                {
-                    'url': flashvars['hd'].replace('?noProxy=1', ''),
-                    'ext': 'flv',
-                    'format_id': 'flv-high',
-                    'quality': 1,
-                }
-            ])
-            thumbnail = flashvars['urlWallpaper']
-        else:
-            thumbnail = self._og_search_thumbnail(webpage)
+        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)
 
         self._sort_formats(formats)
 

+        description = self._html_search_regex(
+            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*<', webpage, 'view count', fatal=False))
+        thumbnail = self._search_regex(
+            r"poster'?\s*:\s*([\"'])(?P<url>(?:(?!\1).)+)\1", webpage,
+            'thumbnail', fatal=False, group='url')
+

         return {
             'id': video_id,
         return {
             'id': video_id,

+            'display_id': display_id,

             'title': title,
             'description': description,
             'thumbnail': thumbnail,
             'title': title,
             'description': description,
             'thumbnail': thumbnail,