New upstream version 2020.03.24

[youtubedl] / youtube_dl / extractor / howstuffworks.py

diff --git a/youtube_dl/extractor/howstuffworks.py b/youtube_dl/extractor/howstuffworks.py
index fccc238840887fd70ed56b2f642c47ea6aa4f43e..cf90ab3c9d976faabed8891fa4c1211853732f24 100644 (file)
--- a/youtube_dl/extractor/howstuffworks.py
+++ b/youtube_dl/extractor/howstuffworks.py
@@ -1,113 +1,76 @@
 from __future__ import unicode_literals
 
 from __future__ import unicode_literals
 

-import re
-import json
-import random
-import string
-

 from .common import InfoExtractor
 from .common import InfoExtractor

-from ..utils import find_xpath_attr
+from ..utils import (
+    find_xpath_attr,
+    int_or_none,
+    js_to_json,
+    unescapeHTML,
+    determine_ext,
+)

 
 
 class HowStuffWorksIE(InfoExtractor):
 
 
 class HowStuffWorksIE(InfoExtractor):

-    _VALID_URL = r'https?://[\da-z-]+\.howstuffworks\.com/(?:[^/]+/)*\d+-(?P<id>.+?)-video\.htm'
+    _VALID_URL = r'https?://[\da-z-]+\.(?:howstuffworks|stuff(?:(?:youshould|theydontwantyouto)know|toblowyourmind|momnevertoldyou)|(?:brain|car)stuffshow|fwthinking|geniusstuff)\.com/(?:[^/]+/)*(?:\d+-)?(?P<id>.+?)-video\.htm'

     _TESTS = [
         {
     _TESTS = [
         {

-            'url': 'http://adventure.howstuffworks.com/5266-cool-jobs-iditarod-musher-video.htm',
-            'info_dict': {
-                'id': '450221',
-                'display_id': 'cool-jobs-iditarod-musher',
-                'ext': 'flv',
-                'title': 'Cool Jobs - Iditarod Musher',
-                'description': 'md5:82bb58438a88027b8186a1fccb365f90',
-                'thumbnail': 're:^https?://.*\.jpg$',
-            },
-            'params': {
-                # md5 is not consistent
-                'skip_download': True
-            }
-        },
-        {
-            'url': 'http://adventure.howstuffworks.com/7199-survival-zone-food-and-water-in-the-savanna-video.htm',
+            'url': 'http://www.stufftoblowyourmind.com/videos/optical-illusions-video.htm',
+            'md5': '76646a5acc0c92bf7cd66751ca5db94d',

             'info_dict': {
             'info_dict': {

-                'id': '453464',
-                'display_id': 'survival-zone-food-and-water-in-the-savanna',
+                'id': '855410',

                 'ext': 'mp4',
                 'ext': 'mp4',

-                'title': 'Survival Zone: Food and Water In the Savanna',
-                'description': 'md5:7e1c89f6411434970c15fa094170c371',
-                'thumbnail': 're:^https?://.*\.jpg$',
+                'title': 'Your Trickster Brain: Optical Illusions -- Science on the Web',
+                'description': 'md5:e374ff9561f6833ad076a8cc0a5ab2fb',

             },
             },

-            'params': {
-                # md5 is not consistent
-                'skip_download': True
-            }

         },
         {
         },
         {

-            'url': 'http://entertainment.howstuffworks.com/arts/2706-sword-swallowing-1-by-dan-meyer-video.htm',
-            'info_dict': {
-                'id': '440011',
-                'display_id': 'sword-swallowing-1-by-dan-meyer',
-                'ext': 'flv',
-                'title': 'Sword Swallowing #1 by Dan Meyer',
-                'description': 'md5:b2409e88172913e2e7d3d1159b0ef735',
-                'thumbnail': 're:^https?://.*\.jpg$',
-            },
-            'params': {
-                # md5 is not consistent
-                'skip_download': True
-            }
-        },
+            'url': 'http://shows.howstuffworks.com/more-shows/why-does-balloon-stick-to-hair-video.htm',
+            'only_matching': True,
+        }

     ]
 
     def _real_extract(self, url):
     ]
 
     def _real_extract(self, url):

-        mobj = re.match(self._VALID_URL, url)
-        display_id = mobj.group('id')
+        display_id = self._match_id(url)

         webpage = self._download_webpage(url, display_id)
         webpage = self._download_webpage(url, display_id)

+        clip_js = self._search_regex(
+            r'(?s)var clip = ({.*?});', webpage, 'clip info')
+        clip_info = self._parse_json(
+            clip_js, display_id, transform_source=js_to_json)

 
 

-        content_id = self._search_regex(r'var siteSectionId="(\d+)";', webpage, 'content id')
-
-        mp4 = self._search_regex(
-            r'''(?xs)var\s+clip\s*=\s*{\s*
-                .+?\s*
-                content_id\s*:\s*%s\s*,\s*
-                .+?\s*
-                mp4\s*:\s*\[(.*?),?\]\s*
-                };\s*
-                videoData\.push\(clip\);''' % content_id,
-            webpage, 'mp4', fatal=False, default=None)
-
-        smil = self._download_xml(
-            'http://services.media.howstuffworks.com/videos/%s/smil-service.smil' % content_id,
-            content_id, 'Downloading video SMIL')
-
-        http_base = find_xpath_attr(
-            smil,
-            './{0}head/{0}meta'.format('{http://www.w3.org/2001/SMIL20/Language}'),
-            'name',
-            'httpBase').get('content')
-
-        def random_string(str_len=0):
-            return ''.join([random.choice(string.ascii_uppercase) for _ in range(str_len)])
-
-        URL_SUFFIX = '?v=2.11.3&fp=LNX 11,2,202,356&r=%s&g=%s' % (random_string(5), random_string(12))
-
+        video_id = clip_info['content_id']

         formats = []
         formats = []

+        m3u8_url = clip_info.get('m3u8')
+        if m3u8_url and determine_ext(m3u8_url) == 'm3u8':
+            formats.extend(self._extract_m3u8_formats(m3u8_url, video_id, 'mp4', format_id='hls', fatal=True))
+        flv_url = clip_info.get('flv_url')
+        if flv_url:
+            formats.append({
+                'url': flv_url,
+                'format_id': 'flv',
+            })
+        for video in clip_info.get('mp4', []):
+            formats.append({
+                'url': video['src'],
+                'format_id': 'mp4-%s' % video['bitrate'],
+                'vbr': int_or_none(video['bitrate'].rstrip('k')),
+            })
+
+        if not formats:
+            smil = self._download_xml(
+                'http://services.media.howstuffworks.com/videos/%s/smil-service.smil' % video_id,
+                video_id, 'Downloading video SMIL')
+
+            http_base = find_xpath_attr(
+                smil,
+                './{0}head/{0}meta'.format('{http://www.w3.org/2001/SMIL20/Language}'),
+                'name',
+                'httpBase').get('content')
+
+            URL_SUFFIX = '?v=2.11.3&fp=LNX 11,2,202,356&r=A&g=A'

 
 

-        if mp4:
-            for video in json.loads('[%s]' % mp4):
-                bitrate = video['bitrate']
-                fmt = {
-                    'url': video['src'].replace('http://pmd.video.howstuffworks.com', http_base) + URL_SUFFIX,
-                    'format_id': bitrate,
-                }
-                m = re.search(r'(?P<vbr>\d+)[Kk]', bitrate)
-                if m:
-                    fmt['vbr'] = int(m.group('vbr'))
-                formats.append(fmt)
-        else:

             for video in smil.findall(
             for video in smil.findall(

-                    './/{0}body/{0}switch/{0}video'.format('{http://www.w3.org/2001/SMIL20/Language}')):
-                vbr = int(video.attrib['system-bitrate']) / 1000
+                    './{0}body/{0}switch/{0}video'.format('{http://www.w3.org/2001/SMIL20/Language}')):
+                vbr = int_or_none(video.attrib['system-bitrate'], scale=1000)

                 formats.append({
                     'url': '%s/%s%s' % (http_base, video.attrib['src'], URL_SUFFIX),
                     'format_id': '%dk' % vbr,
                 formats.append({
                     'url': '%s/%s%s' % (http_base, video.attrib['src'], URL_SUFFIX),
                     'format_id': '%dk' % vbr,


@@ -116,19 +79,12 @@ class HowStuffWorksIE(InfoExtractor):
 
         self._sort_formats(formats)
 
 
         self._sort_formats(formats)
 

-        title = self._og_search_title(webpage)
-        TITLE_SUFFIX = ' : HowStuffWorks'
-        if title.endswith(TITLE_SUFFIX):
-            title = title[:-len(TITLE_SUFFIX)]
-
-        description = self._og_search_description(webpage)
-        thumbnail = self._og_search_thumbnail(webpage)
-

         return {
         return {

-            'id': content_id,
+            'id': '%s' % video_id,

             'display_id': display_id,
             'display_id': display_id,

-            'title': title,
-            'description': description,
-            'thumbnail': thumbnail,
+            'title': unescapeHTML(clip_info['clip_title']),
+            'description': unescapeHTML(clip_info.get('caption')),
+            'thumbnail': clip_info.get('video_still_url'),
+            'duration': int_or_none(clip_info.get('duration')),

             'formats': formats,
         }
             'formats': formats,
         }