Merge tag 'upstream/2014.01.17.2'

[youtubedl] / youtube_dl / extractor / appletrailers.py

diff --git a/youtube_dl/extractor/appletrailers.py b/youtube_dl/extractor/appletrailers.py
index 8b191c19636087d89fe8505964292bcc6db3ba2b..922cede056690bac963cdb2f896eb7b9254680af 100644 (file)
--- a/youtube_dl/extractor/appletrailers.py
+++ b/youtube_dl/extractor/appletrailers.py
@@ -1,147 +1,121 @@
+from __future__ import unicode_literals
+

 import re
 import re

-import xml.etree.ElementTree
+import json

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

+    compat_urlparse,

     determine_ext,
 )
 
 
 class AppleTrailersIE(InfoExtractor):
     determine_ext,
 )
 
 
 class AppleTrailersIE(InfoExtractor):

-    _VALID_URL = r'https?://(?:www\.)?trailers.apple.com/trailers/(?P<company>[^/]+)/(?P<movie>[^/]+)'
+    _VALID_URL = r'https?://(?:www\.)?trailers\.apple\.com/trailers/(?P<company>[^/]+)/(?P<movie>[^/]+)'

     _TEST = {
     _TEST = {

-        u"url": u"http://trailers.apple.com/trailers/wb/manofsteel/",
-        u"playlist": [
+        "url": "http://trailers.apple.com/trailers/wb/manofsteel/",
+        "playlist": [

             {
             {

-                u"file": u"manofsteel-trailer4.mov",
-                u"md5": u"11874af099d480cc09e103b189805d5f",
-                u"info_dict": {
-                    u"duration": 111,
-                    u"thumbnail": u"http://trailers.apple.com/trailers/wb/manofsteel/images/thumbnail_11624.jpg",
-                    u"title": u"Trailer 4",
-                    u"upload_date": u"20130523",
-                    u"uploader_id": u"wb",
+                "file": "manofsteel-trailer4.mov",
+                "md5": "d97a8e575432dbcb81b7c3acb741f8a8",
+                "info_dict": {
+                    "duration": 111,
+                    "title": "Trailer 4",
+                    "upload_date": "20130523",
+                    "uploader_id": "wb",

                 },
             },
             {
                 },
             },
             {

-                u"file": u"manofsteel-trailer3.mov",
-                u"md5": u"07a0a262aae5afe68120eed61137ab34",
-                u"info_dict": {
-                    u"duration": 182,
-                    u"thumbnail": u"http://trailers.apple.com/trailers/wb/manofsteel/images/thumbnail_10793.jpg",
-                    u"title": u"Trailer 3",
-                    u"upload_date": u"20130417",
-                    u"uploader_id": u"wb",
+                "file": "manofsteel-trailer3.mov",
+                "md5": "b8017b7131b721fb4e8d6f49e1df908c",
+                "info_dict": {
+                    "duration": 182,
+                    "title": "Trailer 3",
+                    "upload_date": "20130417",
+                    "uploader_id": "wb",

                 },
             },
             {
                 },
             },
             {

-                u"file": u"manofsteel-trailer.mov",
-                u"md5": u"e401fde0813008e3307e54b6f384cff1",
-                u"info_dict": {
-                    u"duration": 148,
-                    u"thumbnail": u"http://trailers.apple.com/trailers/wb/manofsteel/images/thumbnail_8703.jpg",
-                    u"title": u"Trailer",
-                    u"upload_date": u"20121212",
-                    u"uploader_id": u"wb",
+                "file": "manofsteel-trailer.mov",
+                "md5": "d0f1e1150989b9924679b441f3404d48",
+                "info_dict": {
+                    "duration": 148,
+                    "title": "Trailer",
+                    "upload_date": "20121212",
+                    "uploader_id": "wb",

                 },
             },
             {
                 },
             },
             {

-                u"file": u"manofsteel-teaser.mov",
-                u"md5": u"76b392f2ae9e7c98b22913c10a639c97",
-                u"info_dict": {
-                    u"duration": 93,
-                    u"thumbnail": u"http://trailers.apple.com/trailers/wb/manofsteel/images/thumbnail_6899.jpg",
-                    u"title": u"Teaser",
-                    u"upload_date": u"20120721",
-                    u"uploader_id": u"wb",
+                "file": "manofsteel-teaser.mov",
+                "md5": "5fe08795b943eb2e757fa95cb6def1cb",
+                "info_dict": {
+                    "duration": 93,
+                    "title": "Teaser",
+                    "upload_date": "20120721",
+                    "uploader_id": "wb",

                 },
             }
         ]
     }
 
                 },
             }
         ]
     }
 

+    _JSON_RE = r'iTunes.playURL\((.*?)\);'
+

     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
         movie = mobj.group('movie')
         uploader_id = mobj.group('company')
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
         movie = mobj.group('movie')
         uploader_id = mobj.group('company')
 

-        playlist_url = url.partition(u'?')[0] + u'/includes/playlists/web.inc'
-        playlist_snippet = self._download_webpage(playlist_url, movie)
-        playlist_cleaned = re.sub(r'(?s)<script>.*?</script>', u'', playlist_snippet)
-        playlist_html = u'<html>' + playlist_cleaned + u'</html>'
+        playlist_url = compat_urlparse.urljoin(url, u'includes/playlists/itunes.inc')
+        def fix_html(s):
+            s = re.sub(r'(?s)<script[^<]*?>.*?</script>', u'', s)
+            s = re.sub(r'<img ([^<]*?)>', r'<img \1/>', s)
+            # The ' in the onClick attributes are not escaped, it couldn't be parsed
+            # like: http://trailers.apple.com/trailers/wb/gravity/
+            def _clean_json(m):
+                return u'iTunes.playURL(%s);' % m.group(1).replace('\'', '&#39;')
+            s = re.sub(self._JSON_RE, _clean_json, s)
+            s = u'<html>' + s + u'</html>'
+            return s
+        doc = self._download_xml(playlist_url, movie, transform_source=fix_html)

 
 

-        size_cache = {}
-
-        doc = xml.etree.ElementTree.fromstring(playlist_html)

         playlist = []
         for li in doc.findall('./div/ul/li'):
         playlist = []
         for li in doc.findall('./div/ul/li'):

-            title = li.find('.//h3').text
+            on_click = li.find('.//a').attrib['onClick']
+            trailer_info_json = self._search_regex(self._JSON_RE,
+                on_click, u'trailer info')
+            trailer_info = json.loads(trailer_info_json)
+            title = trailer_info['title']

             video_id = movie + '-' + re.sub(r'[^a-zA-Z0-9]', '', title).lower()
             thumbnail = li.find('.//img').attrib['src']
             video_id = movie + '-' + re.sub(r'[^a-zA-Z0-9]', '', title).lower()
             thumbnail = li.find('.//img').attrib['src']

+            upload_date = trailer_info['posted'].replace('-', '')

 
 

-            date_el = li.find('.//p')
-            upload_date = None
-            m = re.search(r':\s?(?P<month>[0-9]{2})/(?P<day>[0-9]{2})/(?P<year>[0-9]{2})', date_el.text)
-            if m:
-                upload_date = u'20' + m.group('year') + m.group('month') + m.group('day')
-            runtime_el = date_el.find('./br')
-            m = re.search(r':\s?(?P<minutes>[0-9]+):(?P<seconds>[0-9]{1,2})', runtime_el.tail)
+            runtime = trailer_info['runtime']
+            m = re.search(r'(?P<minutes>[0-9]+):(?P<seconds>[0-9]{1,2})', runtime)

             duration = None
             if m:
                 duration = 60 * int(m.group('minutes')) + int(m.group('seconds'))
 
             duration = None
             if m:
                 duration = 60 * int(m.group('minutes')) + int(m.group('seconds'))
 

-            formats = []
-            for formats_el in li.findall('.//a'):
-                if formats_el.attrib['class'] != 'OverlayPanel':
-                    continue
-                target = formats_el.attrib['target']
-
-                format_code = formats_el.text
-                if 'Automatic' in format_code:
-                    continue
-
-                size_q = formats_el.attrib['href']
-                size_id = size_q.rpartition('#videos-')[2]
-                if size_id not in size_cache:
-                    size_url = url + size_q
-                    sizepage_html = self._download_webpage(
-                        size_url, movie,
-                        note=u'Downloading size info %s' % size_id,
-                        errnote=u'Error while downloading size info %s' % size_id,
-                    )
-                    _doc = xml.etree.ElementTree.fromstring(sizepage_html)
-                    size_cache[size_id] = _doc
+            first_url = trailer_info['url']
+            trailer_id = first_url.split('/')[-1].rpartition('_')[0].lower()
+            settings_json_url = compat_urlparse.urljoin(url, 'includes/settings/%s.json' % trailer_id)
+            settings_json = self._download_webpage(settings_json_url, trailer_id, u'Downloading settings json')
+            settings = json.loads(settings_json)

 
 

-                sizepage_doc = size_cache[size_id]
-                links = sizepage_doc.findall('.//{http://www.w3.org/1999/xhtml}ul/{http://www.w3.org/1999/xhtml}li/{http://www.w3.org/1999/xhtml}a')
-                for vid_a in links:
-                    href = vid_a.get('href')
-                    if not href.endswith(target):
-                        continue
-                    detail_q = href.partition('#')[0]
-                    detail_url = url + '/' + detail_q
-
-                    m = re.match(r'includes/(?P<detail_id>[^/]+)/', detail_q)
-                    detail_id = m.group('detail_id')
-
-                    detail_html = self._download_webpage(
-                        detail_url, movie,
-                        note=u'Downloading detail %s %s' % (detail_id, size_id),
-                        errnote=u'Error while downloading detail %s %s' % (detail_id, size_id)
-                    )
-                    detail_doc = xml.etree.ElementTree.fromstring(detail_html)
-                    movie_link_el = detail_doc.find('.//{http://www.w3.org/1999/xhtml}a')
-                    assert movie_link_el.get('class') == 'movieLink'
-                    movie_link = movie_link_el.get('href').partition('?')[0].replace('_', '_h')
-                    ext = determine_ext(movie_link)
-                    assert ext == 'mov'
-
-                    formats.append({
-                        'format': format_code,
-                        'ext': ext,
-                        'url': movie_link,
-                    })
-
-            info = {
+            formats = []
+            for format in settings['metadata']['sizes']:
+                # The src is a file pointing to the real video file
+                format_url = re.sub(r'_(\d*p.mov)', r'_h\1', format['src'])
+                formats.append({
+                    'url': format_url,
+                    'ext': determine_ext(format_url),
+                    'format': format['type'],
+                    'width': format['width'],
+                    'height': int(format['height']),
+                })
+
+            self._sort_formats(formats)
+
+            playlist.append({

                 '_type': 'video',
                 'id': video_id,
                 'title': title,
                 '_type': 'video',
                 'id': video_id,
                 'title': title,


@@ -152,12 +126,7 @@ class AppleTrailersIE(InfoExtractor):
                 'upload_date': upload_date,
                 'uploader_id': uploader_id,
                 'user_agent': 'QuickTime compatible (youtube-dl)',
                 'upload_date': upload_date,
                 'uploader_id': uploader_id,
                 'user_agent': 'QuickTime compatible (youtube-dl)',

-            }
-            # TODO: Remove when #980 has been merged
-            info['url'] = formats[-1]['url']
-            info['ext'] = formats[-1]['ext']
-
-            playlist.append(info)
+            })

 
         return {
             '_type': 'playlist',
 
         return {
             '_type': 'playlist',