X-Git-Url: http://git.bitcoin.ninja/index.cgi?a=blobdiff_plain;f=youtube_dl%2Fextractor%2Fwat.py;h=a584e08966ac57354c51a71d7fee520d7ce67df8;hb=27ace98f5159efdf6942e69ec73c5840cce26fbf;hp=aba4159e9b856d627bc5fe4cff1f45a8259ca432;hpb=705f6f35bc13dd8d2ad3df9b1d22f6cad362afcc;p=youtube-dl

diff --git a/youtube_dl/extractor/wat.py b/youtube_dl/extractor/wat.py
index aba4159e9..a584e0896 100644
--- a/youtube_dl/extractor/wat.py
+++ b/youtube_dl/extractor/wat.py
@@ -1,48 +1,72 @@
-import json
+# coding: utf-8
+from __future__ import unicode_literals
+
 import re
 
 from .common import InfoExtractor
-
 from ..utils import (
-    compat_urllib_parse,
+    unified_strdate,
 )
 
 
 class WatIE(InfoExtractor):
-    _VALID_URL=r'http://www.wat.tv/.*-(?P<shortID>.*?)_.*?.html'
+    _VALID_URL = r'http://www\.wat\.tv/.*-(?P<shortID>.*?)_.*?\.html'
     IE_NAME = 'wat.tv'
     _TEST = {
-        u'url': u'http://www.wat.tv/video/world-war-philadelphia-vost-6bv55_2fjr7_.html',
-        u'file': u'6bv55.mp4',
-        u'md5': u'0a4fe7870f31eaeabb5e25fd8da8414a',
-        u'info_dict': {
-            u"title": u"World War Z - Philadelphia VOST"
-        }
+        'url': 'http://www.wat.tv/video/world-war-philadelphia-vost-6bv55_2fjr7_.html',
+        'info_dict': {
+            'id': '10631273',
+            'ext': 'mp4',
+            'title': 'World War Z - Philadelphia VOST',
+            'description': 'La menace est partout. Que se passe-t-il à Philadelphia ?\r\nWORLD WAR Z, avec Brad Pitt, au cinéma le 3 juillet.\r\nhttp://www.worldwarz.fr',
+        },
+        'params': {
+            # Sometimes wat serves the whole file with the --test option
+            'skip_download': True,
+        },
     }
 
+    def download_video_info(self, real_id):
+        # 'contentv4' is used in the website, but it also returns the related
+        # videos, we don't need them
+        info = self._download_json('http://www.wat.tv/interface/contentv3/' + real_id, real_id)
+        return info['media']
+
     def _real_extract(self, url):
+        def real_id_for_chapter(chapter):
+            return chapter['tc_start'].split('-')[0]
         mobj = re.match(self._VALID_URL, url)
         short_id = mobj.group('shortID')
+        webpage = self._download_webpage(url, short_id)
+        real_id = self._search_regex(r'xtpage = ".*-(.*?)";', webpage, 'real id')
 
-        player_data = compat_urllib_parse.urlencode({'shortVideoId': short_id,
-                                                     'html5': '1'})
-        player_info = self._download_webpage('http://www.wat.tv/player?' + player_data,
-                                             short_id, u'Downloading player info')
-        player = json.loads(player_info)['player']
-        html5_player = self._html_search_regex(r'iframe src="(.*?)"', player,
-                                               'html5 player')
-        player_webpage = self._download_webpage(html5_player, short_id,
-                                                u'Downloading player webpage')
-
-        video_url = self._search_regex(r'urlhtml5 : "(.*?)"', player_webpage,
-                                       'video url')
-        title = self._search_regex(r'contentTitle : "(.*?)"', player_webpage,
-                                   'title')
-        thumbnail = self._search_regex(r'previewMedia : "(.*?)"', player_webpage,
-                                       'thumbnail')
-        return {'id': short_id,
-                'url': video_url,
-                'ext': 'mp4',
-                'title': title,
-                'thumbnail': thumbnail,
-                }
+        video_info = self.download_video_info(real_id)
+        chapters = video_info['chapters']
+        first_chapter = chapters[0]
+
+        if real_id_for_chapter(first_chapter) != real_id:
+            self.to_screen('Multipart video detected')
+            chapter_urls = []
+            for chapter in chapters:
+                chapter_id = real_id_for_chapter(chapter)
+                # Yes, when we this chapter is processed by WatIE,
+                # it will download the info again
+                chapter_info = self.download_video_info(chapter_id)
+                chapter_urls.append(chapter_info['url'])
+            entries = [self.url_result(chapter_url) for chapter_url in chapter_urls]
+            return self.playlist_result(entries, real_id, video_info['title'])
+
+        upload_date = None
+        if 'date_diffusion' in first_chapter:
+            upload_date = unified_strdate(first_chapter['date_diffusion'])
+        # Otherwise we can continue and extract just one part, we have to use
+        # the short id for getting the video url
+        return {
+            'id': real_id,
+            'url': 'http://wat.tv/get/android5/%s.mp4' % real_id,
+            'title': first_chapter['title'],
+            'thumbnail': first_chapter['preview'],
+            'description': first_chapter['description'],
+            'view_count': video_info['views'],
+            'upload_date': upload_date,
+        }