X-Git-Url: http://git.bitcoin.ninja/index.cgi?a=blobdiff_plain;f=youtube_dl%2Fextractor%2Fyam.py;h=ef553554736884ea4b43cb424028b5302c6a872e;hb=98708e6cbdc8b94723b30341b714a94905c70c7c;hp=511c8b67c0cfb3b32e05156b83fc2f30dad08064;hpb=7105440cecf82aff295df4f32575f6c8b64b3c2d;p=youtube-dl

diff --git a/youtube_dl/extractor/yam.py b/youtube_dl/extractor/yam.py
index 511c8b67c..ef5535547 100644
--- a/youtube_dl/extractor/yam.py
+++ b/youtube_dl/extractor/yam.py
@@ -1,14 +1,21 @@
 # coding: utf-8
 from __future__ import unicode_literals
 
+import re
+
 from .common import InfoExtractor
 from ..compat import compat_urlparse
-from ..utils import month_by_abbreviation
-import re
+from ..utils import (
+    float_or_none,
+    month_by_abbreviation,
+    ExtractorError,
+    get_element_by_attribute,
+)
 
 
 class YamIE(InfoExtractor):
-    _VALID_URL = r'http://mymedia.yam.com/m/(?P<id>\d+)'
+    IE_DESC = '蕃薯藤yam天空部落'
+    _VALID_URL = r'https?://mymedia\.yam\.com/m/(?P<id>\d+)'
 
     _TESTS = [{
         # An audio hosted on Yam
@@ -18,28 +25,62 @@ class YamIE(InfoExtractor):
             'id': '2283921',
             'ext': 'mp3',
             'title': '發現 - 趙薇 京華煙雲主題曲',
+            'description': '發現 - 趙薇 京華煙雲主題曲',
             'uploader_id': 'princekt',
             'upload_date': '20080807',
             'duration': 313.0,
         }
     }, {
         # An external video hosted on YouTube
-        'url': 'http://mymedia.yam.com/m/3598173',
-        'md5': '0238ceec479c654e8c2f1223755bf3e9',
+        'url': 'http://mymedia.yam.com/m/3599430',
+        'md5': '03127cf10d8f35d120a9e8e52e3b17c6',
         'info_dict': {
-            'id': 'pJ2Deys283c',
+            'id': 'CNpEoQlrIgA',
             'ext': 'mp4',
-            'upload_date': '20150202',
+            'upload_date': '20150306',
             'uploader': '新莊社大瑜伽社',
-            'description': 'md5:f5cc72f0baf259a70fb731654b0d2eff',
+            'description': 'md5:11e2e405311633ace874f2e6226c8b17',
             'uploader_id': '2323agoy',
-            'title': '外婆的澎湖灣KTV-潘安邦',
+            'title': '20090412陽明山二子坪-1',
+        },
+        'skip': 'Video does not exist',
+    }, {
+        'url': 'http://mymedia.yam.com/m/3598173',
+        'info_dict': {
+            'id': '3598173',
+            'ext': 'mp4',
+        },
+        'skip': 'cause Yam system error',
+    }, {
+        'url': 'http://mymedia.yam.com/m/3599437',
+        'info_dict': {
+            'id': '3599437',
+            'ext': 'mp4',
+        },
+        'skip': 'invalid YouTube URL',
+    }, {
+        'url': 'http://mymedia.yam.com/m/2373534',
+        'md5': '7ff74b91b7a817269d83796f8c5890b1',
+        'info_dict': {
+            'id': '2373534',
+            'ext': 'mp3',
+            'title': '林俊傑&蔡卓妍-小酒窩',
+            'description': 'md5:904003395a0fcce6cfb25028ff468420',
+            'upload_date': '20080928',
+            'uploader_id': 'onliner2',
         }
     }]
 
     def _real_extract(self, url):
-        media_id = self._match_id(url)
-        page = self._download_webpage(url, media_id)
+        video_id = self._match_id(url)
+        page = self._download_webpage(url, video_id)
+
+        # Check for errors
+        system_msg = self._html_search_regex(
+            r'系統訊息(?:<br>|\n|\r)*([^<>]+)<br>', page, 'system message',
+            default=None)
+        if system_msg:
+            raise ExtractorError(system_msg, expected=True)
 
         # Is it hosted externally on YouTube?
         youtube_url = self._html_search_regex(
@@ -48,23 +89,35 @@ class YamIE(InfoExtractor):
         if youtube_url:
             return self.url_result(youtube_url, 'Youtube')
 
+        title = self._html_search_regex(
+            r'<h1[^>]+class="heading"[^>]*>\s*(.+)\s*</h1>', page, 'title')
+
         api_page = self._download_webpage(
-            'http://mymedia.yam.com/api/a/?pID=' + media_id, media_id)
+            'http://mymedia.yam.com/api/a/?pID=' + video_id, video_id,
+            note='Downloading API page')
         api_result_obj = compat_urlparse.parse_qs(api_page)
 
-        author = self._html_search_regex(
-            r'<!-- 發表作者 -->：[\n ]+<a href="/([a-z]+)"', page, 'author')
-        mobj = re.search(r'<!-- 發表於 -->(?P<mon>[A-Z][a-z]{2})  ' +
+        info_table = get_element_by_attribute('class', 'info', page)
+        uploader_id = self._html_search_regex(
+            r'<!-- 發表作者 -->：[\n ]+<a href="/([a-z0-9]+)"',
+            info_table, 'uploader id', fatal=False)
+        mobj = re.search(r'<!-- 發表於 -->(?P<mon>[A-Z][a-z]{2})\s+' +
                          r'(?P<day>\d{1,2}), (?P<year>\d{4})', page)
-        upload_date = '%s%02d%02d' % (mobj.group('year'),
-                                      month_by_abbreviation(mobj.group('mon')),
-                                      int(mobj.group('day')))
+        if mobj:
+            upload_date = '%s%02d%02d' % (
+                mobj.group('year'),
+                month_by_abbreviation(mobj.group('mon')),
+                int(mobj.group('day')))
+        else:
+            upload_date = None
+        duration = float_or_none(api_result_obj['totaltime'][0], scale=1000)
 
         return {
-            'id': media_id,
+            'id': video_id,
             'url': api_result_obj['mp3file'][0],
-            'title': self._html_search_meta('description', page),
-            'duration': float(api_result_obj['totaltime'][0]) / 1000.0,
-            'uploader_id': author,
+            'title': title,
+            'description': self._html_search_meta('description', page),
+            'duration': duration,
+            'uploader_id': uploader_id,
             'upload_date': upload_date,
         }