d/rules: Run offline tests.

[youtubedl] / youtube_dl / extractor / naver.py

diff --git a/youtube_dl/extractor/naver.py b/youtube_dl/extractor/naver.py
index 9df236d69850bdc8e97fcb595e81b675c970ad1b..bb3d944133d6a1e2685779b86a7565ad9b0985f0 100644 (file)
--- a/youtube_dl/extractor/naver.py
+++ b/youtube_dl/extractor/naver.py
@@ -1,73 +1,128 @@
-# encoding: utf-8
-import re
-import xml.etree.ElementTree
+# coding: utf-8
+from __future__ import unicode_literals

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

-    compat_urllib_parse,

     ExtractorError,
     ExtractorError,

+    int_or_none,
+    update_url_query,

 )
 
 
 class NaverIE(InfoExtractor):
 )
 
 
 class NaverIE(InfoExtractor):

-    _VALID_URL = r'https?://tvcast\.naver\.com/v/(?P<id>\d+)'
+    _VALID_URL = r'https?://(?:m\.)?tv(?:cast)?\.naver\.com/v/(?P<id>\d+)'

 
 

-    _TEST = {
-        u'url': u'http://tvcast.naver.com/v/81652',
-        u'file': u'81652.mp4',
-        u'info_dict': {
-            u'title': u'[9월 모의고사 해설강의][수학_김상희] 수학 A형 16~20번',
-            u'description': u'합격불변의 법칙 메가스터디 | 메가스터디 수학 김상희 선생님이 9월 모의고사 수학A형 16번에서 20번까지 해설강의를 공개합니다.',
-            u'upload_date': u'20130903',
+    _TESTS = [{
+        'url': 'http://tv.naver.com/v/81652',
+        'info_dict': {
+            'id': '81652',
+            'ext': 'mp4',
+            'title': '[9월 모의고사 해설강의][수학_김상희] 수학 A형 16~20번',
+            'description': '합격불변의 법칙 메가스터디 | 메가스터디 수학 김상희 선생님이 9월 모의고사 수학A형 16번에서 20번까지 해설강의를 공개합니다.',
+            'upload_date': '20130903',

         },
         },

-    }
+    }, {
+        'url': 'http://tv.naver.com/v/395837',
+        'md5': '638ed4c12012c458fefcddfd01f173cd',
+        'info_dict': {
+            'id': '395837',
+            'ext': 'mp4',
+            'title': '9년이 지나도 아픈 기억, 전효성의 아버지',
+            'description': 'md5:5bf200dcbf4b66eb1b350d1eb9c753f7',
+            'upload_date': '20150519',
+        },
+        'skip': 'Georestricted',
+    }, {
+        'url': 'http://tvcast.naver.com/v/81652',
+        'only_matching': True,
+    }]

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

-        mobj = re.match(self._VALID_URL, url)
-        video_id = mobj.group(1)
+        video_id = self._match_id(url)

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

-        m_id = re.search(r'var rmcPlayer = new nhn.rmcnmv.RMCVideoPlayer\("(.+?)", "(.+?)"',
-            webpage)
-        if m_id is None:
-            raise ExtractorError(u'couldn\'t extract vid and key')
-        vid = m_id.group(1)
-        key = m_id.group(2)
-        query = compat_urllib_parse.urlencode({'vid': vid, 'inKey': key,})
-        query_urls = compat_urllib_parse.urlencode({
-            'masterVid': vid,
-            'protocol': 'p2p',
-            'inKey': key,
-        })
-        info_xml = self._download_webpage(
-            'http://serviceapi.rmcnmv.naver.com/flash/videoInfo.nhn?' + query,
-            video_id, u'Downloading video info')
-        urls_xml = self._download_webpage(
-            'http://serviceapi.rmcnmv.naver.com/flash/playableEncodingOption.nhn?' + query_urls,
-            video_id, u'Downloading video formats info')
-        info = xml.etree.ElementTree.fromstring(info_xml.encode('utf-8'))
-        urls = xml.etree.ElementTree.fromstring(urls_xml.encode('utf-8'))

 
 

+        vid = self._search_regex(
+            r'videoId["\']\s*:\s*(["\'])(?P<value>(?:(?!\1).)+)\1', webpage,
+            'video id', fatal=None, group='value')
+        in_key = self._search_regex(
+            r'inKey["\']\s*:\s*(["\'])(?P<value>(?:(?!\1).)+)\1', webpage,
+            'key', default=None, group='value')
+
+        if not vid or not in_key:
+            error = self._html_search_regex(
+                r'(?s)<div class="(?:nation_error|nation_box|error_box)">\s*(?:<!--.*?-->)?\s*<p class="[^"]+">(?P<msg>.+?)</p>\s*</div>',
+                webpage, 'error', default=None)
+            if error:
+                raise ExtractorError(error, expected=True)
+            raise ExtractorError('couldn\'t extract vid and key')
+        video_data = self._download_json(
+            'http://play.rmcnmv.naver.com/vod/play/v2.0/' + vid,
+            video_id, query={
+                'key': in_key,
+            })
+        meta = video_data['meta']
+        title = meta['subject']

         formats = []
         formats = []

-        for format_el in urls.findall('EncodingOptions/EncodingOption'):
-            domain = format_el.find('Domain').text
-            if domain.startswith('rtmp'):
+
+        def extract_formats(streams, stream_type, query={}):
+            for stream in streams:
+                stream_url = stream.get('source')
+                if not stream_url:
+                    continue
+                stream_url = update_url_query(stream_url, query)
+                encoding_option = stream.get('encodingOption', {})
+                bitrate = stream.get('bitrate', {})
+                formats.append({
+                    'format_id': '%s_%s' % (stream.get('type') or stream_type, encoding_option.get('id') or encoding_option.get('name')),
+                    'url': stream_url,
+                    'width': int_or_none(encoding_option.get('width')),
+                    'height': int_or_none(encoding_option.get('height')),
+                    'vbr': int_or_none(bitrate.get('video')),
+                    'abr': int_or_none(bitrate.get('audio')),
+                    'filesize': int_or_none(stream.get('size')),
+                    'protocol': 'm3u8_native' if stream_type == 'HLS' else None,
+                })
+
+        extract_formats(video_data.get('videos', {}).get('list', []), 'H264')
+        for stream_set in video_data.get('streams', []):
+            query = {}
+            for param in stream_set.get('keys', []):
+                query[param['name']] = param['value']
+            stream_type = stream_set.get('type')
+            videos = stream_set.get('videos')
+            if videos:
+                extract_formats(videos, stream_type, query)
+            elif stream_type == 'HLS':
+                stream_url = stream_set.get('source')
+                if not stream_url:
+                    continue
+                formats.extend(self._extract_m3u8_formats(
+                    update_url_query(stream_url, query), video_id,
+                    'mp4', 'm3u8_native', m3u8_id=stream_type, fatal=False))
+        self._sort_formats(formats)
+
+        subtitles = {}
+        for caption in video_data.get('captions', {}).get('list', []):
+            caption_url = caption.get('source')
+            if not caption_url:

                 continue
                 continue

-            formats.append({
-                'url': domain + format_el.find('uri').text,
-                'ext': 'mp4',
-                'width': int(format_el.find('width').text),
-                'height': int(format_el.find('height').text),
+            subtitles.setdefault(caption.get('language') or caption.get('locale'), []).append({
+                'url': caption_url,

             })
 
             })
 

-        info = {
+        upload_date = self._search_regex(
+            r'<span[^>]+class="date".*?(\d{4}\.\d{2}\.\d{2})',
+            webpage, 'upload date', fatal=False)
+        if upload_date:
+            upload_date = upload_date.replace('.', '')
+
+        return {

             'id': video_id,
             'id': video_id,

-            'title': info.find('Subject').text,
+            'title': title,

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

+            'subtitles': subtitles,

             'description': self._og_search_description(webpage),
             'description': self._og_search_description(webpage),

-            'thumbnail': self._og_search_thumbnail(webpage),
-            'upload_date': info.find('WriteDate').text.replace('.', ''),
-            'view_count': int(info.find('PlayCount').text),
+            'thumbnail': meta.get('cover', {}).get('source') or self._og_search_thumbnail(webpage),
+            'view_count': int_or_none(meta.get('count')),
+            'upload_date': upload_date,

         }
         }

-        # TODO: Remove when #980 has been merged
-        info.update(formats[-1])
-        return info