Update changelog

[youtubedl] / youtube_dl / extractor / afreecatv.py

diff --git a/youtube_dl/extractor/afreecatv.py b/youtube_dl/extractor/afreecatv.py
index 75b36699363609876c755d4c120ec195aa81ec3a..df2a3fc4a2a24d2feeb81083d2414d8dc09b682b 100644 (file)
--- a/youtube_dl/extractor/afreecatv.py
+++ b/youtube_dl/extractor/afreecatv.py
@@ -4,20 +4,17 @@ from __future__ import unicode_literals
 import re
 
 from .common import InfoExtractor
 import re
 
 from .common import InfoExtractor

-from ..compat import (
-    compat_urllib_parse_urlparse,
-    compat_urlparse,
-)
+from ..compat import compat_xpath

 from ..utils import (
 from ..utils import (

+    determine_ext,

     ExtractorError,
     int_or_none,
     ExtractorError,
     int_or_none,

-    update_url_query,
-    xpath_element,

     xpath_text,
 )
 
 
 class AfreecaTVIE(InfoExtractor):
     xpath_text,
 )
 
 
 class AfreecaTVIE(InfoExtractor):

+    IE_NAME = 'afreecatv'

     IE_DESC = 'afreecatv.com'
     _VALID_URL = r'''(?x)
                     https?://
     IE_DESC = 'afreecatv.com'
     _VALID_URL = r'''(?x)
                     https?://


@@ -42,7 +39,8 @@ class AfreecaTVIE(InfoExtractor):
             'uploader': 'dailyapril',
             'uploader_id': 'dailyapril',
             'upload_date': '20160503',
             'uploader': 'dailyapril',
             'uploader_id': 'dailyapril',
             'upload_date': '20160503',

-        }
+        },
+        'skip': 'Video is gone',

     }, {
         'url': 'http://afbbs.afreecatv.com:8080/app/read_ucc_bbs.cgi?nStationNo=16711924&nTitleNo=36153164&szBjId=dailyapril&nBbsNo=18605867',
         'info_dict': {
     }, {
         'url': 'http://afbbs.afreecatv.com:8080/app/read_ucc_bbs.cgi?nStationNo=16711924&nTitleNo=36153164&szBjId=dailyapril&nBbsNo=18605867',
         'info_dict': {


@@ -70,6 +68,93 @@ class AfreecaTVIE(InfoExtractor):
                 'upload_date': '20160502',
             },
         }],
                 'upload_date': '20160502',
             },
         }],

+        'skip': 'Video is gone',
+    }, {
+        'url': 'http://vod.afreecatv.com/PLAYER/STATION/18650793',
+        'info_dict': {
+            'id': '18650793',
+            'ext': 'mp4',
+            'title': '오늘은 다르다! 쏘님의 우월한 위아래~ 댄스리액션!',
+            'thumbnail': r're:^https?://.*\.jpg$',
+            'uploader': '윈아디',
+            'uploader_id': 'badkids',
+            'duration': 107,
+        },
+        'params': {
+            'skip_download': True,
+        },
+    }, {
+        'url': 'http://vod.afreecatv.com/PLAYER/STATION/10481652',
+        'info_dict': {
+            'id': '10481652',
+            'title': "BJ유트루와 함께하는 '팅커벨 메이크업!'",
+            'thumbnail': 're:^https?://(?:video|st)img.afreecatv.com/.*$',
+            'uploader': 'dailyapril',
+            'uploader_id': 'dailyapril',
+            'duration': 6492,
+        },
+        'playlist_count': 2,
+        'playlist': [{
+            'md5': 'd8b7c174568da61d774ef0203159bf97',
+            'info_dict': {
+                'id': '20160502_c4c62b9d_174361386_1',
+                'ext': 'mp4',
+                'title': "BJ유트루와 함께하는 '팅커벨 메이크업!' (part 1)",
+                'thumbnail': 're:^https?://(?:video|st)img.afreecatv.com/.*$',
+                'uploader': 'dailyapril',
+                'uploader_id': 'dailyapril',
+                'upload_date': '20160502',
+                'duration': 3601,
+            },
+        }, {
+            'md5': '58f2ce7f6044e34439ab2d50612ab02b',
+            'info_dict': {
+                'id': '20160502_39e739bb_174361386_2',
+                'ext': 'mp4',
+                'title': "BJ유트루와 함께하는 '팅커벨 메이크업!' (part 2)",
+                'thumbnail': 're:^https?://(?:video|st)img.afreecatv.com/.*$',
+                'uploader': 'dailyapril',
+                'uploader_id': 'dailyapril',
+                'upload_date': '20160502',
+                'duration': 2891,
+            },
+        }],
+        'params': {
+            'skip_download': True,
+        },
+    }, {
+        # non standard key
+        'url': 'http://vod.afreecatv.com/PLAYER/STATION/20515605',
+        'info_dict': {
+            'id': '20170411_BE689A0E_190960999_1_2_h',
+            'ext': 'mp4',
+            'title': '혼자사는여자집',
+            'thumbnail': 're:^https?://(?:video|st)img.afreecatv.com/.*$',
+            'uploader': '♥이슬이',
+            'uploader_id': 'dasl8121',
+            'upload_date': '20170411',
+            'duration': 213,
+        },
+        'params': {
+            'skip_download': True,
+        },
+    }, {
+        # adult video
+        'url': 'http://vod.afreecatv.com/PLAYER/STATION/26542731',
+        'info_dict': {
+            'id': '20171001_F1AE1711_196617479_1',
+            'ext': 'mp4',
+            'title': '[생]서아 초심 찾기 방송 (part 1)',
+            'thumbnail': 're:^https?://(?:video|st)img.afreecatv.com/.*$',
+            'uploader': 'BJ서아',
+            'uploader_id': 'bjdyrksu',
+            'upload_date': '20171001',
+            'duration': 3600,
+            'age_limit': 18,
+        },
+        'params': {
+            'skip_download': True,
+        },

     }, {
         'url': 'http://www.afreecatv.com/player/Player.swf?szType=szBjId=djleegoon&nStationNo=11273158&nBbsNo=13161095&nTitleNo=36327652',
         'only_matching': True,
     }, {
         'url': 'http://www.afreecatv.com/player/Player.swf?szType=szBjId=djleegoon&nStationNo=11273158&nBbsNo=13161095&nTitleNo=36327652',
         'only_matching': True,


@@ -84,42 +169,110 @@ class AfreecaTVIE(InfoExtractor):
         m = re.match(r'^(?P<upload_date>\d{8})_\w+_(?P<part>\d+)$', key)
         if m:
             video_key['upload_date'] = m.group('upload_date')
         m = re.match(r'^(?P<upload_date>\d{8})_\w+_(?P<part>\d+)$', key)
         if m:
             video_key['upload_date'] = m.group('upload_date')

-            video_key['part'] = m.group('part')
+            video_key['part'] = int(m.group('part'))

         return video_key
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
         return video_key
 
     def _real_extract(self, url):
         video_id = self._match_id(url)

-        parsed_url = compat_urllib_parse_urlparse(url)
-        info_url = compat_urlparse.urlunparse(parsed_url._replace(
-            netloc='afbbs.afreecatv.com:8080',
-            path='/api/video/get_video_info.php'))
+
+        webpage = self._download_webpage(url, video_id)
+
+        if re.search(r'alert\(["\']This video has been deleted', webpage):
+            raise ExtractorError(
+                'Video %s has been deleted' % video_id, expected=True)
+
+        station_id = self._search_regex(
+            r'nStationNo\s*=\s*(\d+)', webpage, 'station')
+        bbs_id = self._search_regex(
+            r'nBbsNo\s*=\s*(\d+)', webpage, 'bbs')
+        video_id = self._search_regex(
+            r'nTitleNo\s*=\s*(\d+)', webpage, 'title', default=video_id)

 
         video_xml = self._download_xml(
 
         video_xml = self._download_xml(

-            update_url_query(info_url, {'nTitleNo': video_id}), video_id)
+            'http://afbbs.afreecatv.com:8080/api/video/get_video_info.php',
+            video_id, headers={
+                'Referer': 'http://vod.afreecatv.com/embed.php',
+            }, query={
+                'nTitleNo': video_id,
+                'nStationNo': station_id,
+                'nBbsNo': bbs_id,
+                'partialView': 'SKIP_ADULT',
+            })

 
 

-        if xpath_element(video_xml, './track/video/file') is None:
-            raise ExtractorError('Specified AfreecaTV video does not exist',
-                                 expected=True)
+        flag = xpath_text(video_xml, './track/flag', 'flag', default=None)
+        if flag and flag != 'SUCCEED':
+            raise ExtractorError(
+                '%s said: %s' % (self.IE_NAME, flag), expected=True)
+
+        video_element = video_xml.findall(compat_xpath('./track/video'))[-1]
+        if video_element is None or video_element.text is None:
+            raise ExtractorError(
+                'Video %s video does not exist' % video_id, expected=True)
+
+        video_url = video_element.text.strip()
+
+        title = xpath_text(video_xml, './track/title', 'title', fatal=True)

 
 

-        title = xpath_text(video_xml, './track/title', 'title')

         uploader = xpath_text(video_xml, './track/nickname', 'uploader')
         uploader_id = xpath_text(video_xml, './track/bj_id', 'uploader id')
         uploader = xpath_text(video_xml, './track/nickname', 'uploader')
         uploader_id = xpath_text(video_xml, './track/bj_id', 'uploader id')

-        duration = int_or_none(xpath_text(video_xml, './track/duration',
-                                          'duration'))
+        duration = int_or_none(xpath_text(
+            video_xml, './track/duration', 'duration'))

         thumbnail = xpath_text(video_xml, './track/titleImage', 'thumbnail')
 
         thumbnail = xpath_text(video_xml, './track/titleImage', 'thumbnail')
 

-        entries = []
-        for i, video_file in enumerate(video_xml.findall('./track/video/file')):
-            video_key = self.parse_video_key(video_file.get('key', ''))
-            if not video_key:
-                continue
-            entries.append({
-                'id': '%s_%s' % (video_id, video_key.get('part', i + 1)),
-                'title': title,
-                'upload_date': video_key.get('upload_date'),
-                'duration': int_or_none(video_file.get('duration')),
-                'url': video_file.text,
+        common_entry = {
+            'uploader': uploader,
+            'uploader_id': uploader_id,
+            'thumbnail': thumbnail,
+        }
+
+        info = common_entry.copy()
+        info.update({
+            'id': video_id,
+            'title': title,
+            'duration': duration,
+        })
+
+        if not video_url:
+            entries = []
+            file_elements = video_element.findall(compat_xpath('./file'))
+            one = len(file_elements) == 1
+            for file_num, file_element in enumerate(file_elements, start=1):
+                file_url = file_element.text
+                if not file_url:
+                    continue
+                key = file_element.get('key', '')
+                upload_date = self._search_regex(
+                    r'^(\d{8})_', key, 'upload date', default=None)
+                file_duration = int_or_none(file_element.get('duration'))
+                format_id = key if key else '%s_%s' % (video_id, file_num)
+                if determine_ext(file_url) == 'm3u8':
+                    formats = self._extract_m3u8_formats(
+                        file_url, video_id, 'mp4', entry_protocol='m3u8_native',
+                        m3u8_id='hls',
+                        note='Downloading part %d m3u8 information' % file_num)
+                else:
+                    formats = [{
+                        'url': file_url,
+                        'format_id': 'http',
+                    }]
+                if not formats:
+                    continue
+                self._sort_formats(formats)
+                file_info = common_entry.copy()
+                file_info.update({
+                    'id': format_id,
+                    'title': title if one else '%s (part %d)' % (title, file_num),
+                    'upload_date': upload_date,
+                    'duration': file_duration,
+                    'formats': formats,
+                })
+                entries.append(file_info)
+            entries_info = info.copy()
+            entries_info.update({
+                '_type': 'multi_video',
+                'entries': entries,

             })
             })

+            return entries_info

 
         info = {
             'id': video_id,
 
         info = {
             'id': video_id,


@@ -130,16 +283,17 @@ class AfreecaTVIE(InfoExtractor):
             'thumbnail': thumbnail,
         }
 
             'thumbnail': thumbnail,
         }
 

-        if len(entries) > 1:
-            info['_type'] = 'multi_video'
-            info['entries'] = entries
-        elif len(entries) == 1:
-            info['url'] = entries[0]['url']
-            info['upload_date'] = entries[0].get('upload_date')
+        if determine_ext(video_url) == 'm3u8':
+            info['formats'] = self._extract_m3u8_formats(
+                video_url, video_id, 'mp4', entry_protocol='m3u8_native',
+                m3u8_id='hls')

         else:
         else:

-            raise ExtractorError(
-                'No files found for the specified AfreecaTV video, either'
-                ' the URL is incorrect or the video has been made private.',
-                expected=True)
+            app, playpath = video_url.split('mp4:')
+            info.update({
+                'url': app,
+                'ext': 'flv',
+                'play_path': 'mp4:' + playpath,
+                'rtmp_live': True,  # downloading won't end without this
+            })

 
         return info
 
         return info