]> git.bitcoin.ninja Git - youtube-dl/commitdiff
[Onet,ClipRs] Add new extractor for onet.tv and use it for clip.rs
authorYen Chi Hsuan <yan12125@gmail.com>
Wed, 6 Jul 2016 12:09:05 +0000 (20:09 +0800)
committerYen Chi Hsuan <yan12125@gmail.com>
Wed, 6 Jul 2016 12:09:05 +0000 (20:09 +0800)
Closes #9950

youtube_dl/extractor/cliprs.py
youtube_dl/extractor/extractors.py
youtube_dl/extractor/onet.py [new file with mode: 0644]

index 4f9320ea57cb542f69b6b0be1965bb004fa35574..d55b26d59ff89af6ef9ae7943cb67b4346031f66 100644 (file)
@@ -1,16 +1,10 @@
 # coding: utf-8
 from __future__ import unicode_literals
 
-from .common import InfoExtractor
-from ..utils import (
-    ExtractorError,
-    float_or_none,
-    int_or_none,
-    parse_iso8601,
-)
+from .onet import OnetBaseIE
 
 
-class ClipRsIE(InfoExtractor):
+class ClipRsIE(OnetBaseIE):
     _VALID_URL = r'https?://(?:www\.)?clip\.rs/(?P<id>[^/]+)/\d+'
     _TEST = {
         'url': 'http://www.clip.rs/premijera-frajle-predstavljaju-novi-spot-za-pesmu-moli-me-moli/3732',
@@ -27,64 +21,13 @@ class ClipRsIE(InfoExtractor):
     }
 
     def _real_extract(self, url):
-        video_id = self._match_id(url)
+        display_id = self._match_id(url)
 
-        webpage = self._download_webpage(url, video_id)
+        webpage = self._download_webpage(url, display_id)
 
-        video_id = self._search_regex(
-            r'id=(["\'])mvp:(?P<id>.+?)\1', webpage, 'mvp id', group='id')
+        mvp_id = self._search_mvp_id(webpage)
 
-        response = self._download_json(
-            'http://qi.ckm.onetapi.pl/', video_id,
-            query={
-                'body[id]': video_id,
-                'body[jsonrpc]': '2.0',
-                'body[method]': 'get_asset_detail',
-                'body[params][ID_Publikacji]': video_id,
-                'body[params][Service]': 'www.onet.pl',
-                'content-type': 'application/jsonp',
-                'x-onet-app': 'player.front.onetapi.pl',
-            })
+        info_dict = self._extract_from_id(mvp_id, webpage)
+        info_dict['display_id'] = display_id
 
-        error = response.get('error')
-        if error:
-            raise ExtractorError(
-                '%s said: %s' % (self.IE_NAME, error['message']), expected=True)
-
-        video = response['result'].get('0')
-
-        formats = []
-        for _, formats_dict in video['formats'].items():
-            if not isinstance(formats_dict, dict):
-                continue
-            for format_id, format_list in formats_dict.items():
-                if not isinstance(format_list, list):
-                    continue
-                for f in format_list:
-                    if not f.get('url'):
-                        continue
-                    formats.append({
-                        'url': f['url'],
-                        'format_id': format_id,
-                        'height': int_or_none(f.get('vertical_resolution')),
-                        'width': int_or_none(f.get('horizontal_resolution')),
-                        'abr': float_or_none(f.get('audio_bitrate')),
-                        'vbr': float_or_none(f.get('video_bitrate')),
-                    })
-        self._sort_formats(formats)
-
-        meta = video.get('meta', {})
-
-        title = self._og_search_title(webpage, default=None) or meta['title']
-        description = self._og_search_description(webpage, default=None) or meta.get('description')
-        duration = meta.get('length') or meta.get('lenght')
-        timestamp = parse_iso8601(meta.get('addDate'), ' ')
-
-        return {
-            'id': video_id,
-            'title': title,
-            'description': description,
-            'duration': duration,
-            'timestamp': timestamp,
-            'formats': formats,
-        }
+        return info_dict
index d21d7870a88c19f1b85bad3b2f4235dd8774b84d..e52faa0784453b91062dcc9a1ae3e707641ee24a 100644 (file)
@@ -139,9 +139,9 @@ from .chirbit import (
     ChirbitProfileIE,
 )
 from .cinchcast import CinchcastIE
-from .cliprs import ClipRsIE
 from .clipfish import ClipfishIE
 from .cliphunter import CliphunterIE
+from .cliprs import ClipRsIE
 from .clipsyndicate import ClipsyndicateIE
 from .closertotruth import CloserToTruthIE
 from .cloudy import CloudyIE
@@ -581,6 +581,10 @@ from .nytimes import (
 from .nuvid import NuvidIE
 from .odnoklassniki import OdnoklassnikiIE
 from .oktoberfesttv import OktoberfestTVIE
+from .onet import (
+    OnetIE,
+    OnetChannelIE,
+)
 from .onionstudios import OnionStudiosIE
 from .ooyala import (
     OoyalaIE,
diff --git a/youtube_dl/extractor/onet.py b/youtube_dl/extractor/onet.py
new file mode 100644 (file)
index 0000000..1e0672d
--- /dev/null
@@ -0,0 +1,171 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import re
+
+from .common import InfoExtractor
+from ..utils import (
+    determine_ext,
+    ExtractorError,
+    float_or_none,
+    get_element_by_class,
+    int_or_none,
+    js_to_json,
+    parse_iso8601,
+    remove_start,
+    strip_or_none,
+    url_basename,
+)
+
+
+class OnetBaseIE(InfoExtractor):
+    def _search_mvp_id(self, webpage):
+        return self._search_regex(
+            r'id=(["\'])mvp:(?P<id>.+?)\1', webpage, 'mvp id', group='id')
+
+    def _extract_from_id(self, video_id, webpage):
+        response = self._download_json(
+            'http://qi.ckm.onetapi.pl/', video_id,
+            query={
+                'body[id]': video_id,
+                'body[jsonrpc]': '2.0',
+                'body[method]': 'get_asset_detail',
+                'body[params][ID_Publikacji]': video_id,
+                'body[params][Service]': 'www.onet.pl',
+                'content-type': 'application/jsonp',
+                'x-onet-app': 'player.front.onetapi.pl',
+            })
+
+        error = response.get('error')
+        if error:
+            raise ExtractorError(
+                '%s said: %s' % (self.IE_NAME, error['message']), expected=True)
+
+        video = response['result'].get('0')
+
+        formats = []
+        for _, formats_dict in video['formats'].items():
+            if not isinstance(formats_dict, dict):
+                continue
+            for format_id, format_list in formats_dict.items():
+                if not isinstance(format_list, list):
+                    continue
+                for f in format_list:
+                    video_url = f.get('url')
+                    if not video_url:
+                        continue
+                    ext = determine_ext(video_url)
+                    if format_id == 'ism':
+                        # TODO: Support Microsoft Smooth Streaming
+                        continue
+                    elif ext == 'mpd':
+                        # TODO: Current DASH formats are broken - $Time$ pattern in
+                        # <SegmentTemplate> not implemented yet
+                        # formats.extend(self._extract_mpd_formats(
+                        #    video_url, video_id, mpd_id='dash', fatal=False))
+                        continue
+                    else:
+                        formats.append({
+                            'url': video_url,
+                            'format_id': format_id,
+                            'height': int_or_none(f.get('vertical_resolution')),
+                            'width': int_or_none(f.get('horizontal_resolution')),
+                            'abr': float_or_none(f.get('audio_bitrate')),
+                            'vbr': float_or_none(f.get('video_bitrate')),
+                        })
+        self._sort_formats(formats)
+
+        meta = video.get('meta', {})
+
+        title = self._og_search_title(webpage, default=None) or meta['title']
+        description = self._og_search_description(webpage, default=None) or meta.get('description')
+        duration = meta.get('length') or meta.get('lenght')
+        timestamp = parse_iso8601(meta.get('addDate'), ' ')
+
+        return {
+            'id': video_id,
+            'title': title,
+            'description': description,
+            'duration': duration,
+            'timestamp': timestamp,
+            'formats': formats,
+        }
+
+
+class OnetIE(OnetBaseIE):
+    _VALID_URL = 'https?://(?:www\.)?onet\.tv/[a-z]/[a-z]+/(?P<display_id>[0-9a-z-]+)/(?P<id>[0-9a-z]+)'
+    IE_NAME = 'onet.tv'
+
+    _TEST = {
+        'url': 'http://onet.tv/k/openerfestival/open-er-festival-2016-najdziwniejsze-wymagania-gwiazd/qbpyqc',
+        'info_dict': {
+            'id': 'qbpyqc',
+            'display_id': 'open-er-festival-2016-najdziwniejsze-wymagania-gwiazd',
+            'ext': 'mp4',
+            'title': 'Open\'er Festival 2016: najdziwniejsze wymagania gwiazd',
+            'description': 'Trzy samochody, których nigdy nie użyto, prywatne spa, hotel dekorowany czarnym suknem czy nielegalne używki. Organizatorzy koncertów i festiwali muszą stawać przed nie lada wyzwaniem zapraszając gwia...',
+            'upload_date': '20160705',
+            'timestamp': 1467721580,
+        },
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        display_id, video_id = mobj.group('display_id', 'id')
+
+        webpage = self._download_webpage(url, display_id)
+
+        mvp_id = self._search_mvp_id(webpage)
+
+        info_dict = self._extract_from_id(mvp_id, webpage)
+        info_dict.update({
+            'id': video_id,
+            'display_id': display_id,
+        })
+
+        return info_dict
+
+
+class OnetChannelIE(OnetBaseIE):
+    _VALID_URL = r'https?://(?:www\.)?onet\.tv/[a-z]/(?P<id>[a-z]+)(?:[?#]|$)'
+    IE_NAME = 'onet.tv:channel'
+
+    _TEST = {
+        'url': 'http://onet.tv/k/openerfestival',
+        'info_dict': {
+            'id': 'openerfestival',
+            'title': 'Open\'er Festival Live',
+            'description': 'Dziękujemy, że oglądaliście transmisje. Zobaczcie nasze relacje i wywiady z artystami.',
+        },
+        'playlist_mincount': 46,
+    }
+
+    def _real_extract(self, url):
+        channel_id = self._match_id(url)
+
+        webpage = self._download_webpage(url, channel_id)
+
+        current_clip_info = self._parse_json(self._search_regex(
+            r'var\s+currentClip\s*=\s*({[^}]+})', webpage, 'video info'), channel_id,
+            transform_source=lambda s: js_to_json(re.sub(r'\'\s*\+\s*\'', '', s)))
+        video_id = remove_start(current_clip_info['ckmId'], 'mvp:')
+        video_name = url_basename(current_clip_info['url'])
+
+        if self._downloader.params.get('noplaylist'):
+            self.to_screen(
+                'Downloading just video %s because of --no-playlist' % video_name)
+            return self._extract_from_id(video_id, webpage)
+
+        self.to_screen(
+            'Downloading channel %s - add --no-playlist to just download video %s' % (
+                channel_id, video_name))
+        matches = re.findall(
+            r'<a[^>]+href=[\'"](https?://(?:www\.)?onet\.tv/[a-z]/[a-z]+/[0-9a-z-]+/[0-9a-z]+)',
+            webpage)
+        entries = [
+            self.url_result(video_link, OnetIE.ie_key())
+            for video_link in matches]
+
+        channel_title = strip_or_none(get_element_by_class('o_channelName', webpage))
+        channel_description = strip_or_none(get_element_by_class('o_channelDesc', webpage))
+        return self.playlist_result(entries, channel_id, channel_title, channel_description)