Merge pull request #6790 from remitamine/tele13

[canal13cl] fix info extraction

3 files changed, 82 insertions, 49 deletions

diff --git a/youtube_dl/extractor/__init__.py b/youtube_dl/extractor/__init__.py
index eac50eda5..908581bf7 100644
--- a/youtube_dl/extractor/__init__.py
+++ b/youtube_dl/extractor/__init__.py
@@ -78,7 +78,6 @@ from .camdemy import (
     CamdemyIE,
     CamdemyFolderIE
 )
-from .canal13cl import Canal13clIE
 from .canalplus import CanalplusIE
 from .canalc2 import Canalc2IE
 from .cbs import CBSIE
@@ -648,6 +647,7 @@ from .teachingchannel import TeachingChannelIE
 from .teamcoco import TeamcocoIE
 from .techtalks import TechTalksIE
 from .ted import TEDIE
+from .tele13 import Tele13IE
 from .telebruxelles import TeleBruxellesIE
 from .telecinco import TelecincoIE
 from .telegraaf import TelegraafIE
diff --git a/youtube_dl/extractor/canal13cl.py b/youtube_dl/extractor/canal13cl.py
deleted file mode 100644
index 93241fefe..000000000
--- a/youtube_dl/extractor/canal13cl.py
+++ /dev/null
@@ -1,48 +0,0 @@
-# coding: utf-8
-from __future__ import unicode_literals
-
-import re
-
-from .common import InfoExtractor
-
-
-class Canal13clIE(InfoExtractor):
-    _VALID_URL = r'^http://(?:www\.)?13\.cl/(?:[^/?#]+/)*(?P<id>[^/?#]+)'
-    _TEST = {
-        'url': 'http://www.13.cl/t13/nacional/el-circulo-de-hierro-de-michelle-bachelet-en-su-regreso-a-la-moneda',
-        'md5': '4cb1fa38adcad8fea88487a078831755',
-        'info_dict': {
-            'id': '1403022125',
-            'display_id': 'el-circulo-de-hierro-de-michelle-bachelet-en-su-regreso-a-la-moneda',
-            'ext': 'mp4',
-            'title': 'El "círculo de hierro" de Michelle Bachelet en su regreso a La Moneda',
-            'description': '(Foto: Agencia Uno) En nueve días más, Michelle Bachelet va a asumir por segunda vez como presidenta de la República. Entre aquellos que la acompañarán hay caras que se repiten y otras que se consolidan en su entorno de colaboradores más cercanos.',
-        }
-    }
-
-    def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-        display_id = mobj.group('id')
-
-        webpage = self._download_webpage(url, display_id)
-
-        title = self._html_search_meta(
-            'twitter:title', webpage, 'title', fatal=True)
-        description = self._html_search_meta(
-            'twitter:description', webpage, 'description')
-        url = self._html_search_regex(
-            r'articuloVideo = \"(.*?)\"', webpage, 'url')
-        real_id = self._search_regex(
-            r'[^0-9]([0-9]{7,})[^0-9]', url, 'id', default=display_id)
-        thumbnail = self._html_search_regex(
-            r'articuloImagen = \"(.*?)\"', webpage, 'thumbnail')
-
-        return {
-            'id': real_id,
-            'display_id': display_id,
-            'url': url,
-            'title': title,
-            'description': description,
-            'ext': 'mp4',
-            'thumbnail': thumbnail,
-        }
diff --git a/youtube_dl/extractor/tele13.py b/youtube_dl/extractor/tele13.py
new file mode 100644
index 000000000..a363b4d40
--- /dev/null
+++ b/youtube_dl/extractor/tele13.py
@@ -0,0 +1,81 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+from .common import InfoExtractor
+from .youtube import YoutubeIE
+from ..utils import (
+    js_to_json,
+    qualities,
+    determine_ext,
+)
+
+
+class Tele13IE(InfoExtractor):
+    _VALID_URL = r'^http://(?:www\.)?t13\.cl/videos(?:/[^/]+)+/(?P<id>[\w-]+)'
+    _TESTS = [
+        {
+            'url': 'http://www.t13.cl/videos/actualidad/el-circulo-de-hierro-de-michelle-bachelet-en-su-regreso-a-la-moneda',
+            'md5': '4cb1fa38adcad8fea88487a078831755',
+            'info_dict': {
+                'id': 'el-circulo-de-hierro-de-michelle-bachelet-en-su-regreso-a-la-moneda',
+                'ext': 'mp4',
+                'title': 'El círculo de hierro de Michelle Bachelet en su regreso a La Moneda',
+            },
+            'params': {
+                # HTTP Error 404: Not Found
+                'skip_download': True,
+            },
+        },
+        {
+            'url': 'http://www.t13.cl/videos/mundo/tendencias/video-captan-misteriosa-bola-fuego-cielos-bangkok',
+            'md5': '867adf6a3b3fef932c68a71d70b70946',
+            'info_dict': {
+                'id': 'rOoKv2OMpOw',
+                'ext': 'mp4',
+                'title': 'Shooting star seen on 7-Sep-2015',
+                'description': 'md5:7292ff2a34b2f673da77da222ae77e1e',
+                'uploader': 'Porjai Jaturongkhakun',
+                'upload_date': '20150906',
+                'uploader_id': 'UCnLY_3ezwNcDSC_Wc6suZxw',
+            },
+            'add_ie': ['Youtube'],
+        }
+    ]
+
+    def _real_extract(self, url):
+        display_id = self._match_id(url)
+        webpage = self._download_webpage(url, display_id)
+
+        setup_js = self._search_regex(r"(?s)jwplayer\('player-vivo'\).setup\((\{.*?\})\)", webpage, 'setup code')
+        sources = self._parse_json(self._search_regex(r'sources\s*:\s*(\[[^\]]+\])', setup_js, 'sources'), display_id, js_to_json)
+
+        preference = qualities(['Móvil', 'SD', 'HD'])
+        formats = []
+        urls = []
+        for f in sources:
+            format_url = f['file']
+            if format_url and format_url not in urls:
+                ext = determine_ext(format_url)
+                if ext == 'm3u8':
+                    m3u8_formats = self._extract_m3u8_formats(format_url, display_id, 'mp4', 'm3u8_native', m3u8_id='hls', fatal=False)
+                    if m3u8_formats:
+                        formats.extend(m3u8_formats)
+                elif YoutubeIE.suitable(format_url):
+                    return self.url_result(format_url, 'Youtube')
+                else:
+                    formats.append({
+                        'url': format_url,
+                        'format_id': f.get('label'),
+                        'preference': preference(f.get('label')),
+                        'ext': ext,
+                    })
+                urls.append(format_url)
+        self._sort_formats(formats)
+
+        return {
+            'id': display_id,
+            'title': self._search_regex(r'title\s*:\s*"([^"]+)"', setup_js, 'title'),
+            'description': self._html_search_meta('description', webpage, 'description'),
+            'thumbnail': self._search_regex(r'image\s*:\s*"([^"]+)"', setup_js, 'thumbnail', default=None),
+            'formats': formats,
+        }