[ntvde] Add new extractor (Fixes #4850)

diff --git a/test/test_utils.py b/test/test_utils.py
index ebec7986f2b660da01db37c04bb65ccc283884bd..0ffccd35f0ffbce62ee2d8727101297636488e01 100644 (file)
--- a/test/test_utils.py
+++ b/test/test_utils.py
@@ -238,6 +238,8 @@ class TestUtil(unittest.TestCase):
         self.assertEqual(parse_duration('5 s'), 5)
         self.assertEqual(parse_duration('3 min'), 180)
         self.assertEqual(parse_duration('2.5 hours'), 9000)
+        self.assertEqual(parse_duration('02:03:04'), 7384)
+        self.assertEqual(parse_duration('01:02:03:04'), 93784)
 
     def test_fix_xml_ampersands(self):
         self.assertEqual(
@@ -371,6 +373,16 @@ class TestUtil(unittest.TestCase):
         on = js_to_json('{"abc": true}')
         self.assertEqual(json.loads(on), {'abc': True})
 
+        # Ignore JavaScript code as well
+        on = js_to_json('''{
+            "x": 1,
+            y: "a",
+            z: some.code
+        }''')
+        d = json.loads(on)
+        self.assertEqual(d['x'], 1)
+        self.assertEqual(d['y'], 'a')
+
     def test_clean_html(self):
         self.assertEqual(clean_html('a:\nb'), 'a: b')
         self.assertEqual(clean_html('a:\n   "b"'), 'a:    "b"')

diff --git a/youtube_dl/extractor/__init__.py b/youtube_dl/extractor/__init__.py
index bd707e0286b093ba869263bdb583d3934ed499db..205bd43386c167a792c3613b03ecb818abe37132 100644 (file)
--- a/youtube_dl/extractor/__init__.py
+++ b/youtube_dl/extractor/__init__.py
@@ -317,6 +317,7 @@ from .nrk import (
     NRKIE,
     NRKTVIE,
 )
+from .ntvde import NTVDeIE
 from .ntvru import NTVRuIE
 from .nytimes import NYTimesIE
 from .nuvid import NuvidIE

diff --git a/youtube_dl/extractor/ntvde.py b/youtube_dl/extractor/ntvde.py
new file mode 100644 (file)
index 0000000..a4f628f
--- /dev/null
+++ b/youtube_dl/extractor/ntvde.py
@@ -0,0 +1,65 @@
+# encoding: utf-8
+from __future__ import unicode_literals
+
+from .common import InfoExtractor
+from ..utils import (
+    int_or_none,
+    js_to_json,
+    parse_duration,
+)
+
+
+class NTVDeIE(InfoExtractor):
+    IE_NAME = 'n-tv.de'
+    _VALID_URL = r'https?://(?:www\.)?n-tv\.de/mediathek/videos/[^/?#]+/[^/?#]+-article(?P<id>.+)\.html'
+
+    _TESTS = [{
+        'url': 'http://www.n-tv.de/mediathek/videos/panorama/Schnee-und-Glaette-fuehren-zu-zahlreichen-Unfaellen-und-Staus-article14438086.html',
+        'md5': 'd37b7df1eea32265c51a062499ca488f',
+        'info_dict': {
+            'id': '14438086',
+            'ext': 'mp4',
+            'thumbnail': 're:^https?://.*\.jpg$',
+            'title': 'Schnee und Glätte führen zu zahlreichen Unfällen und Staus',
+            'alt_title': 'Winterchaos auf deutschen Straßen',
+            'description': 'Schnee und Glätte sorgen deutschlandweit für einen chaotischen Start in die Woche: Auf den Straßen kommt es zu kilometerlangen Staus und Dutzenden Glätteunfällen. In Düsseldorf und München wirbelt der Schnee zudem den Flugplan durcheinander. Dutzende Flüge landen zu spät, einige fallen ganz aus.',
+            'duration': 4020,
+            'timestamp': 1422892797,
+            'upload_date': '20150202',
+        },
+    }]
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+        webpage = self._download_webpage(url, video_id)
+
+        info = self._parse_json(self._search_regex(
+            r'(?s)ntv.pageInfo.article =\s(\{.*?\});', webpage, 'info'),
+            video_id, transform_source=js_to_json)
+        timestamp = int_or_none(info.get('publishedDateAsUnixTimeStamp'))
+        vdata = self._parse_json(self._search_regex(
+            r'(?s)\$\(\s*"\#player"\s*\)\s*\.data\(\s*"player",\s*(\{.*?\})\);',
+            webpage, 'player data'),
+            video_id, transform_source=js_to_json)
+        duration = parse_duration(vdata.get('duration'))
+        formats = [{
+            'format_id': 'flash',
+            'url': 'rtmp://fms.n-tv.de/' + vdata['video'],
+        }, {
+            'format_id': 'mp4',
+            'url': 'http://video.n-tv.de' + vdata['videoMp4'],
+        }]
+        m3u8_url = 'http://video.n-tv.de' + vdata['videoM3u8']
+        formats.extend(self._extract_m3u8_formats(m3u8_url, video_id))
+        self._sort_formats(formats)
+
+        return {
+            'id': video_id,
+            'title': info['headline'],
+            'description': info.get('intro'),
+            'alt_title': info.get('kicker'),
+            'timestamp': timestamp,
+            'thumbnail': vdata.get('html5VideoPoster'),
+            'duration': duration,
+            'formats': formats,
+        }

diff --git a/youtube_dl/extractor/ntvru.py b/youtube_dl/extractor/ntvru.py
index 8b9166d486222e1b2f7e05d6e0108294e26d70a5..963bb715a32725bb09361402529cfa69a97027d8 100644 (file)
--- a/youtube_dl/extractor/ntvru.py
+++ b/youtube_dl/extractor/ntvru.py
@@ -10,6 +10,7 @@ from ..utils import (
 
 
 class NTVRuIE(InfoExtractor):
+    IE_NAME = 'ntv.ru'
     _VALID_URL = r'http://(?:www\.)?ntv\.ru/(?P<id>.+)'
 
     _TESTS = [

diff --git a/youtube_dl/utils.py b/youtube_dl/utils.py
index 4ade0554e33597ce8aa965e147d2d73cce6a9fab..251074bf5af626909d4faf643f411faf88a1d18b 100644 (file)
--- a/youtube_dl/utils.py
+++ b/youtube_dl/utils.py
@@ -1275,7 +1275,10 @@ def parse_duration(s):
             (?P<only_hours>[0-9.]+)\s*(?:hours?)|
 
             (?:
-                (?:(?P<hours>[0-9]+)\s*(?:[:h]|hours?)\s*)?
+                (?:
+                    (?:(?P<days>[0-9]+)\s*(?:[:d]|days?)\s*)?
+                    (?P<hours>[0-9]+)\s*(?:[:h]|hours?)\s*
+                )?
                 (?P<mins>[0-9]+)\s*(?:[:m]|mins?|minutes?)\s*
             )?
             (?P<secs>[0-9]+)(?P<ms>\.[0-9]+)?\s*(?:s|secs?|seconds?)?
@@ -1293,6 +1296,8 @@ def parse_duration(s):
         res += int(m.group('mins')) * 60
     if m.group('hours'):
         res += int(m.group('hours')) * 60 * 60
+    if m.group('days'):
+        res += int(m.group('days')) * 24 * 60 * 60
     if m.group('ms'):
         res += float(m.group('ms'))
     return res
@@ -1543,7 +1548,7 @@ def js_to_json(code):
     res = re.sub(r'''(?x)
         "(?:[^"\\]*(?:\\\\|\\")?)*"|
         '(?:[^'\\]*(?:\\\\|\\')?)*'|
-        [a-zA-Z_][a-zA-Z_0-9]*
+        [a-zA-Z_][.a-zA-Z_0-9]*
         ''', fix_kv, code)
     res = re.sub(r',(\s*\])', lambda m: m.group(1), res)
     return res