[abcnews] Added a new extractor (closes #3992)

Related: #6108, #8664, #9459

youtube_dl/extractor/abcnews.py

@@ -0,0 +1,135 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import calendar
+import re
+import time
+
+from .amp import AMPIE
+from .common import InfoExtractor
+from ..compat import compat_urlparse
+
+
+class AbcNewsVideoIE(AMPIE):
+    IE_NAME = 'abcnews:video'
+    _VALID_URL = 'http://abcnews.go.com/[^/]+/video/(?P<display_id>[0-9a-z-]+)-(?P<id>\d+)'
+
+    _TESTS = [{
+        'url': 'http://abcnews.go.com/ThisWeek/video/week-exclusive-irans-foreign-minister-zarif-20411932',
+        'info_dict': {
+            'id': '20411932',
+            'ext': 'mp4',
+            'display_id': 'week-exclusive-irans-foreign-minister-zarif',
+            'title': '\'This Week\' Exclusive: Iran\'s Foreign Minister Zarif',
+            'description': 'George Stephanopoulos goes one-on-one with Iranian Foreign Minister Dr. Javad Zarif.',
+            'duration': 180,
+            'thumbnail': 're:^https?://.*\.jpg$',
+        },
+        'params': {
+            # m3u8 download
+            'skip_download': True,
+        },
+    }, {
+        'url': 'http://abcnews.go.com/2020/video/2020-husband-stands-teacher-jail-student-affairs-26119478',
+        'only_matching': True,
+    }]
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        display_id = mobj.group('display_id')
+        video_id = mobj.group('id')
+        info_dict = self._extract_feed_info(
+            'http://abcnews.go.com/video/itemfeed?id=%s' % video_id)
+        info_dict.update({
+            'id': video_id,
+            'display_id': display_id,
+        })
+        return info_dict
+
+
+class AbcNewsIE(InfoExtractor):
+    IE_NAME = 'abcnews'
+    _VALID_URL = 'https?://abcnews\.go\.com/(?:[^/]+/)+(?P<display_id>[0-9a-z-]+)/story\?id=(?P<id>\d+)'
+
+    _TESTS = [{
+        'url': 'http://abcnews.go.com/Blotter/News/dramatic-video-rare-death-job-america/story?id=10498713#.UIhwosWHLjY',
+        'info_dict': {
+            'id': '10498713',
+            'ext': 'flv',
+            'display_id': 'dramatic-video-rare-death-job-america',
+            'title': 'Occupational Hazards',
+            'description': 'Nightline investigates the dangers that lurk at various jobs.',
+            'thumbnail': 're:^https?://.*\.jpg$',
+            'upload_date': '20100428',
+            'timestamp': 1272412800,
+        },
+        'add_ie': ['AbcNewsVideo'],
+    }, {
+        'url': 'http://abcnews.go.com/Entertainment/justin-timberlake-performs-stop-feeling-eurovision-2016/story?id=39125818',
+        'info_dict': {
+            'id': '39125818',
+            'ext': 'mp4',
+            'display_id': 'justin-timberlake-performs-stop-feeling-eurovision-2016',
+            'title': 'Justin Timberlake Drops Hints For Secret Single',
+            'description': 'Lara Spencer reports the buzziest stories of the day in "GMA" Pop News.',
+            'upload_date': '20160515',
+            'timestamp': 1463329500,
+        },
+        'params': {
+            # m3u8 download
+            'skip_download': True,
+            # The embedded YouTube video is blocked due to copyright issues
+            'playlist_items': '1',
+        },
+        'add_ie': ['AbcNewsVideo'],
+    }, {
+        'url': 'http://abcnews.go.com/Technology/exclusive-apple-ceo-tim-cook-iphone-cracking-software/story?id=37173343',
+        'only_matching': True,
+    }]
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        display_id = mobj.group('display_id')
+        video_id = mobj.group('id')
+
+        webpage = self._download_webpage(url, video_id)
+        video_url = self._search_regex(
+            r'window\.abcnvideo\.url\s*=\s*"([^"]+)"', webpage, 'video URL')
+        full_video_url = compat_urlparse.urljoin(url, video_url)
+
+        youtube_url = self._html_search_regex(
+            r'<iframe[^>]+src="(https://www\.youtube\.com/embed/[^"]+)"',
+            webpage, 'YouTube URL', default=None)
+
+        timestamp = None
+        date_str = self._html_search_regex(
+            r'<span[^>]+class="timestamp">([^<]+)</span>',
+            webpage, 'timestamp', fatal=False)
+        if date_str:
+            tz_offset = 0
+            if date_str.endswith(' ET'):  # Eastern Time
+                tz_offset = -5
+                date_str = date_str[:-3]
+            date_formats = ['%b. %d, %Y', '%b %d, %Y, %I:%M %p']
+            for date_format in date_formats:
+                try:
+                    timestamp = calendar.timegm(time.strptime(date_str.strip(), date_format))
+                except ValueError:
+                    continue
+            if timestamp is not None:
+                timestamp -= tz_offset * 3600
+
+        entry = {
+            '_type': 'url_transparent',
+            'ie_key': AbcNewsVideoIE.ie_key(),
+            'url': full_video_url,
+            'id': video_id,
+            'display_id': display_id,
+            'timestamp': timestamp,
+        }
+
+        if youtube_url:
+            entries = [entry, self.url_result(youtube_url, 'Youtube')]
+            return self.playlist_result(entries)
+
+        return entry

youtube_dl/extractor/amp.py

@@ -52,7 +52,7 @@ class AMPIE(InfoExtractor):
         for media_data in media_content:
             media = media_data['@attributes']
             media_type = media['type']
-            if media_type == 'video/f4m':
+            if media_type in ('video/f4m', 'application/f4m+xml'):
                 formats.extend(self._extract_f4m_formats(
                     media['url'] + '?hdcore=3.4.0&plugin=aasp-3.4.0.132.124',
                     video_id, f4m_id='hds', fatal=False))
@@ -61,7 +61,7 @@ class AMPIE(InfoExtractor):
                     media['url'], video_id, 'mp4', m3u8_id='hls', fatal=False))
             else:
                 formats.append({
-                    'format_id': media_data['media-category']['@attributes']['label'],
+                    'format_id': media_data.get('media-category', {}).get('@attributes', {}).get('label'),
                     'url': media['url'],
                     'tbr': int_or_none(media.get('bitrate')),
                     'filesize': int_or_none(media.get('fileSize')),

youtube_dl/extractor/extractors.py

@@ -3,6 +3,10 @@ from __future__ import unicode_literals
 
 from .abc import ABCIE
 from .abc7news import Abc7NewsIE
+from .abcnews import (
+    AbcNewsIE,
+    AbcNewsVideoIE,
+)
 from .academicearth import AcademicEarthCourseIE
 from .acast import (
     ACastIE,