[gedi] added huffingtonpost, added embeds

diff --git a/youtube_dlc/extractor/extractors.py b/youtube_dlc/extractor/extractors.py
index bee90d74a3e59e89c233f7d420c7996454f5b3cf..01f69c0066063c90692daf6fa247b52213761415 100644 (file)
--- a/youtube_dlc/extractor/extractors.py
+++ b/youtube_dlc/extractor/extractors.py
@@ -414,7 +414,10 @@
 from .gaskrank import GaskrankIE
 from .gazeta import GazetaIE
 from .gdcvault import GDCVaultIE
-from .gedi import GediIE
+from .gedi import (
+    GediIE,
+    GediEmbedsIE,
+)
 from .generic import GenericIE
 from .gfycat import GfycatIE
 from .giantbomb import GiantBombIE

diff --git a/youtube_dlc/extractor/gedi.py b/youtube_dlc/extractor/gedi.py
index 5efc8a6e97cacffefabc7043c5560208e5ab0bb6..f35dfca5ad3b6a7a09e2ed626b5241f3cc409831 100644 (file)
--- a/youtube_dlc/extractor/gedi.py
+++ b/youtube_dlc/extractor/gedi.py
@@ -5,6 +5,11 @@
 
 from .common import InfoExtractor
 from ..compat import compat_str
+from ..utils import (
+    base_url,
+    url_basename,
+    urljoin,
+)
 
 
 class GediBaseIE(InfoExtractor):
@@ -109,6 +114,7 @@ class GediIE(GediBaseIE):
                     (?:
                         (?:espresso\.)?repubblica
                         |lastampa
+                        |huffingtonpost
                         |ilsecoloxix
                         |iltirreno
                         |messaggeroveneto
@@ -175,3 +181,86 @@ class GediIE(GediBaseIE):
             'thumbnail': r're:^https://www\.repstatic\.it/video/photo/.+?-thumb-social-play\.jpg$',
         },
     }]
+
+
+class GediEmbedsIE(GediBaseIE):
+    _VALID_URL = r'''(?x)https?://video\.
+                    (?:
+                        (?:espresso\.)?repubblica
+                        |lastampa
+                        |huffingtonpost
+                        |ilsecoloxix
+                        |iltirreno
+                        |messaggeroveneto
+                        |ilpiccolo
+                        |gazzettadimantova
+                        |mattinopadova
+                        |laprovinciapavese
+                        |tribunatreviso
+                        |nuovavenezia
+                        |gazzettadimodena
+                        |lanuovaferrara
+                        |corrierealpi
+                        |lasentinella
+                    )
+                    (?:\.gelocal)?\.it/embed/.+?/(?P<id>[\d/]+)(?:\?|\&|$)'''
+    _TESTS = [{
+        'url': 'https://video.huffingtonpost.it/embed/politica/cotticelli-non-so-cosa-mi-sia-successo-sto-cercando-di-capire-se-ho-avuto-un-malore/29312/29276?responsive=true&el=video971040871621586700',
+        'md5': '0391c2c83c6506581003aaf0255889c0',
+        'info_dict': {
+            'id': '14772/14870',
+            'ext': 'mp4',
+            'title': 'Festival EMERGENCY, Villa: «La buona informazione aiuta la salute» (14772-14870)',
+            'description': 'md5:2bce954d278248f3c950be355b7c2226',
+            'thumbnail': r're:^https://www\.repstatic\.it/video/photo/.+?-thumb-social-play\.jpg$',
+        },
+    }]
+
+    @staticmethod
+    def _sanitize_urls(urls):
+        # add protocol if missing
+        for i, e in enumerate(urls):
+            if e.startswith('//'):
+                urls[i] = 'https:%s' % e
+        # clean iframes urls
+        for i, e in enumerate(urls):
+            urls[i] = urljoin(base_url(e), url_basename(e))
+        return urls
+
+    @staticmethod
+    def _extract_urls(webpage):
+        entries = [
+            mobj.group('url')
+            for mobj in re.finditer(r'''(?x)
+            (?:
+                data-frame-src=|
+                <iframe[^\n]+src=
+            )
+            (["'])
+                (?P<url>https?://video\.
+                    (?:
+                        (?:espresso\.)?repubblica
+                        |lastampa
+                        |huffingtonpost
+                        |ilsecoloxix
+                        |iltirreno
+                        |messaggeroveneto
+                        |ilpiccolo
+                        |gazzettadimantova
+                        |mattinopadova
+                        |laprovinciapavese
+                        |tribunatreviso
+                        |nuovavenezia
+                        |gazzettadimodena
+                        |lanuovaferrara
+                        |corrierealpi
+                        |lasentinella
+                    )
+                    (?:\.gelocal)?\.it/embed/.+?)
+            \1''', webpage)]
+        return GediEmbedsIE._sanitize_urls(entries)
+
+    @staticmethod
+    def _extract_url(webpage):
+        urls = GediEmbedsIE._extract_urls(webpage)
+        return urls[0] if urls else None

diff --git a/youtube_dlc/extractor/generic.py b/youtube_dlc/extractor/generic.py
index aba06b328e19fad48faee2d83eb42494af672f94..3fab929a83ada1872930ef57cc8fd6087f27cc15 100644 (file)
--- a/youtube_dlc/extractor/generic.py
+++ b/youtube_dlc/extractor/generic.py
@@ -119,6 +119,7 @@
 from .zype import ZypeIE
 from .odnoklassniki import OdnoklassnikiIE
 from .kinja import KinjaEmbedIE
+from .gedi import GediEmbedsIE
 
 
 class GenericIE(InfoExtractor):
@@ -3213,6 +3214,12 @@ def _real_extract(self, url):
             return self.playlist_from_matches(
                 zype_urls, video_id, video_title, ie=ZypeIE.ie_key())
 
+        # Look for RCS media group embeds
+        gedi_urls = GediEmbedsIE._extract_urls(webpage)
+        if gedi_urls:
+            return self.playlist_from_matches(
+                gedi_urls, video_id, video_title, ie=GediEmbedsIE.ie_key())
+
         # Look for HTML5 media
         entries = self._parse_html5_media_entries(url, webpage, video_id, m3u8_id='hls')
         if entries: