[blogger] Add extractor (#1629)

Authored by: pabs3

diff --git a/yt_dlp/extractor/blogger.py b/yt_dlp/extractor/blogger.py
new file mode 100644 (file)
index 0000000..dba131c
--- /dev/null
+++ b/yt_dlp/extractor/blogger.py
@@ -0,0 +1,54 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import re
+
+from ..utils import (
+    mimetype2ext,
+    parse_duration,
+    parse_qs,
+    str_or_none,
+    traverse_obj,
+)
+from .common import InfoExtractor
+
+
+class BloggerIE(InfoExtractor):
+    IE_NAME = 'blogger.com'
+    _VALID_URL = r'https?://(?:www\.)?blogger\.com/video\.g\?token=(?P<id>.+)'
+    _VALID_EMBED = r'''<iframe[^>]+src=["']((?:https?:)?//(?:www\.)?blogger\.com/video\.g\?token=[^"']+)["']'''
+    _TESTS = [{
+        'url': 'https://www.blogger.com/video.g?token=AD6v5dzEe9hfcARr5Hlq1WTkYy6t-fXH3BBahVhGvVHe5szdEUBEloSEDSTA8-b111089KbfWuBvTN7fnbxMtymsHhXAXwVvyzHH4Qch2cfLQdGxKQrrEuFpC1amSl_9GuLWODjPgw',
+        'md5': 'f1bc19b6ea1b0fd1d81e84ca9ec467ac',
+        'info_dict': {
+            'id': 'BLOGGER-video-3c740e3a49197e16-796',
+            'title': 'BLOGGER-video-3c740e3a49197e16-796',
+            'ext': 'mp4',
+            'thumbnail': r're:^https?://.*',
+            'duration': 76.068,
+        }
+    }]
+
+    @staticmethod
+    def _extract_urls(webpage):
+        return re.findall(BloggerIE._VALID_EMBED, webpage)
+
+    def _real_extract(self, url):
+        token_id = self._match_id(url)
+        webpage = self._download_webpage(url, token_id)
+        data_json = self._search_regex(r'var\s+VIDEO_CONFIG\s*=\s*(\{.*)', webpage, 'JSON data')
+        data = self._parse_json(data_json.encode('utf-8').decode('unicode_escape'), token_id)
+        streams = data['streams']
+        formats = [{
+            'ext': mimetype2ext(traverse_obj(parse_qs(stream['play_url']), ('mime', 0))),
+            'url': stream['play_url'],
+            'format_id': str_or_none(stream.get('format_id')),
+        } for stream in streams]
+
+        return {
+            'id': data.get('iframe_id', token_id),
+            'title': data.get('iframe_id', token_id),
+            'formats': formats,
+            'thumbnail': data.get('thumbnail'),
+            'duration': parse_duration(traverse_obj(parse_qs(streams[0]['play_url']), ('dur', 0))),
+        }

diff --git a/yt_dlp/extractor/extractors.py b/yt_dlp/extractor/extractors.py
index 89c61312d4f3e75b80d22a755d59df9d1ce19d67..75cb0b2ab11ed512fe9447caf84aa3e9ac1f9446 100644 (file)
--- a/yt_dlp/extractor/extractors.py
+++ b/yt_dlp/extractor/extractors.py
@@ -166,6 +166,7 @@
     BleacherReportIE,
     BleacherReportCMSIE,
 )
+from .blogger import BloggerIE
 from .bloomberg import BloombergIE
 from .bokecc import BokeCCIE
 from .bongacams import BongaCamsIE

diff --git a/yt_dlp/extractor/generic.py b/yt_dlp/extractor/generic.py
index 3374c1c200f0f1d8c240c7edb04234ca732d366c..d6631e2f3b85202d90d3cad046101021cbf37694 100644 (file)
--- a/yt_dlp/extractor/generic.py
+++ b/yt_dlp/extractor/generic.py
@@ -136,6 +136,7 @@
 from .simplecast import SimplecastIE
 from .wimtv import WimTVIE
 from .tvp import TVPEmbedIE
+from .blogger import BloggerIE
 
 
 class GenericIE(InfoExtractor):
@@ -2173,6 +2174,17 @@ class GenericIE(InfoExtractor):
                 'skip_download': True,
             },
         },
+        {
+            # blogger embed
+            'url': 'https://blog.tomeuvizoso.net/2019/01/a-panfrost-milestone.html',
+            'md5': 'f1bc19b6ea1b0fd1d81e84ca9ec467ac',
+            'info_dict': {
+                'id': 'BLOGGER-video-3c740e3a49197e16-796',
+                'ext': 'mp4',
+                'title': 'Blogger',
+                'thumbnail': r're:^https?://.*',
+            },
+        },
         # {
         #     # TODO: find another test
         #     # http://schema.org/VideoObject
@@ -3216,6 +3228,11 @@ def _real_extract(self, url):
         if onionstudios_url:
             return self.url_result(onionstudios_url)
 
+        # Look for Blogger embeds
+        blogger_urls = BloggerIE._extract_urls(webpage)
+        if blogger_urls:
+            return self.playlist_from_matches(blogger_urls, video_id, video_title, ie=BloggerIE.ie_key())
+
         # Look for ViewLift embeds
         viewlift_url = ViewLiftEmbedIE._extract_url(webpage)
         if viewlift_url: