From 36251bece0c8b48704ec8b7a8cb6e8987745fdfc Mon Sep 17 00:00:00 2001
From: Paul Wise <pabs3@bonedaddy.net>
Date: Tue, 8 Jan 2019 17:28:37 +0800
Subject: [PATCH] [blogger] Add support for videos and embeds

There is no discernable title field in any of the JSON or URLs,
and the title is mandatory so 'Blogger' is used as a substitute.
---
 youtube_dl/extractor/blogger.py    | 68 ++++++++++++++++++++++++++++++
 youtube_dl/extractor/extractors.py |  1 +
 youtube_dl/extractor/generic.py    | 17 ++++++++
 3 files changed, 86 insertions(+)
 create mode 100644 youtube_dl/extractor/blogger.py

diff --git a/youtube_dl/extractor/blogger.py b/youtube_dl/extractor/blogger.py
new file mode 100644
index 000000000..bb871435c
--- /dev/null
+++ b/youtube_dl/extractor/blogger.py
@@ -0,0 +1,68 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import json
+import re
+
+from .common import InfoExtractor
+from ..utils import (
+    compat_urllib_parse_urlparse as urlparse,
+    compat_parse_qs as qsparse,
+    float_or_none,
+    str_or_none,
+)
+
+
+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',
+            'ext': 'mp4',
+            'title': 'Blogger',
+            'thumbnail': r're:^https?://.*',
+        }
+    }]
+
+    @staticmethod
+    def _extract_url(webpage):
+        urls = BloggerIE._extract_urls(webpage)
+        return urls[0] if urls else None
+
+    @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 block')
+        data_json = data_json.encode('utf-8').decode('unicode_escape')
+        data = json.loads(data_json)
+        iframe_id = data.get('iframe_id', token_id)
+        thumbnail = data.get('thumbnail')
+        streams = data['streams']
+        formats = [{
+            'ext':
+                qsparse(
+                    urlparse(stream['play_url']).query
+                ).get('mime')[0].replace('video/', ''),
+            'url': stream['play_url'],
+            'format_id': str_or_none(stream.get('format_id')),
+        } for stream in streams]
+
+        return {
+            'id': iframe_id,
+            'title': 'Blogger',
+            'formats': formats,
+            'thumbnail': thumbnail,
+            'duration':
+                float_or_none(qsparse(
+                    urlparse(streams[0]['play_url']).query
+                ).get('dur')[0]),
+        }
diff --git a/youtube_dl/extractor/extractors.py b/youtube_dl/extractor/extractors.py
index 4b3092028..d7ce18d18 100644
--- a/youtube_dl/extractor/extractors.py
+++ b/youtube_dl/extractor/extractors.py
@@ -118,6 +118,7 @@ from .bleacherreport import (
     BleacherReportCMSIE,
 )
 from .blinkx import BlinkxIE
+from .blogger import BloggerIE
 from .bloomberg import BloombergIE
 from .bokecc import BokeCCIE
 from .bostonglobe import BostonGlobeIE
diff --git a/youtube_dl/extractor/generic.py b/youtube_dl/extractor/generic.py
index 355067a50..06eb3c78b 100644
--- a/youtube_dl/extractor/generic.py
+++ b/youtube_dl/extractor/generic.py
@@ -119,6 +119,7 @@ from .expressen import ExpressenIE
 from .zype import ZypeIE
 from .odnoklassniki import OdnoklassnikiIE
 from .kinja import KinjaEmbedIE
+from .blogger import BloggerIE
 
 
 class GenericIE(InfoExtractor):
@@ -2151,6 +2152,17 @@ 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
@@ -2943,6 +2955,11 @@ class GenericIE(InfoExtractor):
         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: