fix/speedup ci

Original author: dstftw <dstftw@gmail.com>

.gitignore

@@ -15,6 +15,7 @@ haruhi-dl.1
 haruhi-dl.bash-completion
 haruhi-dl.fish
 haruhi_dl/extractor/lazy_extractors.py
+haruhi_dl/extractor_artifacts/
 haruhi-dl
 haruhi-dl.exe
 haruhi-dl.tar.gz

.gitlab-ci.yml

@@ -1,5 +1,6 @@
 default:
   before_script:
+    - sed -i "s@dl-cdn.alpinelinux.org@alpine.sakamoto.pl@g" /etc/apk/repositories
     - apk add bash
     - pip install nose
 

ChangeLog

@@ -1,3 +1,20 @@
+version 2021.08.01
+Extractor
+* [youtube] fixed agegate
+* [niconico] dmc downloader from youtube-dlp
+* [peertube] new URL schemas
+
+version 2021.06.20
+Core
+* [playwright] fixed headlessness
++ [playwright] option to force a specific browser
+
+Extractor
+* [tiktok] fix empty video lists
+* [youtube] fix and speed-up age-gate circumvention
+* [youtube] fix videos with JS-like syntax
+
+
 version 2021.06.01
 Core
 * merging formats by codecs

devscripts/ExtractCryptoYT.sh

@@ -1,6 +1,6 @@
 #!/bin/bash
 data="$(curl -s "https://www.youtube.com/s/player/$1/player_ias.vflset/en_GB/base.js")"
-func="$(grep -P '[a-z]\=a\.split.*a\.join' <<< "$data")"
+func="$(grep -P '[a-z]\=a\.split\([\"'"'"']{2}.*a\.join' <<< "$data")"
 echo "full extracted function: $func"
 
 obfuscatedName="$(grep -Poh '\(""\);[A-Za-z]+' <<< "$func" | sed -s 's/("");//')"

devscripts/prerelease_codegen.py

@@ -0,0 +1,32 @@
+# this is intended to speed-up some extractors,
+# which sometimes need to extract some data that doesn't change very much often,
+# but it does on random times, like youtube's signature "crypto" or soundcloud's client id
+
+import os
+from os.path import dirname as dirn
+import sys
+sys.path.insert(0, dirn(dirn((os.path.abspath(__file__)))))
+
+from haruhi_dl import HaruhiDL
+from haruhi_dl.utils import (
+    ExtractorError,
+)
+
+hdl = HaruhiDL(params={
+    'quiet': True,
+})
+artifact_dir = os.path.join(dirn(dirn((os.path.abspath(__file__)))), 'haruhi_dl', 'extractor_artifacts')
+if not os.path.exists(artifact_dir):
+    os.mkdir(artifact_dir)
+
+for ie_name in (
+    'Youtube',
+    'Soundcloud',
+):
+    ie = hdl.get_info_extractor(ie_name)
+    try:
+        file_contents = ie._generate_prerelease_file()
+        with open(os.path.join(artifact_dir, ie_name.lower() + '.py'), 'w') as file:
+            file.write(file_contents)
+    except ExtractorError as err:
+        print(err)

devscripts/release.sh

@@ -1,141 +1,24 @@
 #!/bin/bash
 
-# IMPORTANT: the following assumptions are made
-# * the GH repo is on the origin remote
-# * the gh-pages branch is named so locally
-# * the git config user.signingkey is properly set
-
-# You will need
-# pip install coverage nose rsa wheel
-
-# TODO
-# release notes
-# make hash on local files
-
-set -e
-
-skip_tests=true
-gpg_sign_commits=""
-buildserver='localhost:8142'
-
-while true
-do
-case "$1" in
-    --run-tests)
-        skip_tests=false
-        shift
-    ;;
-    --gpg-sign-commits|-S)
-        gpg_sign_commits="-S"
-        shift
-    ;;
-    --buildserver)
-        buildserver="$2"
-        shift 2
-    ;;
-    --*)
-        echo "ERROR: unknown option $1"
-        exit 1
-    ;;
-    *)
-        break
-    ;;
-esac
-done
-
-if [ -z "$1" ]; then echo "ERROR: specify version number like this: $0 1994.09.06"; exit 1; fi
-version="$1"
-major_version=$(echo "$version" | sed -n 's#^\([0-9]*\.[0-9]*\.[0-9]*\).*#\1#p')
-if test "$major_version" '!=' "$(date '+%Y.%m.%d')"; then
-    echo "$version does not start with today's date!"
-    exit 1
+if [[ "$(basename $(pwd))" == 'devscripts' ]]; then
+	cd ..
 fi
 
-if [ ! -z "`git tag | grep "$version"`" ]; then echo 'ERROR: version already present'; exit 1; fi
-if [ ! -z "`git status --porcelain | grep -v CHANGELOG`" ]; then echo 'ERROR: the working directory is not clean; commit or stash changes'; exit 1; fi
-useless_files=$(find haruhi_dl -type f -not -name '*.py')
-if [ ! -z "$useless_files" ]; then echo "ERROR: Non-.py files in haruhi_dl: $useless_files"; exit 1; fi
-if [ ! -f "updates_key.pem" ]; then echo 'ERROR: updates_key.pem missing'; exit 1; fi
-if ! type pandoc >/dev/null 2>/dev/null; then echo 'ERROR: pandoc is missing'; exit 1; fi
-if ! python3 -c 'import rsa' 2>/dev/null; then echo 'ERROR: python3-rsa is missing'; exit 1; fi
-if ! python3 -c 'import wheel' 2>/dev/null; then echo 'ERROR: wheel is missing'; exit 1; fi
+v="$(date "+%Y.%m.%d")"
 
-read -p "Is ChangeLog up to date? (y/n) " -n 1
-if [[ ! $REPLY =~ ^[Yy]$ ]]; then exit 1; fi
-
-/bin/echo -e "\n### First of all, testing..."
-make clean
-if $skip_tests ; then
-    echo 'SKIPPING TESTS'
-else
-    nosetests --verbose --with-coverage --cover-package=haruhi_dl --cover-html test --stop || exit 1
+if [[ "$(grep "'$v" haruhi_dl/version.py)" != '' ]]; then #' is this the first release of the day?
+	if [[ "$(grep -Poh '[0-9]{4}\.[0-9]{2}\.[0-9]{2}\.[0-9]' haruhi_dl/version.py)" != '' ]]; then # so, 2nd or nth?
+		v="$v.$(($(cat haruhi_dl/version.py | grep -Poh '[0-9]{4}\.[0-9]{2}\.[0-9]{2}\.[0-9]' | grep -Poh '[0-9]+$')+1))"
+	else
+		v="$v.1"
+	fi
 fi
 
-/bin/echo -e "\n### Changing version in version.py..."
-sed -i "s/__version__ = '.*'/__version__ = '$version'/" haruhi_dl/version.py
+sed "s/__version__ = '.*'/__version__ = '$v'/g" -i haruhi_dl/version.py
 
-/bin/echo -e "\n### Changing version in ChangeLog..."
-sed -i "s/<unreleased>/$version/" ChangeLog
-
-/bin/echo -e "\n### Committing documentation, templates and haruhi_dl/version.py..."
-make README.md CONTRIBUTING.md issuetemplates supportedsites
-git add README.md CONTRIBUTING.md .github/ISSUE_TEMPLATE/1_broken_site.md .github/ISSUE_TEMPLATE/2_site_support_request.md .github/ISSUE_TEMPLATE/3_site_feature_request.md .github/ISSUE_TEMPLATE/4_bug_report.md .github/ISSUE_TEMPLATE/5_feature_request.md .github/ISSUE_TEMPLATE/6_question.md docs/supportedsites.md haruhi_dl/version.py ChangeLog
-git commit $gpg_sign_commits -m "release $version"
-
-/bin/echo -e "\n### Now tagging, signing and pushing..."
-git tag -s -m "Release $version" "$version"
-git show "$version"
-read -p "Is it good, can I push? (y/n) " -n 1
-if [[ ! $REPLY =~ ^[Yy]$ ]]; then exit 1; fi
-echo
-MASTER=$(git rev-parse --abbrev-ref HEAD)
-git push origin $MASTER:master
-git push origin "$version"
-
-/bin/echo -e "\n### OK, now it is time to build the binaries..."
-REV=$(git rev-parse HEAD)
-make haruhi-dl haruhi-dl.tar.gz
-read -p "VM running? (y/n) " -n 1
-wget "http://$buildserver/build/ytdl-org/haruhi-dl/haruhi-dl.exe?rev=$REV" -O haruhi-dl.exe
-mkdir -p "build/$version"
-mv haruhi-dl haruhi-dl.exe "build/$version"
-mv haruhi-dl.tar.gz "build/$version/haruhi-dl-$version.tar.gz"
-RELEASE_FILES="haruhi-dl haruhi-dl.exe haruhi-dl-$version.tar.gz"
-(cd build/$version/ && md5sum $RELEASE_FILES > MD5SUMS)
-(cd build/$version/ && sha1sum $RELEASE_FILES > SHA1SUMS)
-(cd build/$version/ && sha256sum $RELEASE_FILES > SHA2-256SUMS)
-(cd build/$version/ && sha512sum $RELEASE_FILES > SHA2-512SUMS)
-
-/bin/echo -e "\n### Signing and uploading the new binaries to GitHub..."
-for f in $RELEASE_FILES; do gpg --passphrase-repeat 5 --detach-sig "build/$version/$f"; done
-
-ROOT=$(pwd)
-python devscripts/create-github-release.py ChangeLog $version "$ROOT/build/$version"
-
-#ssh ytdl@yt-dl.org "sh html/update_latest.sh $version"
-
-/bin/echo -e "\n### Now switching to gh-pages..."
-git clone --branch gh-pages --single-branch . build/gh-pages
-(
-    set -e
-    ORIGIN_URL=$(git config --get remote.origin.url)
-    cd build/gh-pages
-    "$ROOT/devscripts/gh-pages/add-version.py" $version
-    "$ROOT/devscripts/gh-pages/update-feed.py"
-    "$ROOT/devscripts/gh-pages/sign-versions.py" < "$ROOT/updates_key.pem"
-    "$ROOT/devscripts/gh-pages/generate-download.py"
-    "$ROOT/devscripts/gh-pages/update-copyright.py"
-    "$ROOT/devscripts/gh-pages/update-sites.py"
-    git add *.html *.html.in update
-    git commit $gpg_sign_commits -m "release $version"
-    git push "$ROOT" gh-pages
-    git push "$ORIGIN_URL" gh-pages
-)
-rm -rf build
-
-make pypi-files
-echo "Uploading to PyPi ..."
-python setup.py sdist bdist_wheel upload
-make clean
-
-/bin/echo -e "\n### DONE!"
+python3 setup.py build_lazy_extractors
+python3 devscripts/prerelease_codegen.py
+rm -R build dist
+python3 setup.py sdist bdist_wheel
+python3 -m twine upload dist/*
+devscripts/wine-py2exe.sh setup.py

haruhi_dl/__init__.py

@@ -178,6 +178,8 @@ def _real_main(argv=None):
         parser.error('Unsupported TV Provider, use --ap-list-mso to get a list of supported TV Providers')
     if opts.force_generic_extractor and opts.force_use_mastodon:
         parser.error('force either generic extractor or Mastodon')
+    if opts.force_playwright_browser not in ('firefox', 'chromium', 'webkit', None):
+        parser.error('invalid browser forced, must be on of: firefox, chromium, webkit')
 
     def parse_retries(retries):
         if retries in ('inf', 'infinite'):
@@ -424,6 +426,7 @@ def _real_main(argv=None):
         'headless_playwright': opts.headless_playwright,
         'sleep_interval': opts.sleep_interval,
         'max_sleep_interval': opts.max_sleep_interval,
+        'force_playwright_browser': opts.force_playwright_browser,
         'external_downloader': opts.external_downloader,
         'list_thumbnails': opts.list_thumbnails,
         'playlist_items': opts.playlist_items,

haruhi_dl/downloader/__init__.py

@@ -1,5 +1,18 @@
 from __future__ import unicode_literals
 
+from ..utils import (
+    determine_protocol,
+)
+
+
+def _get_real_downloader(info_dict, protocol=None, *args, **kwargs):
+    info_copy = info_dict.copy()
+    if protocol:
+        info_copy['protocol'] = protocol
+    return get_suitable_downloader(info_copy, *args, **kwargs)
+
+
+# Some of these require _get_real_downloader
 from .common import FileDownloader
 from .f4m import F4mFD
 from .hls import HlsFD
@@ -8,16 +21,13 @@ from .rtmp import RtmpFD
 from .dash import DashSegmentsFD
 from .rtsp import RtspFD
 from .ism import IsmFD
+from .niconico import NiconicoDmcFD
 from .external import (
     get_external_downloader,
     Aria2cFD,
     FFmpegFD,
 )
 
-from ..utils import (
-    determine_protocol,
-)
-
 PROTOCOL_MAP = {
     'rtmp': RtmpFD,
     'm3u8_native': HlsFD,
@@ -28,6 +38,7 @@ PROTOCOL_MAP = {
     'http_dash_segments': DashSegmentsFD,
     'ism': IsmFD,
     'bittorrent': Aria2cFD,
+    'niconico_dmc': NiconicoDmcFD,
 }
 
 

haruhi_dl/downloader/niconico.py

@@ -0,0 +1,55 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import threading
+
+from .common import FileDownloader
+from ..downloader import _get_real_downloader
+from ..extractor.niconico import NiconicoIE
+from ..compat import compat_urllib_request
+
+
+class NiconicoDmcFD(FileDownloader):
+    """ Downloading niconico douga from DMC with heartbeat """
+
+    FD_NAME = 'niconico_dmc'
+
+    def real_download(self, filename, info_dict):
+        self.to_screen('[%s] Downloading from DMC' % self.FD_NAME)
+
+        ie = NiconicoIE(self.hdl)
+        info_dict, heartbeat_info_dict = ie._get_heartbeat_info(info_dict)
+
+        fd = _get_real_downloader(info_dict, params=self.params)(self.hdl, self.params)
+
+        success = download_complete = False
+        timer = [None]
+        heartbeat_lock = threading.Lock()
+        heartbeat_url = heartbeat_info_dict['url']
+        heartbeat_data = heartbeat_info_dict['data'].encode()
+        heartbeat_interval = heartbeat_info_dict.get('interval', 30)
+
+        def heartbeat():
+            try:
+                compat_urllib_request.urlopen(url=heartbeat_url, data=heartbeat_data)
+            except Exception:
+                self.to_screen('[%s] Heartbeat failed' % self.FD_NAME)
+
+            with heartbeat_lock:
+                if not download_complete:
+                    timer[0] = threading.Timer(heartbeat_interval, heartbeat)
+                    timer[0].start()
+
+        heartbeat_info_dict['ping']()
+        self.to_screen('[%s] Heartbeat with %d second interval ...' % (self.FD_NAME, heartbeat_interval))
+        try:
+            heartbeat()
+            if type(fd).__name__ == 'HlsFD':
+                info_dict.update(ie._extract_m3u8_formats(info_dict['url'], info_dict['id'])[0])
+            success = fd.real_download(filename, info_dict)
+        finally:
+            if heartbeat_lock:
+                with heartbeat_lock:
+                    timer[0].cancel()
+                    download_complete = True
+            return success

haruhi_dl/extractor/appleconnect.py

@@ -9,10 +9,10 @@ from ..utils import (
 
 
 class AppleConnectIE(InfoExtractor):
-    _VALID_URL = r'https?://itunes\.apple\.com/\w{0,2}/?post/idsa\.(?P<id>[\w-]+)'
-    _TEST = {
+    _VALID_URL = r'https?://itunes\.apple\.com/\w{0,2}/?post/(?:id)?sa\.(?P<id>[\w-]+)'
+    _TESTS = [{
         'url': 'https://itunes.apple.com/us/post/idsa.4ab17a39-2720-11e5-96c5-a5b38f6c42d3',
-        'md5': 'e7c38568a01ea45402570e6029206723',
+        'md5': 'c1d41f72c8bcaf222e089434619316e4',
         'info_dict': {
             'id': '4ab17a39-2720-11e5-96c5-a5b38f6c42d3',
             'ext': 'm4v',
@@ -22,7 +22,10 @@ class AppleConnectIE(InfoExtractor):
             'upload_date': '20150710',
             'timestamp': 1436545535,
         },
-    }
+    }, {
+        'url': 'https://itunes.apple.com/us/post/sa.0fe0229f-2457-11e5-9f40-1bb645f2d5d9',
+        'only_matching': True,
+    }]
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
@@ -36,7 +39,7 @@ class AppleConnectIE(InfoExtractor):
 
         video_data = self._parse_json(video_json, video_id)
         timestamp = str_to_int(self._html_search_regex(r'data-timestamp="(\d+)"', webpage, 'timestamp'))
-        like_count = str_to_int(self._html_search_regex(r'(\d+) Loves', webpage, 'like count'))
+        like_count = str_to_int(self._html_search_regex(r'(\d+) Loves', webpage, 'like count', default=None))
 
         return {
             'id': video_id,

haruhi_dl/extractor/bilibili.py

@@ -233,7 +233,7 @@ class BiliBiliIE(InfoExtractor):
             webpage)
         if uploader_mobj:
             info.update({
-                'uploader': uploader_mobj.group('name'),
+                'uploader': uploader_mobj.group('name').strip(),
                 'uploader_id': uploader_mobj.group('id'),
             })
         if not info.get('uploader'):

haruhi_dl/extractor/curiositystream.py

@@ -145,7 +145,7 @@ class CuriosityStreamIE(CuriosityStreamBaseIE):
 
 class CuriosityStreamCollectionIE(CuriosityStreamBaseIE):
     IE_NAME = 'curiositystream:collection'
-    _VALID_URL = r'https?://(?:app\.)?curiositystream\.com/(?:collection|series)/(?P<id>\d+)'
+    _VALID_URL = r'https?://(?:app\.)?curiositystream\.com/(?:collections?|series)/(?P<id>\d+)'
     _TESTS = [{
         'url': 'https://app.curiositystream.com/collection/2',
         'info_dict': {
@@ -157,6 +157,9 @@ class CuriosityStreamCollectionIE(CuriosityStreamBaseIE):
     }, {
         'url': 'https://curiositystream.com/series/2',
         'only_matching': True,
+    }, {
+        'url': 'https://curiositystream.com/collections/36',
+        'only_matching': True,
     }]
 
     def _real_extract(self, url):

haruhi_dl/extractor/egghead.py

@@ -22,16 +22,19 @@ class EggheadBaseIE(InfoExtractor):
 class EggheadCourseIE(EggheadBaseIE):
     IE_DESC = 'egghead.io course'
     IE_NAME = 'egghead:course'
-    _VALID_URL = r'https://egghead\.io/courses/(?P<id>[^/?#&]+)'
-    _TEST = {
+    _VALID_URL = r'https://(?:app\.)?egghead\.io/(?:course|playlist)s/(?P<id>[^/?#&]+)'
+    _TESTS = [{
         'url': 'https://egghead.io/courses/professor-frisby-introduces-composable-functional-javascript',
         'playlist_count': 29,
         'info_dict': {
-            'id': '72',
+            'id': '432655',
             'title': 'Professor Frisby Introduces Composable Functional JavaScript',
             'description': 're:(?s)^This course teaches the ubiquitous.*You\'ll start composing functionality before you know it.$',
         },
-    }
+    }, {
+        'url': 'https://app.egghead.io/playlists/professor-frisby-introduces-composable-functional-javascript',
+        'only_matching': True,
+    }]
 
     def _real_extract(self, url):
         playlist_id = self._match_id(url)
@@ -65,7 +68,7 @@ class EggheadCourseIE(EggheadBaseIE):
 class EggheadLessonIE(EggheadBaseIE):
     IE_DESC = 'egghead.io lesson'
     IE_NAME = 'egghead:lesson'
-    _VALID_URL = r'https://egghead\.io/(?:api/v1/)?lessons/(?P<id>[^/?#&]+)'
+    _VALID_URL = r'https://(?:app\.)?egghead\.io/(?:api/v1/)?lessons/(?P<id>[^/?#&]+)'
     _TESTS = [{
         'url': 'https://egghead.io/lessons/javascript-linear-data-flow-with-container-style-types-box',
         'info_dict': {
@@ -88,6 +91,9 @@ class EggheadLessonIE(EggheadBaseIE):
     }, {
         'url': 'https://egghead.io/api/v1/lessons/react-add-redux-to-a-react-application',
         'only_matching': True,
+    }, {
+        'url': 'https://app.egghead.io/lessons/javascript-linear-data-flow-with-container-style-types-box',
+        'only_matching': True,
     }]
 
     def _real_extract(self, url):

haruhi_dl/extractor/extractors.py

@@ -643,10 +643,6 @@ from .linkedin import (
 from .linuxacademy import LinuxAcademyIE
 from .litv import LiTVIE
 from .livejournal import LiveJournalIE
-from .liveleak import (
-    LiveLeakIE,
-    LiveLeakEmbedIE,
-)
 from .livestream import (
     LivestreamIE,
     LivestreamOriginalIE,
@@ -1517,6 +1513,7 @@ from .videomore import (
 )
 from .videopress import VideoPressIE
 from .videotarget import VideoTargetIE
+from .vider import ViderIE
 from .vidio import VidioIE
 from .vidlii import VidLiiIE
 from .vidme import (

haruhi_dl/extractor/facebook.py

@@ -521,7 +521,10 @@ class FacebookIE(InfoExtractor):
                 raise ExtractorError(
                     'The video is not available, Facebook said: "%s"' % m_msg.group(1),
                     expected=True)
-            elif '>You must log in to continue' in webpage:
+            elif any(p in webpage for p in (
+                    '>You must log in to continue',
+                    'id="login_form"',
+                    'id="loginbutton"')):
                 self.raise_login_required()
 
         if not video_data and '/watchparty/' in url:

haruhi_dl/extractor/formula1.py

@@ -5,29 +5,23 @@ from .common import InfoExtractor
 
 
 class Formula1IE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?formula1\.com/(?:content/fom-website/)?en/video/\d{4}/\d{1,2}/(?P<id>.+?)\.html'
-    _TESTS = [{
-        'url': 'http://www.formula1.com/content/fom-website/en/video/2016/5/Race_highlights_-_Spain_2016.html',
-        'md5': '8c79e54be72078b26b89e0e111c0502b',
+    _VALID_URL = r'https?://(?:www\.)?formula1\.com/en/latest/video\.[^.]+\.(?P<id>\d+)\.html'
+    _TEST = {
+        'url': 'https://www.formula1.com/en/latest/video.race-highlights-spain-2016.6060988138001.html',
+        'md5': 'be7d3a8c2f804eb2ab2aa5d941c359f8',
         'info_dict': {
-            'id': 'JvYXJpMzE6pArfHWm5ARp5AiUmD-gibV',
+            'id': '6060988138001',
             'ext': 'mp4',
             'title': 'Race highlights - Spain 2016',
+            'timestamp': 1463332814,
+            'upload_date': '20160515',
+            'uploader_id': '6057949432001',
         },
-        'params': {
-            # m3u8 download
-            'skip_download': True,
-        },
-        'add_ie': ['Ooyala'],
-    }, {
-        'url': 'http://www.formula1.com/en/video/2016/5/Race_highlights_-_Spain_2016.html',
-        'only_matching': True,
-    }]
+        'add_ie': ['BrightcoveNew'],
+    }
+    BRIGHTCOVE_URL_TEMPLATE = 'http://players.brightcove.net/6057949432001/S1WMrhjlh_default/index.html?videoId=%s'
 
     def _real_extract(self, url):
-        display_id = self._match_id(url)
-        webpage = self._download_webpage(url, display_id)
-        ooyala_embed_code = self._search_regex(
-            r'data-videoid="([^"]+)"', webpage, 'ooyala embed code')
+        bc_id = self._match_id(url)
         return self.url_result(
-            'ooyala:%s' % ooyala_embed_code, 'Ooyala', ooyala_embed_code)
+            self.BRIGHTCOVE_URL_TEMPLATE % bc_id, 'BrightcoveNew', bc_id)

haruhi_dl/extractor/generic.py

@@ -84,7 +84,6 @@ from .jwplatform import JWPlatformIE
 from .digiteka import DigitekaIE
 from .arkena import ArkenaIE
 from .instagram import InstagramIE
-from .liveleak import LiveLeakIE
 from .threeqsdn import ThreeQSDNIE
 from .theplatform import ThePlatformIE
 from .kaltura import KalturaIE
@@ -1640,34 +1639,6 @@ class GenericIE(InfoExtractor):
                 'upload_date': '20160409',
             },
         },
-        # LiveLeak embed
-        {
-            'url': 'http://www.wykop.pl/link/3088787/',
-            'md5': '7619da8c820e835bef21a1efa2a0fc71',
-            'info_dict': {
-                'id': '874_1459135191',
-                'ext': 'mp4',
-                'title': 'Man shows poor quality of new apartment building',
-                'description': 'The wall is like a sand pile.',
-                'uploader': 'Lake8737',
-            },
-            'add_ie': [LiveLeakIE.ie_key()],
-            'params': {
-                'force_generic_extractor': True,
-            },
-        },
-        # Another LiveLeak embed pattern (#13336)
-        {
-            'url': 'https://milo.yiannopoulos.net/2017/06/concealed-carry-robbery/',
-            'info_dict': {
-                'id': '2eb_1496309988',
-                'ext': 'mp4',
-                'title': 'Thief robs place where everyone was armed',
-                'description': 'md5:694d73ee79e535953cf2488562288eee',
-                'uploader': 'brazilwtf',
-            },
-            'add_ie': [LiveLeakIE.ie_key()],
-        },
         # Duplicated embedded video URLs
         {
             'url': 'http://www.hudl.com/athlete/2538180/highlights/149298443',
@@ -2744,7 +2715,6 @@ class GenericIE(InfoExtractor):
             SoundcloudEmbedIE,
             TuneInBaseIE,
             JWPlatformIE,
-            LiveLeakIE,
             DBTVIE,
             VideaIE,
             TwentyMinutenIE,

haruhi_dl/extractor/ipla.py

@@ -8,6 +8,7 @@ from .common import InfoExtractor
 from ..utils import (
     int_or_none,
     url_or_none,
+    ExtractorError,
 )
 
 
@@ -79,7 +80,11 @@ class IplaIE(InfoExtractor):
             'Content-type': 'application/json'
         }
 
-        res = self._download_json('http://b2c-mobile.redefine.pl/rpc/navigation/', media_id, data=req, headers=headers)
+        res = self._download_json('https://b2c-mobile.redefine.pl/rpc/navigation/', media_id, data=req, headers=headers)
+        if not res.get('result'):
+            if res['error']['code'] == 13404:
+                raise ExtractorError('Video requires DRM protection', expected=True)
+            raise ExtractorError(f"Ipla said: {res['error']['message']} - {res['error']['data']['userMessage']}")
         return res['result']['mediaItem']
 
     def get_url(self, media_id, source_id):
@@ -93,4 +98,6 @@ class IplaIE(InfoExtractor):
         }
 
         res = self._download_json('https://b2c-mobile.redefine.pl/rpc/drm/', media_id, data=req, headers=headers)
+        if not res.get('result'):
+            raise ExtractorError(f"Ipla said: {res['error']['message']} - {res['error']['data']['userMessage']}")
         return res['result']['url']

haruhi_dl/extractor/liveleak.py

@@ -1,191 +0,0 @@
-from __future__ import unicode_literals
-
-import re
-
-from .common import InfoExtractor
-from ..utils import int_or_none
-
-
-class LiveLeakIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:\w+\.)?liveleak\.com/view\?.*?\b[it]=(?P<id>[\w_]+)'
-    _TESTS = [{
-        'url': 'http://www.liveleak.com/view?i=757_1364311680',
-        'md5': '0813c2430bea7a46bf13acf3406992f4',
-        'info_dict': {
-            'id': '757_1364311680',
-            'ext': 'mp4',
-            'description': 'extremely bad day for this guy..!',
-            'uploader': 'ljfriel2',
-            'title': 'Most unlucky car accident',
-            'thumbnail': r're:^https?://.*\.jpg$'
-        }
-    }, {
-        'url': 'http://www.liveleak.com/view?i=f93_1390833151',
-        'md5': 'd3f1367d14cc3c15bf24fbfbe04b9abf',
-        'info_dict': {
-            'id': 'f93_1390833151',
-            'ext': 'mp4',
-            'description': 'German Television Channel NDR does an exclusive interview with Edward Snowden.\r\nUploaded on LiveLeak cause German Television thinks the rest of the world isn\'t intereseted in Edward Snowden.',
-            'uploader': 'ARD_Stinkt',
-            'title': 'German Television does first Edward Snowden Interview (ENGLISH)',
-            'thumbnail': r're:^https?://.*\.jpg$'
-        }
-    }, {
-        # Prochan embed
-        'url': 'http://www.liveleak.com/view?i=4f7_1392687779',
-        'md5': '42c6d97d54f1db107958760788c5f48f',
-        'info_dict': {
-            'id': '4f7_1392687779',
-            'ext': 'mp4',
-            'description': "The guy with the cigarette seems amazingly nonchalant about the whole thing...  I really hope my friends' reactions would be a bit stronger.\r\n\r\nAction-go to 0:55.",
-            'uploader': 'CapObveus',
-            'title': 'Man is Fatally Struck by Reckless Car While Packing up a Moving Truck',
-            'age_limit': 18,
-        },
-        'skip': 'Video is dead',
-    }, {
-        # Covers https://github.com/ytdl-org/youtube-dl/pull/5983
-        # Multiple resolutions
-        'url': 'http://www.liveleak.com/view?i=801_1409392012',
-        'md5': 'c3a449dbaca5c0d1825caecd52a57d7b',
-        'info_dict': {
-            'id': '801_1409392012',
-            'ext': 'mp4',
-            'description': 'Happened on 27.7.2014. \r\nAt 0:53 you can see people still swimming at near beach.',
-            'uploader': 'bony333',
-            'title': 'Crazy Hungarian tourist films close call waterspout in Croatia',
-            'thumbnail': r're:^https?://.*\.jpg$'
-        }
-    }, {
-        # Covers https://github.com/ytdl-org/youtube-dl/pull/10664#issuecomment-247439521
-        'url': 'http://m.liveleak.com/view?i=763_1473349649',
-        'add_ie': ['Youtube'],
-        'info_dict': {
-            'id': '763_1473349649',
-            'ext': 'mp4',
-            'title': 'Reporters and public officials ignore epidemic of black on asian violence in Sacramento | Colin Flaherty',
-            'description': 'Colin being the warrior he is and showing the injustice Asians in Sacramento are being subjected to.',
-            'uploader': 'Ziz',
-            'upload_date': '20160908',
-            'uploader_id': 'UCEbta5E_jqlZmEJsriTEtnw'
-        },
-        'params': {
-            'skip_download': True,
-        },
-    }, {
-        'url': 'https://www.liveleak.com/view?i=677_1439397581',
-        'info_dict': {
-            'id': '677_1439397581',
-            'title': 'Fuel Depot in China Explosion caught on video',
-        },
-        'playlist_count': 3,
-    }, {
-        'url': 'https://www.liveleak.com/view?t=HvHi_1523016227',
-        'only_matching': True,
-    }, {
-        # No original video
-        'url': 'https://www.liveleak.com/view?t=C26ZZ_1558612804',
-        'only_matching': True,
-    }]
-
-    @staticmethod
-    def _extract_urls(webpage, **kwargs):
-        return re.findall(
-            r'<iframe[^>]+src="(https?://(?:\w+\.)?liveleak\.com/ll_embed\?[^"]*[ift]=[\w_]+[^"]+)"',
-            webpage)
-
-    def _real_extract(self, url):
-        video_id = self._match_id(url)
-        webpage = self._download_webpage(url, video_id)
-
-        video_title = self._og_search_title(webpage).replace('LiveLeak.com -', '').strip()
-        video_description = self._og_search_description(webpage)
-        video_uploader = self._html_search_regex(
-            r'By:.*?(\w+)</a>', webpage, 'uploader', fatal=False)
-        age_limit = int_or_none(self._search_regex(
-            r'you confirm that you are ([0-9]+) years and over.',
-            webpage, 'age limit', default=None))
-        video_thumbnail = self._og_search_thumbnail(webpage)
-
-        entries = self._parse_html5_media_entries(url, webpage, video_id)
-        if not entries:
-            # Maybe an embed?
-            embed_url = self._search_regex(
-                r'<iframe[^>]+src="((?:https?:)?//(?:www\.)?(?:prochan|youtube)\.com/embed[^"]+)"',
-                webpage, 'embed URL')
-            return {
-                '_type': 'url_transparent',
-                'url': embed_url,
-                'id': video_id,
-                'title': video_title,
-                'description': video_description,
-                'uploader': video_uploader,
-                'age_limit': age_limit,
-            }
-
-        for idx, info_dict in enumerate(entries):
-            formats = []
-            for a_format in info_dict['formats']:
-                if not a_format.get('height'):
-                    a_format['height'] = int_or_none(self._search_regex(
-                        r'([0-9]+)p\.mp4', a_format['url'], 'height label',
-                        default=None))
-                formats.append(a_format)
-
-                # Removing '.*.mp4' gives the raw video, which is essentially
-                # the same video without the LiveLeak logo at the top (see
-                # https://github.com/ytdl-org/youtube-dl/pull/4768)
-                orig_url = re.sub(r'\.mp4\.[^.]+', '', a_format['url'])
-                if a_format['url'] != orig_url:
-                    format_id = a_format.get('format_id')
-                    format_id = 'original' + ('-' + format_id if format_id else '')
-                    if self._is_valid_url(orig_url, video_id, format_id):
-                        formats.append({
-                            'format_id': format_id,
-                            'url': orig_url,
-                            'preference': 1,
-                        })
-            self._sort_formats(formats)
-            info_dict['formats'] = formats
-
-            # Don't append entry ID for one-video pages to keep backward compatibility
-            if len(entries) > 1:
-                info_dict['id'] = '%s_%s' % (video_id, idx + 1)
-            else:
-                info_dict['id'] = video_id
-
-            info_dict.update({
-                'title': video_title,
-                'description': video_description,
-                'uploader': video_uploader,
-                'age_limit': age_limit,
-                'thumbnail': video_thumbnail,
-            })
-
-        return self.playlist_result(entries, video_id, video_title)
-
-
-class LiveLeakEmbedIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?liveleak\.com/ll_embed\?.*?\b(?P<kind>[ift])=(?P<id>[\w_]+)'
-
-    # See generic.py for actual test cases
-    _TESTS = [{
-        'url': 'https://www.liveleak.com/ll_embed?i=874_1459135191',
-        'only_matching': True,
-    }, {
-        'url': 'https://www.liveleak.com/ll_embed?f=ab065df993c1',
-        'only_matching': True,
-    }]
-
-    def _real_extract(self, url):
-        kind, video_id = re.match(self._VALID_URL, url).groups()
-
-        if kind == 'f':
-            webpage = self._download_webpage(url, video_id)
-            liveleak_url = self._search_regex(
-                r'(?:logourl\s*:\s*|window\.open\()(?P<q1>[\'"])(?P<url>%s)(?P=q1)' % LiveLeakIE._VALID_URL,
-                webpage, 'LiveLeak URL', group='url')
-        else:
-            liveleak_url = 'http://www.liveleak.com/view?%s=%s' % (kind, video_id)
-
-        return self.url_result(liveleak_url, ie=LiveLeakIE.ie_key())

haruhi_dl/extractor/niconico.py

@@ -1,25 +1,28 @@
+
 # coding: utf-8
 from __future__ import unicode_literals
 
-import datetime
-import functools
+import re
 import json
-import math
+import datetime
 
 from .common import InfoExtractor
+from ..postprocessor.ffmpeg import FFmpegPostProcessor
 from ..compat import (
+    compat_str,
     compat_parse_qs,
     compat_urllib_parse_urlparse,
 )
 from ..utils import (
-    determine_ext,
     dict_get,
     ExtractorError,
-    float_or_none,
-    InAdvancePagedList,
     int_or_none,
+    float_or_none,
+    OnDemandPagedList,
     parse_duration,
     parse_iso8601,
+    PostProcessingError,
+    str_or_none,
     remove_start,
     try_get,
     unified_timestamp,
@@ -34,7 +37,7 @@ class NiconicoIE(InfoExtractor):
 
     _TESTS = [{
         'url': 'http://www.nicovideo.jp/watch/sm22312215',
-        'md5': 'd1a75c0823e2f629128c43e1212760f9',
+        'md5': 'a5bad06f1347452102953f323c69da34s',
         'info_dict': {
             'id': 'sm22312215',
             'ext': 'mp4',
@@ -162,6 +165,11 @@ class NiconicoIE(InfoExtractor):
     _VALID_URL = r'https?://(?:www\.|secure\.|sp\.)?nicovideo\.jp/watch/(?P<id>(?:[a-z]{2})?[0-9]+)'
     _NETRC_MACHINE = 'niconico'
 
+    _API_HEADERS = {
+        'X-Frontend-ID': '6',
+        'X-Frontend-Version': '0'
+    }
+
     def _real_initialize(self):
         self._login()
 
@@ -188,40 +196,92 @@ class NiconicoIE(InfoExtractor):
             if compat_parse_qs(parts.query).get('message', [None])[0] == 'cant_login':
                 login_ok = False
         if not login_ok:
-            self._downloader.report_warning('unable to log in: bad username or password')
+            self.report_warning('unable to log in: bad username or password')
         return login_ok
 
-    def _extract_format_for_quality(self, api_data, video_id, audio_quality, video_quality):
-        def yesno(boolean):
-            return 'yes' if boolean else 'no'
+    def _get_heartbeat_info(self, info_dict):
 
-        session_api_data = api_data['video']['dmcInfo']['session_api']
-        session_api_endpoint = session_api_data['urls'][0]
+        video_id, video_src_id, audio_src_id = info_dict['url'].split(':')[1].split('/')
 
-        format_id = '-'.join(map(lambda s: remove_start(s['id'], 'archive_'), [video_quality, audio_quality]))
+        api_data = (
+            info_dict.get('_api_data')
+            or self._parse_json(
+                self._html_search_regex(
+                    'data-api-data="([^"]+)"',
+                    self._download_webpage('http://www.nicovideo.jp/watch/' + video_id, video_id),
+                    'API data', default='{}'),
+                video_id))
+
+        session_api_data = try_get(api_data, lambda x: x['media']['delivery']['movie']['session'])
+        session_api_endpoint = try_get(session_api_data, lambda x: x['urls'][0])
+
+        def ping():
+            status = try_get(
+                self._download_json(
+                    'https://nvapi.nicovideo.jp/v1/2ab0cbaa/watch', video_id,
+                    query={'t': try_get(api_data, lambda x: x['media']['delivery']['trackingId'])},
+                    note='Acquiring permission for downloading video',
+                    headers=self._API_HEADERS),
+                lambda x: x['meta']['status'])
+            if status != 200:
+                self.report_warning('Failed to acquire permission for playing video. The video may not download.')
+
+        yesno = lambda x: 'yes' if x else 'no'
+
+        # m3u8 (encryption)
+        if try_get(api_data, lambda x: x['media']['delivery']['encryption']) is not None:
+            protocol = 'm3u8'
+            encryption = self._parse_json(session_api_data['token'], video_id)['hls_encryption']
+            session_api_http_parameters = {
+                'parameters': {
+                    'hls_parameters': {
+                        'encryption': {
+                            encryption: {
+                                'encrypted_key': try_get(api_data, lambda x: x['media']['delivery']['encryption']['encryptedKey']),
+                                'key_uri': try_get(api_data, lambda x: x['media']['delivery']['encryption']['keyUri'])
+                            }
+                        },
+                        'transfer_preset': '',
+                        'use_ssl': yesno(session_api_endpoint['isSsl']),
+                        'use_well_known_port': yesno(session_api_endpoint['isWellKnownPort']),
+                        'segment_duration': 6000,
+                    }
+                }
+            }
+        # http
+        else:
+            protocol = 'http'
+            session_api_http_parameters = {
+                'parameters': {
+                    'http_output_download_parameters': {
+                        'use_ssl': yesno(session_api_endpoint['isSsl']),
+                        'use_well_known_port': yesno(session_api_endpoint['isWellKnownPort']),
+                    }
+                }
+            }
 
         session_response = self._download_json(
             session_api_endpoint['url'], video_id,
             query={'_format': 'json'},
             headers={'Content-Type': 'application/json'},
-            note='Downloading JSON metadata for %s' % format_id,
+            note='Downloading JSON metadata for %s' % info_dict['format_id'],
             data=json.dumps({
                 'session': {
                     'client_info': {
-                        'player_id': session_api_data['player_id'],
+                        'player_id': session_api_data.get('playerId'),
                     },
                     'content_auth': {
-                        'auth_type': session_api_data['auth_types'][session_api_data['protocols'][0]],
-                        'content_key_timeout': session_api_data['content_key_timeout'],
+                        'auth_type': try_get(session_api_data, lambda x: x['authTypes'][session_api_data['protocols'][0]]),
+                        'content_key_timeout': session_api_data.get('contentKeyTimeout'),
                         'service_id': 'nicovideo',
-                        'service_user_id': session_api_data['service_user_id']
+                        'service_user_id': session_api_data.get('serviceUserId')
                     },
-                    'content_id': session_api_data['content_id'],
+                    'content_id': session_api_data.get('contentId'),
                     'content_src_id_sets': [{
                         'content_src_ids': [{
                             'src_id_to_mux': {
-                                'audio_src_ids': [audio_quality['id']],
-                                'video_src_ids': [video_quality['id']],
+                                'audio_src_ids': [audio_src_id],
+                                'video_src_ids': [video_src_id],
                             }
                         }]
                     }],
@@ -229,52 +289,81 @@ class NiconicoIE(InfoExtractor):
                     'content_uri': '',
                     'keep_method': {
                         'heartbeat': {
-                            'lifetime': session_api_data['heartbeat_lifetime']
+                            'lifetime': session_api_data.get('heartbeatLifetime')
                         }
                     },
-                    'priority': session_api_data['priority'],
+                    'priority': session_api_data.get('priority'),
                     'protocol': {
                         'name': 'http',
                         'parameters': {
-                            'http_parameters': {
-                                'parameters': {
-                                    'http_output_download_parameters': {
-                                        'use_ssl': yesno(session_api_endpoint['is_ssl']),
-                                        'use_well_known_port': yesno(session_api_endpoint['is_well_known_port']),
-                                    }
-                                }
-                            }
+                            'http_parameters': session_api_http_parameters
                         }
                     },
-                    'recipe_id': session_api_data['recipe_id'],
+                    'recipe_id': session_api_data.get('recipeId'),
                     'session_operation_auth': {
                         'session_operation_auth_by_signature': {
-                            'signature': session_api_data['signature'],
-                            'token': session_api_data['token'],
+                            'signature': session_api_data.get('signature'),
+                            'token': session_api_data.get('token'),
                         }
                     },
                     'timing_constraint': 'unlimited'
                 }
             }).encode())
 
-        resolution = video_quality.get('resolution', {})
+        info_dict['url'] = session_response['data']['session']['content_uri']
+        info_dict['protocol'] = protocol
+
+        # get heartbeat info
+        heartbeat_info_dict = {
+            'url': session_api_endpoint['url'] + '/' + session_response['data']['session']['id'] + '?_format=json&_method=PUT',
+            'data': json.dumps(session_response['data']),
+            # interval, convert milliseconds to seconds, then halve to make a buffer.
+            'interval': float_or_none(session_api_data.get('heartbeatLifetime'), scale=3000),
+            'ping': ping
+        }
+
+        return info_dict, heartbeat_info_dict
+
+    def _extract_format_for_quality(self, api_data, video_id, audio_quality, video_quality):
+        def parse_format_id(id_code):
+            mobj = re.match(r'''(?x)
+                    (?:archive_)?
+                    (?:(?P<codec>[^_]+)_)?
+                    (?:(?P<br>[\d]+)kbps_)?
+                    (?:(?P<res>[\d+]+)p_)?
+                ''', '%s_' % id_code)
+            return mobj.groupdict() if mobj else {}
+
+        protocol = 'niconico_dmc'
+        format_id = '-'.join(map(lambda s: remove_start(s['id'], 'archive_'), [video_quality, audio_quality]))
+        vdict = parse_format_id(video_quality['id'])
+        adict = parse_format_id(audio_quality['id'])
+        resolution = try_get(video_quality, lambda x: x['metadata']['resolution'], dict) or {'height': vdict.get('res')}
+        vbr = try_get(video_quality, lambda x: x['metadata']['bitrate'], float)
 
         return {
-            'url': session_response['data']['session']['content_uri'],
+            'url': '%s:%s/%s/%s' % (protocol, video_id, video_quality['id'], audio_quality['id']),
             'format_id': format_id,
+            'format_note': 'DMC %s' % try_get(video_quality, lambda x: x['metadata']['label'], compat_str),
             'ext': 'mp4',  # Session API are used in HTML5, which always serves mp4
-            'abr': float_or_none(audio_quality.get('bitrate'), 1000),
-            'vbr': float_or_none(video_quality.get('bitrate'), 1000),
-            'height': resolution.get('height'),
-            'width': resolution.get('width'),
+            'vcodec': vdict.get('codec'),
+            'acodec': adict.get('codec'),
+            'vbr': float_or_none(vbr, 1000) or float_or_none(vdict.get('br')),
+            'abr': float_or_none(audio_quality.get('bitrate'), 1000) or float_or_none(adict.get('br')),
+            'height': int_or_none(resolution.get('height', vdict.get('res'))),
+            'width': int_or_none(resolution.get('width')),
+            'quality': -2 if 'low' in format_id else -1,  # Default quality value is -1
+            'protocol': protocol,
+            'http_headers': {
+                'Origin': 'https://www.nicovideo.jp',
+                'Referer': 'https://www.nicovideo.jp/watch/' + video_id,
+            }
         }
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
 
-        # Get video webpage. We are not actually interested in it for normal
-        # cases, but need the cookies in order to be able to download the
-        # info webpage
+        # Get video webpage for API data.
         webpage, handle = self._download_webpage_handle(
             'http://www.nicovideo.jp/watch/' + video_id, video_id)
         if video_id.startswith('so'):
@@ -284,86 +373,136 @@ class NiconicoIE(InfoExtractor):
             'data-api-data="([^"]+)"', webpage,
             'API data', default='{}'), video_id)
 
-        def _format_id_from_url(video_url):
-            return 'economy' if video_real_url.endswith('low') else 'normal'
+        def get_video_info_web(items):
+            return dict_get(api_data['video'], items)
 
-        try:
-            video_real_url = api_data['video']['smileInfo']['url']
-        except KeyError:  # Flash videos
-            # Get flv info
-            flv_info_webpage = self._download_webpage(
-                'http://flapi.nicovideo.jp/api/getflv/' + video_id + '?as3=1',
-                video_id, 'Downloading flv info')
+        # Get video info
+        video_info_xml = self._download_xml(
+            'http://ext.nicovideo.jp/api/getthumbinfo/' + video_id,
+            video_id, note='Downloading video info page')
 
-            flv_info = compat_parse_qs(flv_info_webpage)
-            if 'url' not in flv_info:
-                if 'deleted' in flv_info:
-                    raise ExtractorError('The video has been deleted.',
-                                         expected=True)
-                elif 'closed' in flv_info:
-                    raise ExtractorError('Niconico videos now require logging in',
-                                         expected=True)
-                elif 'error' in flv_info:
-                    raise ExtractorError('%s reports error: %s' % (
-                        self.IE_NAME, flv_info['error'][0]), expected=True)
-                else:
-                    raise ExtractorError('Unable to find video URL')
+        def get_video_info_xml(items):
+            if not isinstance(items, list):
+                items = [items]
+            for item in items:
+                ret = xpath_text(video_info_xml, './/' + item)
+                if ret:
+                    return ret
 
-            video_info_xml = self._download_xml(
-                'http://ext.nicovideo.jp/api/getthumbinfo/' + video_id,
-                video_id, note='Downloading video info page')
+        if get_video_info_xml('error'):
+            error_code = get_video_info_xml('code')
 
-            def get_video_info(items):
-                if not isinstance(items, list):
-                    items = [items]
-                for item in items:
-                    ret = xpath_text(video_info_xml, './/' + item)
-                    if ret:
-                        return ret
+            if error_code == 'DELETED':
+                raise ExtractorError('The video has been deleted.',
+                                     expected=True)
+            elif error_code == 'NOT_FOUND':
+                raise ExtractorError('The video is not found.',
+                                     expected=True)
+            elif error_code == 'COMMUNITY':
+                self.to_screen('%s: The video is community members only.' % video_id)
+            else:
+                raise ExtractorError('%s reports error: %s' % (self.IE_NAME, error_code))
 
-            video_real_url = flv_info['url'][0]
+        # Start extracting video formats
+        formats = []
 
-            extension = get_video_info('movie_type')
-            if not extension:
-                extension = determine_ext(video_real_url)
+        # Get HTML5 videos info
+        quality_info = try_get(api_data, lambda x: x['media']['delivery']['movie'])
+        if not quality_info:
+            raise ExtractorError('The video can\'t be downloaded', expected=True)
 
-            formats = [{
-                'url': video_real_url,
-                'ext': extension,
-                'format_id': _format_id_from_url(video_real_url),
-            }]
-        else:
-            formats = []
+        for audio_quality in quality_info.get('audios') or {}:
+            for video_quality in quality_info.get('videos') or {}:
+                if not audio_quality.get('isAvailable') or not video_quality.get('isAvailable'):
+                    continue
+                formats.append(self._extract_format_for_quality(
+                    api_data, video_id, audio_quality, video_quality))
 
-            dmc_info = api_data['video'].get('dmcInfo')
-            if dmc_info:  # "New" HTML5 videos
-                quality_info = dmc_info['quality']
-                for audio_quality in quality_info['audios']:
-                    for video_quality in quality_info['videos']:
-                        if not audio_quality['available'] or not video_quality['available']:
-                            continue
-                        formats.append(self._extract_format_for_quality(
-                            api_data, video_id, audio_quality, video_quality))
+        # Get flv/swf info
+        timestamp = None
+        video_real_url = try_get(api_data, lambda x: x['video']['smileInfo']['url'])
+        if video_real_url:
+            is_economy = video_real_url.endswith('low')
 
-                self._sort_formats(formats)
-            else:  # "Old" HTML5 videos
-                formats = [{
+            if is_economy:
+                self.report_warning('Site is currently in economy mode! You will only have access to lower quality streams')
+
+            # Invoking ffprobe to determine resolution
+            pp = FFmpegPostProcessor(self._downloader)
+            cookies = self._get_cookies('https://nicovideo.jp').output(header='', sep='; path=/; domain=nicovideo.jp;\n')
+
+            self.to_screen('%s: %s' % (video_id, 'Checking smile format with ffprobe'))
+
+            try:
+                metadata = pp.get_metadata_object(video_real_url, ['-cookies', cookies])
+            except PostProcessingError as err:
+                raise ExtractorError(err.msg, expected=True)
+
+            v_stream = a_stream = {}
+
+            # Some complex swf files doesn't have video stream (e.g. nm4809023)
+            for stream in metadata['streams']:
+                if stream['codec_type'] == 'video':
+                    v_stream = stream
+                elif stream['codec_type'] == 'audio':
+                    a_stream = stream
+
+            # Community restricted videos seem to have issues with the thumb API not returning anything at all
+            filesize = int(
+                (get_video_info_xml('size_high') if not is_economy else get_video_info_xml('size_low'))
+                or metadata['format']['size']
+            )
+            extension = (
+                get_video_info_xml('movie_type')
+                or 'mp4' if 'mp4' in metadata['format']['format_name'] else metadata['format']['format_name']
+            )
+
+            # 'creation_time' tag on video stream of re-encoded SMILEVIDEO mp4 files are '1970-01-01T00:00:00.000000Z'.
+            timestamp = (
+                parse_iso8601(get_video_info_web('first_retrieve'))
+                or unified_timestamp(get_video_info_web('postedDateTime'))
+            )
+            metadata_timestamp = (
+                parse_iso8601(try_get(v_stream, lambda x: x['tags']['creation_time']))
+                or timestamp if extension != 'mp4' else 0
+            )
+
+            # According to compconf, smile videos from pre-2017 are always better quality than their DMC counterparts
+            smile_threshold_timestamp = parse_iso8601('2016-12-08T00:00:00+09:00')
+
+            is_source = timestamp < smile_threshold_timestamp or metadata_timestamp > 0
+
+            # If movie file size is unstable, old server movie is not source movie.
+            if filesize > 1:
+                formats.append({
                     'url': video_real_url,
-                    'ext': 'mp4',
-                    'format_id': _format_id_from_url(video_real_url),
-                }]
+                    'format_id': 'smile' if not is_economy else 'smile_low',
+                    'format_note': 'SMILEVIDEO source' if not is_economy else 'SMILEVIDEO low quality',
+                    'ext': extension,
+                    'container': extension,
+                    'vcodec': v_stream.get('codec_name'),
+                    'acodec': a_stream.get('codec_name'),
+                    # Some complex swf files doesn't have total bit rate metadata (e.g. nm6049209)
+                    'tbr': int_or_none(metadata['format'].get('bit_rate'), scale=1000),
+                    'vbr': int_or_none(v_stream.get('bit_rate'), scale=1000),
+                    'abr': int_or_none(a_stream.get('bit_rate'), scale=1000),
+                    'height': int_or_none(v_stream.get('height')),
+                    'width': int_or_none(v_stream.get('width')),
+                    'source_preference': 5 if not is_economy else -2,
+                    'quality': 5 if is_source and not is_economy else None,
+                    'filesize': filesize
+                })
 
-            def get_video_info(items):
-                return dict_get(api_data['video'], items)
+        self._sort_formats(formats)
 
         # Start extracting information
-        title = get_video_info('title')
-        if not title:
-            title = self._og_search_title(webpage, default=None)
-        if not title:
-            title = self._html_search_regex(
+        title = (
+            get_video_info_xml('title')  # prefer to get the untranslated original title
+            or get_video_info_web(['originalTitle', 'title'])
+            or self._og_search_title(webpage, default=None)
+            or self._html_search_regex(
                 r'<span[^>]+class="videoHeaderTitle"[^>]*>([^<]+)</span>',
-                webpage, 'video title')
+                webpage, 'video title'))
 
         watch_api_data_string = self._html_search_regex(
             r'<div[^>]+id="watchAPIDataContainer"[^>]+>([^<]+)</div>',
@@ -372,14 +511,15 @@ class NiconicoIE(InfoExtractor):
         video_detail = watch_api_data.get('videoDetail', {})
 
         thumbnail = (
-            get_video_info(['thumbnail_url', 'thumbnailURL'])
+            self._html_search_regex(r'<meta property="og:image" content="([^"]+)">', webpage, 'thumbnail data', default=None)
+            or dict_get(  # choose highest from 720p to 240p
+                get_video_info_web('thumbnail'),
+                ['ogp', 'player', 'largeUrl', 'middleUrl', 'url'])
             or self._html_search_meta('image', webpage, 'thumbnail', default=None)
             or video_detail.get('thumbnail'))
 
-        description = get_video_info('description')
+        description = get_video_info_web('description')
 
-        timestamp = (parse_iso8601(get_video_info('first_retrieve'))
-                     or unified_timestamp(get_video_info('postedDateTime')))
         if not timestamp:
             match = self._html_search_meta('datePublished', webpage, 'date published', default=None)
             if match:
@@ -388,19 +528,25 @@ class NiconicoIE(InfoExtractor):
             timestamp = parse_iso8601(
                 video_detail['postedAt'].replace('/', '-'),
                 delimiter=' ', timezone=datetime.timedelta(hours=9))
+        timestamp = timestamp or try_get(api_data, lambda x: parse_iso8601(x['video']['registeredAt']))
 
-        view_count = int_or_none(get_video_info(['view_counter', 'viewCount']))
+        view_count = int_or_none(get_video_info_web(['view_counter', 'viewCount']))
         if not view_count:
             match = self._html_search_regex(
                 r'>Views: <strong[^>]*>([^<]+)</strong>',
                 webpage, 'view count', default=None)
             if match:
                 view_count = int_or_none(match.replace(',', ''))
-        view_count = view_count or video_detail.get('viewCount')
+        view_count = (
+            view_count
+            or video_detail.get('viewCount')
+            or try_get(api_data, lambda x: x['video']['count']['view']))
+
+        comment_count = (
+            int_or_none(get_video_info_web('comment_num'))
+            or video_detail.get('commentCount')
+            or try_get(api_data, lambda x: x['video']['count']['comment']))
 
-        comment_count = (int_or_none(get_video_info('comment_num'))
-                         or video_detail.get('commentCount')
-                         or try_get(api_data, lambda x: x['thread']['commentCount']))
         if not comment_count:
             match = self._html_search_regex(
                 r'>Comments: <strong[^>]*>([^<]+)</strong>',
@@ -409,22 +555,41 @@ class NiconicoIE(InfoExtractor):
                 comment_count = int_or_none(match.replace(',', ''))
 
         duration = (parse_duration(
-            get_video_info('length')
+            get_video_info_web('length')
             or self._html_search_meta(
                 'video:duration', webpage, 'video duration', default=None))
             or video_detail.get('length')
-            or get_video_info('duration'))
+            or get_video_info_web('duration'))
 
-        webpage_url = get_video_info('watch_url') or url
+        webpage_url = get_video_info_web('watch_url') or url
+
+        # for channel movie and community movie
+        channel_id = try_get(
+            api_data,
+            (lambda x: x['channel']['globalId'],
+             lambda x: x['community']['globalId']))
+        channel = try_get(
+            api_data,
+            (lambda x: x['channel']['name'],
+             lambda x: x['community']['name']))
 
         # Note: cannot use api_data.get('owner', {}) because owner may be set to "null"
         # in the JSON, which will cause None to be returned instead of {}.
         owner = try_get(api_data, lambda x: x.get('owner'), dict) or {}
-        uploader_id = get_video_info(['ch_id', 'user_id']) or owner.get('id')
-        uploader = get_video_info(['ch_name', 'user_nickname']) or owner.get('nickname')
+        uploader_id = str_or_none(
+            get_video_info_web(['ch_id', 'user_id'])
+            or owner.get('id')
+            or channel_id
+        )
+        uploader = (
+            get_video_info_web(['ch_name', 'user_nickname'])
+            or owner.get('nickname')
+            or channel
+        )
 
         return {
             'id': video_id,
+            '_api_data': api_data,
             'title': title,
             'formats': formats,
             'thumbnail': thumbnail,
@@ -432,6 +597,8 @@ class NiconicoIE(InfoExtractor):
             'uploader': uploader,
             'timestamp': timestamp,
             'uploader_id': uploader_id,
+            'channel': channel,
+            'channel_id': channel_id,
             'view_count': view_count,
             'comment_count': comment_count,
             'duration': duration,
@@ -440,7 +607,7 @@ class NiconicoIE(InfoExtractor):
 
 
 class NiconicoPlaylistIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?nicovideo\.jp/(?:user/\d+/)?mylist/(?P<id>\d+)'
+    _VALID_URL = r'https?://(?:www\.)?nicovideo\.jp/(?:user/\d+/|my/)?mylist/(?P<id>\d+)'
 
     _TESTS = [{
         'url': 'http://www.nicovideo.jp/mylist/27411728',
@@ -456,60 +623,77 @@ class NiconicoPlaylistIE(InfoExtractor):
         'url': 'https://www.nicovideo.jp/user/805442/mylist/27411728',
         'only_matching': True,
     }]
-    _PAGE_SIZE = 100
 
-    def _call_api(self, list_id, resource, query):
-        return self._download_json(
-            'https://nvapi.nicovideo.jp/v2/mylists/' + list_id, list_id,
-            'Downloading %s JSON metatdata' % resource, query=query,
-            headers={'X-Frontend-Id': 6})['data']['mylist']
-
-    def _parse_owner(self, item):
-        owner = item.get('owner') or {}
-        if owner:
-            return {
-                'uploader': owner.get('name'),
-                'uploader_id': owner.get('id'),
-            }
-        return {}
-
-    def _fetch_page(self, list_id, page):
-        page += 1
-        items = self._call_api(list_id, 'page %d' % page, {
-            'page': page,
-            'pageSize': self._PAGE_SIZE,
-        })['items']
-        for item in items:
-            video = item.get('video') or {}
-            video_id = video.get('id')
-            if not video_id:
-                continue
-            count = video.get('count') or {}
-            get_count = lambda x: int_or_none(count.get(x))
-            info = {
-                '_type': 'url',
-                'id': video_id,
-                'title': video.get('title'),
-                'url': 'https://www.nicovideo.jp/watch/' + video_id,
-                'description': video.get('shortDescription'),
-                'duration': int_or_none(video.get('duration')),
-                'view_count': get_count('view'),
-                'comment_count': get_count('comment'),
-                'ie_key': NiconicoIE.ie_key(),
-            }
-            info.update(self._parse_owner(video))
-            yield info
+    _API_HEADERS = {
+        'X-Frontend-ID': '6',
+        'X-Frontend-Version': '0'
+    }
 
     def _real_extract(self, url):
         list_id = self._match_id(url)
-        mylist = self._call_api(list_id, 'list', {
-            'pageSize': 1,
-        })
-        entries = InAdvancePagedList(
-            functools.partial(self._fetch_page, list_id),
-            math.ceil(mylist['totalItemCount'] / self._PAGE_SIZE),
-            self._PAGE_SIZE)
-        result = self.playlist_result(
-            entries, list_id, mylist.get('name'), mylist.get('description'))
-        result.update(self._parse_owner(mylist))
-        return result
+
+        def get_page_data(pagenum, pagesize):
+            return self._download_json(
+                'http://nvapi.nicovideo.jp/v2/mylists/' + list_id, list_id,
+                query={'page': 1 + pagenum, 'pageSize': pagesize},
+                headers=self._API_HEADERS).get('data').get('mylist')
+
+        data = get_page_data(0, 1)
+        title = data.get('name')
+        description = data.get('description')
+        uploader = data.get('owner').get('name')
+        uploader_id = data.get('owner').get('id')
+
+        def pagefunc(pagenum):
+            data = get_page_data(pagenum, 25)
+            return ({
+                '_type': 'url',
+                'url': 'http://www.nicovideo.jp/watch/' + item.get('watchId'),
+            } for item in data.get('items'))
+
+        return {
+            '_type': 'playlist',
+            'id': list_id,
+            'title': title,
+            'description': description,
+            'uploader': uploader,
+            'uploader_id': uploader_id,
+            'entries': OnDemandPagedList(pagefunc, 25),
+        }
+
+
+class NiconicoUserIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?nicovideo\.jp/user/(?P<id>\d+)/?(?:$|[#?])'
+    _TEST = {
+        'url': 'https://www.nicovideo.jp/user/419948',
+        'info_dict': {
+            'id': '419948',
+        },
+        'playlist_mincount': 101,
+    }
+    _API_URL = "https://nvapi.nicovideo.jp/v1/users/%s/videos?sortKey=registeredAt&sortOrder=desc&pageSize=%s&page=%s"
+    _PAGE_SIZE = 100
+
+    _API_HEADERS = {
+        'X-Frontend-ID': '6',
+        'X-Frontend-Version': '0'
+    }
+
+    def _entries(self, list_id, ):
+        total_count = 1
+        count = page_num = 0
+        while count < total_count:
+            json_parsed = self._download_json(
+                self._API_URL % (list_id, self._PAGE_SIZE, page_num + 1), list_id,
+                headers=self._API_HEADERS,
+                note='Downloading JSON metadata%s' % (' page %d' % page_num if page_num else ''))
+            if not page_num:
+                total_count = int_or_none(json_parsed['data'].get('totalCount'))
+            for entry in json_parsed["data"]["items"]:
+                count += 1
+                yield self.url_result('https://www.nicovideo.jp/watch/%s' % entry['id'])
+            page_num += 1
+
+    def _real_extract(self, url):
+        list_id = self._match_id(url)
+        return self.playlist_result(self._entries(list_id), list_id, ie=NiconicoIE.ie_key())

haruhi_dl/extractor/nrk.py

@@ -58,7 +58,7 @@ class NRKBaseIE(InfoExtractor):
 
     def _call_api(self, path, video_id, item=None, note=None, fatal=True, query=None):
         return self._download_json(
-            urljoin('http://psapi.nrk.no/', path),
+            urljoin('https://psapi.nrk.no/', path),
             video_id, note or 'Downloading %s JSON' % item,
             fatal=fatal, query=query,
             headers={'Accept-Encoding': 'gzip, deflate, br'})

haruhi_dl/extractor/orf.py

@@ -98,6 +98,9 @@ class ORFTVthekIE(InfoExtractor):
                 elif ext == 'f4m':
                     formats.extend(self._extract_f4m_formats(
                         src, video_id, f4m_id=format_id, fatal=False))
+                elif ext == 'mpd':
+                    formats.extend(self._extract_mpd_formats(
+                        src, video_id, mpd_id=format_id, fatal=False))
                 else:
                     formats.append({
                         'format_id': format_id,
@@ -140,6 +143,25 @@ class ORFTVthekIE(InfoExtractor):
                 })
 
             upload_date = unified_strdate(sd.get('created_date'))
+
+            thumbnails = []
+            preview = sd.get('preview_image_url')
+            if preview:
+                thumbnails.append({
+                    'id': 'preview',
+                    'url': preview,
+                    'preference': 0,
+                })
+            image = sd.get('image_full_url')
+            if not image and len(data_jsb) == 1:
+                image = self._og_search_thumbnail(webpage)
+            if image:
+                thumbnails.append({
+                    'id': 'full',
+                    'url': image,
+                    'preference': 1,
+                })
+
             entries.append({
                 '_type': 'video',
                 'id': video_id,
@@ -149,7 +171,7 @@ class ORFTVthekIE(InfoExtractor):
                 'description': sd.get('description'),
                 'duration': int_or_none(sd.get('duration_in_seconds')),
                 'upload_date': upload_date,
-                'thumbnail': sd.get('image_full_url'),
+                'thumbnails': thumbnails,
             })
 
         return {

haruhi_dl/extractor/peertube.py

@@ -21,7 +21,7 @@ from ..utils import (
 
 
 class PeerTubeBaseExtractor(SelfhostedInfoExtractor):
-    _UUID_RE = r'[\da-fA-F]{8}-[\da-fA-F]{4}-[\da-fA-F]{4}-[\da-fA-F]{4}-[\da-fA-F]{12}'
+    _UUID_RE = r'[\da-zA-Z]{22}|[\da-fA-F]{8}-[\da-fA-F]{4}-[\da-fA-F]{4}-[\da-fA-F]{4}-[\da-fA-F]{12}'
     _API_BASE = 'https://%s/api/v1/%s/%s/%s'
     _SH_VALID_CONTENT_STRINGS = (
         '<title>PeerTube<',
@@ -180,16 +180,16 @@ class PeerTubeBaseExtractor(SelfhostedInfoExtractor):
 
 class PeerTubeSHIE(PeerTubeBaseExtractor):
     _VALID_URL = r'peertube:(?P<host>[^:]+):(?P<id>%s)' % (PeerTubeBaseExtractor._UUID_RE)
-    _SH_VALID_URL = r'https?://(?P<host>[^/]+)/(?:videos/(?:watch|embed)|api/v\d/videos)/(?P<id>%s)' % (PeerTubeBaseExtractor._UUID_RE)
+    _SH_VALID_URL = r'https?://(?P<host>[^/]+)/(?:videos/(?:watch|embed)|api/v\d/videos|w)/(?P<id>%s)' % (PeerTubeBaseExtractor._UUID_RE)
 
     _TESTS = [{
         'url': 'https://framatube.org/videos/watch/9c9de5e8-0a1e-484a-b099-e80766180a6d',
-        'md5': '9bed8c0137913e17b86334e5885aacff',
+        'md5': '8563064d245a4be5705bddb22bb00a28',
         'info_dict': {
             'id': '9c9de5e8-0a1e-484a-b099-e80766180a6d',
             'ext': 'mp4',
             'title': 'What is PeerTube?',
-            'description': 'md5:3fefb8dde2b189186ce0719fda6f7b10',
+            'description': 'md5:96adbaf219b4d41747bfc5937df0b017',
             'thumbnail': r're:https?://.*\.(?:jpg|png)',
             'timestamp': 1538391166,
             'upload_date': '20181001',
@@ -220,6 +220,27 @@ class PeerTubeSHIE(PeerTubeBaseExtractor):
             'upload_date': '20200420',
             'uploader': 'Drew DeVault',
         }
+    }, {
+        # new url scheme since PeerTube 3.3
+        'url': 'https://peertube2.cpy.re/w/3fbif9S3WmtTP8gGsC5HBd',
+        'info_dict': {
+            'id': '122d093a-1ede-43bd-bd34-59d2931ffc5e',
+            'ext': 'mp4',
+            'title': 'E2E tests',
+            'uploader_id': '37855',
+            'timestamp': 1589276219,
+            'upload_date': '20200512',
+            'uploader': 'chocobozzz',
+        },
+    }, {
+        'url': 'https://peertube2.cpy.re/w/122d093a-1ede-43bd-bd34-59d2931ffc5e',
+        'only_matching': True,
+    }, {
+        'url': 'https://peertube2.cpy.re/api/v1/videos/3fbif9S3WmtTP8gGsC5HBd',
+        'only_matching': True,
+    }, {
+        'url': 'peertube:peertube2.cpy.re:3fbif9S3WmtTP8gGsC5HBd',
+        'only_matching': True,
     }, {
         'url': 'https://peertube.tamanoir.foucry.net/videos/watch/0b04f13d-1e18-4f1d-814e-4979aa7c9c44',
         'only_matching': True,
@@ -289,7 +310,7 @@ class PeerTubeSHIE(PeerTubeBaseExtractor):
 
         description = None
         if webpage:
-            description = self._og_search_description(webpage)
+            description = self._og_search_description(webpage, default=None)
         if not description:
             full_description = self._call_api(
                 host, 'videos', video_id, 'description', note='Downloading description JSON',
@@ -305,7 +326,7 @@ class PeerTubeSHIE(PeerTubeBaseExtractor):
 
 class PeerTubePlaylistSHIE(PeerTubeBaseExtractor):
     _VALID_URL = r'peertube:playlist:(?P<host>[^:]+):(?P<id>.+)'
-    _SH_VALID_URL = r'https?://(?P<host>[^/]+)/(?:videos/(?:watch|embed)/playlist|api/v\d/video-playlists)/(?P<id>%s)' % (PeerTubeBaseExtractor._UUID_RE)
+    _SH_VALID_URL = r'https?://(?P<host>[^/]+)/(?:videos/(?:watch|embed)/playlist|api/v\d/video-playlists|w/p)/(?P<id>%s)' % (PeerTubeBaseExtractor._UUID_RE)
 
     _TESTS = [{
         'url': 'https://video.internet-czas-dzialac.pl/videos/watch/playlist/3c81b894-acde-4539-91a2-1748b208c14c?playlistPosition=1',
@@ -316,6 +337,9 @@ class PeerTubePlaylistSHIE(PeerTubeBaseExtractor):
             'uploader': 'Internet. Czas działać!',
         },
         'playlist_mincount': 14,
+    }, {
+        'url': 'https://peertube2.cpy.re/w/p/hrAdcvjkMMkHJ28upnoN21',
+        'only_matching': True,
     }]
 
     def _selfhosted_extract(self, url, webpage=None):
@@ -352,18 +376,21 @@ class PeerTubePlaylistSHIE(PeerTubeBaseExtractor):
 
 class PeerTubeChannelSHIE(PeerTubeBaseExtractor):
     _VALID_URL = r'peertube:channel:(?P<host>[^:]+):(?P<id>.+)'
-    _SH_VALID_URL = r'https?://(?P<host>[^/]+)/(?:api/v\d/)?video-channels/(?P<id>[^/?#]+)(?:/videos)?'
+    _SH_VALID_URL = r'https?://(?P<host>[^/]+)/(?:(?:api/v\d/)?video-channels|c)/(?P<id>[^/?#]+)(?:/videos)?'
 
     _TESTS = [{
         'url': 'https://video.internet-czas-dzialac.pl/video-channels/internet_czas_dzialac/videos',
         'info_dict': {
             'id': '2',
-            'title': 'internet_czas_dzialac',
-            'description': 'md5:4d2e215ea0d9ae4501a556ef6e9a5308',
+            'title': 'Internet. Czas działać!',
+            'description': 'md5:ac35d70f6625b04b189e0b4b76e62e17',
             'uploader_id': 3,
             'uploader': 'Internet. Czas działać!',
         },
         'playlist_mincount': 14,
+    }, {
+        'url': 'https://video.internet-czas-dzialac.pl/c/internet_czas_dzialac',
+        'only_matching': True,
     }]
 
     def _selfhosted_extract(self, url, webpage=None):
@@ -401,18 +428,21 @@ class PeerTubeChannelSHIE(PeerTubeBaseExtractor):
 
 class PeerTubeAccountSHIE(PeerTubeBaseExtractor):
     _VALID_URL = r'peertube:account:(?P<host>[^:]+):(?P<id>.+)'
-    _SH_VALID_URL = r'https?://(?P<host>[^/]+)/(?:api/v\d/)?accounts/(?P<id>[^/?#]+)(?:/video(?:s|-channels))?'
+    _SH_VALID_URL = r'https?://(?P<host>[^/]+)/(?:(?:api/v\d/)?accounts|a)/(?P<id>[^/?#]+)(?:/video(?:s|-channels))?'
 
     _TESTS = [{
         'url': 'https://video.internet-czas-dzialac.pl/accounts/icd/video-channels',
         'info_dict': {
             'id': '3',
-            'description': 'md5:ab3c9b934dd39030eea1c9fe76079870',
+            'description': 'md5:ac35d70f6625b04b189e0b4b76e62e17',
             'uploader': 'Internet. Czas działać!',
             'title': 'Internet. Czas działać!',
             'uploader_id': 3,
         },
         'playlist_mincount': 14,
+    }, {
+        'url': 'https://video.internet-czas-dzialac.pl/a/icd',
+        'only_matching': True,
     }]
 
     def _selfhosted_extract(self, url, webpage=None):

haruhi_dl/extractor/polskieradio.py

@@ -91,6 +91,14 @@ class PolskieRadioIE(PolskieRadioBaseExtractor):
                 'upload_date': '20201116',
             },
         }]
+    }, {
+        # PR4 audition - other frontend
+        'url': 'https://www.polskieradio.pl/10/6071/Artykul/2610977,Poglos-29-pazdziernika-godz-2301',
+        'info_dict': {
+            'id': '2610977',
+            'ext': 'mp3',
+            'title': 'Pogłos 29 października godz. 23:01',
+        },
     }, {
         'url': 'http://polskieradio.pl/9/305/Artykul/1632955,Bardzo-popularne-slowo-remis',
         'only_matching': True,
@@ -113,24 +121,34 @@ class PolskieRadioIE(PolskieRadioBaseExtractor):
 
         content = self._search_regex(
             r'(?s)<div[^>]+class="\s*this-article\s*"[^>]*>(.+?)<div[^>]+class="tags"[^>]*>',
-            webpage, 'content')
+            webpage, 'content', default=None)
 
         timestamp = unified_timestamp(self._html_search_regex(
             r'(?s)<span[^>]+id="datetime2"[^>]*>(.+?)</span>',
-            webpage, 'timestamp', fatal=False))
+            webpage, 'timestamp', default=None))
 
-        thumbnail_url = self._og_search_thumbnail(webpage)
+        thumbnail_url = self._og_search_thumbnail(webpage, default=None)
 
         title = self._og_search_title(webpage).strip()
 
+        description = strip_or_none(self._og_search_description(webpage, default=None))
+        
+        if not content:
+            return {
+                'id': playlist_id,
+                'url': 'https:' + self._search_regex(r"source:\s*'(//static\.prsa\.pl/[^']+)'", webpage, 'audition record url'),
+                'title': title,
+                'description': description,
+                'timestamp': timestamp,
+                'thumbnail': thumbnail_url,
+            }
+
         entries = self._extract_webpage_player_entries(content, playlist_id, {
             'title': title,
             'timestamp': timestamp,
             'thumbnail': thumbnail_url,
         })
 
-        description = strip_or_none(self._og_search_description(webpage))
-
         return self.playlist_result(entries, playlist_id, title, description)
 
 

haruhi_dl/extractor/pornhub.py

@@ -31,6 +31,7 @@ from ..utils import (
 class PornHubBaseIE(InfoExtractor):
     _REQUIRES_PLAYWRIGHT = True
     _NETRC_MACHINE = 'pornhub'
+    _PORNHUB_HOST_RE = r'(?:(?P<host>pornhub(?:premium)?\.(?:com|net|org))|pornhubthbh7ap3u\.onion)'
 
     def _download_webpage_handle(self, *args, **kwargs):
         def dl(*args, **kwargs):
@@ -125,11 +126,13 @@ class PornHubIE(PornHubBaseIE):
     _VALID_URL = r'''(?x)
                     https?://
                         (?:
-                            (?:[^/]+\.)?(?P<host>pornhub(?:premium)?\.(?:com|net|org))/(?:(?:view_video\.php|video/show)\?viewkey=|embed/)|
+                            (?:[^/]+\.)?
+                            %s
+                            /(?:(?:view_video\.php|video/show)\?viewkey=|embed/)|
                             (?:www\.)?thumbzilla\.com/video/
                         )
                         (?P<id>[\da-z]+)
-                    '''
+                    ''' % PornHubBaseIE._PORNHUB_HOST_RE
     _TESTS = [{
         'url': 'http://www.pornhub.com/view_video.php?viewkey=648719015',
         'info_dict': {
@@ -238,6 +241,13 @@ class PornHubIE(PornHubBaseIE):
     }, {
         'url': 'https://www.pornhubpremium.com/view_video.php?viewkey=ph5f75b0f4b18e3',
         'only_matching': True,
+    }, {
+        # geo restricted
+        'url': 'https://www.pornhub.com/view_video.php?viewkey=ph5a9813bfa7156',
+        'only_matching': True,
+    }, {
+        'url': 'http://pornhubthbh7ap3u.onion/view_video.php?viewkey=ph5a9813bfa7156',
+        'only_matching': True,
     }]
 
     @staticmethod
@@ -277,6 +287,11 @@ class PornHubIE(PornHubBaseIE):
                 'PornHub said: %s' % error_msg,
                 expected=True, video_id=video_id)
 
+        if any(re.search(p, webpage) for p in (
+                r'class=["\']geoBlocked["\']',
+                r'>\s*This content is unavailable in your country')):
+            self.raise_geo_restricted()
+
         # video_title from flashvars contains whitespace instead of non-ASCII (see
         # http://www.pornhub.com/view_video.php?viewkey=1331683002), not relying
         # on that anymore.
@@ -410,17 +425,14 @@ class PornHubIE(PornHubBaseIE):
                     format_url, video_id, 'mp4', entry_protocol='m3u8_native',
                     m3u8_id='hls', fatal=False))
                 return
-            tbr = None
-            mobj = re.search(r'(?P<height>\d+)[pP]?_(?P<tbr>\d+)[kK]', format_url)
-            if mobj:
-                if not height:
-                    height = int(mobj.group('height'))
-                tbr = int(mobj.group('tbr'))
+            if not height:
+                height = int_or_none(self._search_regex(
+                    r'(?P<height>\d+)[pP]?_\d+[kK]', format_url, 'height',
+                    default=None))
             formats.append({
                 'url': format_url,
                 'format_id': '%dp' % height if height else None,
                 'height': height,
-                'tbr': tbr,
             })
 
         for video_url, height in video_urls:
@@ -442,7 +454,8 @@ class PornHubIE(PornHubBaseIE):
                         add_format(video_url, height)
                 continue
             add_format(video_url)
-        self._sort_formats(formats)
+        self._sort_formats(
+            formats, field_preference=('height', 'width', 'fps', 'format_id'))
 
         video_uploader = self._html_search_regex(
             r'(?s)From:&nbsp;.+?<(?:a\b[^>]+\bhref=["\']/(?:(?:user|channel)s|model|pornstar)/|span\b[^>]+\bclass=["\']username)[^>]+>(.+?)<',
@@ -511,7 +524,7 @@ class PornHubPlaylistBaseIE(PornHubBaseIE):
 
 
 class PornHubUserIE(PornHubPlaylistBaseIE):
-    _VALID_URL = r'(?P<url>https?://(?:[^/]+\.)?(?P<host>pornhub(?:premium)?\.(?:com|net|org))/(?:(?:user|channel)s|model|pornstar)/(?P<id>[^/?#&]+))(?:[?#&]|/(?!videos)|$)'
+    _VALID_URL = r'(?P<url>https?://(?:[^/]+\.)?%s/(?:(?:user|channel)s|model|pornstar)/(?P<id>[^/?#&]+))(?:[?#&]|/(?!videos)|$)' % PornHubBaseIE._PORNHUB_HOST_RE
     _TESTS = [{
         'url': 'https://www.pornhub.com/model/zoe_ph',
         'playlist_mincount': 118,
@@ -540,6 +553,9 @@ class PornHubUserIE(PornHubPlaylistBaseIE):
         # Same as before, multi page
         'url': 'https://www.pornhubpremium.com/pornstar/lily-labeau',
         'only_matching': True,
+    }, {
+        'url': 'https://pornhubthbh7ap3u.onion/model/zoe_ph',
+        'only_matching': True,
     }]
 
     def _real_extract(self, url):
@@ -615,7 +631,7 @@ class PornHubPagedPlaylistBaseIE(PornHubPlaylistBaseIE):
 
 
 class PornHubPagedVideoListIE(PornHubPagedPlaylistBaseIE):
-    _VALID_URL = r'https?://(?:[^/]+\.)?(?P<host>pornhub(?:premium)?\.(?:com|net))/(?P<id>(?:[^/]+/)*[^/?#&]+)'
+    _VALID_URL = r'https?://(?:[^/]+\.)?%s/(?P<id>(?:[^/]+/)*[^/?#&]+)' % PornHubBaseIE._PORNHUB_HOST_RE
     _TESTS = [{
         'url': 'https://www.pornhub.com/model/zoe_ph/videos',
         'only_matching': True,
@@ -720,6 +736,9 @@ class PornHubPagedVideoListIE(PornHubPagedPlaylistBaseIE):
     }, {
         'url': 'https://de.pornhub.com/playlist/4667351',
         'only_matching': True,
+    }, {
+        'url': 'https://pornhubthbh7ap3u.onion/model/zoe_ph/videos',
+        'only_matching': True,
     }]
 
     @classmethod
@@ -730,7 +749,7 @@ class PornHubPagedVideoListIE(PornHubPagedPlaylistBaseIE):
 
 
 class PornHubUserVideosUploadIE(PornHubPagedPlaylistBaseIE):
-    _VALID_URL = r'(?P<url>https?://(?:[^/]+\.)?(?P<host>pornhub(?:premium)?\.(?:com|net|org))/(?:(?:user|channel)s|model|pornstar)/(?P<id>[^/]+)/videos/upload)'
+    _VALID_URL = r'(?P<url>https?://(?:[^/]+\.)?%s/(?:(?:user|channel)s|model|pornstar)/(?P<id>[^/]+)/videos/upload)' % PornHubBaseIE._PORNHUB_HOST_RE
     _TESTS = [{
         'url': 'https://www.pornhub.com/pornstar/jenny-blighe/videos/upload',
         'info_dict': {
@@ -740,4 +759,7 @@ class PornHubUserVideosUploadIE(PornHubPagedPlaylistBaseIE):
     }, {
         'url': 'https://www.pornhub.com/model/zoe_ph/videos/upload',
         'only_matching': True,
+    }, {
+        'url': 'http://pornhubthbh7ap3u.onion/pornstar/jenny-blighe/videos/upload',
+        'only_matching': True,
     }]

haruhi_dl/extractor/soundcloud.py

@@ -30,6 +30,10 @@ from ..utils import (
     url_or_none,
     urlhandle_detect_ext,
 )
+try:
+    from ..extractor_artifacts.soundcloud import prerelease_client_id
+except ImportError:
+    prerelease_client_id = None
 
 
 class SoundcloudEmbedIE(InfoExtractor):
@@ -289,6 +293,10 @@ class SoundcloudIE(InfoExtractor):
                     return
         raise ExtractorError('Unable to extract client id')
 
+    def _generate_prerelease_file(self):
+        self._update_client_id()
+        return 'prerelease_client_id = {!r}\n'.format(self._CLIENT_ID)
+
     def _download_json(self, *args, **kwargs):
         non_fatal = kwargs.get('fatal') is False
         if non_fatal:
@@ -310,7 +318,7 @@ class SoundcloudIE(InfoExtractor):
                 raise
 
     def _real_initialize(self):
-        self._CLIENT_ID = self._downloader.cache.load('soundcloud', 'client_id') or 'YUKXoArFcqrlQn9tfNHvvyfnDISj04zk'
+        self._CLIENT_ID = self._downloader.cache.load('soundcloud', 'client_id') or prerelease_client_id or 'YUKXoArFcqrlQn9tfNHvvyfnDISj04zk'
 
     @classmethod
     def _resolv_url(cls, url):

haruhi_dl/extractor/tiktok.py

@@ -158,7 +158,7 @@ class TikTokPlaywrightBaseIE(TikTokBaseIE):
     def _dedup_by_ids(self, items):
         deduped = []
         dedids = []
-        for item in deduped:
+        for item in items:
             if item['id'] in dedids:
                 continue
             dedids.append(item['id'])

haruhi_dl/extractor/umg.py

@@ -28,7 +28,7 @@ class UMGDeIE(InfoExtractor):
     def _real_extract(self, url):
         video_id = self._match_id(url)
         video_data = self._download_json(
-            'https://api.universal-music.de/graphql',
+            'https://graphql.universal-music.de/',
             video_id, query={
                 'query': '''{
   universalMusic(channel:16) {
@@ -56,11 +56,9 @@ class UMGDeIE(InfoExtractor):
         formats = []
 
         def add_m3u8_format(format_id):
-            m3u8_formats = self._extract_m3u8_formats(
+            formats.extend(self._extract_m3u8_formats(
                 hls_url_template % format_id, video_id, 'mp4',
-                'm3u8_native', m3u8_id='hls', fatal='False')
-            if m3u8_formats and m3u8_formats[0].get('height'):
-                formats.extend(m3u8_formats)
+                'm3u8_native', m3u8_id='hls', fatal=False))
 
         for f in video_data.get('formats', []):
             f_url = f.get('url')

haruhi_dl/extractor/vider.py

@@ -0,0 +1,37 @@
+from .common import InfoExtractor
+
+
+class ViderIE(InfoExtractor):
+    _VALID_URL = r'https?://vider\.(?:pl|info)/(?:vid/\+f|embed/video/)(?P<id>[a-z\d]+)'
+    _TESTS = [{
+        'url': 'https://vider.info/vid/+fsx51se',
+        'info_dict': {
+            'id': 'sx51se',
+            'ext': 'mp4',
+            'title': 'Big Buck Bunny',
+            'upload_date': '20210906',
+            'timestamp': 1630927351,
+        },
+    }]
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+
+        webpage = self._download_webpage(f'https://vider.info/vid/+f{video_id}', video_id)
+
+        json_ld = self._parse_json(
+            self._search_regex(
+                r'(?s)<script type="application/ld\+json">(.+?)</script>',
+                webpage, 'JSON-LD'), video_id)
+        info_dict = self._json_ld(json_ld, video_id)
+        # generated SEO junk
+        info_dict['description'] = None
+        info_dict['id'] = video_id
+        info_dict['formats'] = [{
+            'url': self._search_regex(r'\?file=(.+)', json_ld['embedUrl'], 'video url'),
+            'http_headers': {
+                'Referer': 'https://vider.info/',
+            },
+        }]
+
+        return info_dict

haruhi_dl/extractor/youporn.py

@@ -4,13 +4,12 @@ import re
 
 from .common import InfoExtractor
 from ..utils import (
+    extract_attributes,
     int_or_none,
     str_to_int,
-    unescapeHTML,
     unified_strdate,
     url_or_none,
 )
-from ..aes import aes_decrypt_text
 
 
 class YouPornIE(InfoExtractor):
@@ -34,6 +33,7 @@ class YouPornIE(InfoExtractor):
             'tags': list,
             'age_limit': 18,
         },
+        'skip': 'This video has been disabled',
     }, {
         # Unknown uploader
         'url': 'http://www.youporn.com/watch/561726/big-tits-awesome-brunette-on-amazing-webcam-show/?from=related3&al=2&from_id=561726&pos=4',
@@ -78,6 +78,40 @@ class YouPornIE(InfoExtractor):
         video_id = mobj.group('id')
         display_id = mobj.group('display_id') or video_id
 
+        definitions = self._download_json(
+            'https://www.youporn.com/api/video/media_definitions/%s/' % video_id,
+            display_id)
+
+        formats = []
+        for definition in definitions:
+            if not isinstance(definition, dict):
+                continue
+            video_url = url_or_none(definition.get('videoUrl'))
+            if not video_url:
+                continue
+            f = {
+                'url': video_url,
+                'filesize': int_or_none(definition.get('videoSize')),
+            }
+            height = int_or_none(definition.get('quality'))
+            # Video URL's path looks like this:
+            #  /201012/17/505835/720p_1500k_505835/YouPorn%20-%20Sex%20Ed%20Is%20It%20Safe%20To%20Masturbate%20Daily.mp4
+            #  /201012/17/505835/vl_240p_240k_505835/YouPorn%20-%20Sex%20Ed%20Is%20It%20Safe%20To%20Masturbate%20Daily.mp4
+            #  /videos/201703/11/109285532/1080P_4000K_109285532.mp4
+            # We will benefit from it by extracting some metadata
+            mobj = re.search(r'(?P<height>\d{3,4})[pP]_(?P<bitrate>\d+)[kK]_\d+', video_url)
+            if mobj:
+                if not height:
+                    height = int(mobj.group('height'))
+                bitrate = int(mobj.group('bitrate'))
+                f.update({
+                    'format_id': '%dp-%dk' % (height, bitrate),
+                    'tbr': bitrate,
+                })
+            f['height'] = height
+            formats.append(f)
+        self._sort_formats(formats)
+
         webpage = self._download_webpage(
             'http://www.youporn.com/watch/%s' % video_id, display_id,
             headers={'Cookie': 'age_verified=1'})
@@ -88,65 +122,6 @@ class YouPornIE(InfoExtractor):
             webpage, default=None) or self._html_search_meta(
             'title', webpage, fatal=True)
 
-        links = []
-
-        # Main source
-        definitions = self._parse_json(
-            self._search_regex(
-                r'mediaDefinition\s*[=:]\s*(\[.+?\])\s*[;,]', webpage,
-                'media definitions', default='[]'),
-            video_id, fatal=False)
-        if definitions:
-            for definition in definitions:
-                if not isinstance(definition, dict):
-                    continue
-                video_url = url_or_none(definition.get('videoUrl'))
-                if video_url:
-                    links.append(video_url)
-
-        # Fallback #1, this also contains extra low quality 180p format
-        for _, link in re.findall(r'<a[^>]+href=(["\'])(http(?:(?!\1).)+\.mp4(?:(?!\1).)*)\1[^>]+title=["\']Download [Vv]ideo', webpage):
-            links.append(link)
-
-        # Fallback #2 (unavailable as at 22.06.2017)
-        sources = self._search_regex(
-            r'(?s)sources\s*:\s*({.+?})', webpage, 'sources', default=None)
-        if sources:
-            for _, link in re.findall(r'[^:]+\s*:\s*(["\'])(http.+?)\1', sources):
-                links.append(link)
-
-        # Fallback #3 (unavailable as at 22.06.2017)
-        for _, link in re.findall(
-                r'(?:videoSrc|videoIpadUrl|html5PlayerSrc)\s*[:=]\s*(["\'])(http.+?)\1', webpage):
-            links.append(link)
-
-        # Fallback #4, encrypted links (unavailable as at 22.06.2017)
-        for _, encrypted_link in re.findall(
-                r'encryptedQuality\d{3,4}URL\s*=\s*(["\'])([\da-zA-Z+/=]+)\1', webpage):
-            links.append(aes_decrypt_text(encrypted_link, title, 32).decode('utf-8'))
-
-        formats = []
-        for video_url in set(unescapeHTML(link) for link in links):
-            f = {
-                'url': video_url,
-            }
-            # Video URL's path looks like this:
-            #  /201012/17/505835/720p_1500k_505835/YouPorn%20-%20Sex%20Ed%20Is%20It%20Safe%20To%20Masturbate%20Daily.mp4
-            #  /201012/17/505835/vl_240p_240k_505835/YouPorn%20-%20Sex%20Ed%20Is%20It%20Safe%20To%20Masturbate%20Daily.mp4
-            #  /videos/201703/11/109285532/1080P_4000K_109285532.mp4
-            # We will benefit from it by extracting some metadata
-            mobj = re.search(r'(?P<height>\d{3,4})[pP]_(?P<bitrate>\d+)[kK]_\d+', video_url)
-            if mobj:
-                height = int(mobj.group('height'))
-                bitrate = int(mobj.group('bitrate'))
-                f.update({
-                    'format_id': '%dp-%dk' % (height, bitrate),
-                    'height': height,
-                    'tbr': bitrate,
-                })
-            formats.append(f)
-        self._sort_formats(formats)
-
         description = self._html_search_regex(
             r'(?s)<div[^>]+\bid=["\']description["\'][^>]*>(.+?)</div>',
             webpage, 'description',
@@ -169,13 +144,12 @@ class YouPornIE(InfoExtractor):
 
         age_limit = self._rta_search(webpage)
 
-        average_rating = int_or_none(self._search_regex(
-            r'<div[^>]+class=["\']videoRatingPercentage["\'][^>]*>(\d+)%</div>',
-            webpage, 'average rating', fatal=False))
-
-        view_count = str_to_int(self._search_regex(
-            r'(?s)<div[^>]+class=(["\']).*?\bvideoInfoViews\b.*?\1[^>]*>.*?(?P<count>[\d,.]+)<',
-            webpage, 'view count', fatal=False, group='count'))
+        view_count = None
+        views = self._search_regex(
+            r'(<div[^>]+\bclass=["\']js_videoInfoViews["\']>)', webpage,
+            'views', default=None)
+        if views:
+            view_count = str_to_int(extract_attributes(views).get('data-value'))
         comment_count = str_to_int(self._search_regex(
             r'>All [Cc]omments? \(([\d,.]+)\)',
             webpage, 'comment count', default=None))
@@ -201,7 +175,6 @@ class YouPornIE(InfoExtractor):
             'duration': duration,
             'uploader': uploader,
             'upload_date': upload_date,
-            'average_rating': average_rating,
             'view_count': view_count,
             'comment_count': comment_count,
             'categories': categories,

haruhi_dl/extractor/youtube.py

@@ -4,6 +4,7 @@ from __future__ import unicode_literals
 from datetime import datetime
 import json
 import hashlib
+from inspect import getsource
 import random
 import re
 import time
@@ -28,7 +29,6 @@ from ..utils import (
     float_or_none,
     get_element_by_id,
     int_or_none,
-    list_geoblocked_countres,
     mimetype2ext,
     parse_codecs,
     parse_duration,
@@ -44,7 +44,12 @@ from ..utils import (
     uppercase_escape,
     url_or_none,
     urlencode_postdata,
+    GeoRestrictedError,
 )
+try:
+    from ..extractor_artifacts.youtube import _decrypt_signature_protected
+except ImportError:
+    _decrypt_signature_protected = None
 
 
 class YoutubeBaseInfoExtractor(InfoExtractor):
@@ -567,24 +572,6 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
                 'format': '141/bestaudio[ext=m4a]',
             },
         },
-        # JS player signature function name containing $
-        {
-            'url': 'https://www.youtube.com/watch?v=nfWlot6h_JM',
-            'info_dict': {
-                'id': 'nfWlot6h_JM',
-                'ext': 'm4a',
-                'title': 'Taylor Swift - Shake It Off',
-                'description': 'md5:9dc0bd58efe700594b54f7d82bed0bac',
-                'duration': 242,
-                'uploader': 'TaylorSwiftVEVO',
-                'uploader_id': 'TaylorSwiftVEVO',
-                'upload_date': '20140818',
-            },
-            'params': {
-                'youtube_include_dash_manifest': True,
-                'format': '141/bestaudio[ext=m4a]',
-            },
-        },
         # Normal age-gate video (No vevo, embed allowed)
         {
             'url': 'https://youtube.com/watch?v=HtVdAasjOgU',
@@ -636,24 +623,6 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
                 'DASH manifest missing',
             ]
         },
-        # Olympics (https://github.com/ytdl-org/youtube-dl/issues/4431)
-        {
-            'url': 'lqQg6PlCWgI',
-            'info_dict': {
-                'id': 'lqQg6PlCWgI',
-                'ext': 'mp4',
-                'duration': 6085,
-                'upload_date': '20150827',
-                'uploader_id': 'olympic',
-                'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/olympic',
-                'description': 'HO09  - Women -  GER-AUS - Hockey - 31 July 2012 - London 2012 Olympic Games',
-                'uploader': 'Olympic',
-                'title': 'Hockey - Women -  GER-AUS - London 2012 Olympic Games',
-            },
-            'params': {
-                'skip_download': 'requires avconv',
-            }
-        },
         # Non-square pixels
         {
             'url': 'https://www.youtube.com/watch?v=_b-2C3KPAM0',
@@ -879,26 +848,6 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
             'url': 'https://www.youtubekids.com/watch?v=BnC-cpUCdns',
             'only_matching': True,
         },
-        {
-            # invalid -> valid video id redirection
-            'url': 'DJztXj2GPfl',
-            'info_dict': {
-                'id': 'DJztXj2GPfk',
-                'ext': 'mp4',
-                'title': 'Panjabi MC - Mundian To Bach Ke (The Dictator Soundtrack)',
-                'description': 'md5:bf577a41da97918e94fa9798d9228825',
-                'upload_date': '20090125',
-                'uploader': 'Prochorowka',
-                'uploader_id': 'Prochorowka',
-                'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/Prochorowka',
-                'artist': 'Panjabi MC',
-                'track': 'Beware of the Boys (Mundian to Bach Ke) - Motivo Hi-Lectro Remix',
-                'album': 'Beware of the Boys (Mundian To Bach Ke)',
-            },
-            'params': {
-                'skip_download': True,
-            },
-        },
         {
             # empty description results in an empty string
             'url': 'https://www.youtube.com/watch?v=x41yOUIvK2k',
@@ -919,6 +868,10 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
             'url': 'https://youtube.com/shorts/7awd-y_DTQY',
             'only_matching': True,
         },
+        {
+            'url': 'https://www.youtube.com/video/2NDLF-k2PwA',
+            'only_matching': True,
+        }
     ]
 
     _VALID_SIG_VALUE_RE = r'^AO[a-zA-Z0-9_-]+=*$'
@@ -953,7 +906,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
             raise ExtractorError('Cannot identify player %r' % player_url)
         return id_m.group('id')
 
-    def _extract_signature_function(self, video_id, player_url, example_sig):
+    def _extract_signature_function(self, video_id, player_url):
         player_id = self._extract_player_info(player_url)
 
         # Read from filesystem cache
@@ -1064,33 +1017,45 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
                 '    return %s\n') % (signature_id_tuple, expr_code)
         self.to_screen('Extracted signature function:\n' + code)
 
-    def mess(self, a, b):
+    @staticmethod
+    def mess(a, b):
         c = a[0]
         a[0] = a[b % len(a)]
         a[b % len(a)] = c
         return a
 
-    def _decrypt_signature_protected(self, s):
-        a = list(s)
-        a = a[3:]
-        a = self.mess(a, 63)
-        a = self.mess(a, 20)
-        a = a[1:]
-        a.reverse()
-        a = self.mess(a, 44)
-        a = a[1:]
-        a.reverse()
-        return "".join(a)
-
     def _full_signature_handling(self, sig, player_url, video_id):
-        signature = self._decrypt_signature_protected(sig)
-        if re.match(self._VALID_SIG_VALUE_RE, signature):
-            return signature
+        if _decrypt_signature_protected:
+            signature = _decrypt_signature_protected(sig)
+            if re.match(self._VALID_SIG_VALUE_RE, signature):
+                return signature
         if self._downloader.params.get('verbose'):
             self.to_screen("Built-in signature decryption failed, trying dynamic")
-        sig_decrypt_stack = self._extract_signature_function(video_id, player_url, sig)
+        sig_decrypt_stack = self._extract_signature_function(video_id, player_url)
         return self._do_decrypt_signature(sig, sig_decrypt_stack)
 
+    def _generate_prerelease_file(self):
+        # It's Monday, so I'm in a bad mood, but at least my sailor uniform is super cute!
+        video_id = 'ieQ1rAIjzXc'
+        self._set_consent()
+        webpage = self._download_webpage('https://www.youtube.com/watch?v=%s' % video_id, video_id)
+        player_url = self._search_regex(r'"jsUrl":"(/s/player/.*?/player_ias.vflset/.*?/base.js)', webpage, 'player url')
+        sig_decrypt_stack = self._extract_signature_function(video_id, player_url)
+        func = re.sub(r'(?m)^    ', '', getsource(self.mess).replace('@staticmethod', ''))
+        func += '\n\ndef _decrypt_signature_protected(sig):\n'
+        stack = ['a = list(sig)']
+        for fun in sig_decrypt_stack:
+            if fun[0] == 'splice':
+                stack.append(f'a = a[{fun[1]}:]')
+            elif fun[0] == 'reverse':
+                stack.append('a.reverse()')
+            elif fun[0] == 'mess':
+                stack.append(f'a = mess(a, {fun[1]})')
+            else:
+                raise ExtractorError('Unknown stack action: %s' % (fun[0]))
+        stack.append("return ''.join(a)")
+        return func + '\n'.join(map(lambda x: ' ' * 4 + x, stack)) + '\n'
+
     def _get_subtitles(self, video_id, webpage):
         try:
             subs_doc = self._download_xml(
@@ -1132,7 +1097,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
             # to be implemented in future that will replace this workaround (see
             # https://github.com/ytdl-org/youtube-dl/issues/7468,
             # https://github.com/ytdl-org/youtube-dl/pull/7599)
-            r';ytplayer\.config\s*=\s*({.+?});ytplayer',
+            r';ytplayer\.config\s*=\s*({.+?});\s*ytplayer',
             r';ytplayer\.config\s*=\s*({.+?});',
         )
         config = self._search_regex(
@@ -1473,33 +1438,35 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
 
         # Get video info
         video_info = {}
-        embed_webpage = None
         if (self._og_search_property('restrictions:age', video_webpage, default=None) == '18+'
                 or re.search(r'player-age-gate-content">', video_webpage) is not None):
             age_gate = True
-            # We simulate the access to the video from www.youtube.com/v/{video_id}
-            # this can be viewed without login into Youtube
-            url = proto + '://www.youtube.com/embed/%s' % video_id
-            embed_webpage = self._download_webpage(url, video_id, 'Downloading embed webpage')
-            data = compat_urllib_parse_urlencode({
-                'video_id': video_id,
-                'eurl': 'https://youtube.googleapis.com/v/' + video_id,
-                'html5': 1,
-            })
-            video_info_url = proto + '://www.youtube.com/get_video_info?' + data
             try:
-                video_info_webpage = self._download_webpage(
-                    video_info_url, video_id,
-                    note='Refetching age-gated info webpage',
-                    errnote='unable to download video info webpage')
+                yti1_player = self._download_webpage(
+                    proto + '://www.youtube.com/youtubei/v1/player', video_id,
+                    headers={
+                        'User-Agent': 'Mozilla/5.0 (SMART-TV; Linux; Tizen 4.0) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/56.0.2924.0 Safari/537.36',
+                        'Content-Type': 'application/json',
+                        'X-Goog-Api-Key': self._YOUTUBE_API_KEY,
+                    },
+                    data=bytes(json.dumps({
+                        'context': {
+                            'client': {
+                                'clientName': 'WEB',
+                                'clientVersion': '2.20210721.00.00',
+                                'clientScreen': 'EMBED',
+                            },
+                        },
+                        'videoId': video_id,
+                    }).encode('utf-8')),
+                    note='Downloading age-gated player info',
+                    errnote='unable to download video info')
             except ExtractorError:
-                video_info_webpage = None
-            if video_info_webpage:
-                video_info = compat_parse_qs(video_info_webpage)
-                pl_response = video_info.get('player_response', [None])[0]
-                player_response = extract_player_response(pl_response, video_id)
+                yti1_player = None
+            if yti1_player:
+                player_response = extract_player_response(yti1_player, video_id)
                 add_dash_mpd(video_info)
-                view_count = extract_view_count(video_info)
+                view_count = extract_view_count(video_id)
         else:
             age_gate = False
             # Try looking directly into the video webpage
@@ -1522,9 +1489,11 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
                     player_response = extract_player_response(args.get('player_response'), video_id)
             if not player_response:
                 player_response = extract_player_response(
-                    self._search_regex(
+                    self._search_regex((
+                        # js-like syntax
+                        r'(?:window(?:\["|\.)|var )ytInitialPlayerResponse(?:"])?\s*=\s*({.+?(?!\\)});(?:if \(ytcsi|var [a-zA-Z\_])',
                         r'(?:window(?:\["|\.)|var )ytInitialPlayerResponse(?:"])?\s*=\s*({.+?(?!\\)});',
-                        video_webpage, 'ytInitialPlayerResponse', fatal=False), video_id)
+                    ), video_webpage, 'ytInitialPlayerResponse', fatal=False), video_id)
             if not video_info or self._downloader.params.get('youtube_include_dash_manifest', True):
                 add_dash_mpd_pr(player_response)
 
@@ -1722,24 +1691,16 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
                         ASSETS_RE = r'"jsUrl":"(/s/player/.*?/player_ias.vflset/.*?/base.js)'
 
                         player_url = self._search_regex(
-                            ASSETS_RE,
-                            embed_webpage if age_gate else video_webpage, '', default=player_url)
+                            ASSETS_RE, video_webpage, '', default=player_url)
 
                         if not player_url and not age_gate:
                             # We need the embed website after all
-                            if embed_webpage is None:
-                                embed_url = proto + '://www.youtube.com/embed/%s' % video_id
-                                embed_webpage = self._download_webpage(
-                                    embed_url, video_id, 'Downloading embed webpage')
+                            embed_url = proto + '://www.youtube.com/embed/%s' % video_id
+                            embed_webpage = self._download_webpage(
+                                embed_url, video_id, 'Downloading embed webpage')
                             player_url = self._search_regex(
                                 ASSETS_RE, embed_webpage, 'JS player URL')
 
-                        # if player_url is None:
-                        #    player_url_json = self._search_regex(
-                        #        r'ytplayer\.config.*?"url"\s*:\s*("[^"]+")',
-                        #        video_webpage, 'age gate player URL')
-                        #    player_url = json.loads(player_url_json)
-
                     if 'sig' in url_data:
                         url += '&signature=' + url_data['sig'][0]
                     elif 's' in url_data:
@@ -1871,12 +1832,11 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
                              or ', who has blocked it on copyright grounds' in error_desc
                              or 'It is not available in your country.' in error_desc
                              or ', who has blocked it in your country on copyright grounds.' in error_desc):
-                        raise ExtractorError(
-                            list_geoblocked_countres(
-                                self._search_regex(
-                                    r'<meta itemprop="regionsAllowed" content="((?:(?:[A-Z]{2},)*[A-Z]{2})?)">',
-                                    video_webpage, 'allowed region list').split(',')),
-                            expected=True)
+                        raise GeoRestrictedError(
+                            error_desc,
+                            countries=self._search_regex(
+                                r'<meta itemprop="regionsAllowed" content="((?:(?:[A-Z]{2},)*[A-Z]{2})?)">',
+                                video_webpage, 'allowed region list').split(','))
                     if error_desc and 'Playback on other websites has been disabled' in error_desc:
                         raise ExtractorError(
                             'Embeds disabled for this video, account (with passed credit card or photo ID check, if in EU/EEA/CH/UK) is required',
@@ -2285,8 +2245,9 @@ class YoutubeBaseListInfoExtractor(YoutubeBaseInfoExtractor):
         webpage = self._download_webpage(url, list_id,
                                          note='Downloading %s page #1 (webpage)' % (self._LIST_NAME))
         return self._parse_json(
-            self._search_regex(
-                r'(?:window(?:\["|\.)|var )ytInitialData(?:"])?\s*=\s*({.+});',
+            self._search_regex((
+                r'(?:window(?:\["|\.)|var )ytInitialData(?:"])?\s*=\s*({.+});</script>',
+                r'(?:window(?:\["|\.)|var )ytInitialData(?:"])?\s*=\s*({.+});'),
                 webpage, 'initial data JSON'), 'initial data JSON'), webpage
 
     def _real_extract(self, url, results=None, query=None):

haruhi_dl/options.py

@@ -604,6 +604,10 @@ def parseOpts(overrideArguments=None):
             'Upper bound of a range for randomized sleep before each download '
             '(maximum possible number of seconds to sleep). Must only be used '
             'along with --min-sleep-interval.'))
+    workarounds.add_option(
+        '--force-playwright-browser', dest='force_playwright_browser',
+        help=('Force use a selected browser with extractors using Playwright.'
+              'Must be one of: firefox, chromium, webkit.'))
 
     verbosity = optparse.OptionGroup(parser, 'Verbosity / Simulation Options')
     verbosity.add_option(
@@ -709,7 +713,7 @@ def parseOpts(overrideArguments=None):
         help='Do NOT contact the haruhi-dl server for debugging')
     verbosity.add_option(
         '--no-headless-playwright',
-        dest='headless_playwright', action='store_false', default=False,
+        dest='headless_playwright', action='store_false', default=True,
         help='Show browsers automated with Playwright on the screen')
 
     filesystem = optparse.OptionGroup(parser, 'Filesystem Options')

haruhi_dl/playwright.py

@@ -1,7 +1,5 @@
 # coding: utf-8
-from __future__ import unicode_literals
-
-from .compat import compat_cookiejar_Cookie
+from http.cookiejar import Cookie
 from .utils import (
     ExtractorError,
     is_outdated_version,
@@ -87,15 +85,17 @@ class PlaywrightHelper():
     def _set_cookies_from_browser(self, cookies):
         for cookie in cookies:
             self._extractor._downloader.cookiejar.set_cookie(
-                compat_cookiejar_Cookie(0, cookie['name'], cookie['value'], cookie.get('port'), False,
-                                        cookie['domain'], False, cookie['domain'].startswith('.'),
-                                        cookie['path'], cookie['path'] != '/',
-                                        cookie['secure'], cookie['expires'],
-                                        False, None, None, None))
+                Cookie(0, cookie['name'], cookie['value'], cookie.get('port'), False,
+                       cookie['domain'], False, cookie['domain'].startswith('.'),
+                       cookie['path'], cookie['path'] != '/',
+                       cookie['secure'], cookie['expires'],
+                       False, None, None, None))
 
     def open_page(self, url, display_id, browser_used='firefox', note='Opening page in %(browser)s', html=None):
         pw = self.pw()
         self.pw_instance = pw
+        if self._extractor._downloader.params.get('force_playwright_browser') is not None:
+            browser_used = self._extractor._downloader.params.get('force_playwright_browser')
         browser = {
             'firefox': pw.firefox,
             'chromium': pw.chromium,

haruhi_dl/postprocessor/ffmpeg.py

@@ -231,7 +231,10 @@ class FFmpegPostProcessor(PostProcessor):
         stdout, stderr = p.communicate()
         if p.returncode != 0:
             stderr = stderr.decode('utf-8', 'replace')
-            msg = stderr.strip().split('\n')[-1]
+            msgs = stderr.strip().split('\n')
+            msg = msgs[-1]
+            if self._downloader.params.get('verbose', False):
+                self._downloader.to_screen('[debug] ' + '\n'.join(msgs[:-1]))
             raise FFmpegPostProcessorError(msg)
         self.try_utime(out_path, oldest_mtime, oldest_mtime)
 

haruhi_dl/utils.py

@@ -5775,284 +5775,3 @@ def clean_podcast_url(url):
                 st\.fm # https://podsights.com/docs/
             )/e
         )/''', '', url)
-
-
-# http://country.io/names.json
-country_list = {
-    "BD": "Bangladesh",
-    "BE": "Belgium",
-    "BF": "Burkina Faso",
-    "BG": "Bulgaria",
-    "BA": "Bosnia and Herzegovina",
-    "BB": "Barbados",
-    "WF": "Wallis and Futuna",
-    "BL": "Saint Barthelemy",
-    "BM": "Bermuda",
-    "BN": "Brunei",
-    "BO": "Bolivia",
-    "BH": "Bahrain",
-    "BI": "Burundi",
-    "BJ": "Benin",
-    "BT": "Bhutan",
-    "JM": "Jamaica",
-    "BV": "Bouvet Island",
-    "BW": "Botswana",
-    "WS": "Samoa",
-    "BQ": "Bonaire, Saint Eustatius and Saba ",
-    "BR": "Brazil",
-    "BS": "Bahamas",
-    "JE": "Jersey",
-    "BY": "Belarus",
-    "BZ": "Belize",
-    "RU": "Russia",
-    "RW": "Rwanda",
-    "RS": "Serbia",
-    "TL": "East Timor",
-    "RE": "Reunion",
-    "TM": "Turkmenistan",
-    "TJ": "Tajikistan",
-    "RO": "Romania",
-    "TK": "Tokelau",
-    "GW": "Guinea-Bissau",
-    "GU": "Guam",
-    "GT": "Guatemala",
-    "GS": "South Georgia and the South Sandwich Islands",
-    "GR": "Greece",
-    "GQ": "Equatorial Guinea",
-    "GP": "Guadeloupe",
-    "JP": "Japan",
-    "GY": "Guyana",
-    "GG": "Guernsey",
-    "GF": "French Guiana",
-    "GE": "Georgia",
-    "GD": "Grenada",
-    "GB": "United Kingdom",
-    "GA": "Gabon",
-    "SV": "El Salvador",
-    "GN": "Guinea",
-    "GM": "Gambia",
-    "GL": "Greenland",
-    "GI": "Gibraltar",
-    "GH": "Ghana",
-    "OM": "Oman",
-    "TN": "Tunisia",
-    "JO": "Jordan",
-    "HR": "Croatia",
-    "HT": "Haiti",
-    "HU": "Hungary",
-    "HK": "Hong Kong",
-    "HN": "Honduras",
-    "HM": "Heard Island and McDonald Islands",
-    "VE": "Venezuela",
-    "PR": "Puerto Rico",
-    "PS": "Palestinian Territory",
-    "PW": "Palau",
-    "PT": "Portugal",
-    "SJ": "Svalbard and Jan Mayen",
-    "PY": "Paraguay",
-    "IQ": "Iraq",
-    "PA": "Panama",
-    "PF": "French Polynesia",
-    "PG": "Papua New Guinea",
-    "PE": "Peru",
-    "PK": "Pakistan",
-    "PH": "Philippines",
-    "PN": "Pitcairn",
-    "PL": "Poland",
-    "PM": "Saint Pierre and Miquelon",
-    "ZM": "Zambia",
-    "EH": "Western Sahara",
-    "EE": "Estonia",
-    "EG": "Egypt",
-    "ZA": "South Africa",
-    "EC": "Ecuador",
-    "IT": "Italy",
-    "VN": "Vietnam",
-    "SB": "Solomon Islands",
-    "ET": "Ethiopia",
-    "SO": "Somalia",
-    "ZW": "Zimbabwe",
-    "SA": "Saudi Arabia",
-    "ES": "Spain",
-    "ER": "Eritrea",
-    "ME": "Montenegro",
-    "MD": "Moldova",
-    "MG": "Madagascar",
-    "MF": "Saint Martin",
-    "MA": "Morocco",
-    "MC": "Monaco",
-    "UZ": "Uzbekistan",
-    "MM": "Myanmar",
-    "ML": "Mali",
-    "MO": "Macao",
-    "MN": "Mongolia",
-    "MH": "Marshall Islands",
-    "MK": "Macedonia",
-    "MU": "Mauritius",
-    "MT": "Malta",
-    "MW": "Malawi",
-    "MV": "Maldives",
-    "MQ": "Martinique",
-    "MP": "Northern Mariana Islands",
-    "MS": "Montserrat",
-    "MR": "Mauritania",
-    "IM": "Isle of Man",
-    "UG": "Uganda",
-    "TZ": "Tanzania",
-    "MY": "Malaysia",
-    "MX": "Mexico",
-    "IL": "Israel",
-    "FR": "France",
-    "IO": "British Indian Ocean Territory",
-    "SH": "Saint Helena",
-    "FI": "Finland",
-    "FJ": "Fiji",
-    "FK": "Falkland Islands",
-    "FM": "Micronesia",
-    "FO": "Faroe Islands",
-    "NI": "Nicaragua",
-    "NL": "Netherlands",
-    "NO": "Norway",
-    "NA": "Namibia",
-    "VU": "Vanuatu",
-    "NC": "New Caledonia",
-    "NE": "Niger",
-    "NF": "Norfolk Island",
-    "NG": "Nigeria",
-    "NZ": "New Zealand",
-    "NP": "Nepal",
-    "NR": "Nauru",
-    "NU": "Niue",
-    "CK": "Cook Islands",
-    "XK": "Kosovo",
-    "CI": "Ivory Coast",
-    "CH": "Switzerland",
-    "CO": "Colombia",
-    "CN": "China",
-    "CM": "Cameroon",
-    "CL": "Chile",
-    "CC": "Cocos Islands",
-    "CA": "Canada",
-    "CG": "Republic of the Congo",
-    "CF": "Central African Republic",
-    "CD": "Democratic Republic of the Congo",
-    "CZ": "Czech Republic",
-    "CY": "Cyprus",
-    "CX": "Christmas Island",
-    "CR": "Costa Rica",
-    "CW": "Curacao",
-    "CV": "Cape Verde",
-    "CU": "Cuba",
-    "SZ": "Swaziland",
-    "SY": "Syria",
-    "SX": "Sint Maarten",
-    "KG": "Kyrgyzstan",
-    "KE": "Kenya",
-    "SS": "South Sudan",
-    "SR": "Suriname",
-    "KI": "Kiribati",
-    "KH": "Cambodia",
-    "KN": "Saint Kitts and Nevis",
-    "KM": "Comoros",
-    "ST": "Sao Tome and Principe",
-    "SK": "Slovakia",
-    "KR": "South Korea",
-    "SI": "Slovenia",
-    "KP": "North Korea",
-    "KW": "Kuwait",
-    "SN": "Senegal",
-    "SM": "San Marino",
-    "SL": "Sierra Leone",
-    "SC": "Seychelles",
-    "KZ": "Kazakhstan",
-    "KY": "Cayman Islands",
-    "SG": "Singapore",
-    "SE": "Sweden",
-    "SD": "Sudan",
-    "DO": "Dominican Republic",
-    "DM": "Dominica",
-    "DJ": "Djibouti",
-    "DK": "Denmark",
-    "VG": "British Virgin Islands",
-    "DE": "Germany",
-    "YE": "Yemen",
-    "DZ": "Algeria",
-    "US": "United States",
-    "UY": "Uruguay",
-    "YT": "Mayotte",
-    "UM": "United States Minor Outlying Islands",
-    "LB": "Lebanon",
-    "LC": "Saint Lucia",
-    "LA": "Laos",
-    "TV": "Tuvalu",
-    "TW": "Taiwan",
-    "TT": "Trinidad and Tobago",
-    "TR": "Turkey",
-    "LK": "Sri Lanka",
-    "LI": "Liechtenstein",
-    "LV": "Latvia",
-    "TO": "Tonga",
-    "LT": "Lithuania",
-    "LU": "Luxembourg",
-    "LR": "Liberia",
-    "LS": "Lesotho",
-    "TH": "Thailand",
-    "TF": "French Southern Territories",
-    "TG": "Togo",
-    "TD": "Chad",
-    "TC": "Turks and Caicos Islands",
-    "LY": "Libya",
-    "VA": "Vatican",
-    "VC": "Saint Vincent and the Grenadines",
-    "AE": "United Arab Emirates",
-    "AD": "Andorra",
-    "AG": "Antigua and Barbuda",
-    "AF": "Afghanistan",
-    "AI": "Anguilla",
-    "VI": "U.S. Virgin Islands",
-    "IS": "Iceland",
-    "IR": "Iran",
-    "AM": "Armenia",
-    "AL": "Albania",
-    "AO": "Angola",
-    "AQ": "Antarctica",
-    "AS": "American Samoa",
-    "AR": "Argentina",
-    "AU": "Australia",
-    "AT": "Austria",
-    "AW": "Aruba",
-    "IN": "India",
-    "AX": "Aland Islands",
-    "AZ": "Azerbaijan",
-    "IE": "Ireland",
-    "ID": "Indonesia",
-    "UA": "Ukraine",
-    "QA": "Qatar",
-    "MZ": "Mozambique"
-}
-
-
-def list_countries():
-    return country_list.keys()
-
-
-def list_geoblocked_countres(allowed_countries, reverse=False):
-    geoblocked = []
-    geounlocked = []
-    for country in list_countries():
-        if country in allowed_countries or (country not in allowed_countries and reverse is True):
-            geounlocked.append(country)
-        elif country not in allowed_countries or (country in allowed_countries and reverse is True):
-            geoblocked.append(country)
-    if len(geounlocked) == 0:
-        return 'This video is blocked in all countries'
-    if len(geoblocked) <= 10:
-        return 'This video is blocked in these countries: %s' % ', '.join(sorted(country_list[cnt] for cnt in geoblocked))
-    if len(geounlocked) <= 10:
-        geoblocked.sort()
-        return 'This video is only available in these countries: %s' % ', '.join(country_list[cnt] for cnt in geounlocked)
-    if len(geoblocked) >= len(geounlocked):
-        geounlocked.sort()
-        return 'This video is only available in these countries: %s' % ', '.join(geounlocked)
-    geoblocked.sort()
-    return 'This video is blocked in these countries: %s' % ', '.join(geoblocked)

haruhi_dl/version.py

@@ -1,6 +1,6 @@
 from __future__ import unicode_literals
 
-__version__ = '2021.06.01'
+__version__ = '2021.08.01'
 
 if __name__ == '__main__':
     print(__version__)

setup.py

@@ -115,7 +115,7 @@ setup(
     packages=[
         'haruhi_dl',
         'haruhi_dl.extractor', 'haruhi_dl.downloader',
-        'haruhi_dl.postprocessor'],
+        'haruhi_dl.postprocessor', 'haruhi_dl.extractor_artifacts'],
 
     # Provokes warning on most systems (why?!)
     # test_suite = 'nose.collector',