mirror of
https://github.com/yt-dlp/yt-dlp.git
synced 2024-12-01 02:49:31 +00:00
b827ee921f
* [scrippsnetworks] Add new extractor(closes #19857)(closes #22981) * [teachable] Improve locked lessons detection (#23528) * [teachable] Fail with error message if no video URL found * [extractors] add missing import for ScrippsNetworksIE * [brightcove] cache brightcove player policy keys * [prosiebensat1] improve geo restriction handling(closes #23571) * [soundcloud] automatically update client id on failing requests * [spankbang] Fix extraction (closes #23307, closes #23423, closes #23444) * [spankbang] Improve removed video detection (#23423) * [brightcove] update policy key on failing requests * [pornhub] Fix extraction and add support for m3u8 formats (closes #22749, closes #23082) * [pornhub] Improve locked videos detection (closes #22449, closes #22780) * [brightcove] invalidate policy key cache on failing requests * [soundcloud] fix client id extraction for non fatal requests * [ChangeLog] Actualize [ci skip] * [devscripts/create-github-release] Switch to using PAT for authentication Basic authentication will be deprecated soon * release 2020.01.01 * [redtube] Detect private videos (#23518) * [vice] improve extraction(closes #23631) * [devscripts/create-github-release] Remove unused import * [wistia] improve format extraction and extract subtitles(closes #22590) * [nrktv:seriebase] Fix extraction (closes #23625) (#23537) * [discovery] fix anonymous token extraction(closes #23650) * [scrippsnetworks] add support for www.discovery.com videos * [scrippsnetworks] correct test case URL * [dctp] fix format extraction(closes #23656) * [pandatv] Remove extractor (#23630) * [naver] improve extraction - improve geo-restriction handling - extract automatic captions - extract uploader metadata - extract VLive HLS formats * [naver] improve metadata extraction * [cloudflarestream] improve extraction - add support for bytehighway.net domain - add support for signed URLs - extract thumbnail * [cloudflarestream] import embed URL extraction * [lego] fix extraction and extract subtitle(closes #23687) * [safari] Fix kaltura session extraction (closes #23679) (#23670) * [orf:fm4] Fix extraction (#23599) * [orf:radio] Clean description and improve extraction * [twitter] add support for promo_video_website cards(closes #23711) * [vodplatform] add support for embed.kwikmotion.com domain * [ndr:base:embed] Improve thumbnails extraction (closes #23731) * [canvas] Add support for new API endpoint and update tests (closes #17680, closes #18629) * [travis] Add flake8 job (#23720) * [yourporn] Fix extraction (closes #21645, closes #22255, closes #23459) * [ChangeLog] Actualize [ci skip] * release 2020.01.15 * [soundcloud] Restore previews extraction (closes #23739) * [orf:tvthek] Improve geo restricted videos detection (closes #23741) * [zype] improve extraction - extract subtitles(closes #21258) - support URLs with alternative keys/tokens(#21258) - extract more metadata * [americastestkitchen] fix extraction * [nbc] add support for nbc multi network URLs(closes #23049) * [ard] improve extraction(closes #23761) - simplify extraction - extract age limit and series - bypass geo-restriction * [ivi:compilation] Fix entries extraction (closes #23770) * [24video] Add support for 24video.vip (closes #23753) * [businessinsider] Fix jwplatform id extraction (closes #22929) (#22954) * [ard] add a missing condition * [azmedien] fix extraction(closes #23783) * [voicerepublic] fix extraction * [stretchinternet] fix extraction(closes #4319) * [youtube] Fix sigfunc name extraction (closes #23819) * [ChangeLog] Actualize [ci skip] * release 2020.01.24 * [soundcloud] imporve private playlist/set tracks extraction https://github.com/ytdl-org/youtube-dl/issues/3707#issuecomment-577873539 * [svt] fix article extraction(closes #22897)(closes #22919) * [svt] fix series extraction(closes #22297) * [viewlift] improve extraction - fix extraction(closes #23851) - add add support for authentication - add support for more domains * [vimeo] fix album extraction(closes #23864) * [tva] Relax _VALID_URL (closes #23903) * [tv5mondeplus] Fix extraction (closes #23907, closes #23911) * [twitch:stream] Lowercase channel id for stream request (closes #23917) * [sportdeutschland] Update to new sportdeutschland API They switched to SSL, but under a different host AND path... Remove the old test cases because these videos have become unavailable. * [popcorntimes] Add extractor (closes #23949) * [thisoldhouse] fix extraction(closes #23951) * [toggle] Add support for mewatch.sg (closes #23895) (#23930) * [compat] Introduce compat_realpath (refs #23991) * [update] Fix updating via symlinks (closes #23991) * [nytimes] improve format sorting(closes #24010) * [abc:iview] Support 720p (#22907) (#22921) * [nova:embed] Fix extraction (closes #23672) * [nova:embed] Improve (closes #23690) * [nova] Improve extraction (refs #23690) * [jpopsuki] Remove extractor (closes #23858) * [YoutubeDL] Fix playlist entry indexing with --playlist-items (closes #10591, closes #10622) * [test_YoutubeDL] Fix get_ids * [test_YoutubeDL] Add tests for #10591 (closes #23873) * [24video] Add support for porn.24video.net (closes #23779, closes #23784) * [npr] Add support for streams (closes #24042) * [ChangeLog] Actualize [ci skip] * release 2020.02.16 * [tv2dk:bornholm:play] Fix extraction (#24076) * [imdb] Fix extraction (closes #23443) * [wistia] Add support for multiple generic embeds (closes #8347, closes #11385) * [teachable] Add support for multiple videos per lecture (closes #24101) * [pornhd] Fix extraction (closes #24128) * [options] Remove duplicate short option -v for --version (#24162) * [extractor/common] Convert ISM manifest to unicode before processing on python 2 (#24152) * [YoutubeDL] Force redirect URL to unicode on python 2 * Remove no longer needed compat_str around geturl * [youjizz] Fix extraction (closes #24181) * [test_subtitles] Remove obsolete test * [zdf:channel] Fix tests * [zapiks] Fix test * [xtube] Fix metadata extraction (closes #21073, closes #22455) * [xtube:user] Fix test * [telecinco] Fix extraction (refs #24195) * [telecinco] Add support for article opening videos * [franceculture] Fix extraction (closes #24204) * [xhamster] Fix extraction (closes #24205) * [ChangeLog] Actualize [ci skip] * release 2020.03.01 * [vimeo] Fix subtitles URLs (#24209) * [servus] Add support for new URL schema (closes #23475, closes #23583, closes #24142) * [youtube:playlist] Fix tests (closes #23872) (#23885) * [peertube] Improve extraction * [peertube] Fix issues and improve extraction (closes #23657) * [pornhub] Improve title extraction (closes #24184) * [vimeo] fix showcase password protected video extraction(closes #24224) * [youtube] Fix age-gated videos support without login (closes #24248) * [youtube] Fix tests * [ChangeLog] Actualize [ci skip] * release 2020.03.06 * [nhk] update API version(closes #24270) * [youtube] Improve extraction in 429 error conditions (closes #24283) * [youtube] Improve age-gated videos extraction in 429 error conditions (refs #24283) * [youtube] Remove outdated code Additional get_video_info requests don't seem to provide any extra itags any longer * [README.md] Clarify 429 error * [pornhub] Add support for pornhubpremium.com (#24288) * [utils] Add support for cookies with spaces used instead of tabs * [ChangeLog] Actualize [ci skip] * release 2020.03.08 * Revert "[utils] Add support for cookies with spaces used instead of tabs" According to [1] TABs must be used as separators between fields. Files produces by some tools with spaces as separators are considered malformed. 1. https://curl.haxx.se/docs/http-cookies.html This reverts commitcff99c91d1
. * [utils] Add reference to cookie file format * Revert "[vimeo] fix showcase password protected video extraction(closes #24224)" This reverts commit12ee431676
. * [nhk] Relax _VALID_URL (#24329) * [nhk] Remove obsolete rtmp formats (closes #24329) * [nhk] Update m3u8 URL and use native hls (#24329) * [ndr] Fix extraction (closes #24326) * [xtube] Fix formats extraction (closes #24348) * [xtube] Fix typo * [hellporno] Fix extraction (closes #24399) * [cbc:watch] Add support for authentication * [cbc:watch] Fix authenticated device token caching (closes #19160) * [soundcloud] fix download url extraction(closes #24394) * [limelight] remove disabled API requests(closes #24255) * [bilibili] Add support for new URL schema with BV ids (closes #24439, closes #24442) * [bilibili] Add support for player.bilibili.com (closes #24402) * [teachable] Extract chapter metadata (closes #24421) * [generic] Look for teachable embeds before wistia * [teachable] Update upskillcourses domain New version does not use teachable platform any longer * [teachable] Update gns3 domain * [teachable] Update test * [ChangeLog] Actualize [ci skip] * [ChangeLog] Actualize [ci skip] * release 2020.03.24 * [spankwire] Fix extraction (closes #18924, closes #20648) * [spankwire] Add support for generic embeds (refs #24633) * [youporn] Add support form generic embeds * [mofosex] Add support for generic embeds (closes #24633) * [tele5] Fix extraction (closes #24553) * [extractor/common] Skip malformed ISM manifest XMLs while extracting ISM formats (#24667) * [tv4] Fix ISM formats extraction (closes #24667) * [twitch:clips] Extend _VALID_URL (closes #24290) (#24642) * [motherless] Fix extraction (closes #24699) * [nova:embed] Fix extraction (closes #24700) * [youtube] Skip broken multifeed videos (closes #24711) * [soundcloud] Extract AAC format * [soundcloud] Improve AAC format extraction (closes #19173, closes #24708) * [thisoldhouse] Fix video id extraction (closes #24548) Added support for: with of without "www." and either ".chorus.build" or ".com" It now validated correctly on older URL's ``` <iframe src="https://thisoldhouse.chorus.build/videos/zype/5e33baec27d2e50001d5f52f ``` and newer ones ``` <iframe src="https://www.thisoldhouse.com/videos/zype/5e2b70e95216cc0001615120 ``` * [thisoldhouse] Improve video id extraction (closes #24549) * [youtube] Fix DRM videos detection (refs #24736) * [options] Clarify doc on --exec command (closes #19087) (#24883) * [prosiebensat1] Improve extraction and remove 7tv.de support (#24948) * [prosiebensat1] Extract series metadata * [tenplay] Relax _VALID_URL (closes #25001) * [tvplay] fix Viafree extraction(closes #15189)(closes #24473)(closes #24789) * [yahoo] fix GYAO Player extraction and relax title URL regex(closes #24178)(closes #24778) * [youtube] Use redirected video id if any (closes #25063) * [youtube] Improve player id extraction and add tests * [extractor/common] Extract multiple JSON-LD entries * [crunchyroll] Fix and improve extraction (closes #25096, closes #25060) * [ChangeLog] Actualize [ci skip] * release 2020.05.03 * [puhutv] Remove no longer available HTTP formats (closes #25124) * [utils] Improve cookie files support + Add support for UTF-8 in cookie files * Skip malformed cookie file entries instead of crashing (invalid entry len, invalid expires at) * [dailymotion] Fix typo * [compat] Introduce compat_cookiejar_Cookie * [extractor/common] Use compat_cookiejar_Cookie for _set_cookie (closes #23256, closes #24776) To always ensure cookie name and value are bytestrings on python 2. * [orf] Add support for more radio stations (closes #24938) (#24968) * [uol] fix extraction(closes #22007) * [downloader/http] Finish downloading once received data length matches expected Always do this if possible, i.e. if Content-Length or expected length is known, not only in test. This will save unnecessary last extra loop trying to read 0 bytes. * [downloader/http] Request last data block of exact remaining size Always request last data block of exact size remaining to download if possible not the current block size. * [iprima] Improve extraction (closes #25138) * [youtube] Improve signature cipher extraction (closes #25188) * [ChangeLog] Actualize [ci skip] * release 2020.05.08 * [spike] fix Bellator mgid extraction(closes #25195) * [bbccouk] PEP8 * [mailru] Fix extraction (closes #24530) (#25239) * [README.md] flake8 HTTPS URL (#25230) * [youtube] Add support for yewtu.be (#25226) * [soundcloud] reduce API playlist page limit(closes #25274) * [vimeo] improve format extraction and sorting(closes #25285) * [redtube] Improve title extraction (#25208) * [indavideo] Switch to HTTPS for API request (#25191) * [utils] Fix file permissions in write_json_file (closes #12471) (#25122) * [redtube] Improve formats extraction and extract m3u8 formats (closes #25311, closes #25321) * [ard] Improve _VALID_URL (closes #25134) (#25198) * [giantbomb] Extend _VALID_URL (#25222) * [postprocessor/ffmpeg] Embed series metadata with --add-metadata * [youtube] Add support for more invidious instances (#25417) * [ard:beta] Extend _VALID_URL (closes #25405) * [ChangeLog] Actualize [ci skip] * release 2020.05.29 * [jwplatform] Improve embeds extraction (closes #25467) * [periscope] Fix untitled broadcasts (#25482) * [twitter:broadcast] Add untitled periscope broadcast test * [malltv] Add support for sk.mall.tv (#25445) * [brightcove] Fix subtitles extraction (closes #25540) * [brightcove] Sort imports * [twitch] Pass v5 accept header and fix thumbnails extraction (closes #25531) * [twitch:stream] Fix extraction (closes #25528) * [twitch:stream] Expect 400 and 410 HTTP errors from API * [tele5] Prefer jwplatform over nexx (closes #25533) * [jwplatform] Add support for bypass geo restriction * [tele5] Bypass geo restriction * [ChangeLog] Actualize [ci skip] * release 2020.06.06 * [kaltura] Add support for multiple embeds on a webpage (closes #25523) * [youtube] Extract chapters from JSON (closes #24819) * [facebook] Support single-video ID links I stumbled upon this at https://www.facebook.com/bwfbadminton/posts/10157127020046316 . No idea how prevalent it is yet. * [youtube] Fix playlist and feed extraction (closes #25675) * [youtube] Fix thumbnails extraction and remove uploader id extraction warning (closes #25676) * [youtube] Fix upload date extraction * [youtube] Improve view count extraction * [youtube] Fix uploader id and uploader URL extraction * [ChangeLog] Actualize [ci skip] * release 2020.06.16 * [youtube] Fix categories and improve tags extraction * [youtube] Force old layout (closes #25682, closes #25683, closes #25680, closes #25686) * [ChangeLog] Actualize [ci skip] * release 2020.06.16.1 * [brightcove] Improve embed detection (closes #25674) * [bellmedia] add support for cp24.com clip URLs(closes #25764) * [youtube:playlists] Extend _VALID_URL (closes #25810) * [youtube] Prevent excess HTTP 301 (#25786) * [wistia] Restrict embed regex (closes #25969) * [youtube] Improve description extraction (closes #25937) (#25980) * [youtube] Fix sigfunc name extraction (closes #26134, closes #26135, closes #26136, closes #26137) * [ChangeLog] Actualize [ci skip] * release 2020.07.28 * [xhamster] Extend _VALID_URL (closes #25789) (#25804) * [xhamster] Fix extraction (closes #26157) (#26254) * [xhamster] Extend _VALID_URL (closes #25927) Co-authored-by: Remita Amine <remitamine@gmail.com> Co-authored-by: Sergey M․ <dstftw@gmail.com> Co-authored-by: nmeum <soeren+github@soeren-tempel.net> Co-authored-by: Roxedus <me@roxedus.dev> Co-authored-by: Singwai Chan <c.singwai@gmail.com> Co-authored-by: cdarlint <cdarlint@users.noreply.github.com> Co-authored-by: Johannes N <31795504+jonolt@users.noreply.github.com> Co-authored-by: jnozsc <jnozsc@gmail.com> Co-authored-by: Moritz Patelscheck <moritz.patelscheck@campus.tu-berlin.de> Co-authored-by: PB <3854688+uno20001@users.noreply.github.com> Co-authored-by: Philipp Hagemeister <phihag@phihag.de> Co-authored-by: Xaver Hellauer <software@hellauer.bayern> Co-authored-by: d2au <d2au.dev@gmail.com> Co-authored-by: Jan 'Yenda' Trmal <jtrmal@gmail.com> Co-authored-by: jxu <7989982+jxu@users.noreply.github.com> Co-authored-by: Martin Ström <name@my-domain.se> Co-authored-by: The Hatsune Daishi <nao20010128@gmail.com> Co-authored-by: tsia <github@tsia.de> Co-authored-by: 3risian <59593325+3risian@users.noreply.github.com> Co-authored-by: Tristan Waddington <tristan.waddington@gmail.com> Co-authored-by: Devon Meunier <devon.meunier@gmail.com> Co-authored-by: Felix Stupp <felix.stupp@outlook.com> Co-authored-by: tom <tomster954@gmail.com> Co-authored-by: AndrewMBL <62922222+AndrewMBL@users.noreply.github.com> Co-authored-by: willbeaufoy <will@willbeaufoy.net> Co-authored-by: Philipp Stehle <anderschwiedu@googlemail.com> Co-authored-by: hh0rva1h <61889859+hh0rva1h@users.noreply.github.com> Co-authored-by: comsomisha <shmelev1996@mail.ru> Co-authored-by: TotalCaesar659 <14265316+TotalCaesar659@users.noreply.github.com> Co-authored-by: Juan Francisco Cantero Hurtado <iam@juanfra.info> Co-authored-by: Dave Loyall <dave@the-good-guys.net> Co-authored-by: tlsssl <63866177+tlsssl@users.noreply.github.com> Co-authored-by: Rob <ankenyr@gmail.com> Co-authored-by: Michael Klein <github@a98shuttle.de> Co-authored-by: JordanWeatherby <47519158+JordanWeatherby@users.noreply.github.com> Co-authored-by: striker.sh <19488257+strikersh@users.noreply.github.com> Co-authored-by: Matej Dujava <mdujava@gmail.com> Co-authored-by: Glenn Slayden <5589855+glenn-slayden@users.noreply.github.com> Co-authored-by: MRWITEK <mrvvitek@gmail.com> Co-authored-by: JChris246 <43832407+JChris246@users.noreply.github.com> Co-authored-by: TheRealDude2 <the.real.dude@gmx.de>
925 lines
36 KiB
Python
925 lines
36 KiB
Python
#!/usr/bin/env python
|
|
# coding: utf-8
|
|
|
|
from __future__ import unicode_literals
|
|
|
|
# Allow direct execution
|
|
import os
|
|
import sys
|
|
import unittest
|
|
sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
|
|
|
|
import copy
|
|
|
|
from test.helper import FakeYDL, assertRegexpMatches
|
|
from youtube_dl import YoutubeDL
|
|
from youtube_dl.compat import compat_str, compat_urllib_error
|
|
from youtube_dl.extractor import YoutubeIE
|
|
from youtube_dl.extractor.common import InfoExtractor
|
|
from youtube_dl.postprocessor.common import PostProcessor
|
|
from youtube_dl.utils import ExtractorError, match_filter_func
|
|
|
|
TEST_URL = 'http://localhost/sample.mp4'
|
|
|
|
|
|
class YDL(FakeYDL):
|
|
def __init__(self, *args, **kwargs):
|
|
super(YDL, self).__init__(*args, **kwargs)
|
|
self.downloaded_info_dicts = []
|
|
self.msgs = []
|
|
|
|
def process_info(self, info_dict):
|
|
self.downloaded_info_dicts.append(info_dict)
|
|
|
|
def to_screen(self, msg):
|
|
self.msgs.append(msg)
|
|
|
|
|
|
def _make_result(formats, **kwargs):
|
|
res = {
|
|
'formats': formats,
|
|
'id': 'testid',
|
|
'title': 'testttitle',
|
|
'extractor': 'testex',
|
|
'extractor_key': 'TestEx',
|
|
}
|
|
res.update(**kwargs)
|
|
return res
|
|
|
|
|
|
class TestFormatSelection(unittest.TestCase):
|
|
def test_prefer_free_formats(self):
|
|
# Same resolution => download webm
|
|
ydl = YDL()
|
|
ydl.params['prefer_free_formats'] = True
|
|
formats = [
|
|
{'ext': 'webm', 'height': 460, 'url': TEST_URL},
|
|
{'ext': 'mp4', 'height': 460, 'url': TEST_URL},
|
|
]
|
|
info_dict = _make_result(formats)
|
|
yie = YoutubeIE(ydl)
|
|
yie._sort_formats(info_dict['formats'])
|
|
ydl.process_ie_result(info_dict)
|
|
downloaded = ydl.downloaded_info_dicts[0]
|
|
self.assertEqual(downloaded['ext'], 'webm')
|
|
|
|
# Different resolution => download best quality (mp4)
|
|
ydl = YDL()
|
|
ydl.params['prefer_free_formats'] = True
|
|
formats = [
|
|
{'ext': 'webm', 'height': 720, 'url': TEST_URL},
|
|
{'ext': 'mp4', 'height': 1080, 'url': TEST_URL},
|
|
]
|
|
info_dict['formats'] = formats
|
|
yie = YoutubeIE(ydl)
|
|
yie._sort_formats(info_dict['formats'])
|
|
ydl.process_ie_result(info_dict)
|
|
downloaded = ydl.downloaded_info_dicts[0]
|
|
self.assertEqual(downloaded['ext'], 'mp4')
|
|
|
|
# No prefer_free_formats => prefer mp4 and flv for greater compatibility
|
|
ydl = YDL()
|
|
ydl.params['prefer_free_formats'] = False
|
|
formats = [
|
|
{'ext': 'webm', 'height': 720, 'url': TEST_URL},
|
|
{'ext': 'mp4', 'height': 720, 'url': TEST_URL},
|
|
{'ext': 'flv', 'height': 720, 'url': TEST_URL},
|
|
]
|
|
info_dict['formats'] = formats
|
|
yie = YoutubeIE(ydl)
|
|
yie._sort_formats(info_dict['formats'])
|
|
ydl.process_ie_result(info_dict)
|
|
downloaded = ydl.downloaded_info_dicts[0]
|
|
self.assertEqual(downloaded['ext'], 'mp4')
|
|
|
|
ydl = YDL()
|
|
ydl.params['prefer_free_formats'] = False
|
|
formats = [
|
|
{'ext': 'flv', 'height': 720, 'url': TEST_URL},
|
|
{'ext': 'webm', 'height': 720, 'url': TEST_URL},
|
|
]
|
|
info_dict['formats'] = formats
|
|
yie = YoutubeIE(ydl)
|
|
yie._sort_formats(info_dict['formats'])
|
|
ydl.process_ie_result(info_dict)
|
|
downloaded = ydl.downloaded_info_dicts[0]
|
|
self.assertEqual(downloaded['ext'], 'flv')
|
|
|
|
def test_format_selection(self):
|
|
formats = [
|
|
{'format_id': '35', 'ext': 'mp4', 'preference': 1, 'url': TEST_URL},
|
|
{'format_id': 'example-with-dashes', 'ext': 'webm', 'preference': 1, 'url': TEST_URL},
|
|
{'format_id': '45', 'ext': 'webm', 'preference': 2, 'url': TEST_URL},
|
|
{'format_id': '47', 'ext': 'webm', 'preference': 3, 'url': TEST_URL},
|
|
{'format_id': '2', 'ext': 'flv', 'preference': 4, 'url': TEST_URL},
|
|
]
|
|
info_dict = _make_result(formats)
|
|
|
|
ydl = YDL({'format': '20/47'})
|
|
ydl.process_ie_result(info_dict.copy())
|
|
downloaded = ydl.downloaded_info_dicts[0]
|
|
self.assertEqual(downloaded['format_id'], '47')
|
|
|
|
ydl = YDL({'format': '20/71/worst'})
|
|
ydl.process_ie_result(info_dict.copy())
|
|
downloaded = ydl.downloaded_info_dicts[0]
|
|
self.assertEqual(downloaded['format_id'], '35')
|
|
|
|
ydl = YDL()
|
|
ydl.process_ie_result(info_dict.copy())
|
|
downloaded = ydl.downloaded_info_dicts[0]
|
|
self.assertEqual(downloaded['format_id'], '2')
|
|
|
|
ydl = YDL({'format': 'webm/mp4'})
|
|
ydl.process_ie_result(info_dict.copy())
|
|
downloaded = ydl.downloaded_info_dicts[0]
|
|
self.assertEqual(downloaded['format_id'], '47')
|
|
|
|
ydl = YDL({'format': '3gp/40/mp4'})
|
|
ydl.process_ie_result(info_dict.copy())
|
|
downloaded = ydl.downloaded_info_dicts[0]
|
|
self.assertEqual(downloaded['format_id'], '35')
|
|
|
|
ydl = YDL({'format': 'example-with-dashes'})
|
|
ydl.process_ie_result(info_dict.copy())
|
|
downloaded = ydl.downloaded_info_dicts[0]
|
|
self.assertEqual(downloaded['format_id'], 'example-with-dashes')
|
|
|
|
def test_format_selection_audio(self):
|
|
formats = [
|
|
{'format_id': 'audio-low', 'ext': 'webm', 'preference': 1, 'vcodec': 'none', 'url': TEST_URL},
|
|
{'format_id': 'audio-mid', 'ext': 'webm', 'preference': 2, 'vcodec': 'none', 'url': TEST_URL},
|
|
{'format_id': 'audio-high', 'ext': 'flv', 'preference': 3, 'vcodec': 'none', 'url': TEST_URL},
|
|
{'format_id': 'vid', 'ext': 'mp4', 'preference': 4, 'url': TEST_URL},
|
|
]
|
|
info_dict = _make_result(formats)
|
|
|
|
ydl = YDL({'format': 'bestaudio'})
|
|
ydl.process_ie_result(info_dict.copy())
|
|
downloaded = ydl.downloaded_info_dicts[0]
|
|
self.assertEqual(downloaded['format_id'], 'audio-high')
|
|
|
|
ydl = YDL({'format': 'worstaudio'})
|
|
ydl.process_ie_result(info_dict.copy())
|
|
downloaded = ydl.downloaded_info_dicts[0]
|
|
self.assertEqual(downloaded['format_id'], 'audio-low')
|
|
|
|
formats = [
|
|
{'format_id': 'vid-low', 'ext': 'mp4', 'preference': 1, 'url': TEST_URL},
|
|
{'format_id': 'vid-high', 'ext': 'mp4', 'preference': 2, 'url': TEST_URL},
|
|
]
|
|
info_dict = _make_result(formats)
|
|
|
|
ydl = YDL({'format': 'bestaudio/worstaudio/best'})
|
|
ydl.process_ie_result(info_dict.copy())
|
|
downloaded = ydl.downloaded_info_dicts[0]
|
|
self.assertEqual(downloaded['format_id'], 'vid-high')
|
|
|
|
def test_format_selection_audio_exts(self):
|
|
formats = [
|
|
{'format_id': 'mp3-64', 'ext': 'mp3', 'abr': 64, 'url': 'http://_', 'vcodec': 'none'},
|
|
{'format_id': 'ogg-64', 'ext': 'ogg', 'abr': 64, 'url': 'http://_', 'vcodec': 'none'},
|
|
{'format_id': 'aac-64', 'ext': 'aac', 'abr': 64, 'url': 'http://_', 'vcodec': 'none'},
|
|
{'format_id': 'mp3-32', 'ext': 'mp3', 'abr': 32, 'url': 'http://_', 'vcodec': 'none'},
|
|
{'format_id': 'aac-32', 'ext': 'aac', 'abr': 32, 'url': 'http://_', 'vcodec': 'none'},
|
|
]
|
|
|
|
info_dict = _make_result(formats)
|
|
ydl = YDL({'format': 'best'})
|
|
ie = YoutubeIE(ydl)
|
|
ie._sort_formats(info_dict['formats'])
|
|
ydl.process_ie_result(copy.deepcopy(info_dict))
|
|
downloaded = ydl.downloaded_info_dicts[0]
|
|
self.assertEqual(downloaded['format_id'], 'aac-64')
|
|
|
|
ydl = YDL({'format': 'mp3'})
|
|
ie = YoutubeIE(ydl)
|
|
ie._sort_formats(info_dict['formats'])
|
|
ydl.process_ie_result(copy.deepcopy(info_dict))
|
|
downloaded = ydl.downloaded_info_dicts[0]
|
|
self.assertEqual(downloaded['format_id'], 'mp3-64')
|
|
|
|
ydl = YDL({'prefer_free_formats': True})
|
|
ie = YoutubeIE(ydl)
|
|
ie._sort_formats(info_dict['formats'])
|
|
ydl.process_ie_result(copy.deepcopy(info_dict))
|
|
downloaded = ydl.downloaded_info_dicts[0]
|
|
self.assertEqual(downloaded['format_id'], 'ogg-64')
|
|
|
|
def test_format_selection_video(self):
|
|
formats = [
|
|
{'format_id': 'dash-video-low', 'ext': 'mp4', 'preference': 1, 'acodec': 'none', 'url': TEST_URL},
|
|
{'format_id': 'dash-video-high', 'ext': 'mp4', 'preference': 2, 'acodec': 'none', 'url': TEST_URL},
|
|
{'format_id': 'vid', 'ext': 'mp4', 'preference': 3, 'url': TEST_URL},
|
|
]
|
|
info_dict = _make_result(formats)
|
|
|
|
ydl = YDL({'format': 'bestvideo'})
|
|
ydl.process_ie_result(info_dict.copy())
|
|
downloaded = ydl.downloaded_info_dicts[0]
|
|
self.assertEqual(downloaded['format_id'], 'dash-video-high')
|
|
|
|
ydl = YDL({'format': 'worstvideo'})
|
|
ydl.process_ie_result(info_dict.copy())
|
|
downloaded = ydl.downloaded_info_dicts[0]
|
|
self.assertEqual(downloaded['format_id'], 'dash-video-low')
|
|
|
|
ydl = YDL({'format': 'bestvideo[format_id^=dash][format_id$=low]'})
|
|
ydl.process_ie_result(info_dict.copy())
|
|
downloaded = ydl.downloaded_info_dicts[0]
|
|
self.assertEqual(downloaded['format_id'], 'dash-video-low')
|
|
|
|
formats = [
|
|
{'format_id': 'vid-vcodec-dot', 'ext': 'mp4', 'preference': 1, 'vcodec': 'avc1.123456', 'acodec': 'none', 'url': TEST_URL},
|
|
]
|
|
info_dict = _make_result(formats)
|
|
|
|
ydl = YDL({'format': 'bestvideo[vcodec=avc1.123456]'})
|
|
ydl.process_ie_result(info_dict.copy())
|
|
downloaded = ydl.downloaded_info_dicts[0]
|
|
self.assertEqual(downloaded['format_id'], 'vid-vcodec-dot')
|
|
|
|
def test_format_selection_string_ops(self):
|
|
formats = [
|
|
{'format_id': 'abc-cba', 'ext': 'mp4', 'url': TEST_URL},
|
|
{'format_id': 'zxc-cxz', 'ext': 'webm', 'url': TEST_URL},
|
|
]
|
|
info_dict = _make_result(formats)
|
|
|
|
# equals (=)
|
|
ydl = YDL({'format': '[format_id=abc-cba]'})
|
|
ydl.process_ie_result(info_dict.copy())
|
|
downloaded = ydl.downloaded_info_dicts[0]
|
|
self.assertEqual(downloaded['format_id'], 'abc-cba')
|
|
|
|
# does not equal (!=)
|
|
ydl = YDL({'format': '[format_id!=abc-cba]'})
|
|
ydl.process_ie_result(info_dict.copy())
|
|
downloaded = ydl.downloaded_info_dicts[0]
|
|
self.assertEqual(downloaded['format_id'], 'zxc-cxz')
|
|
|
|
ydl = YDL({'format': '[format_id!=abc-cba][format_id!=zxc-cxz]'})
|
|
self.assertRaises(ExtractorError, ydl.process_ie_result, info_dict.copy())
|
|
|
|
# starts with (^=)
|
|
ydl = YDL({'format': '[format_id^=abc]'})
|
|
ydl.process_ie_result(info_dict.copy())
|
|
downloaded = ydl.downloaded_info_dicts[0]
|
|
self.assertEqual(downloaded['format_id'], 'abc-cba')
|
|
|
|
# does not start with (!^=)
|
|
ydl = YDL({'format': '[format_id!^=abc]'})
|
|
ydl.process_ie_result(info_dict.copy())
|
|
downloaded = ydl.downloaded_info_dicts[0]
|
|
self.assertEqual(downloaded['format_id'], 'zxc-cxz')
|
|
|
|
ydl = YDL({'format': '[format_id!^=abc][format_id!^=zxc]'})
|
|
self.assertRaises(ExtractorError, ydl.process_ie_result, info_dict.copy())
|
|
|
|
# ends with ($=)
|
|
ydl = YDL({'format': '[format_id$=cba]'})
|
|
ydl.process_ie_result(info_dict.copy())
|
|
downloaded = ydl.downloaded_info_dicts[0]
|
|
self.assertEqual(downloaded['format_id'], 'abc-cba')
|
|
|
|
# does not end with (!$=)
|
|
ydl = YDL({'format': '[format_id!$=cba]'})
|
|
ydl.process_ie_result(info_dict.copy())
|
|
downloaded = ydl.downloaded_info_dicts[0]
|
|
self.assertEqual(downloaded['format_id'], 'zxc-cxz')
|
|
|
|
ydl = YDL({'format': '[format_id!$=cba][format_id!$=cxz]'})
|
|
self.assertRaises(ExtractorError, ydl.process_ie_result, info_dict.copy())
|
|
|
|
# contains (*=)
|
|
ydl = YDL({'format': '[format_id*=bc-cb]'})
|
|
ydl.process_ie_result(info_dict.copy())
|
|
downloaded = ydl.downloaded_info_dicts[0]
|
|
self.assertEqual(downloaded['format_id'], 'abc-cba')
|
|
|
|
# does not contain (!*=)
|
|
ydl = YDL({'format': '[format_id!*=bc-cb]'})
|
|
ydl.process_ie_result(info_dict.copy())
|
|
downloaded = ydl.downloaded_info_dicts[0]
|
|
self.assertEqual(downloaded['format_id'], 'zxc-cxz')
|
|
|
|
ydl = YDL({'format': '[format_id!*=abc][format_id!*=zxc]'})
|
|
self.assertRaises(ExtractorError, ydl.process_ie_result, info_dict.copy())
|
|
|
|
ydl = YDL({'format': '[format_id!*=-]'})
|
|
self.assertRaises(ExtractorError, ydl.process_ie_result, info_dict.copy())
|
|
|
|
def test_youtube_format_selection(self):
|
|
order = [
|
|
'38', '37', '46', '22', '45', '35', '44', '18', '34', '43', '6', '5', '17', '36', '13',
|
|
# Apple HTTP Live Streaming
|
|
'96', '95', '94', '93', '92', '132', '151',
|
|
# 3D
|
|
'85', '84', '102', '83', '101', '82', '100',
|
|
# Dash video
|
|
'137', '248', '136', '247', '135', '246',
|
|
'245', '244', '134', '243', '133', '242', '160',
|
|
# Dash audio
|
|
'141', '172', '140', '171', '139',
|
|
]
|
|
|
|
def format_info(f_id):
|
|
info = YoutubeIE._formats[f_id].copy()
|
|
|
|
# XXX: In real cases InfoExtractor._parse_mpd_formats() fills up 'acodec'
|
|
# and 'vcodec', while in tests such information is incomplete since
|
|
# commit a6c2c24479e5f4827ceb06f64d855329c0a6f593
|
|
# test_YoutubeDL.test_youtube_format_selection is broken without
|
|
# this fix
|
|
if 'acodec' in info and 'vcodec' not in info:
|
|
info['vcodec'] = 'none'
|
|
elif 'vcodec' in info and 'acodec' not in info:
|
|
info['acodec'] = 'none'
|
|
|
|
info['format_id'] = f_id
|
|
info['url'] = 'url:' + f_id
|
|
return info
|
|
formats_order = [format_info(f_id) for f_id in order]
|
|
|
|
info_dict = _make_result(list(formats_order), extractor='youtube')
|
|
ydl = YDL({'format': 'bestvideo+bestaudio'})
|
|
yie = YoutubeIE(ydl)
|
|
yie._sort_formats(info_dict['formats'])
|
|
ydl.process_ie_result(info_dict)
|
|
downloaded = ydl.downloaded_info_dicts[0]
|
|
self.assertEqual(downloaded['format_id'], '137+141')
|
|
self.assertEqual(downloaded['ext'], 'mp4')
|
|
|
|
info_dict = _make_result(list(formats_order), extractor='youtube')
|
|
ydl = YDL({'format': 'bestvideo[height>=999999]+bestaudio/best'})
|
|
yie = YoutubeIE(ydl)
|
|
yie._sort_formats(info_dict['formats'])
|
|
ydl.process_ie_result(info_dict)
|
|
downloaded = ydl.downloaded_info_dicts[0]
|
|
self.assertEqual(downloaded['format_id'], '38')
|
|
|
|
info_dict = _make_result(list(formats_order), extractor='youtube')
|
|
ydl = YDL({'format': 'bestvideo/best,bestaudio'})
|
|
yie = YoutubeIE(ydl)
|
|
yie._sort_formats(info_dict['formats'])
|
|
ydl.process_ie_result(info_dict)
|
|
downloaded_ids = [info['format_id'] for info in ydl.downloaded_info_dicts]
|
|
self.assertEqual(downloaded_ids, ['137', '141'])
|
|
|
|
info_dict = _make_result(list(formats_order), extractor='youtube')
|
|
ydl = YDL({'format': '(bestvideo[ext=mp4],bestvideo[ext=webm])+bestaudio'})
|
|
yie = YoutubeIE(ydl)
|
|
yie._sort_formats(info_dict['formats'])
|
|
ydl.process_ie_result(info_dict)
|
|
downloaded_ids = [info['format_id'] for info in ydl.downloaded_info_dicts]
|
|
self.assertEqual(downloaded_ids, ['137+141', '248+141'])
|
|
|
|
info_dict = _make_result(list(formats_order), extractor='youtube')
|
|
ydl = YDL({'format': '(bestvideo[ext=mp4],bestvideo[ext=webm])[height<=720]+bestaudio'})
|
|
yie = YoutubeIE(ydl)
|
|
yie._sort_formats(info_dict['formats'])
|
|
ydl.process_ie_result(info_dict)
|
|
downloaded_ids = [info['format_id'] for info in ydl.downloaded_info_dicts]
|
|
self.assertEqual(downloaded_ids, ['136+141', '247+141'])
|
|
|
|
info_dict = _make_result(list(formats_order), extractor='youtube')
|
|
ydl = YDL({'format': '(bestvideo[ext=none]/bestvideo[ext=webm])+bestaudio'})
|
|
yie = YoutubeIE(ydl)
|
|
yie._sort_formats(info_dict['formats'])
|
|
ydl.process_ie_result(info_dict)
|
|
downloaded_ids = [info['format_id'] for info in ydl.downloaded_info_dicts]
|
|
self.assertEqual(downloaded_ids, ['248+141'])
|
|
|
|
for f1, f2 in zip(formats_order, formats_order[1:]):
|
|
info_dict = _make_result([f1, f2], extractor='youtube')
|
|
ydl = YDL({'format': 'best/bestvideo'})
|
|
yie = YoutubeIE(ydl)
|
|
yie._sort_formats(info_dict['formats'])
|
|
ydl.process_ie_result(info_dict)
|
|
downloaded = ydl.downloaded_info_dicts[0]
|
|
self.assertEqual(downloaded['format_id'], f1['format_id'])
|
|
|
|
info_dict = _make_result([f2, f1], extractor='youtube')
|
|
ydl = YDL({'format': 'best/bestvideo'})
|
|
yie = YoutubeIE(ydl)
|
|
yie._sort_formats(info_dict['formats'])
|
|
ydl.process_ie_result(info_dict)
|
|
downloaded = ydl.downloaded_info_dicts[0]
|
|
self.assertEqual(downloaded['format_id'], f1['format_id'])
|
|
|
|
def test_audio_only_extractor_format_selection(self):
|
|
# For extractors with incomplete formats (all formats are audio-only or
|
|
# video-only) best and worst should fallback to corresponding best/worst
|
|
# video-only or audio-only formats (as per
|
|
# https://github.com/ytdl-org/youtube-dl/pull/5556)
|
|
formats = [
|
|
{'format_id': 'low', 'ext': 'mp3', 'preference': 1, 'vcodec': 'none', 'url': TEST_URL},
|
|
{'format_id': 'high', 'ext': 'mp3', 'preference': 2, 'vcodec': 'none', 'url': TEST_URL},
|
|
]
|
|
info_dict = _make_result(formats)
|
|
|
|
ydl = YDL({'format': 'best'})
|
|
ydl.process_ie_result(info_dict.copy())
|
|
downloaded = ydl.downloaded_info_dicts[0]
|
|
self.assertEqual(downloaded['format_id'], 'high')
|
|
|
|
ydl = YDL({'format': 'worst'})
|
|
ydl.process_ie_result(info_dict.copy())
|
|
downloaded = ydl.downloaded_info_dicts[0]
|
|
self.assertEqual(downloaded['format_id'], 'low')
|
|
|
|
def test_format_not_available(self):
|
|
formats = [
|
|
{'format_id': 'regular', 'ext': 'mp4', 'height': 360, 'url': TEST_URL},
|
|
{'format_id': 'video', 'ext': 'mp4', 'height': 720, 'acodec': 'none', 'url': TEST_URL},
|
|
]
|
|
info_dict = _make_result(formats)
|
|
|
|
# This must fail since complete video-audio format does not match filter
|
|
# and extractor does not provide incomplete only formats (i.e. only
|
|
# video-only or audio-only).
|
|
ydl = YDL({'format': 'best[height>360]'})
|
|
self.assertRaises(ExtractorError, ydl.process_ie_result, info_dict.copy())
|
|
|
|
def test_format_selection_issue_10083(self):
|
|
# See https://github.com/ytdl-org/youtube-dl/issues/10083
|
|
formats = [
|
|
{'format_id': 'regular', 'height': 360, 'url': TEST_URL},
|
|
{'format_id': 'video', 'height': 720, 'acodec': 'none', 'url': TEST_URL},
|
|
{'format_id': 'audio', 'vcodec': 'none', 'url': TEST_URL},
|
|
]
|
|
info_dict = _make_result(formats)
|
|
|
|
ydl = YDL({'format': 'best[height>360]/bestvideo[height>360]+bestaudio'})
|
|
ydl.process_ie_result(info_dict.copy())
|
|
self.assertEqual(ydl.downloaded_info_dicts[0]['format_id'], 'video+audio')
|
|
|
|
def test_invalid_format_specs(self):
|
|
def assert_syntax_error(format_spec):
|
|
ydl = YDL({'format': format_spec})
|
|
info_dict = _make_result([{'format_id': 'foo', 'url': TEST_URL}])
|
|
self.assertRaises(SyntaxError, ydl.process_ie_result, info_dict)
|
|
|
|
assert_syntax_error('bestvideo,,best')
|
|
assert_syntax_error('+bestaudio')
|
|
assert_syntax_error('bestvideo+')
|
|
assert_syntax_error('/')
|
|
|
|
def test_format_filtering(self):
|
|
formats = [
|
|
{'format_id': 'A', 'filesize': 500, 'width': 1000},
|
|
{'format_id': 'B', 'filesize': 1000, 'width': 500},
|
|
{'format_id': 'C', 'filesize': 1000, 'width': 400},
|
|
{'format_id': 'D', 'filesize': 2000, 'width': 600},
|
|
{'format_id': 'E', 'filesize': 3000},
|
|
{'format_id': 'F'},
|
|
{'format_id': 'G', 'filesize': 1000000},
|
|
]
|
|
for f in formats:
|
|
f['url'] = 'http://_/'
|
|
f['ext'] = 'unknown'
|
|
info_dict = _make_result(formats)
|
|
|
|
ydl = YDL({'format': 'best[filesize<3000]'})
|
|
ydl.process_ie_result(info_dict)
|
|
downloaded = ydl.downloaded_info_dicts[0]
|
|
self.assertEqual(downloaded['format_id'], 'D')
|
|
|
|
ydl = YDL({'format': 'best[filesize<=3000]'})
|
|
ydl.process_ie_result(info_dict)
|
|
downloaded = ydl.downloaded_info_dicts[0]
|
|
self.assertEqual(downloaded['format_id'], 'E')
|
|
|
|
ydl = YDL({'format': 'best[filesize <= ? 3000]'})
|
|
ydl.process_ie_result(info_dict)
|
|
downloaded = ydl.downloaded_info_dicts[0]
|
|
self.assertEqual(downloaded['format_id'], 'F')
|
|
|
|
ydl = YDL({'format': 'best [filesize = 1000] [width>450]'})
|
|
ydl.process_ie_result(info_dict)
|
|
downloaded = ydl.downloaded_info_dicts[0]
|
|
self.assertEqual(downloaded['format_id'], 'B')
|
|
|
|
ydl = YDL({'format': 'best [filesize = 1000] [width!=450]'})
|
|
ydl.process_ie_result(info_dict)
|
|
downloaded = ydl.downloaded_info_dicts[0]
|
|
self.assertEqual(downloaded['format_id'], 'C')
|
|
|
|
ydl = YDL({'format': '[filesize>?1]'})
|
|
ydl.process_ie_result(info_dict)
|
|
downloaded = ydl.downloaded_info_dicts[0]
|
|
self.assertEqual(downloaded['format_id'], 'G')
|
|
|
|
ydl = YDL({'format': '[filesize<1M]'})
|
|
ydl.process_ie_result(info_dict)
|
|
downloaded = ydl.downloaded_info_dicts[0]
|
|
self.assertEqual(downloaded['format_id'], 'E')
|
|
|
|
ydl = YDL({'format': '[filesize<1MiB]'})
|
|
ydl.process_ie_result(info_dict)
|
|
downloaded = ydl.downloaded_info_dicts[0]
|
|
self.assertEqual(downloaded['format_id'], 'G')
|
|
|
|
ydl = YDL({'format': 'all[width>=400][width<=600]'})
|
|
ydl.process_ie_result(info_dict)
|
|
downloaded_ids = [info['format_id'] for info in ydl.downloaded_info_dicts]
|
|
self.assertEqual(downloaded_ids, ['B', 'C', 'D'])
|
|
|
|
ydl = YDL({'format': 'best[height<40]'})
|
|
try:
|
|
ydl.process_ie_result(info_dict)
|
|
except ExtractorError:
|
|
pass
|
|
self.assertEqual(ydl.downloaded_info_dicts, [])
|
|
|
|
def test_default_format_spec(self):
|
|
ydl = YDL({'simulate': True})
|
|
self.assertEqual(ydl._default_format_spec({}), 'bestvideo+bestaudio/best')
|
|
|
|
ydl = YDL({})
|
|
self.assertEqual(ydl._default_format_spec({'is_live': True}), 'best/bestvideo+bestaudio')
|
|
|
|
ydl = YDL({'simulate': True})
|
|
self.assertEqual(ydl._default_format_spec({'is_live': True}), 'bestvideo+bestaudio/best')
|
|
|
|
ydl = YDL({'outtmpl': '-'})
|
|
self.assertEqual(ydl._default_format_spec({}), 'best/bestvideo+bestaudio')
|
|
|
|
ydl = YDL({})
|
|
self.assertEqual(ydl._default_format_spec({}, download=False), 'bestvideo+bestaudio/best')
|
|
self.assertEqual(ydl._default_format_spec({'is_live': True}), 'best/bestvideo+bestaudio')
|
|
|
|
|
|
class TestYoutubeDL(unittest.TestCase):
|
|
def test_subtitles(self):
|
|
def s_formats(lang, autocaption=False):
|
|
return [{
|
|
'ext': ext,
|
|
'url': 'http://localhost/video.%s.%s' % (lang, ext),
|
|
'_auto': autocaption,
|
|
} for ext in ['vtt', 'srt', 'ass']]
|
|
subtitles = dict((l, s_formats(l)) for l in ['en', 'fr', 'es'])
|
|
auto_captions = dict((l, s_formats(l, True)) for l in ['it', 'pt', 'es'])
|
|
info_dict = {
|
|
'id': 'test',
|
|
'title': 'Test',
|
|
'url': 'http://localhost/video.mp4',
|
|
'subtitles': subtitles,
|
|
'automatic_captions': auto_captions,
|
|
'extractor': 'TEST',
|
|
}
|
|
|
|
def get_info(params={}):
|
|
params.setdefault('simulate', True)
|
|
ydl = YDL(params)
|
|
ydl.report_warning = lambda *args, **kargs: None
|
|
return ydl.process_video_result(info_dict, download=False)
|
|
|
|
result = get_info()
|
|
self.assertFalse(result.get('requested_subtitles'))
|
|
self.assertEqual(result['subtitles'], subtitles)
|
|
self.assertEqual(result['automatic_captions'], auto_captions)
|
|
|
|
result = get_info({'writesubtitles': True})
|
|
subs = result['requested_subtitles']
|
|
self.assertTrue(subs)
|
|
self.assertEqual(set(subs.keys()), set(['en']))
|
|
self.assertTrue(subs['en'].get('data') is None)
|
|
self.assertEqual(subs['en']['ext'], 'ass')
|
|
|
|
result = get_info({'writesubtitles': True, 'subtitlesformat': 'foo/srt'})
|
|
subs = result['requested_subtitles']
|
|
self.assertEqual(subs['en']['ext'], 'srt')
|
|
|
|
result = get_info({'writesubtitles': True, 'subtitleslangs': ['es', 'fr', 'it']})
|
|
subs = result['requested_subtitles']
|
|
self.assertTrue(subs)
|
|
self.assertEqual(set(subs.keys()), set(['es', 'fr']))
|
|
|
|
result = get_info({'writesubtitles': True, 'writeautomaticsub': True, 'subtitleslangs': ['es', 'pt']})
|
|
subs = result['requested_subtitles']
|
|
self.assertTrue(subs)
|
|
self.assertEqual(set(subs.keys()), set(['es', 'pt']))
|
|
self.assertFalse(subs['es']['_auto'])
|
|
self.assertTrue(subs['pt']['_auto'])
|
|
|
|
result = get_info({'writeautomaticsub': True, 'subtitleslangs': ['es', 'pt']})
|
|
subs = result['requested_subtitles']
|
|
self.assertTrue(subs)
|
|
self.assertEqual(set(subs.keys()), set(['es', 'pt']))
|
|
self.assertTrue(subs['es']['_auto'])
|
|
self.assertTrue(subs['pt']['_auto'])
|
|
|
|
def test_add_extra_info(self):
|
|
test_dict = {
|
|
'extractor': 'Foo',
|
|
}
|
|
extra_info = {
|
|
'extractor': 'Bar',
|
|
'playlist': 'funny videos',
|
|
}
|
|
YDL.add_extra_info(test_dict, extra_info)
|
|
self.assertEqual(test_dict['extractor'], 'Foo')
|
|
self.assertEqual(test_dict['playlist'], 'funny videos')
|
|
|
|
def test_prepare_filename(self):
|
|
info = {
|
|
'id': '1234',
|
|
'ext': 'mp4',
|
|
'width': None,
|
|
'height': 1080,
|
|
'title1': '$PATH',
|
|
'title2': '%PATH%',
|
|
}
|
|
|
|
def fname(templ):
|
|
ydl = YoutubeDL({'outtmpl': templ})
|
|
return ydl.prepare_filename(info)
|
|
self.assertEqual(fname('%(id)s.%(ext)s'), '1234.mp4')
|
|
self.assertEqual(fname('%(id)s-%(width)s.%(ext)s'), '1234-NA.mp4')
|
|
# Replace missing fields with 'NA'
|
|
self.assertEqual(fname('%(uploader_date)s-%(id)s.%(ext)s'), 'NA-1234.mp4')
|
|
self.assertEqual(fname('%(height)d.%(ext)s'), '1080.mp4')
|
|
self.assertEqual(fname('%(height)6d.%(ext)s'), ' 1080.mp4')
|
|
self.assertEqual(fname('%(height)-6d.%(ext)s'), '1080 .mp4')
|
|
self.assertEqual(fname('%(height)06d.%(ext)s'), '001080.mp4')
|
|
self.assertEqual(fname('%(height) 06d.%(ext)s'), ' 01080.mp4')
|
|
self.assertEqual(fname('%(height) 06d.%(ext)s'), ' 01080.mp4')
|
|
self.assertEqual(fname('%(height)0 6d.%(ext)s'), ' 01080.mp4')
|
|
self.assertEqual(fname('%(height)0 6d.%(ext)s'), ' 01080.mp4')
|
|
self.assertEqual(fname('%(height) 0 6d.%(ext)s'), ' 01080.mp4')
|
|
self.assertEqual(fname('%%'), '%')
|
|
self.assertEqual(fname('%%%%'), '%%')
|
|
self.assertEqual(fname('%%(height)06d.%(ext)s'), '%(height)06d.mp4')
|
|
self.assertEqual(fname('%(width)06d.%(ext)s'), 'NA.mp4')
|
|
self.assertEqual(fname('%(width)06d.%%(ext)s'), 'NA.%(ext)s')
|
|
self.assertEqual(fname('%%(width)06d.%(ext)s'), '%(width)06d.mp4')
|
|
self.assertEqual(fname('Hello %(title1)s'), 'Hello $PATH')
|
|
self.assertEqual(fname('Hello %(title2)s'), 'Hello %PATH%')
|
|
|
|
def test_format_note(self):
|
|
ydl = YoutubeDL()
|
|
self.assertEqual(ydl._format_note({}), '')
|
|
assertRegexpMatches(self, ydl._format_note({
|
|
'vbr': 10,
|
|
}), r'^\s*10k$')
|
|
assertRegexpMatches(self, ydl._format_note({
|
|
'fps': 30,
|
|
}), r'^30fps$')
|
|
|
|
def test_postprocessors(self):
|
|
filename = 'post-processor-testfile.mp4'
|
|
audiofile = filename + '.mp3'
|
|
|
|
class SimplePP(PostProcessor):
|
|
def run(self, info):
|
|
with open(audiofile, 'wt') as f:
|
|
f.write('EXAMPLE')
|
|
return [info['filepath']], info
|
|
|
|
def run_pp(params, PP):
|
|
with open(filename, 'wt') as f:
|
|
f.write('EXAMPLE')
|
|
ydl = YoutubeDL(params)
|
|
ydl.add_post_processor(PP())
|
|
ydl.post_process(filename, {'filepath': filename})
|
|
|
|
run_pp({'keepvideo': True}, SimplePP)
|
|
self.assertTrue(os.path.exists(filename), '%s doesn\'t exist' % filename)
|
|
self.assertTrue(os.path.exists(audiofile), '%s doesn\'t exist' % audiofile)
|
|
os.unlink(filename)
|
|
os.unlink(audiofile)
|
|
|
|
run_pp({'keepvideo': False}, SimplePP)
|
|
self.assertFalse(os.path.exists(filename), '%s exists' % filename)
|
|
self.assertTrue(os.path.exists(audiofile), '%s doesn\'t exist' % audiofile)
|
|
os.unlink(audiofile)
|
|
|
|
class ModifierPP(PostProcessor):
|
|
def run(self, info):
|
|
with open(info['filepath'], 'wt') as f:
|
|
f.write('MODIFIED')
|
|
return [], info
|
|
|
|
run_pp({'keepvideo': False}, ModifierPP)
|
|
self.assertTrue(os.path.exists(filename), '%s doesn\'t exist' % filename)
|
|
os.unlink(filename)
|
|
|
|
def test_match_filter(self):
|
|
class FilterYDL(YDL):
|
|
def __init__(self, *args, **kwargs):
|
|
super(FilterYDL, self).__init__(*args, **kwargs)
|
|
self.params['simulate'] = True
|
|
|
|
def process_info(self, info_dict):
|
|
super(YDL, self).process_info(info_dict)
|
|
|
|
def _match_entry(self, info_dict, incomplete):
|
|
res = super(FilterYDL, self)._match_entry(info_dict, incomplete)
|
|
if res is None:
|
|
self.downloaded_info_dicts.append(info_dict)
|
|
return res
|
|
|
|
first = {
|
|
'id': '1',
|
|
'url': TEST_URL,
|
|
'title': 'one',
|
|
'extractor': 'TEST',
|
|
'duration': 30,
|
|
'filesize': 10 * 1024,
|
|
'playlist_id': '42',
|
|
'uploader': "變態妍字幕版 太妍 тест",
|
|
'creator': "тест ' 123 ' тест--",
|
|
}
|
|
second = {
|
|
'id': '2',
|
|
'url': TEST_URL,
|
|
'title': 'two',
|
|
'extractor': 'TEST',
|
|
'duration': 10,
|
|
'description': 'foo',
|
|
'filesize': 5 * 1024,
|
|
'playlist_id': '43',
|
|
'uploader': "тест 123",
|
|
}
|
|
videos = [first, second]
|
|
|
|
def get_videos(filter_=None):
|
|
ydl = FilterYDL({'match_filter': filter_})
|
|
for v in videos:
|
|
ydl.process_ie_result(v, download=True)
|
|
return [v['id'] for v in ydl.downloaded_info_dicts]
|
|
|
|
res = get_videos()
|
|
self.assertEqual(res, ['1', '2'])
|
|
|
|
def f(v):
|
|
if v['id'] == '1':
|
|
return None
|
|
else:
|
|
return 'Video id is not 1'
|
|
res = get_videos(f)
|
|
self.assertEqual(res, ['1'])
|
|
|
|
f = match_filter_func('duration < 30')
|
|
res = get_videos(f)
|
|
self.assertEqual(res, ['2'])
|
|
|
|
f = match_filter_func('description = foo')
|
|
res = get_videos(f)
|
|
self.assertEqual(res, ['2'])
|
|
|
|
f = match_filter_func('description =? foo')
|
|
res = get_videos(f)
|
|
self.assertEqual(res, ['1', '2'])
|
|
|
|
f = match_filter_func('filesize > 5KiB')
|
|
res = get_videos(f)
|
|
self.assertEqual(res, ['1'])
|
|
|
|
f = match_filter_func('playlist_id = 42')
|
|
res = get_videos(f)
|
|
self.assertEqual(res, ['1'])
|
|
|
|
f = match_filter_func('uploader = "變態妍字幕版 太妍 тест"')
|
|
res = get_videos(f)
|
|
self.assertEqual(res, ['1'])
|
|
|
|
f = match_filter_func('uploader != "變態妍字幕版 太妍 тест"')
|
|
res = get_videos(f)
|
|
self.assertEqual(res, ['2'])
|
|
|
|
f = match_filter_func('creator = "тест \' 123 \' тест--"')
|
|
res = get_videos(f)
|
|
self.assertEqual(res, ['1'])
|
|
|
|
f = match_filter_func("creator = 'тест \\' 123 \\' тест--'")
|
|
res = get_videos(f)
|
|
self.assertEqual(res, ['1'])
|
|
|
|
f = match_filter_func(r"creator = 'тест \' 123 \' тест--' & duration > 30")
|
|
res = get_videos(f)
|
|
self.assertEqual(res, [])
|
|
|
|
def test_playlist_items_selection(self):
|
|
entries = [{
|
|
'id': compat_str(i),
|
|
'title': compat_str(i),
|
|
'url': TEST_URL,
|
|
} for i in range(1, 5)]
|
|
playlist = {
|
|
'_type': 'playlist',
|
|
'id': 'test',
|
|
'entries': entries,
|
|
'extractor': 'test:playlist',
|
|
'extractor_key': 'test:playlist',
|
|
'webpage_url': 'http://example.com',
|
|
}
|
|
|
|
def get_downloaded_info_dicts(params):
|
|
ydl = YDL(params)
|
|
# make a deep copy because the dictionary and nested entries
|
|
# can be modified
|
|
ydl.process_ie_result(copy.deepcopy(playlist))
|
|
return ydl.downloaded_info_dicts
|
|
|
|
def get_ids(params):
|
|
return [int(v['id']) for v in get_downloaded_info_dicts(params)]
|
|
|
|
result = get_ids({})
|
|
self.assertEqual(result, [1, 2, 3, 4])
|
|
|
|
result = get_ids({'playlistend': 10})
|
|
self.assertEqual(result, [1, 2, 3, 4])
|
|
|
|
result = get_ids({'playlistend': 2})
|
|
self.assertEqual(result, [1, 2])
|
|
|
|
result = get_ids({'playliststart': 10})
|
|
self.assertEqual(result, [])
|
|
|
|
result = get_ids({'playliststart': 2})
|
|
self.assertEqual(result, [2, 3, 4])
|
|
|
|
result = get_ids({'playlist_items': '2-4'})
|
|
self.assertEqual(result, [2, 3, 4])
|
|
|
|
result = get_ids({'playlist_items': '2,4'})
|
|
self.assertEqual(result, [2, 4])
|
|
|
|
result = get_ids({'playlist_items': '10'})
|
|
self.assertEqual(result, [])
|
|
|
|
result = get_ids({'playlist_items': '3-10'})
|
|
self.assertEqual(result, [3, 4])
|
|
|
|
result = get_ids({'playlist_items': '2-4,3-4,3'})
|
|
self.assertEqual(result, [2, 3, 4])
|
|
|
|
# Tests for https://github.com/ytdl-org/youtube-dl/issues/10591
|
|
# @{
|
|
result = get_downloaded_info_dicts({'playlist_items': '2-4,3-4,3'})
|
|
self.assertEqual(result[0]['playlist_index'], 2)
|
|
self.assertEqual(result[1]['playlist_index'], 3)
|
|
|
|
result = get_downloaded_info_dicts({'playlist_items': '2-4,3-4,3'})
|
|
self.assertEqual(result[0]['playlist_index'], 2)
|
|
self.assertEqual(result[1]['playlist_index'], 3)
|
|
self.assertEqual(result[2]['playlist_index'], 4)
|
|
|
|
result = get_downloaded_info_dicts({'playlist_items': '4,2'})
|
|
self.assertEqual(result[0]['playlist_index'], 4)
|
|
self.assertEqual(result[1]['playlist_index'], 2)
|
|
# @}
|
|
|
|
def test_urlopen_no_file_protocol(self):
|
|
# see https://github.com/ytdl-org/youtube-dl/issues/8227
|
|
ydl = YDL()
|
|
self.assertRaises(compat_urllib_error.URLError, ydl.urlopen, 'file:///etc/passwd')
|
|
|
|
def test_do_not_override_ie_key_in_url_transparent(self):
|
|
ydl = YDL()
|
|
|
|
class Foo1IE(InfoExtractor):
|
|
_VALID_URL = r'foo1:'
|
|
|
|
def _real_extract(self, url):
|
|
return {
|
|
'_type': 'url_transparent',
|
|
'url': 'foo2:',
|
|
'ie_key': 'Foo2',
|
|
'title': 'foo1 title',
|
|
'id': 'foo1_id',
|
|
}
|
|
|
|
class Foo2IE(InfoExtractor):
|
|
_VALID_URL = r'foo2:'
|
|
|
|
def _real_extract(self, url):
|
|
return {
|
|
'_type': 'url',
|
|
'url': 'foo3:',
|
|
'ie_key': 'Foo3',
|
|
}
|
|
|
|
class Foo3IE(InfoExtractor):
|
|
_VALID_URL = r'foo3:'
|
|
|
|
def _real_extract(self, url):
|
|
return _make_result([{'url': TEST_URL}], title='foo3 title')
|
|
|
|
ydl.add_info_extractor(Foo1IE(ydl))
|
|
ydl.add_info_extractor(Foo2IE(ydl))
|
|
ydl.add_info_extractor(Foo3IE(ydl))
|
|
ydl.extract_info('foo1:')
|
|
downloaded = ydl.downloaded_info_dicts[0]
|
|
self.assertEqual(downloaded['url'], TEST_URL)
|
|
self.assertEqual(downloaded['title'], 'foo1 title')
|
|
self.assertEqual(downloaded['id'], 'testid')
|
|
self.assertEqual(downloaded['extractor'], 'testex')
|
|
self.assertEqual(downloaded['extractor_key'], 'TestEx')
|
|
|
|
|
|
if __name__ == '__main__':
|
|
unittest.main()
|