12345678910111213141516171819202122232425262728293031323334353637383940414243444546474849505152535455565758596061626364656667686970717273747576777879808182838485868788899091929394959697989910010110210310410510610710810911011111211311411511611711811912012112212312412512612712812913013113213313413513613713813914014114214314414514614714814915015115215315415515615715815916016116216316416516616716816917017117217317417517617717817918018118218318418518618718818919019119219319419519619719819920020120220320420520620720820921021121221321421521621721821922022122222322422522622722822923023123223323423523623723823924024124224324424524624724824925025125225325425525625725825926026126226326426526626726826927027127227327427527627727827928028128228328428528628728828929029129229329429529629729829930030130230330430530630730830931031131231331431531631731831932032132232332432532632732832933033133233333433533633733833934034134234334434534634734834935035135235335435535635735835936036136236336436536636736836937037137237337437537637737837938038138238338438538638738838939039139239339439539639739839940040140240340440540640740840941041141241341441541641741841942042142242342442542642742842943043143243343443543643743843944044144244344444544644744844945045145245345445545645745845946046146246346446546646746846947047147247347447547647747847948048148248348448548648748848949049149249349449549649749849950050150250350450550650750850951051151251351451551651751851952052152252352452552652752852953053153253353453553653753853954054154254354454554654754854955055155255355455555655755855956056156256356456556656756856957057157257357457557657757857958058158258358458558658758858959059159259359459559659759859960060160260360460560660760860961061161261361461561661761861962062162262362462562662762862963063163263363463563663763863964064164264364464564664764864965065165265365465565665765865966066166266366466566666766866967067167267367467567667767867968068168268368468568668768868969069169269369469569669769869970070170270370470570670770870971071171271371471571671771871972072172272372472572672772872973073173273373473573673773873974074174274374474574674774874975075175275375475575675775875976076176276376476576676776876977077177277377477577677777877978078178278378478578678778878979079179279379479579679779879980080180280380480580680780880981081181281381481581681781881982082182282382482582682782882983083183283383483583683783883984084184284384484584684784884985085185285385485585685785885986086186286386486586686786886987087187287387487587687787887988088188288388488588688788888989089189289389489589689789889990090190290390490590690790890991091191291391491591691791891992092192292392492592692792892993093193293393493593693793893994094194294394494594694794894995095195295395495595695795895996096196296396496596696796896997097197297397497597697797897998098198298398498598698798898999099199299399499599699799899910001001100210031004100510061007100810091010101110121013101410151016101710181019102010211022102310241025102610271028102910301031103210331034103510361037103810391040104110421043104410451046104710481049105010511052105310541055105610571058105910601061106210631064106510661067106810691070107110721073107410751076107710781079108010811082108310841085108610871088108910901091109210931094109510961097109810991100110111021103110411051106110711081109111011111112111311141115111611171118111911201121112211231124112511261127112811291130113111321133113411351136113711381139114011411142114311441145114611471148114911501151115211531154115511561157115811591160116111621163116411651166116711681169117011711172117311741175117611771178117911801181118211831184118511861187118811891190119111921193119411951196119711981199120012011202120312041205120612071208120912101211121212131214121512161217121812191220122112221223122412251226122712281229123012311232123312341235123612371238123912401241124212431244124512461247124812491250125112521253125412551256125712581259126012611262126312641265126612671268126912701271127212731274127512761277127812791280128112821283128412851286128712881289129012911292129312941295129612971298129913001301130213031304130513061307130813091310131113121313131413151316131713181319132013211322132313241325132613271328132913301331133213331334133513361337133813391340134113421343134413451346134713481349135013511352135313541355135613571358135913601361136213631364136513661367136813691370137113721373137413751376137713781379138013811382138313841385138613871388138913901391139213931394139513961397139813991400140114021403140414051406140714081409141014111412141314141415141614171418141914201421142214231424142514261427142814291430143114321433143414351436143714381439144014411442144314441445144614471448144914501451145214531454145514561457145814591460146114621463146414651466146714681469147014711472147314741475147614771478147914801481148214831484148514861487148814891490149114921493149414951496149714981499150015011502150315041505150615071508150915101511151215131514151515161517151815191520152115221523152415251526152715281529153015311532153315341535153615371538153915401541154215431544154515461547154815491550155115521553155415551556155715581559156015611562156315641565156615671568156915701571157215731574157515761577157815791580158115821583158415851586158715881589159015911592159315941595159615971598159916001601160216031604160516061607160816091610161116121613161416151616161716181619162016211622162316241625162616271628162916301631163216331634163516361637163816391640164116421643164416451646164716481649165016511652165316541655165616571658165916601661166216631664166516661667166816691670167116721673167416751676167716781679168016811682168316841685168616871688168916901691169216931694169516961697169816991700170117021703170417051706170717081709171017111712171317141715171617171718171917201721172217231724172517261727172817291730173117321733173417351736173717381739174017411742174317441745174617471748174917501751175217531754175517561757175817591760176117621763176417651766176717681769177017711772177317741775177617771778177917801781178217831784178517861787178817891790179117921793179417951796179717981799180018011802180318041805180618071808180918101811181218131814181518161817181818191820182118221823182418251826182718281829183018311832183318341835183618371838183918401841184218431844184518461847184818491850185118521853185418551856185718581859186018611862186318641865186618671868186918701871187218731874187518761877187818791880188118821883188418851886188718881889189018911892189318941895189618971898189919001901190219031904190519061907190819091910191119121913191419151916191719181919192019211922192319241925192619271928192919301931193219331934193519361937193819391940194119421943194419451946194719481949195019511952195319541955195619571958195919601961196219631964196519661967196819691970197119721973197419751976197719781979198019811982198319841985198619871988198919901991199219931994199519961997199819992000200120022003200420052006200720082009201020112012201320142015201620172018201920202021202220232024202520262027202820292030203120322033203420352036203720382039204020412042204320442045204620472048204920502051205220532054205520562057205820592060206120622063206420652066206720682069207020712072207320742075207620772078207920802081208220832084208520862087208820892090209120922093209420952096209720982099210021012102210321042105210621072108210921102111211221132114211521162117211821192120212121222123212421252126212721282129213021312132213321342135213621372138213921402141214221432144214521462147214821492150215121522153215421552156215721582159216021612162216321642165216621672168216921702171217221732174217521762177217821792180218121822183218421852186218721882189219021912192219321942195219621972198219922002201220222032204220522062207220822092210221122122213221422152216221722182219222022212222222322242225222622272228222922302231223222332234223522362237223822392240224122422243224422452246224722482249225022512252225322542255225622572258225922602261226222632264226522662267226822692270227122722273227422752276227722782279228022812282228322842285228622872288228922902291229222932294229522962297229822992300230123022303230423052306230723082309231023112312231323142315231623172318231923202321232223232324232523262327232823292330233123322333233423352336233723382339234023412342234323442345234623472348234923502351235223532354235523562357235823592360236123622363236423652366236723682369237023712372237323742375237623772378237923802381238223832384238523862387238823892390239123922393239423952396239723982399240024012402240324042405240624072408240924102411241224132414241524162417241824192420242124222423242424252426242724282429243024312432243324342435243624372438243924402441244224432444244524462447244824492450245124522453245424552456245724582459246024612462246324642465246624672468246924702471247224732474247524762477247824792480248124822483248424852486248724882489249024912492249324942495249624972498249925002501250225032504250525062507250825092510251125122513251425152516251725182519252025212522252325242525252625272528252925302531253225332534253525362537253825392540254125422543254425452546254725482549255025512552255325542555255625572558255925602561256225632564256525662567256825692570257125722573257425752576257725782579258025812582258325842585258625872588258925902591259225932594259525962597259825992600260126022603260426052606260726082609261026112612261326142615261626172618261926202621262226232624262526262627262826292630263126322633263426352636263726382639264026412642264326442645264626472648264926502651265226532654265526562657265826592660266126622663266426652666266726682669267026712672267326742675267626772678267926802681268226832684268526862687268826892690269126922693269426952696269726982699270027012702270327042705270627072708270927102711271227132714271527162717271827192720272127222723272427252726272727282729273027312732273327342735273627372738273927402741274227432744274527462747274827492750275127522753275427552756275727582759276027612762276327642765276627672768276927702771277227732774277527762777277827792780278127822783278427852786278727882789279027912792279327942795279627972798279928002801280228032804280528062807280828092810281128122813281428152816281728182819282028212822282328242825282628272828282928302831283228332834283528362837283828392840284128422843284428452846284728482849285028512852285328542855285628572858285928602861286228632864286528662867286828692870287128722873287428752876287728782879288028812882288328842885288628872888288928902891289228932894289528962897289828992900290129022903290429052906290729082909291029112912291329142915291629172918291929202921292229232924292529262927292829292930293129322933293429352936293729382939294029412942294329442945294629472948294929502951295229532954295529562957295829592960296129622963296429652966296729682969297029712972297329742975297629772978297929802981298229832984298529862987298829892990299129922993299429952996299729982999300030013002300330043005300630073008300930103011301230133014301530163017301830193020302130223023302430253026302730283029303030313032303330343035303630373038303930403041304230433044304530463047304830493050305130523053305430553056305730583059306030613062306330643065306630673068306930703071307230733074307530763077307830793080308130823083308430853086308730883089309030913092309330943095309630973098309931003101310231033104310531063107310831093110311131123113311431153116311731183119312031213122312331243125312631273128312931303131313231333134313531363137313831393140314131423143314431453146314731483149315031513152315331543155315631573158315931603161316231633164316531663167316831693170317131723173317431753176317731783179318031813182318331843185318631873188318931903191319231933194319531963197319831993200320132023203320432053206320732083209321032113212321332143215321632173218321932203221322232233224322532263227322832293230323132323233323432353236323732383239324032413242324332443245324632473248324932503251325232533254325532563257325832593260326132623263326432653266326732683269327032713272327332743275327632773278327932803281328232833284328532863287328832893290329132923293329432953296329732983299330033013302330333043305330633073308330933103311331233133314331533163317331833193320332133223323332433253326332733283329333033313332333333343335333633373338333933403341334233433344334533463347334833493350335133523353335433553356335733583359336033613362336333643365336633673368336933703371337233733374337533763377337833793380338133823383338433853386338733883389339033913392339333943395339633973398339934003401340234033404340534063407340834093410341134123413341434153416341734183419342034213422342334243425342634273428342934303431343234333434343534363437343834393440344134423443344434453446344734483449345034513452345334543455345634573458345934603461346234633464346534663467346834693470347134723473347434753476347734783479348034813482348334843485348634873488348934903491349234933494349534963497349834993500350135023503350435053506350735083509351035113512351335143515351635173518351935203521352235233524352535263527352835293530353135323533353435353536353735383539354035413542354335443545354635473548354935503551355235533554355535563557355835593560356135623563356435653566356735683569357035713572357335743575357635773578357935803581358235833584358535863587358835893590359135923593359435953596359735983599360036013602360336043605360636073608360936103611361236133614361536163617361836193620362136223623362436253626362736283629363036313632363336343635363636373638363936403641364236433644364536463647364836493650365136523653365436553656365736583659366036613662366336643665366636673668366936703671367236733674367536763677367836793680368136823683368436853686368736883689369036913692369336943695369636973698369937003701370237033704370537063707370837093710371137123713371437153716371737183719372037213722372337243725372637273728372937303731373237333734373537363737373837393740374137423743374437453746374737483749375037513752375337543755375637573758375937603761376237633764376537663767376837693770377137723773377437753776377737783779378037813782378337843785378637873788378937903791379237933794379537963797379837993800380138023803380438053806380738083809381038113812381338143815381638173818381938203821382238233824382538263827382838293830383138323833383438353836383738383839384038413842384338443845384638473848384938503851385238533854385538563857385838593860386138623863386438653866386738683869387038713872387338743875387638773878387938803881388238833884388538863887388838893890389138923893389438953896389738983899390039013902390339043905390639073908390939103911391239133914391539163917391839193920392139223923392439253926392739283929393039313932393339343935393639373938393939403941394239433944394539463947394839493950395139523953395439553956395739583959396039613962396339643965396639673968396939703971397239733974397539763977397839793980398139823983398439853986398739883989399039913992399339943995399639973998399940004001400240034004400540064007400840094010401140124013401440154016401740184019402040214022402340244025402640274028 |
- import collections
- import contextlib
- import datetime
- import errno
- import fileinput
- import functools
- import io
- import itertools
- import json
- import locale
- import operator
- import os
- import random
- import re
- import shutil
- import subprocess
- import sys
- import tempfile
- import time
- import tokenize
- import traceback
- import unicodedata
- import urllib.request
- from string import ascii_letters
- from .cache import Cache
- from .compat import compat_os_name, compat_shlex_quote
- from .cookies import load_cookies
- from .downloader import FFmpegFD, get_suitable_downloader, shorten_protocol_name
- from .downloader.rtmp import rtmpdump_version
- from .extractor import gen_extractor_classes, get_info_extractor
- from .extractor.common import UnsupportedURLIE
- from .extractor.openload import PhantomJSwrapper
- from .minicurses import format_text
- from .postprocessor import _PLUGIN_CLASSES as plugin_postprocessors
- from .postprocessor import (
- EmbedThumbnailPP,
- FFmpegFixupDuplicateMoovPP,
- FFmpegFixupDurationPP,
- FFmpegFixupM3u8PP,
- FFmpegFixupM4aPP,
- FFmpegFixupStretchedPP,
- FFmpegFixupTimestampPP,
- FFmpegMergerPP,
- FFmpegPostProcessor,
- FFmpegVideoConvertorPP,
- MoveFilesAfterDownloadPP,
- get_postprocessor,
- )
- from .utils import (
- DEFAULT_OUTTMPL,
- IDENTITY,
- LINK_TEMPLATES,
- MEDIA_EXTENSIONS,
- NO_DEFAULT,
- NUMBER_RE,
- OUTTMPL_TYPES,
- POSTPROCESS_WHEN,
- STR_FORMAT_RE_TMPL,
- STR_FORMAT_TYPES,
- ContentTooShortError,
- DateRange,
- DownloadCancelled,
- DownloadError,
- EntryNotInPlaylist,
- ExistingVideoReached,
- ExtractorError,
- FormatSorter,
- GeoRestrictedError,
- HEADRequest,
- ISO3166Utils,
- LazyList,
- MaxDownloadsReached,
- Namespace,
- PagedList,
- PerRequestProxyHandler,
- PlaylistEntries,
- Popen,
- PostProcessingError,
- ReExtractInfo,
- RejectedVideoReached,
- SameFileError,
- UnavailableVideoError,
- UserNotLive,
- YoutubeDLCookieProcessor,
- YoutubeDLHandler,
- YoutubeDLRedirectHandler,
- age_restricted,
- args_to_str,
- bug_reports_message,
- date_from_str,
- deprecation_warning,
- determine_ext,
- determine_protocol,
- encode_compat_str,
- encodeFilename,
- error_to_compat_str,
- escapeHTML,
- expand_path,
- filter_dict,
- float_or_none,
- format_bytes,
- format_decimal_suffix,
- format_field,
- formatSeconds,
- get_compatible_ext,
- get_domain,
- int_or_none,
- iri_to_uri,
- is_path_like,
- join_nonempty,
- locked_file,
- make_archive_id,
- make_dir,
- make_HTTPS_handler,
- merge_headers,
- network_exceptions,
- number_of_digits,
- orderedSet,
- orderedSet_from_options,
- parse_filesize,
- preferredencoding,
- prepend_extension,
- register_socks_protocols,
- remove_terminal_sequences,
- render_table,
- replace_extension,
- sanitize_filename,
- sanitize_path,
- sanitize_url,
- sanitized_Request,
- std_headers,
- str_or_none,
- strftime_or_none,
- subtitles_filename,
- supports_terminal_sequences,
- system_identifier,
- timetuple_from_msec,
- to_high_limit_path,
- traverse_obj,
- try_call,
- try_get,
- url_basename,
- variadic,
- version_tuple,
- windows_enable_vt_mode,
- write_json_file,
- write_string,
- )
- from .version import RELEASE_GIT_HEAD, VARIANT, __version__
- if compat_os_name == 'nt':
- import ctypes
- class YoutubeDL:
- """YoutubeDL class.
- YoutubeDL objects are the ones responsible of downloading the
- actual video file and writing it to disk if the user has requested
- it, among some other tasks. In most cases there should be one per
- program. As, given a video URL, the downloader doesn't know how to
- extract all the needed information, task that InfoExtractors do, it
- has to pass the URL to one of them.
- For this, YoutubeDL objects have a method that allows
- InfoExtractors to be registered in a given order. When it is passed
- a URL, the YoutubeDL object handles it to the first InfoExtractor it
- finds that reports being able to handle it. The InfoExtractor extracts
- all the information about the video or videos the URL refers to, and
- YoutubeDL process the extracted information, possibly using a File
- Downloader to download the video.
- YoutubeDL objects accept a lot of parameters. In order not to saturate
- the object constructor with arguments, it receives a dictionary of
- options instead. These options are available through the params
- attribute for the InfoExtractors to use. The YoutubeDL also
- registers itself as the downloader in charge for the InfoExtractors
- that are added to it, so this is a "mutual registration".
- Available options:
- username: Username for authentication purposes.
- password: Password for authentication purposes.
- videopassword: Password for accessing a video.
- ap_mso: Adobe Pass multiple-system operator identifier.
- ap_username: Multiple-system operator account username.
- ap_password: Multiple-system operator account password.
- usenetrc: Use netrc for authentication instead.
- verbose: Print additional info to stdout.
- quiet: Do not print messages to stdout.
- no_warnings: Do not print out anything for warnings.
- forceprint: A dict with keys WHEN mapped to a list of templates to
- print to stdout. The allowed keys are video or any of the
- items in utils.POSTPROCESS_WHEN.
- For compatibility, a single list is also accepted
- print_to_file: A dict with keys WHEN (same as forceprint) mapped to
- a list of tuples with (template, filename)
- forcejson: Force printing info_dict as JSON.
- dump_single_json: Force printing the info_dict of the whole playlist
- (or video) as a single JSON line.
- force_write_download_archive: Force writing download archive regardless
- of 'skip_download' or 'simulate'.
- simulate: Do not download the video files. If unset (or None),
- simulate only if listsubtitles, listformats or list_thumbnails is used
- format: Video format code. see "FORMAT SELECTION" for more details.
- You can also pass a function. The function takes 'ctx' as
- argument and returns the formats to download.
- See "build_format_selector" for an implementation
- allow_unplayable_formats: Allow unplayable formats to be extracted and downloaded.
- ignore_no_formats_error: Ignore "No video formats" error. Usefull for
- extracting metadata even if the video is not actually
- available for download (experimental)
- format_sort: A list of fields by which to sort the video formats.
- See "Sorting Formats" for more details.
- format_sort_force: Force the given format_sort. see "Sorting Formats"
- for more details.
- prefer_free_formats: Whether to prefer video formats with free containers
- over non-free ones of same quality.
- allow_multiple_video_streams: Allow multiple video streams to be merged
- into a single file
- allow_multiple_audio_streams: Allow multiple audio streams to be merged
- into a single file
- check_formats Whether to test if the formats are downloadable.
- Can be True (check all), False (check none),
- 'selected' (check selected formats),
- or None (check only if requested by extractor)
- paths: Dictionary of output paths. The allowed keys are 'home'
- 'temp' and the keys of OUTTMPL_TYPES (in utils.py)
- outtmpl: Dictionary of templates for output names. Allowed keys
- are 'default' and the keys of OUTTMPL_TYPES (in utils.py).
- For compatibility with youtube-dl, a single string can also be used
- outtmpl_na_placeholder: Placeholder for unavailable meta fields.
- restrictfilenames: Do not allow "&" and spaces in file names
- trim_file_name: Limit length of filename (extension excluded)
- windowsfilenames: Force the filenames to be windows compatible
- ignoreerrors: Do not stop on download/postprocessing errors.
- Can be 'only_download' to ignore only download errors.
- Default is 'only_download' for CLI, but False for API
- skip_playlist_after_errors: Number of allowed failures until the rest of
- the playlist is skipped
- allowed_extractors: List of regexes to match against extractor names that are allowed
- overwrites: Overwrite all video and metadata files if True,
- overwrite only non-video files if None
- and don't overwrite any file if False
- For compatibility with youtube-dl,
- "nooverwrites" may also be used instead
- playlist_items: Specific indices of playlist to download.
- playlistrandom: Download playlist items in random order.
- lazy_playlist: Process playlist entries as they are received.
- matchtitle: Download only matching titles.
- rejecttitle: Reject downloads for matching titles.
- logger: Log messages to a logging.Logger instance.
- logtostderr: Print everything to stderr instead of stdout.
- consoletitle: Display progress in console window's titlebar.
- writedescription: Write the video description to a .description file
- writeinfojson: Write the video description to a .info.json file
- clean_infojson: Remove private fields from the infojson
- getcomments: Extract video comments. This will not be written to disk
- unless writeinfojson is also given
- writeannotations: Write the video annotations to a .annotations.xml file
- writethumbnail: Write the thumbnail image to a file
- allow_playlist_files: Whether to write playlists' description, infojson etc
- also to disk when using the 'write*' options
- write_all_thumbnails: Write all thumbnail formats to files
- writelink: Write an internet shortcut file, depending on the
- current platform (.url/.webloc/.desktop)
- writeurllink: Write a Windows internet shortcut file (.url)
- writewebloclink: Write a macOS internet shortcut file (.webloc)
- writedesktoplink: Write a Linux internet shortcut file (.desktop)
- writesubtitles: Write the video subtitles to a file
- writeautomaticsub: Write the automatically generated subtitles to a file
- listsubtitles: Lists all available subtitles for the video
- subtitlesformat: The format code for subtitles
- subtitleslangs: List of languages of the subtitles to download (can be regex).
- The list may contain "all" to refer to all the available
- subtitles. The language can be prefixed with a "-" to
- exclude it from the requested languages, e.g. ['all', '-live_chat']
- keepvideo: Keep the video file after post-processing
- daterange: A DateRange object, download only if the upload_date is in the range.
- skip_download: Skip the actual download of the video file
- cachedir: Location of the cache files in the filesystem.
- False to disable filesystem cache.
- noplaylist: Download single video instead of a playlist if in doubt.
- age_limit: An integer representing the user's age in years.
- Unsuitable videos for the given age are skipped.
- min_views: An integer representing the minimum view count the video
- must have in order to not be skipped.
- Videos without view count information are always
- downloaded. None for no limit.
- max_views: An integer representing the maximum view count.
- Videos that are more popular than that are not
- downloaded.
- Videos without view count information are always
- downloaded. None for no limit.
- download_archive: A set, or the name of a file where all downloads are recorded.
- Videos already present in the file are not downloaded again.
- break_on_existing: Stop the download process after attempting to download a
- file that is in the archive.
- break_on_reject: Stop the download process when encountering a video that
- has been filtered out.
- break_per_url: Whether break_on_reject and break_on_existing
- should act on each input URL as opposed to for the entire queue
- cookiefile: File name or text stream from where cookies should be read and dumped to
- cookiesfrombrowser: A tuple containing the name of the browser, the profile
- name/path from where cookies are loaded, the name of the keyring,
- and the container name, e.g. ('chrome', ) or
- ('vivaldi', 'default', 'BASICTEXT') or ('firefox', 'default', None, 'Meta')
- legacyserverconnect: Explicitly allow HTTPS connection to servers that do not
- support RFC 5746 secure renegotiation
- nocheckcertificate: Do not verify SSL certificates
- client_certificate: Path to client certificate file in PEM format. May include the private key
- client_certificate_key: Path to private key file for client certificate
- client_certificate_password: Password for client certificate private key, if encrypted.
- If not provided and the key is encrypted, hypervideo will ask interactively
- prefer_insecure: Use HTTP instead of HTTPS to retrieve information.
- (Only supported by some extractors)
- http_headers: A dictionary of custom headers to be used for all requests
- proxy: URL of the proxy server to use
- geo_verification_proxy: URL of the proxy to use for IP address verification
- on geo-restricted sites.
- socket_timeout: Time to wait for unresponsive hosts, in seconds
- bidi_workaround: Work around buggy terminals without bidirectional text
- support, using fridibi
- debug_printtraffic:Print out sent and received HTTP traffic
- default_search: Prepend this string if an input url is not valid.
- 'auto' for elaborate guessing
- encoding: Use this encoding instead of the system-specified.
- extract_flat: Whether to resolve and process url_results further
- * False: Always process (default)
- * True: Never process
- * 'in_playlist': Do not process inside playlist/multi_video
- * 'discard': Always process, but don't return the result
- from inside playlist/multi_video
- * 'discard_in_playlist': Same as "discard", but only for
- playlists (not multi_video)
- wait_for_video: If given, wait for scheduled streams to become available.
- The value should be a tuple containing the range
- (min_secs, max_secs) to wait between retries
- postprocessors: A list of dictionaries, each with an entry
- * key: The name of the postprocessor. See
- hypervideo_dl/postprocessor/__init__.py for a list.
- * when: When to run the postprocessor. Allowed values are
- the entries of utils.POSTPROCESS_WHEN
- Assumed to be 'post_process' if not given
- progress_hooks: A list of functions that get called on download
- progress, with a dictionary with the entries
- * status: One of "downloading", "error", or "finished".
- Check this first and ignore unknown values.
- * info_dict: The extracted info_dict
- If status is one of "downloading", or "finished", the
- following properties may also be present:
- * filename: The final filename (always present)
- * tmpfilename: The filename we're currently writing to
- * downloaded_bytes: Bytes on disk
- * total_bytes: Size of the whole file, None if unknown
- * total_bytes_estimate: Guess of the eventual file size,
- None if unavailable.
- * elapsed: The number of seconds since download started.
- * eta: The estimated time in seconds, None if unknown
- * speed: The download speed in bytes/second, None if
- unknown
- * fragment_index: The counter of the currently
- downloaded video fragment.
- * fragment_count: The number of fragments (= individual
- files that will be merged)
- Progress hooks are guaranteed to be called at least once
- (with status "finished") if the download is successful.
- postprocessor_hooks: A list of functions that get called on postprocessing
- progress, with a dictionary with the entries
- * status: One of "started", "processing", or "finished".
- Check this first and ignore unknown values.
- * postprocessor: Name of the postprocessor
- * info_dict: The extracted info_dict
- Progress hooks are guaranteed to be called at least twice
- (with status "started" and "finished") if the processing is successful.
- merge_output_format: "/" separated list of extensions to use when merging formats.
- final_ext: Expected final extension; used to detect when the file was
- already downloaded and converted
- fixup: Automatically correct known faults of the file.
- One of:
- - "never": do nothing
- - "warn": only emit a warning
- - "detect_or_warn": check whether we can do anything
- about it, warn otherwise (default)
- source_address: Client-side IP address to bind to.
- sleep_interval_requests: Number of seconds to sleep between requests
- during extraction
- sleep_interval: Number of seconds to sleep before each download when
- used alone or a lower bound of a range for randomized
- sleep before each download (minimum possible number
- of seconds to sleep) when used along with
- max_sleep_interval.
- max_sleep_interval:Upper bound of a range for randomized sleep before each
- download (maximum possible number of seconds to sleep).
- Must only be used along with sleep_interval.
- Actual sleep time will be a random float from range
- [sleep_interval; max_sleep_interval].
- sleep_interval_subtitles: Number of seconds to sleep before each subtitle download
- listformats: Print an overview of available video formats and exit.
- list_thumbnails: Print a table of all thumbnails and exit.
- match_filter: A function that gets called for every video with the signature
- (info_dict, *, incomplete: bool) -> Optional[str]
- For backward compatibility with youtube-dl, the signature
- (info_dict) -> Optional[str] is also allowed.
- - If it returns a message, the video is ignored.
- - If it returns None, the video is downloaded.
- - If it returns utils.NO_DEFAULT, the user is interactively
- asked whether to download the video.
- match_filter_func in utils.py is one example for this.
- no_color: Do not emit color codes in output.
- geo_bypass: Bypass geographic restriction via faking X-Forwarded-For
- HTTP header
- geo_bypass_country:
- Two-letter ISO 3166-2 country code that will be used for
- explicit geographic restriction bypassing via faking
- X-Forwarded-For HTTP header
- geo_bypass_ip_block:
- IP range in CIDR notation that will be used similarly to
- geo_bypass_country
- external_downloader: A dictionary of protocol keys and the executable of the
- external downloader to use for it. The allowed protocols
- are default|http|ftp|m3u8|dash|rtsp|rtmp|mms.
- Set the value to 'native' to use the native downloader
- compat_opts: Compatibility options. See "Differences in default behavior".
- The following options do not work when used through the API:
- filename, abort-on-error, multistreams, no-live-chat, format-sort
- no-clean-infojson, no-playlist-metafiles, no-keep-subs, no-attach-info-json.
- Refer __init__.py for their implementation
- progress_template: Dictionary of templates for progress outputs.
- Allowed keys are 'download', 'postprocess',
- 'download-title' (console title) and 'postprocess-title'.
- The template is mapped on a dictionary with keys 'progress' and 'info'
- retry_sleep_functions: Dictionary of functions that takes the number of attempts
- as argument and returns the time to sleep in seconds.
- Allowed keys are 'http', 'fragment', 'file_access'
- download_ranges: A callback function that gets called for every video with
- the signature (info_dict, ydl) -> Iterable[Section].
- Only the returned sections will be downloaded.
- Each Section is a dict with the following keys:
- * start_time: Start time of the section in seconds
- * end_time: End time of the section in seconds
- * title: Section title (Optional)
- * index: Section number (Optional)
- force_keyframes_at_cuts: Re-encode the video when downloading ranges to get precise cuts
- noprogress: Do not print the progress bar
- live_from_start: Whether to download livestreams videos from the start
- The following parameters are not used by YoutubeDL itself, they are used by
- the downloader (see hypervideo_dl/downloader/common.py):
- nopart, updatetime, buffersize, ratelimit, throttledratelimit, min_filesize,
- max_filesize, test, noresizebuffer, retries, file_access_retries, fragment_retries,
- continuedl, xattr_set_filesize, hls_use_mpegts, http_chunk_size,
- external_downloader_args, concurrent_fragment_downloads.
- The following options are used by the post processors:
- ffmpeg_location: Location of the ffmpeg/avconv binary; either the path
- to the binary or its containing directory.
- postprocessor_args: A dictionary of postprocessor/executable keys (in lower case)
- and a list of additional command-line arguments for the
- postprocessor/executable. The dict can also have "PP+EXE" keys
- which are used when the given exe is used by the given PP.
- Use 'default' as the name for arguments to passed to all PP
- For compatibility with youtube-dl, a single list of args
- can also be used
- The following options are used by the extractors:
- extractor_retries: Number of times to retry for known errors
- dynamic_mpd: Whether to process dynamic DASH manifests (default: True)
- hls_split_discontinuity: Split HLS playlists to different formats at
- discontinuities such as ad breaks (default: False)
- extractor_args: A dictionary of arguments to be passed to the extractors.
- See "EXTRACTOR ARGUMENTS" for details.
- E.g. {'youtube': {'skip': ['dash', 'hls']}}
- mark_watched: Mark videos watched (even with --simulate). Only for YouTube
- The following options are deprecated and may be removed in the future:
- force_generic_extractor: Force downloader to use the generic extractor
- - Use allowed_extractors = ['generic', 'default']
- playliststart: - Use playlist_items
- Playlist item to start at.
- playlistend: - Use playlist_items
- Playlist item to end at.
- playlistreverse: - Use playlist_items
- Download playlist items in reverse order.
- forceurl: - Use forceprint
- Force printing final URL.
- forcetitle: - Use forceprint
- Force printing title.
- forceid: - Use forceprint
- Force printing ID.
- forcethumbnail: - Use forceprint
- Force printing thumbnail URL.
- forcedescription: - Use forceprint
- Force printing description.
- forcefilename: - Use forceprint
- Force printing final filename.
- forceduration: - Use forceprint
- Force printing duration.
- allsubtitles: - Use subtitleslangs = ['all']
- Downloads all the subtitles of the video
- (requires writesubtitles or writeautomaticsub)
- include_ads: - Doesn't work
- Download ads as well
- call_home: - Not implemented
- Boolean, true iff we are allowed to contact the
- hypervideo servers for debugging.
- post_hooks: - Register a custom postprocessor
- A list of functions that get called as the final step
- for each video file, after all postprocessors have been
- called. The filename will be passed as the only argument.
- hls_prefer_native: - Use external_downloader = {'m3u8': 'native'} or {'m3u8': 'ffmpeg'}.
- Use the native HLS downloader instead of ffmpeg/avconv
- if True, otherwise use ffmpeg/avconv if False, otherwise
- use downloader suggested by extractor if None.
- prefer_ffmpeg: - avconv support is deprecated
- If False, use avconv instead of ffmpeg if both are available,
- otherwise prefer ffmpeg.
- youtube_include_dash_manifest: - Use extractor_args
- If True (default), DASH manifests and related
- data will be downloaded and processed by extractor.
- You can reduce network I/O by disabling it if you don't
- care about DASH. (only for youtube)
- youtube_include_hls_manifest: - Use extractor_args
- If True (default), HLS manifests and related
- data will be downloaded and processed by extractor.
- You can reduce network I/O by disabling it if you don't
- care about HLS. (only for youtube)
- """
- _NUMERIC_FIELDS = {
- 'width', 'height', 'asr', 'audio_channels', 'fps',
- 'tbr', 'abr', 'vbr', 'filesize', 'filesize_approx',
- 'timestamp', 'release_timestamp',
- 'duration', 'view_count', 'like_count', 'dislike_count', 'repost_count',
- 'average_rating', 'comment_count', 'age_limit',
- 'start_time', 'end_time',
- 'chapter_number', 'season_number', 'episode_number',
- 'track_number', 'disc_number', 'release_year',
- }
- _format_fields = {
- # NB: Keep in sync with the docstring of extractor/common.py
- 'url', 'manifest_url', 'manifest_stream_number', 'ext', 'format', 'format_id', 'format_note',
- 'width', 'height', 'aspect_ratio', 'resolution', 'dynamic_range', 'tbr', 'abr', 'acodec', 'asr', 'audio_channels',
- 'vbr', 'fps', 'vcodec', 'container', 'filesize', 'filesize_approx', 'rows', 'columns',
- 'player_url', 'protocol', 'fragment_base_url', 'fragments', 'is_from_start',
- 'preference', 'language', 'language_preference', 'quality', 'source_preference',
- 'http_headers', 'stretched_ratio', 'no_resume', 'has_drm', 'downloader_options',
- 'page_url', 'app', 'play_path', 'tc_url', 'flash_version', 'rtmp_live', 'rtmp_conn', 'rtmp_protocol', 'rtmp_real_time'
- }
- _format_selection_exts = {
- 'audio': set(MEDIA_EXTENSIONS.common_audio),
- 'video': set(MEDIA_EXTENSIONS.common_video + ('3gp', )),
- 'storyboards': set(MEDIA_EXTENSIONS.storyboards),
- }
- def __init__(self, params=None, auto_init=True):
- """Create a FileDownloader object with the given options.
- @param auto_init Whether to load the default extractors and print header (if verbose).
- Set to 'no_verbose_header' to not print the header
- """
- if params is None:
- params = {}
- self.params = params
- self._ies = {}
- self._ies_instances = {}
- self._pps = {k: [] for k in POSTPROCESS_WHEN}
- self._printed_messages = set()
- self._first_webpage_request = True
- self._post_hooks = []
- self._progress_hooks = []
- self._postprocessor_hooks = []
- self._download_retcode = 0
- self._num_downloads = 0
- self._num_videos = 0
- self._playlist_level = 0
- self._playlist_urls = set()
- self.cache = Cache(self)
- windows_enable_vt_mode()
- stdout = sys.stderr if self.params.get('logtostderr') else sys.stdout
- self._out_files = Namespace(
- out=stdout,
- error=sys.stderr,
- screen=sys.stderr if self.params.get('quiet') else stdout,
- console=None if compat_os_name == 'nt' else next(
- filter(supports_terminal_sequences, (sys.stderr, sys.stdout)), None)
- )
- self._allow_colors = Namespace(**{
- type_: not self.params.get('no_color') and supports_terminal_sequences(stream)
- for type_, stream in self._out_files.items_ if type_ != 'console'
- })
- # The code is left like this to be reused for future deprecations
- MIN_SUPPORTED, MIN_RECOMMENDED = (3, 7), (3, 7)
- current_version = sys.version_info[:2]
- if current_version < MIN_RECOMMENDED:
- msg = ('Support for Python version %d.%d has been deprecated. '
- 'See https://github.com/hypervideo/hypervideo/issues/3764 for more details.'
- '\n You will no longer receive updates on this version')
- if current_version < MIN_SUPPORTED:
- msg = 'Python version %d.%d is no longer supported'
- self.deprecation_warning(
- f'{msg}! Please update to Python %d.%d or above' % (*current_version, *MIN_RECOMMENDED))
- if self.params.get('allow_unplayable_formats'):
- self.report_warning(
- f'You have asked for {self._format_err("UNPLAYABLE", self.Styles.EMPHASIS)} formats to be listed/downloaded. '
- 'This is a developer option intended for debugging. \n'
- ' If you experience any issues while using this option, '
- f'{self._format_err("DO NOT", self.Styles.ERROR)} open a bug report')
- if self.params.get('bidi_workaround', False):
- try:
- import pty
- master, slave = pty.openpty()
- width = shutil.get_terminal_size().columns
- width_args = [] if width is None else ['-w', str(width)]
- sp_kwargs = {'stdin': subprocess.PIPE, 'stdout': slave, 'stderr': self._out_files.error}
- try:
- self._output_process = Popen(['bidiv'] + width_args, **sp_kwargs)
- except OSError:
- self._output_process = Popen(['fribidi', '-c', 'UTF-8'] + width_args, **sp_kwargs)
- self._output_channel = os.fdopen(master, 'rb')
- except OSError as ose:
- if ose.errno == errno.ENOENT:
- self.report_warning(
- 'Could not find fribidi executable, ignoring --bidi-workaround. '
- 'Make sure that fribidi is an executable file in one of the directories in your $PATH.')
- else:
- raise
- self.params['compat_opts'] = set(self.params.get('compat_opts', ()))
- if auto_init and auto_init != 'no_verbose_header':
- self.print_debug_header()
- def check_deprecated(param, option, suggestion):
- if self.params.get(param) is not None:
- self.report_warning(f'{option} is deprecated. Use {suggestion} instead')
- return True
- return False
- if check_deprecated('cn_verification_proxy', '--cn-verification-proxy', '--geo-verification-proxy'):
- if self.params.get('geo_verification_proxy') is None:
- self.params['geo_verification_proxy'] = self.params['cn_verification_proxy']
- check_deprecated('autonumber', '--auto-number', '-o "%(autonumber)s-%(title)s.%(ext)s"')
- check_deprecated('usetitle', '--title', '-o "%(title)s-%(id)s.%(ext)s"')
- check_deprecated('useid', '--id', '-o "%(id)s.%(ext)s"')
- for msg in self.params.get('_warnings', []):
- self.report_warning(msg)
- for msg in self.params.get('_deprecation_warnings', []):
- self.deprecated_feature(msg)
- if 'list-formats' in self.params['compat_opts']:
- self.params['listformats_table'] = False
- if 'overwrites' not in self.params and self.params.get('nooverwrites') is not None:
- # nooverwrites was unnecessarily changed to overwrites
- # in 0c3d0f51778b153f65c21906031c2e091fcfb641
- # This ensures compatibility with both keys
- self.params['overwrites'] = not self.params['nooverwrites']
- elif self.params.get('overwrites') is None:
- self.params.pop('overwrites', None)
- else:
- self.params['nooverwrites'] = not self.params['overwrites']
- if self.params.get('simulate') is None and any((
- self.params.get('list_thumbnails'),
- self.params.get('listformats'),
- self.params.get('listsubtitles'),
- )):
- self.params['simulate'] = 'list_only'
- self.params.setdefault('forceprint', {})
- self.params.setdefault('print_to_file', {})
- # Compatibility with older syntax
- if not isinstance(params['forceprint'], dict):
- self.params['forceprint'] = {'video': params['forceprint']}
- if auto_init:
- self.add_default_info_extractors()
- if (sys.platform != 'win32'
- and sys.getfilesystemencoding() in ['ascii', 'ANSI_X3.4-1968']
- and not self.params.get('restrictfilenames', False)):
- # Unicode filesystem API will throw errors (#1474, #13027)
- self.report_warning(
- 'Assuming --restrict-filenames since file system encoding '
- 'cannot encode all characters. '
- 'Set the LC_ALL environment variable to fix this.')
- self.params['restrictfilenames'] = True
- self._parse_outtmpl()
- # Creating format selector here allows us to catch syntax errors before the extraction
- self.format_selector = (
- self.params.get('format') if self.params.get('format') in (None, '-')
- else self.params['format'] if callable(self.params['format'])
- else self.build_format_selector(self.params['format']))
- # Set http_headers defaults according to std_headers
- self.params['http_headers'] = merge_headers(std_headers, self.params.get('http_headers', {}))
- hooks = {
- 'post_hooks': self.add_post_hook,
- 'progress_hooks': self.add_progress_hook,
- 'postprocessor_hooks': self.add_postprocessor_hook,
- }
- for opt, fn in hooks.items():
- for ph in self.params.get(opt, []):
- fn(ph)
- for pp_def_raw in self.params.get('postprocessors', []):
- pp_def = dict(pp_def_raw)
- when = pp_def.pop('when', 'post_process')
- self.add_post_processor(
- get_postprocessor(pp_def.pop('key'))(self, **pp_def),
- when=when)
- self._setup_opener()
- register_socks_protocols()
- def preload_download_archive(fn):
- """Preload the archive, if any is specified"""
- archive = set()
- if fn is None:
- return archive
- elif not is_path_like(fn):
- return fn
- self.write_debug(f'Loading archive file {fn!r}')
- try:
- with locked_file(fn, 'r', encoding='utf-8') as archive_file:
- for line in archive_file:
- archive.add(line.strip())
- except OSError as ioe:
- if ioe.errno != errno.ENOENT:
- raise
- return archive
- self.archive = preload_download_archive(self.params.get('download_archive'))
- def warn_if_short_id(self, argv):
- # short YouTube ID starting with dash?
- idxs = [
- i for i, a in enumerate(argv)
- if re.match(r'^-[0-9A-Za-z_-]{10}$', a)]
- if idxs:
- correct_argv = (
- ['hypervideo']
- + [a for i, a in enumerate(argv) if i not in idxs]
- + ['--'] + [argv[i] for i in idxs]
- )
- self.report_warning(
- 'Long argument string detected. '
- 'Use -- to separate parameters and URLs, like this:\n%s' %
- args_to_str(correct_argv))
- def add_info_extractor(self, ie):
- """Add an InfoExtractor object to the end of the list."""
- ie_key = ie.ie_key()
- self._ies[ie_key] = ie
- if not isinstance(ie, type):
- self._ies_instances[ie_key] = ie
- ie.set_downloader(self)
- def get_info_extractor(self, ie_key):
- """
- Get an instance of an IE with name ie_key, it will try to get one from
- the _ies list, if there's no instance it will create a new one and add
- it to the extractor list.
- """
- ie = self._ies_instances.get(ie_key)
- if ie is None:
- ie = get_info_extractor(ie_key)()
- self.add_info_extractor(ie)
- return ie
- def add_default_info_extractors(self):
- """
- Add the InfoExtractors returned by gen_extractors to the end of the list
- """
- all_ies = {ie.IE_NAME.lower(): ie for ie in gen_extractor_classes()}
- all_ies['end'] = UnsupportedURLIE()
- try:
- ie_names = orderedSet_from_options(
- self.params.get('allowed_extractors', ['default']), {
- 'all': list(all_ies),
- 'default': [name for name, ie in all_ies.items() if ie._ENABLED],
- }, use_regex=True)
- except re.error as e:
- raise ValueError(f'Wrong regex for allowed_extractors: {e.pattern}')
- for name in ie_names:
- self.add_info_extractor(all_ies[name])
- self.write_debug(f'Loaded {len(ie_names)} extractors')
- def add_post_processor(self, pp, when='post_process'):
- """Add a PostProcessor object to the end of the chain."""
- assert when in POSTPROCESS_WHEN, f'Invalid when={when}'
- self._pps[when].append(pp)
- pp.set_downloader(self)
- def add_post_hook(self, ph):
- """Add the post hook"""
- self._post_hooks.append(ph)
- def add_progress_hook(self, ph):
- """Add the download progress hook"""
- self._progress_hooks.append(ph)
- def add_postprocessor_hook(self, ph):
- """Add the postprocessing progress hook"""
- self._postprocessor_hooks.append(ph)
- for pps in self._pps.values():
- for pp in pps:
- pp.add_progress_hook(ph)
- def _bidi_workaround(self, message):
- if not hasattr(self, '_output_channel'):
- return message
- assert hasattr(self, '_output_process')
- assert isinstance(message, str)
- line_count = message.count('\n') + 1
- self._output_process.stdin.write((message + '\n').encode())
- self._output_process.stdin.flush()
- res = ''.join(self._output_channel.readline().decode()
- for _ in range(line_count))
- return res[:-len('\n')]
- def _write_string(self, message, out=None, only_once=False):
- if only_once:
- if message in self._printed_messages:
- return
- self._printed_messages.add(message)
- write_string(message, out=out, encoding=self.params.get('encoding'))
- def to_stdout(self, message, skip_eol=False, quiet=None):
- """Print message to stdout"""
- if quiet is not None:
- self.deprecation_warning('"YoutubeDL.to_stdout" no longer accepts the argument quiet. '
- 'Use "YoutubeDL.to_screen" instead')
- if skip_eol is not False:
- self.deprecation_warning('"YoutubeDL.to_stdout" no longer accepts the argument skip_eol. '
- 'Use "YoutubeDL.to_screen" instead')
- self._write_string(f'{self._bidi_workaround(message)}\n', self._out_files.out)
- def to_screen(self, message, skip_eol=False, quiet=None, only_once=False):
- """Print message to screen if not in quiet mode"""
- if self.params.get('logger'):
- self.params['logger'].debug(message)
- return
- if (self.params.get('quiet') if quiet is None else quiet) and not self.params.get('verbose'):
- return
- self._write_string(
- '%s%s' % (self._bidi_workaround(message), ('' if skip_eol else '\n')),
- self._out_files.screen, only_once=only_once)
- def to_stderr(self, message, only_once=False):
- """Print message to stderr"""
- assert isinstance(message, str)
- if self.params.get('logger'):
- self.params['logger'].error(message)
- else:
- self._write_string(f'{self._bidi_workaround(message)}\n', self._out_files.error, only_once=only_once)
- def _send_console_code(self, code):
- if compat_os_name == 'nt' or not self._out_files.console:
- return
- self._write_string(code, self._out_files.console)
- def to_console_title(self, message):
- if not self.params.get('consoletitle', False):
- return
- message = remove_terminal_sequences(message)
- if compat_os_name == 'nt':
- if ctypes.windll.kernel32.GetConsoleWindow():
- # c_wchar_p() might not be necessary if `message` is
- # already of type unicode()
- ctypes.windll.kernel32.SetConsoleTitleW(ctypes.c_wchar_p(message))
- else:
- self._send_console_code(f'\033]0;{message}\007')
- def save_console_title(self):
- if not self.params.get('consoletitle') or self.params.get('simulate'):
- return
- self._send_console_code('\033[22;0t') # Save the title on stack
- def restore_console_title(self):
- if not self.params.get('consoletitle') or self.params.get('simulate'):
- return
- self._send_console_code('\033[23;0t') # Restore the title from stack
- def __enter__(self):
- self.save_console_title()
- return self
- def __exit__(self, *args):
- self.restore_console_title()
- if self.params.get('cookiefile') is not None:
- self.cookiejar.save(ignore_discard=True, ignore_expires=True)
- def trouble(self, message=None, tb=None, is_error=True):
- """Determine action to take when a download problem appears.
- Depending on if the downloader has been configured to ignore
- download errors or not, this method may throw an exception or
- not when errors are found, after printing the message.
- @param tb If given, is additional traceback information
- @param is_error Whether to raise error according to ignorerrors
- """
- if message is not None:
- self.to_stderr(message)
- if self.params.get('verbose'):
- if tb is None:
- if sys.exc_info()[0]: # if .trouble has been called from an except block
- tb = ''
- if hasattr(sys.exc_info()[1], 'exc_info') and sys.exc_info()[1].exc_info[0]:
- tb += ''.join(traceback.format_exception(*sys.exc_info()[1].exc_info))
- tb += encode_compat_str(traceback.format_exc())
- else:
- tb_data = traceback.format_list(traceback.extract_stack())
- tb = ''.join(tb_data)
- if tb:
- self.to_stderr(tb)
- if not is_error:
- return
- if not self.params.get('ignoreerrors'):
- if sys.exc_info()[0] and hasattr(sys.exc_info()[1], 'exc_info') and sys.exc_info()[1].exc_info[0]:
- exc_info = sys.exc_info()[1].exc_info
- else:
- exc_info = sys.exc_info()
- raise DownloadError(message, exc_info)
- self._download_retcode = 1
- Styles = Namespace(
- HEADERS='yellow',
- EMPHASIS='light blue',
- FILENAME='green',
- ID='green',
- DELIM='blue',
- ERROR='red',
- WARNING='yellow',
- SUPPRESS='light black',
- )
- def _format_text(self, handle, allow_colors, text, f, fallback=None, *, test_encoding=False):
- text = str(text)
- if test_encoding:
- original_text = text
- # handle.encoding can be None. See https://github.com/hypervideo/hypervideo/issues/2711
- encoding = self.params.get('encoding') or getattr(handle, 'encoding', None) or 'ascii'
- text = text.encode(encoding, 'ignore').decode(encoding)
- if fallback is not None and text != original_text:
- text = fallback
- return format_text(text, f) if allow_colors else text if fallback is None else fallback
- def _format_out(self, *args, **kwargs):
- return self._format_text(self._out_files.out, self._allow_colors.out, *args, **kwargs)
- def _format_screen(self, *args, **kwargs):
- return self._format_text(self._out_files.screen, self._allow_colors.screen, *args, **kwargs)
- def _format_err(self, *args, **kwargs):
- return self._format_text(self._out_files.error, self._allow_colors.error, *args, **kwargs)
- def report_warning(self, message, only_once=False):
- '''
- Print the message to stderr, it will be prefixed with 'WARNING:'
- If stderr is a tty file the 'WARNING:' will be colored
- '''
- if self.params.get('logger') is not None:
- self.params['logger'].warning(message)
- else:
- if self.params.get('no_warnings'):
- return
- self.to_stderr(f'{self._format_err("WARNING:", self.Styles.WARNING)} {message}', only_once)
- def deprecation_warning(self, message, *, stacklevel=0):
- deprecation_warning(
- message, stacklevel=stacklevel + 1, printer=self.report_error, is_error=False)
- def deprecated_feature(self, message):
- if self.params.get('logger') is not None:
- self.params['logger'].warning(f'Deprecated Feature: {message}')
- self.to_stderr(f'{self._format_err("Deprecated Feature:", self.Styles.ERROR)} {message}', True)
- def report_error(self, message, *args, **kwargs):
- '''
- Do the same as trouble, but prefixes the message with 'ERROR:', colored
- in red if stderr is a tty file.
- '''
- self.trouble(f'{self._format_err("ERROR:", self.Styles.ERROR)} {message}', *args, **kwargs)
- def write_debug(self, message, only_once=False):
- '''Log debug message or Print message to stderr'''
- if not self.params.get('verbose', False):
- return
- message = f'[debug] {message}'
- if self.params.get('logger'):
- self.params['logger'].debug(message)
- else:
- self.to_stderr(message, only_once)
- def report_file_already_downloaded(self, file_name):
- """Report file has already been fully downloaded."""
- try:
- self.to_screen('[download] %s has already been downloaded' % file_name)
- except UnicodeEncodeError:
- self.to_screen('[download] The file has already been downloaded')
- def report_file_delete(self, file_name):
- """Report that existing file will be deleted."""
- try:
- self.to_screen('Deleting existing file %s' % file_name)
- except UnicodeEncodeError:
- self.to_screen('Deleting existing file')
- def raise_no_formats(self, info, forced=False, *, msg=None):
- has_drm = info.get('_has_drm')
- ignored, expected = self.params.get('ignore_no_formats_error'), bool(msg)
- msg = msg or has_drm and 'This video is DRM protected' or 'No video formats found!'
- if forced or not ignored:
- raise ExtractorError(msg, video_id=info['id'], ie=info['extractor'],
- expected=has_drm or ignored or expected)
- else:
- self.report_warning(msg)
- def parse_outtmpl(self):
- self.deprecation_warning('"YoutubeDL.parse_outtmpl" is deprecated and may be removed in a future version')
- self._parse_outtmpl()
- return self.params['outtmpl']
- def _parse_outtmpl(self):
- sanitize = IDENTITY
- if self.params.get('restrictfilenames'): # Remove spaces in the default template
- sanitize = lambda x: x.replace(' - ', ' ').replace(' ', '-')
- outtmpl = self.params.setdefault('outtmpl', {})
- if not isinstance(outtmpl, dict):
- self.params['outtmpl'] = outtmpl = {'default': outtmpl}
- outtmpl.update({k: sanitize(v) for k, v in DEFAULT_OUTTMPL.items() if outtmpl.get(k) is None})
- def get_output_path(self, dir_type='', filename=None):
- paths = self.params.get('paths', {})
- assert isinstance(paths, dict), '"paths" parameter must be a dictionary'
- path = os.path.join(
- expand_path(paths.get('home', '').strip()),
- expand_path(paths.get(dir_type, '').strip()) if dir_type else '',
- filename or '')
- return sanitize_path(path, force=self.params.get('windowsfilenames'))
- @staticmethod
- def _outtmpl_expandpath(outtmpl):
- # expand_path translates '%%' into '%' and '$$' into '$'
- # correspondingly that is not what we want since we need to keep
- # '%%' intact for template dict substitution step. Working around
- # with boundary-alike separator hack.
- sep = ''.join([random.choice(ascii_letters) for _ in range(32)])
- outtmpl = outtmpl.replace('%%', f'%{sep}%').replace('$$', f'${sep}$')
- # outtmpl should be expand_path'ed before template dict substitution
- # because meta fields may contain env variables we don't want to
- # be expanded. E.g. for outtmpl "%(title)s.%(ext)s" and
- # title "Hello $PATH", we don't want `$PATH` to be expanded.
- return expand_path(outtmpl).replace(sep, '')
- @staticmethod
- def escape_outtmpl(outtmpl):
- ''' Escape any remaining strings like %s, %abc% etc. '''
- return re.sub(
- STR_FORMAT_RE_TMPL.format('', '(?![%(\0])'),
- lambda mobj: ('' if mobj.group('has_key') else '%') + mobj.group(0),
- outtmpl)
- @classmethod
- def validate_outtmpl(cls, outtmpl):
- ''' @return None or Exception object '''
- outtmpl = re.sub(
- STR_FORMAT_RE_TMPL.format('[^)]*', '[ljhqBUDS]'),
- lambda mobj: f'{mobj.group(0)[:-1]}s',
- cls._outtmpl_expandpath(outtmpl))
- try:
- cls.escape_outtmpl(outtmpl) % collections.defaultdict(int)
- return None
- except ValueError as err:
- return err
- @staticmethod
- def _copy_infodict(info_dict):
- info_dict = dict(info_dict)
- info_dict.pop('__postprocessors', None)
- info_dict.pop('__pending_error', None)
- return info_dict
- def prepare_outtmpl(self, outtmpl, info_dict, sanitize=False):
- """ Make the outtmpl and info_dict suitable for substitution: ydl.escape_outtmpl(outtmpl) % info_dict
- @param sanitize Whether to sanitize the output as a filename.
- For backward compatibility, a function can also be passed
- """
- info_dict.setdefault('epoch', int(time.time())) # keep epoch consistent once set
- info_dict = self._copy_infodict(info_dict)
- info_dict['duration_string'] = ( # %(duration>%H-%M-%S)s is wrong if duration > 24hrs
- formatSeconds(info_dict['duration'], '-' if sanitize else ':')
- if info_dict.get('duration', None) is not None
- else None)
- info_dict['autonumber'] = int(self.params.get('autonumber_start', 1) - 1 + self._num_downloads)
- info_dict['video_autonumber'] = self._num_videos
- if info_dict.get('resolution') is None:
- info_dict['resolution'] = self.format_resolution(info_dict, default=None)
- # For fields playlist_index, playlist_autonumber and autonumber convert all occurrences
- # of %(field)s to %(field)0Nd for backward compatibility
- field_size_compat_map = {
- 'playlist_index': number_of_digits(info_dict.get('__last_playlist_index') or 0),
- 'playlist_autonumber': number_of_digits(info_dict.get('n_entries') or 0),
- 'autonumber': self.params.get('autonumber_size') or 5,
- }
- TMPL_DICT = {}
- EXTERNAL_FORMAT_RE = re.compile(STR_FORMAT_RE_TMPL.format('[^)]*', f'[{STR_FORMAT_TYPES}ljhqBUDS]'))
- MATH_FUNCTIONS = {
- '+': float.__add__,
- '-': float.__sub__,
- }
- # Field is of the form key1.key2...
- # where keys (except first) can be string, int, slice or "{field, ...}"
- FIELD_INNER_RE = r'(?:\w+|%(num)s|%(num)s?(?::%(num)s?){1,2})' % {'num': r'(?:-?\d+)'}
- FIELD_RE = r'\w*(?:\.(?:%(inner)s|{%(field)s(?:,%(field)s)*}))*' % {
- 'inner': FIELD_INNER_RE,
- 'field': rf'\w*(?:\.{FIELD_INNER_RE})*'
- }
- MATH_FIELD_RE = rf'(?:{FIELD_RE}|-?{NUMBER_RE})'
- MATH_OPERATORS_RE = r'(?:%s)' % '|'.join(map(re.escape, MATH_FUNCTIONS.keys()))
- INTERNAL_FORMAT_RE = re.compile(rf'''(?x)
- (?P<negate>-)?
- (?P<fields>{FIELD_RE})
- (?P<maths>(?:{MATH_OPERATORS_RE}{MATH_FIELD_RE})*)
- (?:>(?P<strf_format>.+?))?
- (?P<remaining>
- (?P<alternate>(?<!\\),[^|&)]+)?
- (?:&(?P<replacement>.*?))?
- (?:\|(?P<default>.*?))?
- )$''')
- def _traverse_infodict(fields):
- fields = [f for x in re.split(r'\.({.+?})\.?', fields)
- for f in ([x] if x.startswith('{') else x.split('.'))]
- for i in (0, -1):
- if fields and not fields[i]:
- fields.pop(i)
- for i, f in enumerate(fields):
- if not f.startswith('{'):
- continue
- assert f.endswith('}'), f'No closing brace for {f} in {fields}'
- fields[i] = {k: k.split('.') for k in f[1:-1].split(',')}
- return traverse_obj(info_dict, fields, is_user_input=True, traverse_string=True)
- def get_value(mdict):
- # Object traversal
- value = _traverse_infodict(mdict['fields'])
- # Negative
- if mdict['negate']:
- value = float_or_none(value)
- if value is not None:
- value *= -1
- # Do maths
- offset_key = mdict['maths']
- if offset_key:
- value = float_or_none(value)
- operator = None
- while offset_key:
- item = re.match(
- MATH_FIELD_RE if operator else MATH_OPERATORS_RE,
- offset_key).group(0)
- offset_key = offset_key[len(item):]
- if operator is None:
- operator = MATH_FUNCTIONS[item]
- continue
- item, multiplier = (item[1:], -1) if item[0] == '-' else (item, 1)
- offset = float_or_none(item)
- if offset is None:
- offset = float_or_none(_traverse_infodict(item))
- try:
- value = operator(value, multiplier * offset)
- except (TypeError, ZeroDivisionError):
- return None
- operator = None
- # Datetime formatting
- if mdict['strf_format']:
- value = strftime_or_none(value, mdict['strf_format'].replace('\\,', ','))
- # XXX: Workaround for https://github.com/hypervideo/hypervideo/issues/4485
- if sanitize and value == '':
- value = None
- return value
- na = self.params.get('outtmpl_na_placeholder', 'NA')
- def filename_sanitizer(key, value, restricted=self.params.get('restrictfilenames')):
- return sanitize_filename(str(value), restricted=restricted, is_id=(
- bool(re.search(r'(^|[_.])id(\.|$)', key))
- if 'filename-sanitization' in self.params['compat_opts']
- else NO_DEFAULT))
- sanitizer = sanitize if callable(sanitize) else filename_sanitizer
- sanitize = bool(sanitize)
- def _dumpjson_default(obj):
- if isinstance(obj, (set, LazyList)):
- return list(obj)
- return repr(obj)
- def create_key(outer_mobj):
- if not outer_mobj.group('has_key'):
- return outer_mobj.group(0)
- key = outer_mobj.group('key')
- mobj = re.match(INTERNAL_FORMAT_RE, key)
- initial_field = mobj.group('fields') if mobj else ''
- value, replacement, default = None, None, na
- while mobj:
- mobj = mobj.groupdict()
- default = mobj['default'] if mobj['default'] is not None else default
- value = get_value(mobj)
- replacement = mobj['replacement']
- if value is None and mobj['alternate']:
- mobj = re.match(INTERNAL_FORMAT_RE, mobj['remaining'][1:])
- else:
- break
- fmt = outer_mobj.group('format')
- if fmt == 's' and value is not None and key in field_size_compat_map.keys():
- fmt = f'0{field_size_compat_map[key]:d}d'
- value = default if value is None else value if replacement is None else replacement
- flags = outer_mobj.group('conversion') or ''
- str_fmt = f'{fmt[:-1]}s'
- if fmt[-1] == 'l': # list
- delim = '\n' if '#' in flags else ', '
- value, fmt = delim.join(map(str, variadic(value, allowed_types=(str, bytes)))), str_fmt
- elif fmt[-1] == 'j': # json
- value, fmt = json.dumps(
- value, default=_dumpjson_default,
- indent=4 if '#' in flags else None, ensure_ascii='+' not in flags), str_fmt
- elif fmt[-1] == 'h': # html
- value, fmt = escapeHTML(str(value)), str_fmt
- elif fmt[-1] == 'q': # quoted
- value = map(str, variadic(value) if '#' in flags else [value])
- value, fmt = ' '.join(map(compat_shlex_quote, value)), str_fmt
- elif fmt[-1] == 'B': # bytes
- value = f'%{str_fmt}'.encode() % str(value).encode()
- value, fmt = value.decode('utf-8', 'ignore'), 's'
- elif fmt[-1] == 'U': # unicode normalized
- value, fmt = unicodedata.normalize(
- # "+" = compatibility equivalence, "#" = NFD
- 'NF%s%s' % ('K' if '+' in flags else '', 'D' if '#' in flags else 'C'),
- value), str_fmt
- elif fmt[-1] == 'D': # decimal suffix
- num_fmt, fmt = fmt[:-1].replace('#', ''), 's'
- value = format_decimal_suffix(value, f'%{num_fmt}f%s' if num_fmt else '%d%s',
- factor=1024 if '#' in flags else 1000)
- elif fmt[-1] == 'S': # filename sanitization
- value, fmt = filename_sanitizer(initial_field, value, restricted='#' in flags), str_fmt
- elif fmt[-1] == 'c':
- if value:
- value = str(value)[0]
- else:
- fmt = str_fmt
- elif fmt[-1] not in 'rs': # numeric
- value = float_or_none(value)
- if value is None:
- value, fmt = default, 's'
- if sanitize:
- if fmt[-1] == 'r':
- # If value is an object, sanitize might convert it to a string
- # So we convert it to repr first
- value, fmt = repr(value), str_fmt
- if fmt[-1] in 'csr':
- value = sanitizer(initial_field, value)
- key = '%s\0%s' % (key.replace('%', '%\0'), outer_mobj.group('format'))
- TMPL_DICT[key] = value
- return '{prefix}%({key}){fmt}'.format(key=key, fmt=fmt, prefix=outer_mobj.group('prefix'))
- return EXTERNAL_FORMAT_RE.sub(create_key, outtmpl), TMPL_DICT
- def evaluate_outtmpl(self, outtmpl, info_dict, *args, **kwargs):
- outtmpl, info_dict = self.prepare_outtmpl(outtmpl, info_dict, *args, **kwargs)
- return self.escape_outtmpl(outtmpl) % info_dict
- def _prepare_filename(self, info_dict, *, outtmpl=None, tmpl_type=None):
- assert None in (outtmpl, tmpl_type), 'outtmpl and tmpl_type are mutually exclusive'
- if outtmpl is None:
- outtmpl = self.params['outtmpl'].get(tmpl_type or 'default', self.params['outtmpl']['default'])
- try:
- outtmpl = self._outtmpl_expandpath(outtmpl)
- filename = self.evaluate_outtmpl(outtmpl, info_dict, True)
- if not filename:
- return None
- if tmpl_type in ('', 'temp'):
- final_ext, ext = self.params.get('final_ext'), info_dict.get('ext')
- if final_ext and ext and final_ext != ext and filename.endswith(f'.{final_ext}'):
- filename = replace_extension(filename, ext, final_ext)
- elif tmpl_type:
- force_ext = OUTTMPL_TYPES[tmpl_type]
- if force_ext:
- filename = replace_extension(filename, force_ext, info_dict.get('ext'))
- # https://github.com/blackjack4494/youtube-dlc/issues/85
- trim_file_name = self.params.get('trim_file_name', False)
- if trim_file_name:
- no_ext, *ext = filename.rsplit('.', 2)
- filename = join_nonempty(no_ext[:trim_file_name], *ext, delim='.')
- return filename
- except ValueError as err:
- self.report_error('Error in output template: ' + str(err) + ' (encoding: ' + repr(preferredencoding()) + ')')
- return None
- def prepare_filename(self, info_dict, dir_type='', *, outtmpl=None, warn=False):
- """Generate the output filename"""
- if outtmpl:
- assert not dir_type, 'outtmpl and dir_type are mutually exclusive'
- dir_type = None
- filename = self._prepare_filename(info_dict, tmpl_type=dir_type, outtmpl=outtmpl)
- if not filename and dir_type not in ('', 'temp'):
- return ''
- if warn:
- if not self.params.get('paths'):
- pass
- elif filename == '-':
- self.report_warning('--paths is ignored when an outputting to stdout', only_once=True)
- elif os.path.isabs(filename):
- self.report_warning('--paths is ignored since an absolute path is given in output template', only_once=True)
- if filename == '-' or not filename:
- return filename
- return self.get_output_path(dir_type, filename)
- def _match_entry(self, info_dict, incomplete=False, silent=False):
- """Returns None if the file should be downloaded"""
- _type = info_dict.get('_type', 'video')
- assert incomplete or _type == 'video', 'Only video result can be considered complete'
- video_title = info_dict.get('title', info_dict.get('id', 'entry'))
- def check_filter():
- if _type in ('playlist', 'multi_video'):
- return
- elif _type in ('url', 'url_transparent') and not try_call(
- lambda: self.get_info_extractor(info_dict['ie_key']).is_single_video(info_dict['url'])):
- return
- if 'title' in info_dict:
- # This can happen when we're just evaluating the playlist
- title = info_dict['title']
- matchtitle = self.params.get('matchtitle', False)
- if matchtitle:
- if not re.search(matchtitle, title, re.IGNORECASE):
- return '"' + title + '" title did not match pattern "' + matchtitle + '"'
- rejecttitle = self.params.get('rejecttitle', False)
- if rejecttitle:
- if re.search(rejecttitle, title, re.IGNORECASE):
- return '"' + title + '" title matched reject pattern "' + rejecttitle + '"'
- date = info_dict.get('upload_date')
- if date is not None:
- dateRange = self.params.get('daterange', DateRange())
- if date not in dateRange:
- return f'{date_from_str(date).isoformat()} upload date is not in range {dateRange}'
- view_count = info_dict.get('view_count')
- if view_count is not None:
- min_views = self.params.get('min_views')
- if min_views is not None and view_count < min_views:
- return 'Skipping %s, because it has not reached minimum view count (%d/%d)' % (video_title, view_count, min_views)
- max_views = self.params.get('max_views')
- if max_views is not None and view_count > max_views:
- return 'Skipping %s, because it has exceeded the maximum view count (%d/%d)' % (video_title, view_count, max_views)
- if age_restricted(info_dict.get('age_limit'), self.params.get('age_limit')):
- return 'Skipping "%s" because it is age restricted' % video_title
- match_filter = self.params.get('match_filter')
- if match_filter is not None:
- try:
- ret = match_filter(info_dict, incomplete=incomplete)
- except TypeError:
- # For backward compatibility
- ret = None if incomplete else match_filter(info_dict)
- if ret is NO_DEFAULT:
- while True:
- filename = self._format_screen(self.prepare_filename(info_dict), self.Styles.FILENAME)
- reply = input(self._format_screen(
- f'Download "{filename}"? (Y/n): ', self.Styles.EMPHASIS)).lower().strip()
- if reply in {'y', ''}:
- return None
- elif reply == 'n':
- return f'Skipping {video_title}'
- elif ret is not None:
- return ret
- return None
- if self.in_download_archive(info_dict):
- reason = '%s has already been recorded in the archive' % video_title
- break_opt, break_err = 'break_on_existing', ExistingVideoReached
- else:
- reason = check_filter()
- break_opt, break_err = 'break_on_reject', RejectedVideoReached
- if reason is not None:
- if not silent:
- self.to_screen('[download] ' + reason)
- if self.params.get(break_opt, False):
- raise break_err()
- return reason
- @staticmethod
- def add_extra_info(info_dict, extra_info):
- '''Set the keys from extra_info in info dict if they are missing'''
- for key, value in extra_info.items():
- info_dict.setdefault(key, value)
- def extract_info(self, url, download=True, ie_key=None, extra_info=None,
- process=True, force_generic_extractor=False):
- """
- Extract and return the information dictionary of the URL
- Arguments:
- @param url URL to extract
- Keyword arguments:
- @param download Whether to download videos
- @param process Whether to resolve all unresolved references (URLs, playlist items).
- Must be True for download to work
- @param ie_key Use only the extractor with this key
- @param extra_info Dictionary containing the extra values to add to the info (For internal use only)
- @force_generic_extractor Force using the generic extractor (Deprecated; use ie_key='Generic')
- """
- if extra_info is None:
- extra_info = {}
- if not ie_key and force_generic_extractor:
- ie_key = 'Generic'
- if ie_key:
- ies = {ie_key: self._ies[ie_key]} if ie_key in self._ies else {}
- else:
- ies = self._ies
- for key, ie in ies.items():
- if not ie.suitable(url):
- continue
- if not ie.working():
- self.report_warning('The program functionality for this site has been marked as broken, '
- 'and will probably not work.')
- temp_id = ie.get_temp_id(url)
- if temp_id is not None and self.in_download_archive({'id': temp_id, 'ie_key': key}):
- self.to_screen(f'[{key}] {temp_id}: has already been recorded in the archive')
- if self.params.get('break_on_existing', False):
- raise ExistingVideoReached()
- break
- return self.__extract_info(url, self.get_info_extractor(key), download, extra_info, process)
- else:
- extractors_restricted = self.params.get('allowed_extractors') not in (None, ['default'])
- self.report_error(f'No suitable extractor{format_field(ie_key, None, " (%s)")} found for URL {url}',
- tb=False if extractors_restricted else None)
- def _handle_extraction_exceptions(func):
- @functools.wraps(func)
- def wrapper(self, *args, **kwargs):
- while True:
- try:
- return func(self, *args, **kwargs)
- except (DownloadCancelled, LazyList.IndexError, PagedList.IndexError):
- raise
- except ReExtractInfo as e:
- if e.expected:
- self.to_screen(f'{e}; Re-extracting data')
- else:
- self.to_stderr('\r')
- self.report_warning(f'{e}; Re-extracting data')
- continue
- except GeoRestrictedError as e:
- msg = e.msg
- if e.countries:
- msg += '\nThis video is available in %s.' % ', '.join(
- map(ISO3166Utils.short2full, e.countries))
- msg += '\nYou might want to use a VPN or a proxy server (with --proxy) to workaround.'
- self.report_error(msg)
- except ExtractorError as e: # An error we somewhat expected
- self.report_error(str(e), e.format_traceback())
- except Exception as e:
- if self.params.get('ignoreerrors'):
- self.report_error(str(e), tb=encode_compat_str(traceback.format_exc()))
- else:
- raise
- break
- return wrapper
- def _wait_for_video(self, ie_result={}):
- if (not self.params.get('wait_for_video')
- or ie_result.get('_type', 'video') != 'video'
- or ie_result.get('formats') or ie_result.get('url')):
- return
- format_dur = lambda dur: '%02d:%02d:%02d' % timetuple_from_msec(dur * 1000)[:-1]
- last_msg = ''
- def progress(msg):
- nonlocal last_msg
- full_msg = f'{msg}\n'
- if not self.params.get('noprogress'):
- full_msg = msg + ' ' * (len(last_msg) - len(msg)) + '\r'
- elif last_msg:
- return
- self.to_screen(full_msg, skip_eol=True)
- last_msg = msg
- min_wait, max_wait = self.params.get('wait_for_video')
- diff = try_get(ie_result, lambda x: x['release_timestamp'] - time.time())
- if diff is None and ie_result.get('live_status') == 'is_upcoming':
- diff = round(random.uniform(min_wait, max_wait) if (max_wait and min_wait) else (max_wait or min_wait), 0)
- self.report_warning('Release time of video is not known')
- elif ie_result and (diff or 0) <= 0:
- self.report_warning('Video should already be available according to extracted info')
- diff = min(max(diff or 0, min_wait or 0), max_wait or float('inf'))
- self.to_screen(f'[wait] Waiting for {format_dur(diff)} - Press Ctrl+C to try now')
- wait_till = time.time() + diff
- try:
- while True:
- diff = wait_till - time.time()
- if diff <= 0:
- progress('')
- raise ReExtractInfo('[wait] Wait period ended', expected=True)
- progress(f'[wait] Remaining time until next attempt: {self._format_screen(format_dur(diff), self.Styles.EMPHASIS)}')
- time.sleep(1)
- except KeyboardInterrupt:
- progress('')
- raise ReExtractInfo('[wait] Interrupted by user', expected=True)
- except BaseException as e:
- if not isinstance(e, ReExtractInfo):
- self.to_screen('')
- raise
- @_handle_extraction_exceptions
- def __extract_info(self, url, ie, download, extra_info, process):
- try:
- ie_result = ie.extract(url)
- except UserNotLive as e:
- if process:
- if self.params.get('wait_for_video'):
- self.report_warning(e)
- self._wait_for_video()
- raise
- if ie_result is None: # Finished already (backwards compatibility; listformats and friends should be moved here)
- self.report_warning(f'Extractor {ie.IE_NAME} returned nothing{bug_reports_message()}')
- return
- if isinstance(ie_result, list):
- # Backwards compatibility: old IE result format
- ie_result = {
- '_type': 'compat_list',
- 'entries': ie_result,
- }
- if extra_info.get('original_url'):
- ie_result.setdefault('original_url', extra_info['original_url'])
- self.add_default_extra_info(ie_result, ie, url)
- if process:
- self._wait_for_video(ie_result)
- return self.process_ie_result(ie_result, download, extra_info)
- else:
- return ie_result
- def add_default_extra_info(self, ie_result, ie, url):
- if url is not None:
- self.add_extra_info(ie_result, {
- 'webpage_url': url,
- 'original_url': url,
- })
- webpage_url = ie_result.get('webpage_url')
- if webpage_url:
- self.add_extra_info(ie_result, {
- 'webpage_url_basename': url_basename(webpage_url),
- 'webpage_url_domain': get_domain(webpage_url),
- })
- if ie is not None:
- self.add_extra_info(ie_result, {
- 'extractor': ie.IE_NAME,
- 'extractor_key': ie.ie_key(),
- })
- def process_ie_result(self, ie_result, download=True, extra_info=None):
- """
- Take the result of the ie(may be modified) and resolve all unresolved
- references (URLs, playlist items).
- It will also download the videos if 'download'.
- Returns the resolved ie_result.
- """
- if extra_info is None:
- extra_info = {}
- result_type = ie_result.get('_type', 'video')
- if result_type in ('url', 'url_transparent'):
- ie_result['url'] = sanitize_url(
- ie_result['url'], scheme='http' if self.params.get('prefer_insecure') else 'https')
- if ie_result.get('original_url'):
- extra_info.setdefault('original_url', ie_result['original_url'])
- extract_flat = self.params.get('extract_flat', False)
- if ((extract_flat == 'in_playlist' and 'playlist' in extra_info)
- or extract_flat is True):
- info_copy = ie_result.copy()
- ie = try_get(ie_result.get('ie_key'), self.get_info_extractor)
- if ie and not ie_result.get('id'):
- info_copy['id'] = ie.get_temp_id(ie_result['url'])
- self.add_default_extra_info(info_copy, ie, ie_result['url'])
- self.add_extra_info(info_copy, extra_info)
- info_copy, _ = self.pre_process(info_copy)
- self._fill_common_fields(info_copy, False)
- self.__forced_printings(info_copy, self.prepare_filename(info_copy), incomplete=True)
- self._raise_pending_errors(info_copy)
- if self.params.get('force_write_download_archive', False):
- self.record_download_archive(info_copy)
- return ie_result
- if result_type == 'video':
- self.add_extra_info(ie_result, extra_info)
- ie_result = self.process_video_result(ie_result, download=download)
- self._raise_pending_errors(ie_result)
- additional_urls = (ie_result or {}).get('additional_urls')
- if additional_urls:
- # TODO: Improve MetadataParserPP to allow setting a list
- if isinstance(additional_urls, str):
- additional_urls = [additional_urls]
- self.to_screen(
- '[info] %s: %d additional URL(s) requested' % (ie_result['id'], len(additional_urls)))
- self.write_debug('Additional URLs: "%s"' % '", "'.join(additional_urls))
- ie_result['additional_entries'] = [
- self.extract_info(
- url, download, extra_info=extra_info,
- force_generic_extractor=self.params.get('force_generic_extractor'))
- for url in additional_urls
- ]
- return ie_result
- elif result_type == 'url':
- # We have to add extra_info to the results because it may be
- # contained in a playlist
- return self.extract_info(
- ie_result['url'], download,
- ie_key=ie_result.get('ie_key'),
- extra_info=extra_info)
- elif result_type == 'url_transparent':
- # Use the information from the embedding page
- info = self.extract_info(
- ie_result['url'], ie_key=ie_result.get('ie_key'),
- extra_info=extra_info, download=False, process=False)
- # extract_info may return None when ignoreerrors is enabled and
- # extraction failed with an error, don't crash and return early
- # in this case
- if not info:
- return info
- exempted_fields = {'_type', 'url', 'ie_key'}
- if not ie_result.get('section_end') and ie_result.get('section_start') is None:
- # For video clips, the id etc of the clip extractor should be used
- exempted_fields |= {'id', 'extractor', 'extractor_key'}
- new_result = info.copy()
- new_result.update(filter_dict(ie_result, lambda k, v: v is not None and k not in exempted_fields))
- # Extracted info may not be a video result (i.e.
- # info.get('_type', 'video') != video) but rather an url or
- # url_transparent. In such cases outer metadata (from ie_result)
- # should be propagated to inner one (info). For this to happen
- # _type of info should be overridden with url_transparent. This
- # fixes issue from https://github.com/ytdl-org/youtube-dl/pull/11163.
- if new_result.get('_type') == 'url':
- new_result['_type'] = 'url_transparent'
- return self.process_ie_result(
- new_result, download=download, extra_info=extra_info)
- elif result_type in ('playlist', 'multi_video'):
- # Protect from infinite recursion due to recursively nested playlists
- # (see https://github.com/ytdl-org/youtube-dl/issues/27833)
- webpage_url = ie_result.get('webpage_url') # Playlists maynot have webpage_url
- if webpage_url and webpage_url in self._playlist_urls:
- self.to_screen(
- '[download] Skipping already downloaded playlist: %s'
- % ie_result.get('title') or ie_result.get('id'))
- return
- self._playlist_level += 1
- self._playlist_urls.add(webpage_url)
- self._fill_common_fields(ie_result, False)
- self._sanitize_thumbnails(ie_result)
- try:
- return self.__process_playlist(ie_result, download)
- finally:
- self._playlist_level -= 1
- if not self._playlist_level:
- self._playlist_urls.clear()
- elif result_type == 'compat_list':
- self.report_warning(
- 'Extractor %s returned a compat_list result. '
- 'It needs to be updated.' % ie_result.get('extractor'))
- def _fixup(r):
- self.add_extra_info(r, {
- 'extractor': ie_result['extractor'],
- 'webpage_url': ie_result['webpage_url'],
- 'webpage_url_basename': url_basename(ie_result['webpage_url']),
- 'webpage_url_domain': get_domain(ie_result['webpage_url']),
- 'extractor_key': ie_result['extractor_key'],
- })
- return r
- ie_result['entries'] = [
- self.process_ie_result(_fixup(r), download, extra_info)
- for r in ie_result['entries']
- ]
- return ie_result
- else:
- raise Exception('Invalid result type: %s' % result_type)
- def _ensure_dir_exists(self, path):
- return make_dir(path, self.report_error)
- @staticmethod
- def _playlist_infodict(ie_result, strict=False, **kwargs):
- info = {
- 'playlist_count': ie_result.get('playlist_count'),
- 'playlist': ie_result.get('title') or ie_result.get('id'),
- 'playlist_id': ie_result.get('id'),
- 'playlist_title': ie_result.get('title'),
- 'playlist_uploader': ie_result.get('uploader'),
- 'playlist_uploader_id': ie_result.get('uploader_id'),
- **kwargs,
- }
- if strict:
- return info
- if ie_result.get('webpage_url'):
- info.update({
- 'webpage_url': ie_result['webpage_url'],
- 'webpage_url_basename': url_basename(ie_result['webpage_url']),
- 'webpage_url_domain': get_domain(ie_result['webpage_url']),
- })
- return {
- **info,
- 'playlist_index': 0,
- '__last_playlist_index': max(ie_result['requested_entries'] or (0, 0)),
- 'extractor': ie_result['extractor'],
- 'extractor_key': ie_result['extractor_key'],
- }
- def __process_playlist(self, ie_result, download):
- """Process each entry in the playlist"""
- assert ie_result['_type'] in ('playlist', 'multi_video')
- common_info = self._playlist_infodict(ie_result, strict=True)
- title = common_info.get('playlist') or '<Untitled>'
- if self._match_entry(common_info, incomplete=True) is not None:
- return
- self.to_screen(f'[download] Downloading {ie_result["_type"]}: {title}')
- all_entries = PlaylistEntries(self, ie_result)
- entries = orderedSet(all_entries.get_requested_items(), lazy=True)
- lazy = self.params.get('lazy_playlist')
- if lazy:
- resolved_entries, n_entries = [], 'N/A'
- ie_result['requested_entries'], ie_result['entries'] = None, None
- else:
- entries = resolved_entries = list(entries)
- n_entries = len(resolved_entries)
- ie_result['requested_entries'], ie_result['entries'] = tuple(zip(*resolved_entries)) or ([], [])
- if not ie_result.get('playlist_count'):
- # Better to do this after potentially exhausting entries
- ie_result['playlist_count'] = all_entries.get_full_count()
- extra = self._playlist_infodict(ie_result, n_entries=int_or_none(n_entries))
- ie_copy = collections.ChainMap(ie_result, extra)
- _infojson_written = False
- write_playlist_files = self.params.get('allow_playlist_files', True)
- if write_playlist_files and self.params.get('list_thumbnails'):
- self.list_thumbnails(ie_result)
- if write_playlist_files and not self.params.get('simulate'):
- _infojson_written = self._write_info_json(
- 'playlist', ie_result, self.prepare_filename(ie_copy, 'pl_infojson'))
- if _infojson_written is None:
- return
- if self._write_description('playlist', ie_result,
- self.prepare_filename(ie_copy, 'pl_description')) is None:
- return
- # TODO: This should be passed to ThumbnailsConvertor if necessary
- self._write_thumbnails('playlist', ie_result, self.prepare_filename(ie_copy, 'pl_thumbnail'))
- if lazy:
- if self.params.get('playlistreverse') or self.params.get('playlistrandom'):
- self.report_warning('playlistreverse and playlistrandom are not supported with lazy_playlist', only_once=True)
- elif self.params.get('playlistreverse'):
- entries.reverse()
- elif self.params.get('playlistrandom'):
- random.shuffle(entries)
- self.to_screen(f'[{ie_result["extractor"]}] Playlist {title}: Downloading {n_entries} items'
- f'{format_field(ie_result, "playlist_count", " of %s")}')
- keep_resolved_entries = self.params.get('extract_flat') != 'discard'
- if self.params.get('extract_flat') == 'discard_in_playlist':
- keep_resolved_entries = ie_result['_type'] != 'playlist'
- if keep_resolved_entries:
- self.write_debug('The information of all playlist entries will be held in memory')
- failures = 0
- max_failures = self.params.get('skip_playlist_after_errors') or float('inf')
- for i, (playlist_index, entry) in enumerate(entries):
- if lazy:
- resolved_entries.append((playlist_index, entry))
- if not entry:
- continue
- entry['__x_forwarded_for_ip'] = ie_result.get('__x_forwarded_for_ip')
- if not lazy and 'playlist-index' in self.params.get('compat_opts', []):
- playlist_index = ie_result['requested_entries'][i]
- entry_copy = collections.ChainMap(entry, {
- **common_info,
- 'n_entries': int_or_none(n_entries),
- 'playlist_index': playlist_index,
- 'playlist_autonumber': i + 1,
- })
- if self._match_entry(entry_copy, incomplete=True) is not None:
- # For compatabilty with youtube-dl. See https://github.com/hypervideo/hypervideo/issues/4369
- resolved_entries[i] = (playlist_index, NO_DEFAULT)
- continue
- self.to_screen('[download] Downloading item %s of %s' % (
- self._format_screen(i + 1, self.Styles.ID), self._format_screen(n_entries, self.Styles.EMPHASIS)))
- extra.update({
- 'playlist_index': playlist_index,
- 'playlist_autonumber': i + 1,
- })
- entry_result = self.__process_iterable_entry(entry, download, extra)
- if not entry_result:
- failures += 1
- if failures >= max_failures:
- self.report_error(
- f'Skipping the remaining entries in playlist "{title}" since {failures} items failed extraction')
- break
- if keep_resolved_entries:
- resolved_entries[i] = (playlist_index, entry_result)
- # Update with processed data
- ie_result['entries'] = [e for _, e in resolved_entries if e is not NO_DEFAULT]
- ie_result['requested_entries'] = [i for i, e in resolved_entries if e is not NO_DEFAULT]
- if ie_result['requested_entries'] == try_call(lambda: list(range(1, ie_result['playlist_count'] + 1))):
- # Do not set for full playlist
- ie_result.pop('requested_entries')
- # Write the updated info to json
- if _infojson_written is True and self._write_info_json(
- 'updated playlist', ie_result,
- self.prepare_filename(ie_copy, 'pl_infojson'), overwrite=True) is None:
- return
- ie_result = self.run_all_pps('playlist', ie_result)
- self.to_screen(f'[download] Finished downloading playlist: {title}')
- return ie_result
- @_handle_extraction_exceptions
- def __process_iterable_entry(self, entry, download, extra_info):
- return self.process_ie_result(
- entry, download=download, extra_info=extra_info)
- def _build_format_filter(self, filter_spec):
- " Returns a function to filter the formats according to the filter_spec "
- OPERATORS = {
- '<': operator.lt,
- '<=': operator.le,
- '>': operator.gt,
- '>=': operator.ge,
- '=': operator.eq,
- '!=': operator.ne,
- }
- operator_rex = re.compile(r'''(?x)\s*
- (?P<key>width|height|tbr|abr|vbr|asr|filesize|filesize_approx|fps)\s*
- (?P<op>%s)(?P<none_inclusive>\s*\?)?\s*
- (?P<value>[0-9.]+(?:[kKmMgGtTpPeEzZyY]i?[Bb]?)?)\s*
- ''' % '|'.join(map(re.escape, OPERATORS.keys())))
- m = operator_rex.fullmatch(filter_spec)
- if m:
- try:
- comparison_value = int(m.group('value'))
- except ValueError:
- comparison_value = parse_filesize(m.group('value'))
- if comparison_value is None:
- comparison_value = parse_filesize(m.group('value') + 'B')
- if comparison_value is None:
- raise ValueError(
- 'Invalid value %r in format specification %r' % (
- m.group('value'), filter_spec))
- op = OPERATORS[m.group('op')]
- if not m:
- STR_OPERATORS = {
- '=': operator.eq,
- '^=': lambda attr, value: attr.startswith(value),
- '$=': lambda attr, value: attr.endswith(value),
- '*=': lambda attr, value: value in attr,
- '~=': lambda attr, value: value.search(attr) is not None
- }
- str_operator_rex = re.compile(r'''(?x)\s*
- (?P<key>[a-zA-Z0-9._-]+)\s*
- (?P<negation>!\s*)?(?P<op>%s)\s*(?P<none_inclusive>\?\s*)?
- (?P<quote>["'])?
- (?P<value>(?(quote)(?:(?!(?P=quote))[^\\]|\\.)+|[\w.-]+))
- (?(quote)(?P=quote))\s*
- ''' % '|'.join(map(re.escape, STR_OPERATORS.keys())))
- m = str_operator_rex.fullmatch(filter_spec)
- if m:
- if m.group('op') == '~=':
- comparison_value = re.compile(m.group('value'))
- else:
- comparison_value = re.sub(r'''\\([\\"'])''', r'\1', m.group('value'))
- str_op = STR_OPERATORS[m.group('op')]
- if m.group('negation'):
- op = lambda attr, value: not str_op(attr, value)
- else:
- op = str_op
- if not m:
- raise SyntaxError('Invalid filter specification %r' % filter_spec)
- def _filter(f):
- actual_value = f.get(m.group('key'))
- if actual_value is None:
- return m.group('none_inclusive')
- return op(actual_value, comparison_value)
- return _filter
- def _check_formats(self, formats):
- for f in formats:
- self.to_screen('[info] Testing format %s' % f['format_id'])
- path = self.get_output_path('temp')
- if not self._ensure_dir_exists(f'{path}/'):
- continue
- temp_file = tempfile.NamedTemporaryFile(suffix='.tmp', delete=False, dir=path or None)
- temp_file.close()
- try:
- success, _ = self.dl(temp_file.name, f, test=True)
- except (DownloadError, OSError, ValueError) + network_exceptions:
- success = False
- finally:
- if os.path.exists(temp_file.name):
- try:
- os.remove(temp_file.name)
- except OSError:
- self.report_warning('Unable to delete temporary file "%s"' % temp_file.name)
- if success:
- yield f
- else:
- self.to_screen('[info] Unable to download format %s. Skipping...' % f['format_id'])
- def _default_format_spec(self, info_dict, download=True):
- def can_merge():
- merger = FFmpegMergerPP(self)
- return merger.available and merger.can_merge()
- prefer_best = (
- not self.params.get('simulate')
- and download
- and (
- not can_merge()
- or info_dict.get('is_live') and not self.params.get('live_from_start')
- or self.params['outtmpl']['default'] == '-'))
- compat = (
- prefer_best
- or self.params.get('allow_multiple_audio_streams', False)
- or 'format-spec' in self.params['compat_opts'])
- return (
- 'best/bestvideo+bestaudio' if prefer_best
- else 'bestvideo*+bestaudio/best' if not compat
- else 'bestvideo+bestaudio/best')
- def build_format_selector(self, format_spec):
- def syntax_error(note, start):
- message = (
- 'Invalid format specification: '
- '{}\n\t{}\n\t{}^'.format(note, format_spec, ' ' * start[1]))
- return SyntaxError(message)
- PICKFIRST = 'PICKFIRST'
- MERGE = 'MERGE'
- SINGLE = 'SINGLE'
- GROUP = 'GROUP'
- FormatSelector = collections.namedtuple('FormatSelector', ['type', 'selector', 'filters'])
- allow_multiple_streams = {'audio': self.params.get('allow_multiple_audio_streams', False),
- 'video': self.params.get('allow_multiple_video_streams', False)}
- check_formats = self.params.get('check_formats') == 'selected'
- def _parse_filter(tokens):
- filter_parts = []
- for type, string, start, _, _ in tokens:
- if type == tokenize.OP and string == ']':
- return ''.join(filter_parts)
- else:
- filter_parts.append(string)
- def _remove_unused_ops(tokens):
- # Remove operators that we don't use and join them with the surrounding strings.
- # E.g. 'mp4' '-' 'baseline' '-' '16x9' is converted to 'mp4-baseline-16x9'
- ALLOWED_OPS = ('/', '+', ',', '(', ')')
- last_string, last_start, last_end, last_line = None, None, None, None
- for type, string, start, end, line in tokens:
- if type == tokenize.OP and string == '[':
- if last_string:
- yield tokenize.NAME, last_string, last_start, last_end, last_line
- last_string = None
- yield type, string, start, end, line
- # everything inside brackets will be handled by _parse_filter
- for type, string, start, end, line in tokens:
- yield type, string, start, end, line
- if type == tokenize.OP and string == ']':
- break
- elif type == tokenize.OP and string in ALLOWED_OPS:
- if last_string:
- yield tokenize.NAME, last_string, last_start, last_end, last_line
- last_string = None
- yield type, string, start, end, line
- elif type in [tokenize.NAME, tokenize.NUMBER, tokenize.OP]:
- if not last_string:
- last_string = string
- last_start = start
- last_end = end
- else:
- last_string += string
- if last_string:
- yield tokenize.NAME, last_string, last_start, last_end, last_line
- def _parse_format_selection(tokens, inside_merge=False, inside_choice=False, inside_group=False):
- selectors = []
- current_selector = None
- for type, string, start, _, _ in tokens:
- # ENCODING is only defined in python 3.x
- if type == getattr(tokenize, 'ENCODING', None):
- continue
- elif type in [tokenize.NAME, tokenize.NUMBER]:
- current_selector = FormatSelector(SINGLE, string, [])
- elif type == tokenize.OP:
- if string == ')':
- if not inside_group:
- # ')' will be handled by the parentheses group
- tokens.restore_last_token()
- break
- elif inside_merge and string in ['/', ',']:
- tokens.restore_last_token()
- break
- elif inside_choice and string == ',':
- tokens.restore_last_token()
- break
- elif string == ',':
- if not current_selector:
- raise syntax_error('"," must follow a format selector', start)
- selectors.append(current_selector)
- current_selector = None
- elif string == '/':
- if not current_selector:
- raise syntax_error('"/" must follow a format selector', start)
- first_choice = current_selector
- second_choice = _parse_format_selection(tokens, inside_choice=True)
- current_selector = FormatSelector(PICKFIRST, (first_choice, second_choice), [])
- elif string == '[':
- if not current_selector:
- current_selector = FormatSelector(SINGLE, 'best', [])
- format_filter = _parse_filter(tokens)
- current_selector.filters.append(format_filter)
- elif string == '(':
- if current_selector:
- raise syntax_error('Unexpected "("', start)
- group = _parse_format_selection(tokens, inside_group=True)
- current_selector = FormatSelector(GROUP, group, [])
- elif string == '+':
- if not current_selector:
- raise syntax_error('Unexpected "+"', start)
- selector_1 = current_selector
- selector_2 = _parse_format_selection(tokens, inside_merge=True)
- if not selector_2:
- raise syntax_error('Expected a selector', start)
- current_selector = FormatSelector(MERGE, (selector_1, selector_2), [])
- else:
- raise syntax_error(f'Operator not recognized: "{string}"', start)
- elif type == tokenize.ENDMARKER:
- break
- if current_selector:
- selectors.append(current_selector)
- return selectors
- def _merge(formats_pair):
- format_1, format_2 = formats_pair
- formats_info = []
- formats_info.extend(format_1.get('requested_formats', (format_1,)))
- formats_info.extend(format_2.get('requested_formats', (format_2,)))
- if not allow_multiple_streams['video'] or not allow_multiple_streams['audio']:
- get_no_more = {'video': False, 'audio': False}
- for (i, fmt_info) in enumerate(formats_info):
- if fmt_info.get('acodec') == fmt_info.get('vcodec') == 'none':
- formats_info.pop(i)
- continue
- for aud_vid in ['audio', 'video']:
- if not allow_multiple_streams[aud_vid] and fmt_info.get(aud_vid[0] + 'codec') != 'none':
- if get_no_more[aud_vid]:
- formats_info.pop(i)
- break
- get_no_more[aud_vid] = True
- if len(formats_info) == 1:
- return formats_info[0]
- video_fmts = [fmt_info for fmt_info in formats_info if fmt_info.get('vcodec') != 'none']
- audio_fmts = [fmt_info for fmt_info in formats_info if fmt_info.get('acodec') != 'none']
- the_only_video = video_fmts[0] if len(video_fmts) == 1 else None
- the_only_audio = audio_fmts[0] if len(audio_fmts) == 1 else None
- output_ext = get_compatible_ext(
- vcodecs=[f.get('vcodec') for f in video_fmts],
- acodecs=[f.get('acodec') for f in audio_fmts],
- vexts=[f['ext'] for f in video_fmts],
- aexts=[f['ext'] for f in audio_fmts],
- preferences=(try_call(lambda: self.params['merge_output_format'].split('/'))
- or self.params.get('prefer_free_formats') and ('webm', 'mkv')))
- filtered = lambda *keys: filter(None, (traverse_obj(fmt, *keys) for fmt in formats_info))
- new_dict = {
- 'requested_formats': formats_info,
- 'format': '+'.join(filtered('format')),
- 'format_id': '+'.join(filtered('format_id')),
- 'ext': output_ext,
- 'protocol': '+'.join(map(determine_protocol, formats_info)),
- 'language': '+'.join(orderedSet(filtered('language'))) or None,
- 'format_note': '+'.join(orderedSet(filtered('format_note'))) or None,
- 'filesize_approx': sum(filtered('filesize', 'filesize_approx')) or None,
- 'tbr': sum(filtered('tbr', 'vbr', 'abr')),
- }
- if the_only_video:
- new_dict.update({
- 'width': the_only_video.get('width'),
- 'height': the_only_video.get('height'),
- 'resolution': the_only_video.get('resolution') or self.format_resolution(the_only_video),
- 'fps': the_only_video.get('fps'),
- 'dynamic_range': the_only_video.get('dynamic_range'),
- 'vcodec': the_only_video.get('vcodec'),
- 'vbr': the_only_video.get('vbr'),
- 'stretched_ratio': the_only_video.get('stretched_ratio'),
- 'aspect_ratio': the_only_video.get('aspect_ratio'),
- })
- if the_only_audio:
- new_dict.update({
- 'acodec': the_only_audio.get('acodec'),
- 'abr': the_only_audio.get('abr'),
- 'asr': the_only_audio.get('asr'),
- 'audio_channels': the_only_audio.get('audio_channels')
- })
- return new_dict
- def _check_formats(formats):
- if not check_formats:
- yield from formats
- return
- yield from self._check_formats(formats)
- def _build_selector_function(selector):
- if isinstance(selector, list): # ,
- fs = [_build_selector_function(s) for s in selector]
- def selector_function(ctx):
- for f in fs:
- yield from f(ctx)
- return selector_function
- elif selector.type == GROUP: # ()
- selector_function = _build_selector_function(selector.selector)
- elif selector.type == PICKFIRST: # /
- fs = [_build_selector_function(s) for s in selector.selector]
- def selector_function(ctx):
- for f in fs:
- picked_formats = list(f(ctx))
- if picked_formats:
- return picked_formats
- return []
- elif selector.type == MERGE: # +
- selector_1, selector_2 = map(_build_selector_function, selector.selector)
- def selector_function(ctx):
- for pair in itertools.product(selector_1(ctx), selector_2(ctx)):
- yield _merge(pair)
- elif selector.type == SINGLE: # atom
- format_spec = selector.selector or 'best'
- # TODO: Add allvideo, allaudio etc by generalizing the code with best/worst selector
- if format_spec == 'all':
- def selector_function(ctx):
- yield from _check_formats(ctx['formats'][::-1])
- elif format_spec == 'mergeall':
- def selector_function(ctx):
- formats = list(_check_formats(
- f for f in ctx['formats'] if f.get('vcodec') != 'none' or f.get('acodec') != 'none'))
- if not formats:
- return
- merged_format = formats[-1]
- for f in formats[-2::-1]:
- merged_format = _merge((merged_format, f))
- yield merged_format
- else:
- format_fallback, seperate_fallback, format_reverse, format_idx = False, None, True, 1
- mobj = re.match(
- r'(?P<bw>best|worst|b|w)(?P<type>video|audio|v|a)?(?P<mod>\*)?(?:\.(?P<n>[1-9]\d*))?$',
- format_spec)
- if mobj is not None:
- format_idx = int_or_none(mobj.group('n'), default=1)
- format_reverse = mobj.group('bw')[0] == 'b'
- format_type = (mobj.group('type') or [None])[0]
- not_format_type = {'v': 'a', 'a': 'v'}.get(format_type)
- format_modified = mobj.group('mod') is not None
- format_fallback = not format_type and not format_modified # for b, w
- _filter_f = (
- (lambda f: f.get('%scodec' % format_type) != 'none')
- if format_type and format_modified # bv*, ba*, wv*, wa*
- else (lambda f: f.get('%scodec' % not_format_type) == 'none')
- if format_type # bv, ba, wv, wa
- else (lambda f: f.get('vcodec') != 'none' and f.get('acodec') != 'none')
- if not format_modified # b, w
- else lambda f: True) # b*, w*
- filter_f = lambda f: _filter_f(f) and (
- f.get('vcodec') != 'none' or f.get('acodec') != 'none')
- else:
- if format_spec in self._format_selection_exts['audio']:
- filter_f = lambda f: f.get('ext') == format_spec and f.get('acodec') != 'none'
- elif format_spec in self._format_selection_exts['video']:
- filter_f = lambda f: f.get('ext') == format_spec and f.get('acodec') != 'none' and f.get('vcodec') != 'none'
- seperate_fallback = lambda f: f.get('ext') == format_spec and f.get('vcodec') != 'none'
- elif format_spec in self._format_selection_exts['storyboards']:
- filter_f = lambda f: f.get('ext') == format_spec and f.get('acodec') == 'none' and f.get('vcodec') == 'none'
- else:
- filter_f = lambda f: f.get('format_id') == format_spec # id
- def selector_function(ctx):
- formats = list(ctx['formats'])
- matches = list(filter(filter_f, formats)) if filter_f is not None else formats
- if not matches:
- if format_fallback and ctx['incomplete_formats']:
- # for extractors with incomplete formats (audio only (soundcloud)
- # or video only (imgur)) best/worst will fallback to
- # best/worst {video,audio}-only format
- matches = formats
- elif seperate_fallback and not ctx['has_merged_format']:
- # for compatibility with youtube-dl when there is no pre-merged format
- matches = list(filter(seperate_fallback, formats))
- matches = LazyList(_check_formats(matches[::-1 if format_reverse else 1]))
- try:
- yield matches[format_idx - 1]
- except LazyList.IndexError:
- return
- filters = [self._build_format_filter(f) for f in selector.filters]
- def final_selector(ctx):
- ctx_copy = dict(ctx)
- for _filter in filters:
- ctx_copy['formats'] = list(filter(_filter, ctx_copy['formats']))
- return selector_function(ctx_copy)
- return final_selector
- stream = io.BytesIO(format_spec.encode())
- try:
- tokens = list(_remove_unused_ops(tokenize.tokenize(stream.readline)))
- except tokenize.TokenError:
- raise syntax_error('Missing closing/opening brackets or parenthesis', (0, len(format_spec)))
- class TokenIterator:
- def __init__(self, tokens):
- self.tokens = tokens
- self.counter = 0
- def __iter__(self):
- return self
- def __next__(self):
- if self.counter >= len(self.tokens):
- raise StopIteration()
- value = self.tokens[self.counter]
- self.counter += 1
- return value
- next = __next__
- def restore_last_token(self):
- self.counter -= 1
- parsed_selector = _parse_format_selection(iter(TokenIterator(tokens)))
- return _build_selector_function(parsed_selector)
- def _calc_headers(self, info_dict):
- res = merge_headers(self.params['http_headers'], info_dict.get('http_headers') or {})
- cookies = self._calc_cookies(info_dict['url'])
- if cookies:
- res['Cookie'] = cookies
- if 'X-Forwarded-For' not in res:
- x_forwarded_for_ip = info_dict.get('__x_forwarded_for_ip')
- if x_forwarded_for_ip:
- res['X-Forwarded-For'] = x_forwarded_for_ip
- return res
- def _calc_cookies(self, url):
- pr = sanitized_Request(url)
- self.cookiejar.add_cookie_header(pr)
- return pr.get_header('Cookie')
- def _sort_thumbnails(self, thumbnails):
- thumbnails.sort(key=lambda t: (
- t.get('preference') if t.get('preference') is not None else -1,
- t.get('width') if t.get('width') is not None else -1,
- t.get('height') if t.get('height') is not None else -1,
- t.get('id') if t.get('id') is not None else '',
- t.get('url')))
- def _sanitize_thumbnails(self, info_dict):
- thumbnails = info_dict.get('thumbnails')
- if thumbnails is None:
- thumbnail = info_dict.get('thumbnail')
- if thumbnail:
- info_dict['thumbnails'] = thumbnails = [{'url': thumbnail}]
- if not thumbnails:
- return
- def check_thumbnails(thumbnails):
- for t in thumbnails:
- self.to_screen(f'[info] Testing thumbnail {t["id"]}')
- try:
- self.urlopen(HEADRequest(t['url']))
- except network_exceptions as err:
- self.to_screen(f'[info] Unable to connect to thumbnail {t["id"]} URL {t["url"]!r} - {err}. Skipping...')
- continue
- yield t
- self._sort_thumbnails(thumbnails)
- for i, t in enumerate(thumbnails):
- if t.get('id') is None:
- t['id'] = '%d' % i
- if t.get('width') and t.get('height'):
- t['resolution'] = '%dx%d' % (t['width'], t['height'])
- t['url'] = sanitize_url(t['url'])
- if self.params.get('check_formats') is True:
- info_dict['thumbnails'] = LazyList(check_thumbnails(thumbnails[::-1]), reverse=True)
- else:
- info_dict['thumbnails'] = thumbnails
- def _fill_common_fields(self, info_dict, final=True):
- # TODO: move sanitization here
- if final:
- title = info_dict.get('title', NO_DEFAULT)
- if title is NO_DEFAULT:
- raise ExtractorError('Missing "title" field in extractor result',
- video_id=info_dict['id'], ie=info_dict['extractor'])
- info_dict['fulltitle'] = title
- if not title:
- if title == '':
- self.write_debug('Extractor gave empty title. Creating a generic title')
- else:
- self.report_warning('Extractor failed to obtain "title". Creating a generic title instead')
- info_dict['title'] = f'{info_dict["extractor"].replace(":", "-")} video #{info_dict["id"]}'
- if info_dict.get('duration') is not None:
- info_dict['duration_string'] = formatSeconds(info_dict['duration'])
- for ts_key, date_key in (
- ('timestamp', 'upload_date'),
- ('release_timestamp', 'release_date'),
- ('modified_timestamp', 'modified_date'),
- ):
- if info_dict.get(date_key) is None and info_dict.get(ts_key) is not None:
- # Working around out-of-range timestamp values (e.g. negative ones on Windows,
- # see http://bugs.python.org/issue1646728)
- with contextlib.suppress(ValueError, OverflowError, OSError):
- upload_date = datetime.datetime.utcfromtimestamp(info_dict[ts_key])
- info_dict[date_key] = upload_date.strftime('%Y%m%d')
- live_keys = ('is_live', 'was_live')
- live_status = info_dict.get('live_status')
- if live_status is None:
- for key in live_keys:
- if info_dict.get(key) is False:
- continue
- if info_dict.get(key):
- live_status = key
- break
- if all(info_dict.get(key) is False for key in live_keys):
- live_status = 'not_live'
- if live_status:
- info_dict['live_status'] = live_status
- for key in live_keys:
- if info_dict.get(key) is None:
- info_dict[key] = (live_status == key)
- if live_status == 'post_live':
- info_dict['was_live'] = True
- # Auto generate title fields corresponding to the *_number fields when missing
- # in order to always have clean titles. This is very common for TV series.
- for field in ('chapter', 'season', 'episode'):
- if final and info_dict.get('%s_number' % field) is not None and not info_dict.get(field):
- info_dict[field] = '%s %d' % (field.capitalize(), info_dict['%s_number' % field])
- def _raise_pending_errors(self, info):
- err = info.pop('__pending_error', None)
- if err:
- self.report_error(err, tb=False)
- def sort_formats(self, info_dict):
- formats = self._get_formats(info_dict)
- if not formats:
- return
- # Backward compatibility with InfoExtractor._sort_formats
- field_preference = formats[0].pop('__sort_fields', None)
- if field_preference:
- info_dict['_format_sort_fields'] = field_preference
- formats.sort(key=FormatSorter(
- self, info_dict.get('_format_sort_fields', [])).calculate_preference)
- def process_video_result(self, info_dict, download=True):
- assert info_dict.get('_type', 'video') == 'video'
- self._num_videos += 1
- if 'id' not in info_dict:
- raise ExtractorError('Missing "id" field in extractor result', ie=info_dict['extractor'])
- elif not info_dict.get('id'):
- raise ExtractorError('Extractor failed to obtain "id"', ie=info_dict['extractor'])
- def report_force_conversion(field, field_not, conversion):
- self.report_warning(
- '"%s" field is not %s - forcing %s conversion, there is an error in extractor'
- % (field, field_not, conversion))
- def sanitize_string_field(info, string_field):
- field = info.get(string_field)
- if field is None or isinstance(field, str):
- return
- report_force_conversion(string_field, 'a string', 'string')
- info[string_field] = str(field)
- def sanitize_numeric_fields(info):
- for numeric_field in self._NUMERIC_FIELDS:
- field = info.get(numeric_field)
- if field is None or isinstance(field, (int, float)):
- continue
- report_force_conversion(numeric_field, 'numeric', 'int')
- info[numeric_field] = int_or_none(field)
- sanitize_string_field(info_dict, 'id')
- sanitize_numeric_fields(info_dict)
- if info_dict.get('section_end') and info_dict.get('section_start') is not None:
- info_dict['duration'] = round(info_dict['section_end'] - info_dict['section_start'], 3)
- if (info_dict.get('duration') or 0) <= 0 and info_dict.pop('duration', None):
- self.report_warning('"duration" field is negative, there is an error in extractor')
- chapters = info_dict.get('chapters') or []
- if chapters and chapters[0].get('start_time'):
- chapters.insert(0, {'start_time': 0})
- dummy_chapter = {'end_time': 0, 'start_time': info_dict.get('duration')}
- for idx, (prev, current, next_) in enumerate(zip(
- (dummy_chapter, *chapters), chapters, (*chapters[1:], dummy_chapter)), 1):
- if current.get('start_time') is None:
- current['start_time'] = prev.get('end_time')
- if not current.get('end_time'):
- current['end_time'] = next_.get('start_time')
- if not current.get('title'):
- current['title'] = f'<Untitled Chapter {idx}>'
- if 'playlist' not in info_dict:
- # It isn't part of a playlist
- info_dict['playlist'] = None
- info_dict['playlist_index'] = None
- self._sanitize_thumbnails(info_dict)
- thumbnail = info_dict.get('thumbnail')
- thumbnails = info_dict.get('thumbnails')
- if thumbnail:
- info_dict['thumbnail'] = sanitize_url(thumbnail)
- elif thumbnails:
- info_dict['thumbnail'] = thumbnails[-1]['url']
- if info_dict.get('display_id') is None and 'id' in info_dict:
- info_dict['display_id'] = info_dict['id']
- self._fill_common_fields(info_dict)
- for cc_kind in ('subtitles', 'automatic_captions'):
- cc = info_dict.get(cc_kind)
- if cc:
- for _, subtitle in cc.items():
- for subtitle_format in subtitle:
- if subtitle_format.get('url'):
- subtitle_format['url'] = sanitize_url(subtitle_format['url'])
- if subtitle_format.get('ext') is None:
- subtitle_format['ext'] = determine_ext(subtitle_format['url']).lower()
- automatic_captions = info_dict.get('automatic_captions')
- subtitles = info_dict.get('subtitles')
- info_dict['requested_subtitles'] = self.process_subtitles(
- info_dict['id'], subtitles, automatic_captions)
- self.sort_formats(info_dict)
- formats = self._get_formats(info_dict)
- # or None ensures --clean-infojson removes it
- info_dict['_has_drm'] = any(f.get('has_drm') for f in formats) or None
- if not self.params.get('allow_unplayable_formats'):
- formats = [f for f in formats if not f.get('has_drm')]
- if formats and all(f.get('acodec') == f.get('vcodec') == 'none' for f in formats):
- self.report_warning(
- f'{"This video is DRM protected and " if info_dict["_has_drm"] else ""}'
- 'only images are available for download. Use --list-formats to see them'.capitalize())
- get_from_start = not info_dict.get('is_live') or bool(self.params.get('live_from_start'))
- if not get_from_start:
- info_dict['title'] += ' ' + datetime.datetime.now().strftime('%Y-%m-%d %H:%M')
- if info_dict.get('is_live') and formats:
- formats = [f for f in formats if bool(f.get('is_from_start')) == get_from_start]
- if get_from_start and not formats:
- self.raise_no_formats(info_dict, msg=(
- '--live-from-start is passed, but there are no formats that can be downloaded from the start. '
- 'If you want to download from the current time, use --no-live-from-start'))
- def is_wellformed(f):
- url = f.get('url')
- if not url:
- self.report_warning(
- '"url" field is missing or empty - skipping format, '
- 'there is an error in extractor')
- return False
- if isinstance(url, bytes):
- sanitize_string_field(f, 'url')
- return True
- # Filter out malformed formats for better extraction robustness
- formats = list(filter(is_wellformed, formats or []))
- if not formats:
- self.raise_no_formats(info_dict)
- formats_dict = {}
- # We check that all the formats have the format and format_id fields
- for i, format in enumerate(formats):
- sanitize_string_field(format, 'format_id')
- sanitize_numeric_fields(format)
- format['url'] = sanitize_url(format['url'])
- if not format.get('format_id'):
- format['format_id'] = str(i)
- else:
- # Sanitize format_id from characters used in format selector expression
- format['format_id'] = re.sub(r'[\s,/+\[\]()]', '_', format['format_id'])
- format_id = format['format_id']
- if format_id not in formats_dict:
- formats_dict[format_id] = []
- formats_dict[format_id].append(format)
- # Make sure all formats have unique format_id
- common_exts = set(itertools.chain(*self._format_selection_exts.values()))
- for format_id, ambiguous_formats in formats_dict.items():
- ambigious_id = len(ambiguous_formats) > 1
- for i, format in enumerate(ambiguous_formats):
- if ambigious_id:
- format['format_id'] = '%s-%d' % (format_id, i)
- if format.get('ext') is None:
- format['ext'] = determine_ext(format['url']).lower()
- # Ensure there is no conflict between id and ext in format selection
- # See https://github.com/hypervideo/hypervideo/issues/1282
- if format['format_id'] != format['ext'] and format['format_id'] in common_exts:
- format['format_id'] = 'f%s' % format['format_id']
- for i, format in enumerate(formats):
- if format.get('format') is None:
- format['format'] = '{id} - {res}{note}'.format(
- id=format['format_id'],
- res=self.format_resolution(format),
- note=format_field(format, 'format_note', ' (%s)'),
- )
- if format.get('protocol') is None:
- format['protocol'] = determine_protocol(format)
- if format.get('resolution') is None:
- format['resolution'] = self.format_resolution(format, default=None)
- if format.get('dynamic_range') is None and format.get('vcodec') != 'none':
- format['dynamic_range'] = 'SDR'
- if format.get('aspect_ratio') is None:
- format['aspect_ratio'] = try_call(lambda: round(format['width'] / format['height'], 2))
- if (info_dict.get('duration') and format.get('tbr')
- and not format.get('filesize') and not format.get('filesize_approx')):
- format['filesize_approx'] = int(info_dict['duration'] * format['tbr'] * (1024 / 8))
- # Add HTTP headers, so that external programs can use them from the
- # json output
- full_format_info = info_dict.copy()
- full_format_info.update(format)
- format['http_headers'] = self._calc_headers(full_format_info)
- # Remove private housekeeping stuff
- if '__x_forwarded_for_ip' in info_dict:
- del info_dict['__x_forwarded_for_ip']
- if self.params.get('check_formats') is True:
- formats = LazyList(self._check_formats(formats[::-1]), reverse=True)
- if not formats or formats[0] is not info_dict:
- # only set the 'formats' fields if the original info_dict list them
- # otherwise we end up with a circular reference, the first (and unique)
- # element in the 'formats' field in info_dict is info_dict itself,
- # which can't be exported to json
- info_dict['formats'] = formats
- info_dict, _ = self.pre_process(info_dict)
- if self._match_entry(info_dict, incomplete=self._format_fields) is not None:
- return info_dict
- self.post_extract(info_dict)
- info_dict, _ = self.pre_process(info_dict, 'after_filter')
- # The pre-processors may have modified the formats
- formats = self._get_formats(info_dict)
- list_only = self.params.get('simulate') == 'list_only'
- interactive_format_selection = not list_only and self.format_selector == '-'
- if self.params.get('list_thumbnails'):
- self.list_thumbnails(info_dict)
- if self.params.get('listsubtitles'):
- if 'automatic_captions' in info_dict:
- self.list_subtitles(
- info_dict['id'], automatic_captions, 'automatic captions')
- self.list_subtitles(info_dict['id'], subtitles, 'subtitles')
- if self.params.get('listformats') or interactive_format_selection:
- self.list_formats(info_dict)
- if list_only:
- # Without this printing, -F --print-json will not work
- self.__forced_printings(info_dict, self.prepare_filename(info_dict), incomplete=True)
- return info_dict
- format_selector = self.format_selector
- if format_selector is None:
- req_format = self._default_format_spec(info_dict, download=download)
- self.write_debug('Default format spec: %s' % req_format)
- format_selector = self.build_format_selector(req_format)
- while True:
- if interactive_format_selection:
- req_format = input(
- self._format_screen('\nEnter format selector: ', self.Styles.EMPHASIS))
- try:
- format_selector = self.build_format_selector(req_format)
- except SyntaxError as err:
- self.report_error(err, tb=False, is_error=False)
- continue
- formats_to_download = list(format_selector({
- 'formats': formats,
- 'has_merged_format': any('none' not in (f.get('acodec'), f.get('vcodec')) for f in formats),
- 'incomplete_formats': (
- # All formats are video-only or
- all(f.get('vcodec') != 'none' and f.get('acodec') == 'none' for f in formats)
- # all formats are audio-only
- or all(f.get('vcodec') == 'none' and f.get('acodec') != 'none' for f in formats)),
- }))
- if interactive_format_selection and not formats_to_download:
- self.report_error('Requested format is not available', tb=False, is_error=False)
- continue
- break
- if not formats_to_download:
- if not self.params.get('ignore_no_formats_error'):
- raise ExtractorError(
- 'Requested format is not available. Use --list-formats for a list of available formats',
- expected=True, video_id=info_dict['id'], ie=info_dict['extractor'])
- self.report_warning('Requested format is not available')
- # Process what we can, even without any available formats.
- formats_to_download = [{}]
- requested_ranges = tuple(self.params.get('download_ranges', lambda *_: [{}])(info_dict, self))
- best_format, downloaded_formats = formats_to_download[-1], []
- if download:
- if best_format and requested_ranges:
- def to_screen(*msg):
- self.to_screen(f'[info] {info_dict["id"]}: {" ".join(", ".join(variadic(m)) for m in msg)}')
- to_screen(f'Downloading {len(formats_to_download)} format(s):',
- (f['format_id'] for f in formats_to_download))
- if requested_ranges != ({}, ):
- to_screen(f'Downloading {len(requested_ranges)} time ranges:',
- (f'{c["start_time"]:.1f}-{c["end_time"]:.1f}' for c in requested_ranges))
- max_downloads_reached = False
- for fmt, chapter in itertools.product(formats_to_download, requested_ranges):
- new_info = self._copy_infodict(info_dict)
- new_info.update(fmt)
- offset, duration = info_dict.get('section_start') or 0, info_dict.get('duration') or float('inf')
- end_time = offset + min(chapter.get('end_time', duration), duration)
- if chapter or offset:
- new_info.update({
- 'section_start': offset + chapter.get('start_time', 0),
- # duration may not be accurate. So allow deviations <1sec
- 'section_end': end_time if end_time <= offset + duration + 1 else None,
- 'section_title': chapter.get('title'),
- 'section_number': chapter.get('index'),
- })
- downloaded_formats.append(new_info)
- try:
- self.process_info(new_info)
- except MaxDownloadsReached:
- max_downloads_reached = True
- self._raise_pending_errors(new_info)
- # Remove copied info
- for key, val in tuple(new_info.items()):
- if info_dict.get(key) == val:
- new_info.pop(key)
- if max_downloads_reached:
- break
- write_archive = {f.get('__write_download_archive', False) for f in downloaded_formats}
- assert write_archive.issubset({True, False, 'ignore'})
- if True in write_archive and False not in write_archive:
- self.record_download_archive(info_dict)
- info_dict['requested_downloads'] = downloaded_formats
- info_dict = self.run_all_pps('after_video', info_dict)
- if max_downloads_reached:
- raise MaxDownloadsReached()
- # We update the info dict with the selected best quality format (backwards compatibility)
- info_dict.update(best_format)
- return info_dict
- def process_subtitles(self, video_id, normal_subtitles, automatic_captions):
- """Select the requested subtitles and their format"""
- available_subs, normal_sub_langs = {}, []
- if normal_subtitles and self.params.get('writesubtitles'):
- available_subs.update(normal_subtitles)
- normal_sub_langs = tuple(normal_subtitles.keys())
- if automatic_captions and self.params.get('writeautomaticsub'):
- for lang, cap_info in automatic_captions.items():
- if lang not in available_subs:
- available_subs[lang] = cap_info
- if not available_subs or (
- not self.params.get('writesubtitles')
- and not self.params.get('writeautomaticsub')):
- return None
- all_sub_langs = tuple(available_subs.keys())
- if self.params.get('allsubtitles', False):
- requested_langs = all_sub_langs
- elif self.params.get('subtitleslangs', False):
- try:
- requested_langs = orderedSet_from_options(
- self.params.get('subtitleslangs'), {'all': all_sub_langs}, use_regex=True)
- except re.error as e:
- raise ValueError(f'Wrong regex for subtitlelangs: {e.pattern}')
- elif normal_sub_langs:
- requested_langs = ['en'] if 'en' in normal_sub_langs else normal_sub_langs[:1]
- else:
- requested_langs = ['en'] if 'en' in all_sub_langs else all_sub_langs[:1]
- if requested_langs:
- self.to_screen(f'[info] {video_id}: Downloading subtitles: {", ".join(requested_langs)}')
- formats_query = self.params.get('subtitlesformat', 'best')
- formats_preference = formats_query.split('/') if formats_query else []
- subs = {}
- for lang in requested_langs:
- formats = available_subs.get(lang)
- if formats is None:
- self.report_warning(f'{lang} subtitles not available for {video_id}')
- continue
- for ext in formats_preference:
- if ext == 'best':
- f = formats[-1]
- break
- matches = list(filter(lambda f: f['ext'] == ext, formats))
- if matches:
- f = matches[-1]
- break
- else:
- f = formats[-1]
- self.report_warning(
- 'No subtitle format found matching "%s" for language %s, '
- 'using %s' % (formats_query, lang, f['ext']))
- subs[lang] = f
- return subs
- def _forceprint(self, key, info_dict):
- if info_dict is None:
- return
- info_copy = info_dict.copy()
- info_copy['formats_table'] = self.render_formats_table(info_dict)
- info_copy['thumbnails_table'] = self.render_thumbnails_table(info_dict)
- info_copy['subtitles_table'] = self.render_subtitles_table(info_dict.get('id'), info_dict.get('subtitles'))
- info_copy['automatic_captions_table'] = self.render_subtitles_table(info_dict.get('id'), info_dict.get('automatic_captions'))
- def format_tmpl(tmpl):
- mobj = re.fullmatch(r'([\w.:,]|-\d|(?P<dict>{([\w.:,]|-\d)+}))+=?', tmpl)
- if not mobj:
- return tmpl
- fmt = '%({})s'
- if tmpl.startswith('{'):
- tmpl = f'.{tmpl}'
- if tmpl.endswith('='):
- tmpl, fmt = tmpl[:-1], '{0} = %({0})#j'
- return '\n'.join(map(fmt.format, [tmpl] if mobj.group('dict') else tmpl.split(',')))
- for tmpl in self.params['forceprint'].get(key, []):
- self.to_stdout(self.evaluate_outtmpl(format_tmpl(tmpl), info_copy))
- for tmpl, file_tmpl in self.params['print_to_file'].get(key, []):
- filename = self.prepare_filename(info_dict, outtmpl=file_tmpl)
- tmpl = format_tmpl(tmpl)
- self.to_screen(f'[info] Writing {tmpl!r} to: {filename}')
- if self._ensure_dir_exists(filename):
- with open(filename, 'a', encoding='utf-8') as f:
- f.write(self.evaluate_outtmpl(tmpl, info_copy) + '\n')
- def __forced_printings(self, info_dict, filename, incomplete):
- def print_mandatory(field, actual_field=None):
- if actual_field is None:
- actual_field = field
- if (self.params.get('force%s' % field, False)
- and (not incomplete or info_dict.get(actual_field) is not None)):
- self.to_stdout(info_dict[actual_field])
- def print_optional(field):
- if (self.params.get('force%s' % field, False)
- and info_dict.get(field) is not None):
- self.to_stdout(info_dict[field])
- info_dict = info_dict.copy()
- if filename is not None:
- info_dict['filename'] = filename
- if info_dict.get('requested_formats') is not None:
- # For RTMP URLs, also include the playpath
- info_dict['urls'] = '\n'.join(f['url'] + f.get('play_path', '') for f in info_dict['requested_formats'])
- elif info_dict.get('url'):
- info_dict['urls'] = info_dict['url'] + info_dict.get('play_path', '')
- if (self.params.get('forcejson')
- or self.params['forceprint'].get('video')
- or self.params['print_to_file'].get('video')):
- self.post_extract(info_dict)
- self._forceprint('video', info_dict)
- print_mandatory('title')
- print_mandatory('id')
- print_mandatory('url', 'urls')
- print_optional('thumbnail')
- print_optional('description')
- print_optional('filename')
- if self.params.get('forceduration') and info_dict.get('duration') is not None:
- self.to_stdout(formatSeconds(info_dict['duration']))
- print_mandatory('format')
- if self.params.get('forcejson'):
- self.to_stdout(json.dumps(self.sanitize_info(info_dict)))
- def dl(self, name, info, subtitle=False, test=False):
- if not info.get('url'):
- self.raise_no_formats(info, True)
- if test:
- verbose = self.params.get('verbose')
- params = {
- 'test': True,
- 'quiet': self.params.get('quiet') or not verbose,
- 'verbose': verbose,
- 'noprogress': not verbose,
- 'nopart': True,
- 'skip_unavailable_fragments': False,
- 'keep_fragments': False,
- 'overwrites': True,
- '_no_ytdl_file': True,
- }
- else:
- params = self.params
- fd = get_suitable_downloader(info, params, to_stdout=(name == '-'))(self, params)
- if not test:
- for ph in self._progress_hooks:
- fd.add_progress_hook(ph)
- urls = '", "'.join(
- (f['url'].split(',')[0] + ',<data>' if f['url'].startswith('data:') else f['url'])
- for f in info.get('requested_formats', []) or [info])
- self.write_debug(f'Invoking {fd.FD_NAME} downloader on "{urls}"')
- # Note: Ideally info should be a deep-copied so that hooks cannot modify it.
- # But it may contain objects that are not deep-copyable
- new_info = self._copy_infodict(info)
- if new_info.get('http_headers') is None:
- new_info['http_headers'] = self._calc_headers(new_info)
- return fd.download(name, new_info, subtitle)
- def existing_file(self, filepaths, *, default_overwrite=True):
- existing_files = list(filter(os.path.exists, orderedSet(filepaths)))
- if existing_files and not self.params.get('overwrites', default_overwrite):
- return existing_files[0]
- for file in existing_files:
- self.report_file_delete(file)
- os.remove(file)
- return None
- def process_info(self, info_dict):
- """Process a single resolved IE result. (Modifies it in-place)"""
- assert info_dict.get('_type', 'video') == 'video'
- original_infodict = info_dict
- if 'format' not in info_dict and 'ext' in info_dict:
- info_dict['format'] = info_dict['ext']
- if self._match_entry(info_dict) is not None:
- info_dict['__write_download_archive'] = 'ignore'
- return
- # Does nothing under normal operation - for backward compatibility of process_info
- self.post_extract(info_dict)
- self._num_downloads += 1
- # info_dict['_filename'] needs to be set for backward compatibility
- info_dict['_filename'] = full_filename = self.prepare_filename(info_dict, warn=True)
- temp_filename = self.prepare_filename(info_dict, 'temp')
- files_to_move = {}
- # Forced printings
- self.__forced_printings(info_dict, full_filename, incomplete=('format' not in info_dict))
- def check_max_downloads():
- if self._num_downloads >= float(self.params.get('max_downloads') or 'inf'):
- raise MaxDownloadsReached()
- if self.params.get('simulate'):
- info_dict['__write_download_archive'] = self.params.get('force_write_download_archive')
- check_max_downloads()
- return
- if full_filename is None:
- return
- if not self._ensure_dir_exists(encodeFilename(full_filename)):
- return
- if not self._ensure_dir_exists(encodeFilename(temp_filename)):
- return
- if self._write_description('video', info_dict,
- self.prepare_filename(info_dict, 'description')) is None:
- return
- sub_files = self._write_subtitles(info_dict, temp_filename)
- if sub_files is None:
- return
- files_to_move.update(dict(sub_files))
- thumb_files = self._write_thumbnails(
- 'video', info_dict, temp_filename, self.prepare_filename(info_dict, 'thumbnail'))
- if thumb_files is None:
- return
- files_to_move.update(dict(thumb_files))
- infofn = self.prepare_filename(info_dict, 'infojson')
- _infojson_written = self._write_info_json('video', info_dict, infofn)
- if _infojson_written:
- info_dict['infojson_filename'] = infofn
- # For backward compatibility, even though it was a private field
- info_dict['__infojson_filename'] = infofn
- elif _infojson_written is None:
- return
- # Note: Annotations are deprecated
- annofn = None
- if self.params.get('writeannotations', False):
- annofn = self.prepare_filename(info_dict, 'annotation')
- if annofn:
- if not self._ensure_dir_exists(encodeFilename(annofn)):
- return
- if not self.params.get('overwrites', True) and os.path.exists(encodeFilename(annofn)):
- self.to_screen('[info] Video annotations are already present')
- elif not info_dict.get('annotations'):
- self.report_warning('There are no annotations to write.')
- else:
- try:
- self.to_screen('[info] Writing video annotations to: ' + annofn)
- with open(encodeFilename(annofn), 'w', encoding='utf-8') as annofile:
- annofile.write(info_dict['annotations'])
- except (KeyError, TypeError):
- self.report_warning('There are no annotations to write.')
- except OSError:
- self.report_error('Cannot write annotations file: ' + annofn)
- return
- # Write internet shortcut files
- def _write_link_file(link_type):
- url = try_get(info_dict['webpage_url'], iri_to_uri)
- if not url:
- self.report_warning(
- f'Cannot write internet shortcut file because the actual URL of "{info_dict["webpage_url"]}" is unknown')
- return True
- linkfn = replace_extension(self.prepare_filename(info_dict, 'link'), link_type, info_dict.get('ext'))
- if not self._ensure_dir_exists(encodeFilename(linkfn)):
- return False
- if self.params.get('overwrites', True) and os.path.exists(encodeFilename(linkfn)):
- self.to_screen(f'[info] Internet shortcut (.{link_type}) is already present')
- return True
- try:
- self.to_screen(f'[info] Writing internet shortcut (.{link_type}) to: {linkfn}')
- with open(encodeFilename(to_high_limit_path(linkfn)), 'w', encoding='utf-8',
- newline='\r\n' if link_type == 'url' else '\n') as linkfile:
- template_vars = {'url': url}
- if link_type == 'desktop':
- template_vars['filename'] = linkfn[:-(len(link_type) + 1)]
- linkfile.write(LINK_TEMPLATES[link_type] % template_vars)
- except OSError:
- self.report_error(f'Cannot write internet shortcut {linkfn}')
- return False
- return True
- write_links = {
- 'url': self.params.get('writeurllink'),
- 'webloc': self.params.get('writewebloclink'),
- 'desktop': self.params.get('writedesktoplink'),
- }
- if self.params.get('writelink'):
- link_type = ('webloc' if sys.platform == 'darwin'
- else 'desktop' if sys.platform.startswith('linux')
- else 'url')
- write_links[link_type] = True
- if any(should_write and not _write_link_file(link_type)
- for link_type, should_write in write_links.items()):
- return
- def replace_info_dict(new_info):
- nonlocal info_dict
- if new_info == info_dict:
- return
- info_dict.clear()
- info_dict.update(new_info)
- new_info, files_to_move = self.pre_process(info_dict, 'before_dl', files_to_move)
- replace_info_dict(new_info)
- if self.params.get('skip_download'):
- info_dict['filepath'] = temp_filename
- info_dict['__finaldir'] = os.path.dirname(os.path.abspath(encodeFilename(full_filename)))
- info_dict['__files_to_move'] = files_to_move
- replace_info_dict(self.run_pp(MoveFilesAfterDownloadPP(self, False), info_dict))
- info_dict['__write_download_archive'] = self.params.get('force_write_download_archive')
- else:
- # Download
- info_dict.setdefault('__postprocessors', [])
- try:
- def existing_video_file(*filepaths):
- ext = info_dict.get('ext')
- converted = lambda file: replace_extension(file, self.params.get('final_ext') or ext, ext)
- file = self.existing_file(itertools.chain(*zip(map(converted, filepaths), filepaths)),
- default_overwrite=False)
- if file:
- info_dict['ext'] = os.path.splitext(file)[1][1:]
- return file
- fd, success = None, True
- if info_dict.get('protocol') or info_dict.get('url'):
- fd = get_suitable_downloader(info_dict, self.params, to_stdout=temp_filename == '-')
- if fd is not FFmpegFD and (
- info_dict.get('section_start') or info_dict.get('section_end')):
- msg = ('This format cannot be partially downloaded' if FFmpegFD.available()
- else 'You have requested downloading the video partially, but ffmpeg is not installed')
- self.report_error(f'{msg}. Aborting')
- return
- if info_dict.get('requested_formats') is not None:
- requested_formats = info_dict['requested_formats']
- old_ext = info_dict['ext']
- if self.params.get('merge_output_format') is None:
- if (info_dict['ext'] == 'webm'
- and info_dict.get('thumbnails')
- # check with type instead of pp_key, __name__, or isinstance
- # since we dont want any custom PPs to trigger this
- and any(type(pp) == EmbedThumbnailPP for pp in self._pps['post_process'])): # noqa: E721
- info_dict['ext'] = 'mkv'
- self.report_warning(
- 'webm doesn\'t support embedding a thumbnail, mkv will be used')
- new_ext = info_dict['ext']
- def correct_ext(filename, ext=new_ext):
- if filename == '-':
- return filename
- filename_real_ext = os.path.splitext(filename)[1][1:]
- filename_wo_ext = (
- os.path.splitext(filename)[0]
- if filename_real_ext in (old_ext, new_ext)
- else filename)
- return f'{filename_wo_ext}.{ext}'
- # Ensure filename always has a correct extension for successful merge
- full_filename = correct_ext(full_filename)
- temp_filename = correct_ext(temp_filename)
- dl_filename = existing_video_file(full_filename, temp_filename)
- info_dict['__real_download'] = False
- merger = FFmpegMergerPP(self)
- downloaded = []
- if dl_filename is not None:
- self.report_file_already_downloaded(dl_filename)
- elif fd:
- for f in requested_formats if fd != FFmpegFD else []:
- f['filepath'] = fname = prepend_extension(
- correct_ext(temp_filename, info_dict['ext']),
- 'f%s' % f['format_id'], info_dict['ext'])
- downloaded.append(fname)
- info_dict['url'] = '\n'.join(f['url'] for f in requested_formats)
- success, real_download = self.dl(temp_filename, info_dict)
- info_dict['__real_download'] = real_download
- else:
- if self.params.get('allow_unplayable_formats'):
- self.report_warning(
- 'You have requested merging of multiple formats '
- 'while also allowing unplayable formats to be downloaded. '
- 'The formats won\'t be merged to prevent data corruption.')
- elif not merger.available:
- msg = 'You have requested merging of multiple formats but ffmpeg is not installed'
- if not self.params.get('ignoreerrors'):
- self.report_error(f'{msg}. Aborting due to --abort-on-error')
- return
- self.report_warning(f'{msg}. The formats won\'t be merged')
- if temp_filename == '-':
- reason = ('using a downloader other than ffmpeg' if FFmpegFD.can_merge_formats(info_dict, self.params)
- else 'but the formats are incompatible for simultaneous download' if merger.available
- else 'but ffmpeg is not installed')
- self.report_warning(
- f'You have requested downloading multiple formats to stdout {reason}. '
- 'The formats will be streamed one after the other')
- fname = temp_filename
- for f in requested_formats:
- new_info = dict(info_dict)
- del new_info['requested_formats']
- new_info.update(f)
- if temp_filename != '-':
- fname = prepend_extension(
- correct_ext(temp_filename, new_info['ext']),
- 'f%s' % f['format_id'], new_info['ext'])
- if not self._ensure_dir_exists(fname):
- return
- f['filepath'] = fname
- downloaded.append(fname)
- partial_success, real_download = self.dl(fname, new_info)
- info_dict['__real_download'] = info_dict['__real_download'] or real_download
- success = success and partial_success
- if downloaded and merger.available and not self.params.get('allow_unplayable_formats'):
- info_dict['__postprocessors'].append(merger)
- info_dict['__files_to_merge'] = downloaded
- # Even if there were no downloads, it is being merged only now
- info_dict['__real_download'] = True
- else:
- for file in downloaded:
- files_to_move[file] = None
- else:
- # Just a single file
- dl_filename = existing_video_file(full_filename, temp_filename)
- if dl_filename is None or dl_filename == temp_filename:
- # dl_filename == temp_filename could mean that the file was partially downloaded with --no-part.
- # So we should try to resume the download
- success, real_download = self.dl(temp_filename, info_dict)
- info_dict['__real_download'] = real_download
- else:
- self.report_file_already_downloaded(dl_filename)
- dl_filename = dl_filename or temp_filename
- info_dict['__finaldir'] = os.path.dirname(os.path.abspath(encodeFilename(full_filename)))
- except network_exceptions as err:
- self.report_error('unable to download video data: %s' % error_to_compat_str(err))
- return
- except OSError as err:
- raise UnavailableVideoError(err)
- except (ContentTooShortError, ) as err:
- self.report_error(f'content too short (expected {err.expected} bytes and served {err.downloaded})')
- return
- self._raise_pending_errors(info_dict)
- if success and full_filename != '-':
- def fixup():
- do_fixup = True
- fixup_policy = self.params.get('fixup')
- vid = info_dict['id']
- if fixup_policy in ('ignore', 'never'):
- return
- elif fixup_policy == 'warn':
- do_fixup = 'warn'
- elif fixup_policy != 'force':
- assert fixup_policy in ('detect_or_warn', None)
- if not info_dict.get('__real_download'):
- do_fixup = False
- def ffmpeg_fixup(cndn, msg, cls):
- if not (do_fixup and cndn):
- return
- elif do_fixup == 'warn':
- self.report_warning(f'{vid}: {msg}')
- return
- pp = cls(self)
- if pp.available:
- info_dict['__postprocessors'].append(pp)
- else:
- self.report_warning(f'{vid}: {msg}. Install ffmpeg to fix this automatically')
- stretched_ratio = info_dict.get('stretched_ratio')
- ffmpeg_fixup(stretched_ratio not in (1, None),
- f'Non-uniform pixel ratio {stretched_ratio}',
- FFmpegFixupStretchedPP)
- downloader = get_suitable_downloader(info_dict, self.params) if 'protocol' in info_dict else None
- downloader = downloader.FD_NAME if downloader else None
- ext = info_dict.get('ext')
- postprocessed_by_ffmpeg = info_dict.get('requested_formats') or any((
- isinstance(pp, FFmpegVideoConvertorPP)
- and resolve_recode_mapping(ext, pp.mapping)[0] not in (ext, None)
- ) for pp in self._pps['post_process'])
- if not postprocessed_by_ffmpeg:
- ffmpeg_fixup(ext == 'm4a' and info_dict.get('container') == 'm4a_dash',
- 'writing DASH m4a. Only some players support this container',
- FFmpegFixupM4aPP)
- ffmpeg_fixup(downloader == 'hlsnative' and not self.params.get('hls_use_mpegts')
- or info_dict.get('is_live') and self.params.get('hls_use_mpegts') is None,
- 'Possible MPEG-TS in MP4 container or malformed AAC timestamps',
- FFmpegFixupM3u8PP)
- ffmpeg_fixup(info_dict.get('is_live') and downloader == 'DashSegmentsFD',
- 'Possible duplicate MOOV atoms', FFmpegFixupDuplicateMoovPP)
- ffmpeg_fixup(downloader == 'web_socket_fragment', 'Malformed timestamps detected', FFmpegFixupTimestampPP)
- ffmpeg_fixup(downloader == 'web_socket_fragment', 'Malformed duration detected', FFmpegFixupDurationPP)
- fixup()
- try:
- replace_info_dict(self.post_process(dl_filename, info_dict, files_to_move))
- except PostProcessingError as err:
- self.report_error('Postprocessing: %s' % str(err))
- return
- try:
- for ph in self._post_hooks:
- ph(info_dict['filepath'])
- except Exception as err:
- self.report_error('post hooks: %s' % str(err))
- return
- info_dict['__write_download_archive'] = True
- assert info_dict is original_infodict # Make sure the info_dict was modified in-place
- if self.params.get('force_write_download_archive'):
- info_dict['__write_download_archive'] = True
- check_max_downloads()
- def __download_wrapper(self, func):
- @functools.wraps(func)
- def wrapper(*args, **kwargs):
- try:
- res = func(*args, **kwargs)
- except UnavailableVideoError as e:
- self.report_error(e)
- except DownloadCancelled as e:
- self.to_screen(f'[info] {e}')
- if not self.params.get('break_per_url'):
- raise
- self._num_downloads = 0
- else:
- if self.params.get('dump_single_json', False):
- self.post_extract(res)
- self.to_stdout(json.dumps(self.sanitize_info(res)))
- return wrapper
- def download(self, url_list):
- """Download a given list of URLs."""
- url_list = variadic(url_list) # Passing a single URL is a common mistake
- outtmpl = self.params['outtmpl']['default']
- if (len(url_list) > 1
- and outtmpl != '-'
- and '%' not in outtmpl
- and self.params.get('max_downloads') != 1):
- raise SameFileError(outtmpl)
- for url in url_list:
- self.__download_wrapper(self.extract_info)(
- url, force_generic_extractor=self.params.get('force_generic_extractor', False))
- return self._download_retcode
- def download_with_info_file(self, info_filename):
- with contextlib.closing(fileinput.FileInput(
- [info_filename], mode='r',
- openhook=fileinput.hook_encoded('utf-8'))) as f:
- # FileInput doesn't have a read method, we can't call json.load
- info = self.sanitize_info(json.loads('\n'.join(f)), self.params.get('clean_infojson', True))
- try:
- self.__download_wrapper(self.process_ie_result)(info, download=True)
- except (DownloadError, EntryNotInPlaylist, ReExtractInfo) as e:
- if not isinstance(e, EntryNotInPlaylist):
- self.to_stderr('\r')
- webpage_url = info.get('webpage_url')
- if webpage_url is not None:
- self.report_warning(f'The info failed to download: {e}; trying with URL {webpage_url}')
- return self.download([webpage_url])
- else:
- raise
- return self._download_retcode
- @staticmethod
- def sanitize_info(info_dict, remove_private_keys=False):
- ''' Sanitize the infodict for converting to json '''
- if info_dict is None:
- return info_dict
- info_dict.setdefault('epoch', int(time.time()))
- info_dict.setdefault('_type', 'video')
- info_dict.setdefault('_version', {
- 'version': __version__,
- 'current_git_head': current_git_head(),
- 'release_git_head': RELEASE_GIT_HEAD,
- 'repository': REPOSITORY,
- })
- if remove_private_keys:
- reject = lambda k, v: v is None or k.startswith('__') or k in {
- 'requested_downloads', 'requested_formats', 'requested_subtitles', 'requested_entries',
- 'entries', 'filepath', '_filename', 'infojson_filename', 'original_url', 'playlist_autonumber',
- }
- else:
- reject = lambda k, v: False
- def filter_fn(obj):
- if isinstance(obj, dict):
- return {k: filter_fn(v) for k, v in obj.items() if not reject(k, v)}
- elif isinstance(obj, (list, tuple, set, LazyList)):
- return list(map(filter_fn, obj))
- elif obj is None or isinstance(obj, (str, int, float, bool)):
- return obj
- else:
- return repr(obj)
- return filter_fn(info_dict)
- @staticmethod
- def filter_requested_info(info_dict, actually_filter=True):
- ''' Alias of sanitize_info for backward compatibility '''
- return YoutubeDL.sanitize_info(info_dict, actually_filter)
- def _delete_downloaded_files(self, *files_to_delete, info={}, msg=None):
- for filename in set(filter(None, files_to_delete)):
- if msg:
- self.to_screen(msg % filename)
- try:
- os.remove(filename)
- except OSError:
- self.report_warning(f'Unable to delete file {filename}')
- if filename in info.get('__files_to_move', []): # NB: Delete even if None
- del info['__files_to_move'][filename]
- @staticmethod
- def post_extract(info_dict):
- def actual_post_extract(info_dict):
- if info_dict.get('_type') in ('playlist', 'multi_video'):
- for video_dict in info_dict.get('entries', {}):
- actual_post_extract(video_dict or {})
- return
- post_extractor = info_dict.pop('__post_extractor', None) or (lambda: {})
- info_dict.update(post_extractor())
- actual_post_extract(info_dict or {})
- def run_pp(self, pp, infodict):
- files_to_delete = []
- if '__files_to_move' not in infodict:
- infodict['__files_to_move'] = {}
- try:
- files_to_delete, infodict = pp.run(infodict)
- except PostProcessingError as e:
- # Must be True and not 'only_download'
- if self.params.get('ignoreerrors') is True:
- self.report_error(e)
- return infodict
- raise
- if not files_to_delete:
- return infodict
- if self.params.get('keepvideo', False):
- for f in files_to_delete:
- infodict['__files_to_move'].setdefault(f, '')
- else:
- self._delete_downloaded_files(
- *files_to_delete, info=infodict, msg='Deleting original file %s (pass -k to keep)')
- return infodict
- def run_all_pps(self, key, info, *, additional_pps=None):
- self._forceprint(key, info)
- for pp in (additional_pps or []) + self._pps[key]:
- info = self.run_pp(pp, info)
- return info
- def pre_process(self, ie_info, key='pre_process', files_to_move=None):
- info = dict(ie_info)
- info['__files_to_move'] = files_to_move or {}
- try:
- info = self.run_all_pps(key, info)
- except PostProcessingError as err:
- msg = f'Preprocessing: {err}'
- info.setdefault('__pending_error', msg)
- self.report_error(msg, is_error=False)
- return info, info.pop('__files_to_move', None)
- def post_process(self, filename, info, files_to_move=None):
- """Run all the postprocessors on the given file."""
- info['filepath'] = filename
- info['__files_to_move'] = files_to_move or {}
- info = self.run_all_pps('post_process', info, additional_pps=info.get('__postprocessors'))
- info = self.run_pp(MoveFilesAfterDownloadPP(self), info)
- del info['__files_to_move']
- return self.run_all_pps('after_move', info)
- def _make_archive_id(self, info_dict):
- video_id = info_dict.get('id')
- if not video_id:
- return
- # Future-proof against any change in case
- # and backwards compatibility with prior versions
- extractor = info_dict.get('extractor_key') or info_dict.get('ie_key') # key in a playlist
- if extractor is None:
- url = str_or_none(info_dict.get('url'))
- if not url:
- return
- # Try to find matching extractor for the URL and take its ie_key
- for ie_key, ie in self._ies.items():
- if ie.suitable(url):
- extractor = ie_key
- break
- else:
- return
- return make_archive_id(extractor, video_id)
- def in_download_archive(self, info_dict):
- if not self.archive:
- return False
- vid_ids = [self._make_archive_id(info_dict)]
- vid_ids.extend(info_dict.get('_old_archive_ids') or [])
- return any(id_ in self.archive for id_ in vid_ids)
- def record_download_archive(self, info_dict):
- fn = self.params.get('download_archive')
- if fn is None:
- return
- vid_id = self._make_archive_id(info_dict)
- assert vid_id
- self.write_debug(f'Adding to archive: {vid_id}')
- if is_path_like(fn):
- with locked_file(fn, 'a', encoding='utf-8') as archive_file:
- archive_file.write(vid_id + '\n')
- self.archive.add(vid_id)
- @staticmethod
- def format_resolution(format, default='unknown'):
- if format.get('vcodec') == 'none' and format.get('acodec') != 'none':
- return 'audio only'
- if format.get('resolution') is not None:
- return format['resolution']
- if format.get('width') and format.get('height'):
- return '%dx%d' % (format['width'], format['height'])
- elif format.get('height'):
- return '%sp' % format['height']
- elif format.get('width'):
- return '%dx?' % format['width']
- return default
- def _list_format_headers(self, *headers):
- if self.params.get('listformats_table', True) is not False:
- return [self._format_out(header, self.Styles.HEADERS) for header in headers]
- return headers
- def _format_note(self, fdict):
- res = ''
- if fdict.get('ext') in ['f4f', 'f4m']:
- res += '(unsupported)'
- if fdict.get('language'):
- if res:
- res += ' '
- res += '[%s]' % fdict['language']
- if fdict.get('format_note') is not None:
- if res:
- res += ' '
- res += fdict['format_note']
- if fdict.get('tbr') is not None:
- if res:
- res += ', '
- res += '%4dk' % fdict['tbr']
- if fdict.get('container') is not None:
- if res:
- res += ', '
- res += '%s container' % fdict['container']
- if (fdict.get('vcodec') is not None
- and fdict.get('vcodec') != 'none'):
- if res:
- res += ', '
- res += fdict['vcodec']
- if fdict.get('vbr') is not None:
- res += '@'
- elif fdict.get('vbr') is not None and fdict.get('abr') is not None:
- res += 'video@'
- if fdict.get('vbr') is not None:
- res += '%4dk' % fdict['vbr']
- if fdict.get('fps') is not None:
- if res:
- res += ', '
- res += '%sfps' % fdict['fps']
- if fdict.get('acodec') is not None:
- if res:
- res += ', '
- if fdict['acodec'] == 'none':
- res += 'video only'
- else:
- res += '%-5s' % fdict['acodec']
- elif fdict.get('abr') is not None:
- if res:
- res += ', '
- res += 'audio'
- if fdict.get('abr') is not None:
- res += '@%3dk' % fdict['abr']
- if fdict.get('asr') is not None:
- res += ' (%5dHz)' % fdict['asr']
- if fdict.get('filesize') is not None:
- if res:
- res += ', '
- res += format_bytes(fdict['filesize'])
- elif fdict.get('filesize_approx') is not None:
- if res:
- res += ', '
- res += '~' + format_bytes(fdict['filesize_approx'])
- return res
- def _get_formats(self, info_dict):
- if info_dict.get('formats') is None:
- if info_dict.get('url') and info_dict.get('_type', 'video') == 'video':
- return [info_dict]
- return []
- return info_dict['formats']
- def render_formats_table(self, info_dict):
- formats = self._get_formats(info_dict)
- if not formats:
- return
- if not self.params.get('listformats_table', True) is not False:
- table = [
- [
- format_field(f, 'format_id'),
- format_field(f, 'ext'),
- self.format_resolution(f),
- self._format_note(f)
- ] for f in formats if (f.get('preference') or 0) >= -1000]
- return render_table(['format code', 'extension', 'resolution', 'note'], table, extra_gap=1)
- def simplified_codec(f, field):
- assert field in ('acodec', 'vcodec')
- codec = f.get(field, 'unknown')
- if not codec:
- return 'unknown'
- elif codec != 'none':
- return '.'.join(codec.split('.')[:4])
- if field == 'vcodec' and f.get('acodec') == 'none':
- return 'images'
- elif field == 'acodec' and f.get('vcodec') == 'none':
- return ''
- return self._format_out('audio only' if field == 'vcodec' else 'video only',
- self.Styles.SUPPRESS)
- delim = self._format_out('\u2502', self.Styles.DELIM, '|', test_encoding=True)
- table = [
- [
- self._format_out(format_field(f, 'format_id'), self.Styles.ID),
- format_field(f, 'ext'),
- format_field(f, func=self.format_resolution, ignore=('audio only', 'images')),
- format_field(f, 'fps', '\t%d', func=round),
- format_field(f, 'dynamic_range', '%s', ignore=(None, 'SDR')).replace('HDR', ''),
- format_field(f, 'audio_channels', '\t%s'),
- delim,
- format_field(f, 'filesize', ' \t%s', func=format_bytes) + format_field(f, 'filesize_approx', '~\t%s', func=format_bytes),
- format_field(f, 'tbr', '\t%dk', func=round),
- shorten_protocol_name(f.get('protocol', '')),
- delim,
- simplified_codec(f, 'vcodec'),
- format_field(f, 'vbr', '\t%dk', func=round),
- simplified_codec(f, 'acodec'),
- format_field(f, 'abr', '\t%dk', func=round),
- format_field(f, 'asr', '\t%s', func=format_decimal_suffix),
- join_nonempty(
- self._format_out('UNSUPPORTED', 'light red') if f.get('ext') in ('f4f', 'f4m') else None,
- format_field(f, 'language', '[%s]'),
- join_nonempty(format_field(f, 'format_note'),
- format_field(f, 'container', ignore=(None, f.get('ext'))),
- delim=', '),
- delim=' '),
- ] for f in formats if f.get('preference') is None or f['preference'] >= -1000]
- header_line = self._list_format_headers(
- 'ID', 'EXT', 'RESOLUTION', '\tFPS', 'HDR', 'CH', delim, '\tFILESIZE', '\tTBR', 'PROTO',
- delim, 'VCODEC', '\tVBR', 'ACODEC', '\tABR', '\tASR', 'MORE INFO')
- return render_table(
- header_line, table, hide_empty=True,
- delim=self._format_out('\u2500', self.Styles.DELIM, '-', test_encoding=True))
- def render_thumbnails_table(self, info_dict):
- thumbnails = list(info_dict.get('thumbnails') or [])
- if not thumbnails:
- return None
- return render_table(
- self._list_format_headers('ID', 'Width', 'Height', 'URL'),
- [[t.get('id'), t.get('width') or 'unknown', t.get('height') or 'unknown', t['url']] for t in thumbnails])
- def render_subtitles_table(self, video_id, subtitles):
- def _row(lang, formats):
- exts, names = zip(*((f['ext'], f.get('name') or 'unknown') for f in reversed(formats)))
- if len(set(names)) == 1:
- names = [] if names[0] == 'unknown' else names[:1]
- return [lang, ', '.join(names), ', '.join(exts)]
- if not subtitles:
- return None
- return render_table(
- self._list_format_headers('Language', 'Name', 'Formats'),
- [_row(lang, formats) for lang, formats in subtitles.items()],
- hide_empty=True)
- def __list_table(self, video_id, name, func, *args):
- table = func(*args)
- if not table:
- self.to_screen(f'{video_id} has no {name}')
- return
- self.to_screen(f'[info] Available {name} for {video_id}:')
- self.to_stdout(table)
- def list_formats(self, info_dict):
- self.__list_table(info_dict['id'], 'formats', self.render_formats_table, info_dict)
- def list_thumbnails(self, info_dict):
- self.__list_table(info_dict['id'], 'thumbnails', self.render_thumbnails_table, info_dict)
- def list_subtitles(self, video_id, subtitles, name='subtitles'):
- self.__list_table(video_id, name, self.render_subtitles_table, video_id, subtitles)
- def urlopen(self, req):
- """ Start an HTTP download """
- if isinstance(req, str):
- req = sanitized_Request(req)
- return self._opener.open(req, timeout=self._socket_timeout)
- def print_debug_header(self):
- if not self.params.get('verbose'):
- return
- from . import _IN_CLI # Must be delayed import
- # These imports can be slow. So import them only as needed
- from .extractor.extractors import _LAZY_LOADER
- from .extractor.extractors import _PLUGIN_CLASSES as plugin_extractors
- def get_encoding(stream):
- ret = str(getattr(stream, 'encoding', 'missing (%s)' % type(stream).__name__))
- if not supports_terminal_sequences(stream):
- from .utils import WINDOWS_VT_MODE # Must be imported locally
- ret += ' (No VT)' if WINDOWS_VT_MODE is False else ' (No ANSI)'
- return ret
- encoding_str = 'Encodings: locale %s, fs %s, pref %s, %s' % (
- locale.getpreferredencoding(),
- sys.getfilesystemencoding(),
- self.get_encoding(),
- ', '.join(
- f'{key} {get_encoding(stream)}' for key, stream in self._out_files.items_
- if stream is not None and key != 'console')
- )
- logger = self.params.get('logger')
- if logger:
- write_debug = lambda msg: logger.debug(f'[debug] {msg}')
- write_debug(encoding_str)
- else:
- write_string(f'[debug] {encoding_str}\n', encoding=None)
- write_debug = lambda msg: self._write_string(f'[debug] {msg}\n')
- source = detect_variant()
- if VARIANT not in (None, 'pip'):
- source += '*'
- write_debug(join_nonempty(
- f'{"hypervideo" if REPOSITORY == "hypervideo/hypervideo" else REPOSITORY} version',
- __version__,
- f'[{RELEASE_GIT_HEAD}]' if RELEASE_GIT_HEAD else '',
- '' if source == 'unknown' else f'({source})',
- '' if _IN_CLI else 'API',
- delim=' '))
- if not _IN_CLI:
- write_debug(f'params: {self.params}')
- if not _LAZY_LOADER:
- if os.environ.get('YTDLP_NO_LAZY_EXTRACTORS'):
- write_debug('Lazy loading extractors is forcibly disabled')
- else:
- write_debug('Lazy loading extractors is disabled')
- if plugin_extractors or plugin_postprocessors:
- write_debug('Plugins: %s' % [
- '%s%s' % (klass.__name__, '' if klass.__name__ == name else f' as {name}')
- for name, klass in itertools.chain(plugin_extractors.items(), plugin_postprocessors.items())])
- if self.params['compat_opts']:
- write_debug('Compatibility options: %s' % ', '.join(self.params['compat_opts']))
- if current_git_head():
- write_debug(f'Git HEAD: {current_git_head()}')
- write_debug(system_identifier())
- exe_versions, ffmpeg_features = FFmpegPostProcessor.get_versions_and_features(self)
- ffmpeg_features = {key for key, val in ffmpeg_features.items() if val}
- if ffmpeg_features:
- exe_versions['ffmpeg'] += ' (%s)' % ','.join(sorted(ffmpeg_features))
- exe_versions['rtmpdump'] = rtmpdump_version()
- exe_versions['phantomjs'] = PhantomJSwrapper._version()
- exe_str = ', '.join(
- f'{exe} {v}' for exe, v in sorted(exe_versions.items()) if v
- ) or 'none'
- write_debug('exe versions: %s' % exe_str)
- from .compat.compat_utils import get_package_info
- from .dependencies import available_dependencies
- write_debug('Optional libraries: %s' % (', '.join(sorted({
- join_nonempty(*get_package_info(m)) for m in available_dependencies.values()
- })) or 'none'))
- self._setup_opener()
- proxy_map = {}
- for handler in self._opener.handlers:
- if hasattr(handler, 'proxies'):
- proxy_map.update(handler.proxies)
- write_debug(f'Proxy map: {proxy_map}')
- # Not implemented
- if False and self.params.get('call_home'):
- ipaddr = self.urlopen('https://yt-dl.org/ip').read().decode()
- write_debug('Public IP address: %s' % ipaddr)
- latest_version = self.urlopen(
- 'https://yt-dl.org/latest/version').read().decode()
- if version_tuple(latest_version) > version_tuple(__version__):
- self.report_warning(
- 'You are using an outdated version (newest version: %s)! '
- 'See https://yt-dl.org/update if you need help updating.' %
- latest_version)
- def _setup_opener(self):
- if hasattr(self, '_opener'):
- return
- timeout_val = self.params.get('socket_timeout')
- self._socket_timeout = 20 if timeout_val is None else float(timeout_val)
- opts_cookiesfrombrowser = self.params.get('cookiesfrombrowser')
- opts_cookiefile = self.params.get('cookiefile')
- opts_proxy = self.params.get('proxy')
- self.cookiejar = load_cookies(opts_cookiefile, opts_cookiesfrombrowser, self)
- cookie_processor = YoutubeDLCookieProcessor(self.cookiejar)
- if opts_proxy is not None:
- if opts_proxy == '':
- proxies = {}
- else:
- proxies = {'http': opts_proxy, 'https': opts_proxy}
- else:
- proxies = urllib.request.getproxies()
- # Set HTTPS proxy to HTTP one if given (https://github.com/ytdl-org/youtube-dl/issues/805)
- if 'http' in proxies and 'https' not in proxies:
- proxies['https'] = proxies['http']
- proxy_handler = PerRequestProxyHandler(proxies)
- debuglevel = 1 if self.params.get('debug_printtraffic') else 0
- https_handler = make_HTTPS_handler(self.params, debuglevel=debuglevel)
- ydlh = YoutubeDLHandler(self.params, debuglevel=debuglevel)
- redirect_handler = YoutubeDLRedirectHandler()
- data_handler = urllib.request.DataHandler()
- # When passing our own FileHandler instance, build_opener won't add the
- # default FileHandler and allows us to disable the file protocol, which
- # can be used for malicious purposes (see
- # https://github.com/ytdl-org/youtube-dl/issues/8227)
- file_handler = urllib.request.FileHandler()
- def file_open(*args, **kwargs):
- raise urllib.error.URLError('file:// scheme is explicitly disabled in hypervideo for security reasons')
- file_handler.file_open = file_open
- opener = urllib.request.build_opener(
- proxy_handler, https_handler, cookie_processor, ydlh, redirect_handler, data_handler, file_handler)
- # Delete the default user-agent header, which would otherwise apply in
- # cases where our custom HTTP handler doesn't come into play
- # (See https://github.com/ytdl-org/youtube-dl/issues/1309 for details)
- opener.addheaders = []
- self._opener = opener
- def encode(self, s):
- if isinstance(s, bytes):
- return s # Already encoded
- try:
- return s.encode(self.get_encoding())
- except UnicodeEncodeError as err:
- err.reason = err.reason + '. Check your system encoding configuration or use the --encoding option.'
- raise
- def get_encoding(self):
- encoding = self.params.get('encoding')
- if encoding is None:
- encoding = preferredencoding()
- return encoding
- def _write_info_json(self, label, ie_result, infofn, overwrite=None):
- ''' Write infojson and returns True = written, 'exists' = Already exists, False = skip, None = error '''
- if overwrite is None:
- overwrite = self.params.get('overwrites', True)
- if not self.params.get('writeinfojson'):
- return False
- elif not infofn:
- self.write_debug(f'Skipping writing {label} infojson')
- return False
- elif not self._ensure_dir_exists(infofn):
- return None
- elif not overwrite and os.path.exists(infofn):
- self.to_screen(f'[info] {label.title()} metadata is already present')
- return 'exists'
- self.to_screen(f'[info] Writing {label} metadata as JSON to: {infofn}')
- try:
- write_json_file(self.sanitize_info(ie_result, self.params.get('clean_infojson', True)), infofn)
- return True
- except OSError:
- self.report_error(f'Cannot write {label} metadata to JSON file {infofn}')
- return None
- def _write_description(self, label, ie_result, descfn):
- ''' Write description and returns True = written, False = skip, None = error '''
- if not self.params.get('writedescription'):
- return False
- elif not descfn:
- self.write_debug(f'Skipping writing {label} description')
- return False
- elif not self._ensure_dir_exists(descfn):
- return None
- elif not self.params.get('overwrites', True) and os.path.exists(descfn):
- self.to_screen(f'[info] {label.title()} description is already present')
- elif ie_result.get('description') is None:
- self.report_warning(f'There\'s no {label} description to write')
- return False
- else:
- try:
- self.to_screen(f'[info] Writing {label} description to: {descfn}')
- with open(encodeFilename(descfn), 'w', encoding='utf-8') as descfile:
- descfile.write(ie_result['description'])
- except OSError:
- self.report_error(f'Cannot write {label} description file {descfn}')
- return None
- return True
- def _write_subtitles(self, info_dict, filename):
- ''' Write subtitles to file and return list of (sub_filename, final_sub_filename); or None if error'''
- ret = []
- subtitles = info_dict.get('requested_subtitles')
- if not subtitles or not (self.params.get('writesubtitles') or self.params.get('writeautomaticsub')):
- # subtitles download errors are already managed as troubles in relevant IE
- # that way it will silently go on when used with unsupporting IE
- return ret
- sub_filename_base = self.prepare_filename(info_dict, 'subtitle')
- if not sub_filename_base:
- self.to_screen('[info] Skipping writing video subtitles')
- return ret
- for sub_lang, sub_info in subtitles.items():
- sub_format = sub_info['ext']
- sub_filename = subtitles_filename(filename, sub_lang, sub_format, info_dict.get('ext'))
- sub_filename_final = subtitles_filename(sub_filename_base, sub_lang, sub_format, info_dict.get('ext'))
- existing_sub = self.existing_file((sub_filename_final, sub_filename))
- if existing_sub:
- self.to_screen(f'[info] Video subtitle {sub_lang}.{sub_format} is already present')
- sub_info['filepath'] = existing_sub
- ret.append((existing_sub, sub_filename_final))
- continue
- self.to_screen(f'[info] Writing video subtitles to: {sub_filename}')
- if sub_info.get('data') is not None:
- try:
- # Use newline='' to prevent conversion of newline characters
- # See https://github.com/ytdl-org/youtube-dl/issues/10268
- with open(sub_filename, 'w', encoding='utf-8', newline='') as subfile:
- subfile.write(sub_info['data'])
- sub_info['filepath'] = sub_filename
- ret.append((sub_filename, sub_filename_final))
- continue
- except OSError:
- self.report_error(f'Cannot write video subtitles file {sub_filename}')
- return None
- try:
- sub_copy = sub_info.copy()
- sub_copy.setdefault('http_headers', info_dict.get('http_headers'))
- self.dl(sub_filename, sub_copy, subtitle=True)
- sub_info['filepath'] = sub_filename
- ret.append((sub_filename, sub_filename_final))
- except (DownloadError, ExtractorError, IOError, OSError, ValueError) + network_exceptions as err:
- msg = f'Unable to download video subtitles for {sub_lang!r}: {err}'
- if self.params.get('ignoreerrors') is not True: # False or 'only_download'
- if not self.params.get('ignoreerrors'):
- self.report_error(msg)
- raise DownloadError(msg)
- self.report_warning(msg)
- return ret
- def _write_thumbnails(self, label, info_dict, filename, thumb_filename_base=None):
- ''' Write thumbnails to file and return list of (thumb_filename, final_thumb_filename) '''
- write_all = self.params.get('write_all_thumbnails', False)
- thumbnails, ret = [], []
- if write_all or self.params.get('writethumbnail', False):
- thumbnails = info_dict.get('thumbnails') or []
- multiple = write_all and len(thumbnails) > 1
- if thumb_filename_base is None:
- thumb_filename_base = filename
- if thumbnails and not thumb_filename_base:
- self.write_debug(f'Skipping writing {label} thumbnail')
- return ret
- for idx, t in list(enumerate(thumbnails))[::-1]:
- thumb_ext = (f'{t["id"]}.' if multiple else '') + determine_ext(t['url'], 'jpg')
- thumb_display_id = f'{label} thumbnail {t["id"]}'
- thumb_filename = replace_extension(filename, thumb_ext, info_dict.get('ext'))
- thumb_filename_final = replace_extension(thumb_filename_base, thumb_ext, info_dict.get('ext'))
- existing_thumb = self.existing_file((thumb_filename_final, thumb_filename))
- if existing_thumb:
- self.to_screen('[info] %s is already present' % (
- thumb_display_id if multiple else f'{label} thumbnail').capitalize())
- t['filepath'] = existing_thumb
- ret.append((existing_thumb, thumb_filename_final))
- else:
- self.to_screen(f'[info] Downloading {thumb_display_id} ...')
- try:
- uf = self.urlopen(sanitized_Request(t['url'], headers=t.get('http_headers', {})))
- self.to_screen(f'[info] Writing {thumb_display_id} to: {thumb_filename}')
- with open(encodeFilename(thumb_filename), 'wb') as thumbf:
- shutil.copyfileobj(uf, thumbf)
- ret.append((thumb_filename, thumb_filename_final))
- t['filepath'] = thumb_filename
- except network_exceptions as err:
- thumbnails.pop(idx)
- self.report_warning(f'Unable to download {thumb_display_id}: {err}')
- if ret and not write_all:
- break
- return ret
|