From d47e2518f0a6900f537ef5241c80247d14049fa3 Mon Sep 17 00:00:00 2001 From: Filippo Valsorda Date: Wed, 5 Jun 2013 11:30:34 +0200 Subject: [PATCH 1/5] First commit to show implementation --- Makefile | 8 +- setup.py | 7 +- test/test_all_urls.py | 6 +- test/test_write_info_json.py | 2 +- test/test_youtube_lists.py | 7 +- test/test_youtube_subtitles.py | 4 +- test/tests.json | 466 -- youtube_dl/InfoExtractors.py | 4460 -------------------- youtube_dl/InfoExtractors/Generic.py | 145 + youtube_dl/InfoExtractors/InfoExtractor.py | 218 + youtube_dl/InfoExtractors/Youtube.py | 727 ++++ youtube_dl/InfoExtractors/__init__.py | 28 + 12 files changed, 1142 insertions(+), 4936 deletions(-) delete mode 100755 youtube_dl/InfoExtractors.py create mode 100644 youtube_dl/InfoExtractors/Generic.py create mode 100644 youtube_dl/InfoExtractors/InfoExtractor.py create mode 100644 youtube_dl/InfoExtractors/Youtube.py create mode 100755 youtube_dl/InfoExtractors/__init__.py diff --git a/Makefile b/Makefile index e00f5e650..4550b6d13 100644 --- a/Makefile +++ b/Makefile @@ -40,15 +40,15 @@ tar: youtube-dl.tar.gz pypi-files: youtube-dl.bash-completion README.txt youtube-dl.1 -youtube-dl: youtube_dl/*.py - zip --quiet youtube-dl youtube_dl/*.py +youtube-dl: youtube_dl/*.py youtube_dl/InfoExtractors/*.py + zip --quiet youtube-dl youtube_dl/*.py youtube_dl/InfoExtractors/*.py zip --quiet --junk-paths youtube-dl youtube_dl/__main__.py echo '#!$(PYTHON)' > youtube-dl cat youtube-dl.zip >> youtube-dl rm youtube-dl.zip chmod a+x youtube-dl -README.md: youtube_dl/*.py +README.md: youtube_dl/*.py youtube_dl/InfoExtractors/*.py COLUMNS=80 python -m youtube_dl --help | python devscripts/make_readme.py README.txt: README.md @@ -57,7 +57,7 @@ README.txt: README.md youtube-dl.1: README.md pandoc -s -f markdown -t man README.md -o youtube-dl.1 -youtube-dl.bash-completion: youtube_dl/*.py devscripts/bash-completion.in +youtube-dl.bash-completion: youtube_dl/*.py youtube_dl/InfoExtractors/*.py devscripts/bash-completion.in python devscripts/bash-completion.py bash-completion: youtube-dl.bash-completion diff --git a/setup.py b/setup.py index eb7b1a212..def686122 100644 --- a/setup.py +++ b/setup.py @@ -4,6 +4,7 @@ from __future__ import print_function import pkg_resources import sys +from glob import glob try: from setuptools import setup @@ -23,7 +24,9 @@ py2exe_options = { "compressed": 1, "optimize": 2, "dist_dir": '.', - "dll_excludes": ['w9xpopen.exe'] + "dll_excludes": ['w9xpopen.exe'], + "includes": [ m.replace('\\', '.').replace('/', '.')[:-3] + for m in glob('youtube_dl/InfoExtractors/*.py') ] } py2exe_console = [{ "script": "./youtube_dl/__main__.py", @@ -57,7 +60,7 @@ setup( author = 'Ricardo Garcia', maintainer = 'Philipp Hagemeister', maintainer_email = 'phihag@phihag.de', - packages = ['youtube_dl'], + packages = ['youtube_dl', 'youtube_dl.InfoExtractors'], # Provokes warning on most systems (why?!) #test_suite = 'nose.collector', diff --git a/test/test_all_urls.py b/test/test_all_urls.py index a40360122..367d42bad 100644 --- a/test/test_all_urls.py +++ b/test/test_all_urls.py @@ -7,7 +7,11 @@ import unittest import os sys.path.append(os.path.dirname(os.path.dirname(os.path.abspath(__file__)))) -from youtube_dl.InfoExtractors import YoutubeIE, YoutubePlaylistIE, YoutubeChannelIE +from youtube_dl.InfoExtractors import get_info_extractor + +YoutubeIE = get_info_extractor('Youtube') +YoutubePlaylistIE = get_info_extractor('YoutubePlaylist') +YoutubeChannelIE = get_info_extractor('YoutubeChannel') class TestAllURLsMatching(unittest.TestCase): def test_youtube_playlist_matching(self): diff --git a/test/test_write_info_json.py b/test/test_write_info_json.py index 8134dda37..be1c79724 100644 --- a/test/test_write_info_json.py +++ b/test/test_write_info_json.py @@ -48,7 +48,7 @@ class TestInfoJSON(unittest.TestCase): self.tearDown() def test_info_json(self): - ie = youtube_dl.InfoExtractors.YoutubeIE() + ie = youtube_dl.InfoExtractors.get_info_extractor('Youtube')() fd = FileDownloader(params) fd.add_info_extractor(ie) fd.download([TEST_ID]) diff --git a/test/test_youtube_lists.py b/test/test_youtube_lists.py index 78657b51c..9df10c9e6 100644 --- a/test/test_youtube_lists.py +++ b/test/test_youtube_lists.py @@ -8,9 +8,14 @@ import json import os sys.path.append(os.path.dirname(os.path.dirname(os.path.abspath(__file__)))) -from youtube_dl.InfoExtractors import YoutubeUserIE, YoutubePlaylistIE, YoutubeIE, YoutubeChannelIE from youtube_dl.utils import * from youtube_dl.FileDownloader import FileDownloader +from youtube_dl.InfoExtractors import get_info_extractor + +YoutubeIE = get_info_extractor('Youtube') +YoutubePlaylistIE = get_info_extractor('YoutubePlaylist') +YoutubeChannelIE = get_info_extractor('YoutubeChannel') +YoutubeUserIE = get_info_extractor('YoutubeUser') PARAMETERS_FILE = os.path.join(os.path.dirname(os.path.abspath(__file__)), "parameters.json") with io.open(PARAMETERS_FILE, encoding='utf-8') as pf: diff --git a/test/test_youtube_subtitles.py b/test/test_youtube_subtitles.py index a123e6d72..da1d25d7a 100644 --- a/test/test_youtube_subtitles.py +++ b/test/test_youtube_subtitles.py @@ -10,9 +10,11 @@ import hashlib import os sys.path.append(os.path.dirname(os.path.dirname(os.path.abspath(__file__)))) -from youtube_dl.InfoExtractors import YoutubeIE from youtube_dl.utils import * from youtube_dl import FileDownloader +from youtube_dl.InfoExtractors import get_info_extractor + +YoutubeIE = get_info_extractor('Youtube') PARAMETERS_FILE = os.path.join(os.path.dirname(os.path.abspath(__file__)), "parameters.json") with io.open(PARAMETERS_FILE, encoding='utf-8') as pf: diff --git a/test/tests.json b/test/tests.json index f57ebf1c9..e8dd8d0ae 100644 --- a/test/tests.json +++ b/test/tests.json @@ -11,476 +11,10 @@ "description": "test chars: \"'/\\ä↭𝕐\n\nThis is a test video for youtube-dl.\n\nFor more information, contact phihag@phihag.de ." } }, - { - "name": "Dailymotion", - "md5": "392c4b85a60a90dc4792da41ce3144eb", - "url": "http://www.dailymotion.com/video/x33vw9_tutoriel-de-youtubeur-dl-des-video_tech", - "file": "x33vw9.mp4" - }, - { - "name": "Metacafe", - "add_ie": ["Youtube"], - "url": "http://metacafe.com/watch/yt-_aUehQsCQtM/the_electric_company_short_i_pbs_kids_go/", - "file": "_aUehQsCQtM.flv" - }, - { - "name": "BlipTV", - "md5": "b2d849efcf7ee18917e4b4d9ff37cafe", - "url": "http://blip.tv/cbr/cbr-exclusive-gotham-city-imposters-bats-vs-jokerz-short-3-5796352", - "file": "5779306.m4v" - }, - { - "name": "XVideos", - "md5": "1d0c835822f0a71a7bf011855db929d0", - "url": "http://www.xvideos.com/video939581/funny_porns_by_s_-1", - "file": "939581.flv" - }, - { - "name": "YouPorn", - "md5": "c37ddbaaa39058c76a7e86c6813423c1", - "url": "http://www.youporn.com/watch/505835/sex-ed-is-it-safe-to-masturbate-daily/", - "file": "505835.mp4" - }, - { - "name": "Pornotube", - "md5": "374dd6dcedd24234453b295209aa69b6", - "url": "http://pornotube.com/c/173/m/1689755/Marilyn-Monroe-Bathing", - "file": "1689755.flv" - }, - { - "name": "YouJizz", - "md5": "07e15fa469ba384c7693fd246905547c", - "url": "http://www.youjizz.com/videos/zeichentrick-1-2189178.html", - "file": "2189178.flv" - }, - { - "name": "Vimeo", - "md5": "8879b6cc097e987f02484baf890129e5", - "url": "http://vimeo.com/56015672", - "file": "56015672.mp4", - "info_dict": { - "title": "youtube-dl test video - ★ \" ' 幸 / \\ ä ↭ 𝕐", - "uploader": "Filippo Valsorda", - "uploader_id": "user7108434", - "upload_date": "20121220", - "description": "This is a test case for youtube-dl.\nFor more information, see github.com/rg3/youtube-dl\nTest chars: ★ \" ' 幸 / \\ ä ↭ 𝕐" - } - }, - { - "name": "Soundcloud", - "md5": "ebef0a451b909710ed1d7787dddbf0d7", - "url": "http://soundcloud.com/ethmusic/lostin-powers-she-so-heavy", - "file": "62986583.mp3" - }, - { - "name": "StanfordOpenClassroom", - "md5": "544a9468546059d4e80d76265b0443b8", - "url": "http://openclassroom.stanford.edu/MainFolder/VideoPage.php?course=PracticalUnix&video=intro-environment&speed=100", - "file": "PracticalUnix_intro-environment.mp4" - }, - { - "name": "XNXX", - "md5": "0831677e2b4761795f68d417e0b7b445", - "url": "http://video.xnxx.com/video1135332/lida_naked_funny_actress_5_", - "file": "1135332.flv" - }, - { - "name": "Youku", - "url": "http://v.youku.com/v_show/id_XNDgyMDQ2NTQw.html", - "file": "XNDgyMDQ2NTQw_part00.flv", - "md5": "ffe3f2e435663dc2d1eea34faeff5b5b", - "params": { "test": false } - }, - { - "name": "NBA", - "url": "http://www.nba.com/video/games/nets/2012/12/04/0021200253-okc-bkn-recap.nba/index.html", - "file": "0021200253-okc-bkn-recap.nba.mp4", - "md5": "c0edcfc37607344e2ff8f13c378c88a4" - }, - { - "name": "JustinTV", - "url": "http://www.twitch.tv/thegamedevhub/b/296128360", - "file": "296128360.flv", - "md5": "ecaa8a790c22a40770901460af191c9a" - }, - { - "name": "MyVideo", - "url": "http://www.myvideo.de/watch/8229274/bowling_fail_or_win", - "file": "8229274.flv", - "md5": "2d2753e8130479ba2cb7e0a37002053e" - }, - { - "name": "Escapist", - "url": "http://www.escapistmagazine.com/videos/view/the-escapist-presents/6618-Breaking-Down-Baldurs-Gate", - "file": "6618-Breaking-Down-Baldurs-Gate.mp4", - "md5": "c6793dbda81388f4264c1ba18684a74d" - }, - { - "name": "GooglePlus", - "url": "https://plus.google.com/u/0/108897254135232129896/posts/ZButuJc6CtH", - "file": "ZButuJc6CtH.flv" - }, - { - "name": "FunnyOrDie", - "url": "http://www.funnyordie.com/videos/0732f586d7/heart-shaped-box-literal-video-version", - "file": "0732f586d7.mp4", - "md5": "f647e9e90064b53b6e046e75d0241fbd" - }, - { - "name": "Steam", - "url": "http://store.steampowered.com/video/105600/", - "playlist": [ - { - "file": "81300.flv", - "md5": "f870007cee7065d7c76b88f0a45ecc07", - "info_dict": { - "title": "Terraria 1.1 Trailer" - } - }, - { - "file": "80859.flv", - "md5": "61aaf31a5c5c3041afb58fb83cbb5751", - "info_dict": { - "title": "Terraria Trailer" - } - } - ] - }, - { - "name": "Ustream", - "url": "http://www.ustream.tv/recorded/20274954", - "file": "20274954.flv", - "md5": "088f151799e8f572f84eb62f17d73e5c", - "info_dict": { - "title": "Young Americans for Liberty February 7, 2012 2:28 AM", - "uploader": "Young Americans for Liberty" - } - }, - { - "name": "InfoQ", - "url": "http://www.infoq.com/presentations/A-Few-of-My-Favorite-Python-Things", - "file": "12-jan-pythonthings.mp4", - "info_dict": { - "title": "A Few of My Favorite [Python] Things" - }, - "params": { - "skip_download": true - } - }, - { - "name": "ComedyCentral", - "url": "http://www.thedailyshow.com/watch/thu-december-13-2012/kristen-stewart", - "file": "422212.mp4", - "md5": "4e2f5cb088a83cd8cdb7756132f9739d", - "info_dict": { - "title": "thedailyshow-kristen-stewart part 1" - } - }, - { - "name": "RBMARadio", - "url": "http://www.rbmaradio.com/shows/ford-lopatin-live-at-primavera-sound-2011", - "file": "ford-lopatin-live-at-primavera-sound-2011.mp3", - "md5": "6bc6f9bcb18994b4c983bc3bf4384d95", - "info_dict": { - "title": "Live at Primavera Sound 2011", - "description": "Joel Ford and Daniel \u2019Oneohtrix Point Never\u2019 Lopatin fly their midified pop extravaganza to Spain. Live at Primavera Sound 2011.", - "uploader": "Ford & Lopatin", - "uploader_id": "ford-lopatin", - "location": "Spain" - } - }, - { - "name": "Facebook", - "url": "https://www.facebook.com/photo.php?v=120708114770723", - "file": "120708114770723.mp4", - "md5": "48975a41ccc4b7a581abd68651c1a5a8", - "info_dict": { - "title": "PEOPLE ARE AWESOME 2013", - "duration": 279 - } - }, - { - "name": "EightTracks", - "url": "http://8tracks.com/ytdl/youtube-dl-test-tracks-a", - "playlist": [ - { - "file": "11885610.m4a", - "md5": "96ce57f24389fc8734ce47f4c1abcc55", - "info_dict": { - "title": "youtue-dl project<>\"' - youtube-dl test track 1 \"'/\\\u00e4\u21ad", - "uploader_id": "ytdl" - } - }, - { - "file": "11885608.m4a", - "md5": "4ab26f05c1f7291ea460a3920be8021f", - "info_dict": { - "title": "youtube-dl project - youtube-dl test track 2 \"'/\\\u00e4\u21ad", - "uploader_id": "ytdl" - - } - }, - { - "file": "11885679.m4a", - "md5": "d30b5b5f74217410f4689605c35d1fd7", - "info_dict": { - "title": "youtube-dl project as well - youtube-dl test track 3 \"'/\\\u00e4\u21ad" - } - }, - { - "file": "11885680.m4a", - "md5": "4eb0a669317cd725f6bbd336a29f923a", - "info_dict": { - "title": "youtube-dl project as well - youtube-dl test track 4 \"'/\\\u00e4\u21ad" - } - }, - { - "file": "11885682.m4a", - "md5": "1893e872e263a2705558d1d319ad19e8", - "info_dict": { - "title": "PH - youtube-dl test track 5 \"'/\\\u00e4\u21ad" - } - }, - { - "file": "11885683.m4a", - "md5": "b673c46f47a216ab1741ae8836af5899", - "info_dict": { - "title": "PH - youtube-dl test track 6 \"'/\\\u00e4\u21ad" - } - }, - { - "file": "11885684.m4a", - "md5": "1d74534e95df54986da7f5abf7d842b7", - "info_dict": { - "title": "phihag - youtube-dl test track 7 \"'/\\\u00e4\u21ad" - } - }, - { - "file": "11885685.m4a", - "md5": "f081f47af8f6ae782ed131d38b9cd1c0", - "info_dict": { - "title": "phihag - youtube-dl test track 8 \"'/\\\u00e4\u21ad" - } - } - ] - }, - { - "name": "Keek", - "url": "http://www.keek.com/ytdl/keeks/NODfbab", - "file": "NODfbab.mp4", - "md5": "9b0636f8c0f7614afa4ea5e4c6e57e83", - "info_dict": { - "title": "test chars: \"'/\\ä<>This is a test video for youtube-dl.For more information, contact phihag@phihag.de ." - } - - }, - { - "name": "TED", - "url": "http://www.ted.com/talks/dan_dennett_on_our_consciousness.html", - "file": "102.mp4", - "md5": "7bc087e71d16f18f9b8ab9fa62a8a031", - "info_dict": { - "title": "Dan Dennett: The illusion of consciousness", - "thumbnail": "http://images.ted.com/images/ted/488_389x292.jpg" - } - }, - { - "name": "MySpass", - "url": "http://www.myspass.de/myspass/shows/tvshows/absolute-mehrheit/Absolute-Mehrheit-vom-17022013-Die-Highlights-Teil-2--/11741/", - "file": "11741.mp4", - "md5": "0b49f4844a068f8b33f4b7c88405862b", - "info_dict": { - "title": "Absolute Mehrheit vom 17.02.2013 - Die Highlights, Teil 2" - } - }, { "name": "Generic", "url": "http://www.hodiho.fr/2013/02/regis-plante-sa-jeep.html", "file": "13601338388002.mp4", "md5": "85b90ccc9d73b4acd9138d3af4c27f89" - }, - { - "name": "Spiegel", - "url": "http://www.spiegel.de/video/vulkan-tungurahua-in-ecuador-ist-wieder-aktiv-video-1259285.html", - "file": "1259285.mp4", - "md5": "2c2754212136f35fb4b19767d242f66e", - "info_dict": { - "title": "Vulkanausbruch in Ecuador: Der \"Feuerschlund\" ist wieder aktiv" - } - }, - { - "name": "LiveLeak", - "md5": "0813c2430bea7a46bf13acf3406992f4", - "url": "http://www.liveleak.com/view?i=757_1364311680", - "file": "757_1364311680.mp4", - "info_dict": { - "title": "Most unlucky car accident", - "description": "extremely bad day for this guy..!", - "uploader": "ljfriel2" - } - }, - { - "name": "WorldStarHipHop", - "url": "http://www.worldstarhiphop.com/videos/video.php?v=wshh6a7q1ny0G34ZwuIO", - "file": "wshh6a7q1ny0G34ZwuIO.mp4", - "md5": "9d04de741161603bf7071bbf4e883186", - "info_dict": { - "title": "Video: KO Of The Week: MMA Fighter Gets Knocked Out By Swift Head Kick! " - } - }, - { - "name": "ARD", - "url": "http://www.ardmediathek.de/das-erste/tagesschau-in-100-sek?documentId=14077640", - "file": "14077640.mp4", - "md5": "6ca8824255460c787376353f9e20bbd8", - "info_dict": { - "title": "11.04.2013 09:23 Uhr - Tagesschau in 100 Sekunden" - }, - "skip": "Requires rtmpdump" - }, - { - "name": "Tumblr", - "url": "http://birthdayproject2012.tumblr.com/post/17258355236/a-sample-video-from-leeann-if-you-need-an-idea", - "file": "17258355236.mp4", - "md5": "7c6a514d691b034ccf8567999e9e88a3", - "info_dict": { - "title": "Calling all Pris! - A sample video from LeeAnn. (If you need an idea..." - } - }, - { - "name": "SoundcloudSet", - "url":"https://soundcloud.com/the-concept-band/sets/the-royal-concept-ep", - "playlist":[ - { - "file":"30510138.mp3", - "md5":"f9136bf103901728f29e419d2c70f55d", - "info_dict": { - "title":"D-D-Dance" - } - }, - { - "file":"47127625.mp3", - "md5":"09b6758a018470570f8fd423c9453dd8", - "info_dict": { - "title":"The Royal Concept - Gimme Twice" - } - }, - { - "file":"47127627.mp3", - "md5":"154abd4e418cea19c3b901f1e1306d9c", - "info_dict": { - "title":"Goldrushed" - } - }, - { - "file":"47127629.mp3", - "md5":"2f5471edc79ad3f33a683153e96a79c1", - "info_dict": { - "title":"In the End" - } - }, - { - "file":"47127631.mp3", - "md5":"f9ba87aa940af7213f98949254f1c6e2", - "info_dict": { - "title":"Knocked Up" - } - }, - { - "file":"75206121.mp3", - "md5":"f9d1fe9406717e302980c30de4af9353", - "info_dict": { - "title":"World On Fire" - } - } - ] - }, - { - "name":"Bandcamp", - "url":"http://youtube-dl.bandcamp.com/track/youtube-dl-test-song", - "file":"1812978515.mp3", - "md5":"cdeb30cdae1921719a3cbcab696ef53c", - "info_dict": { - "title":"youtube-dl test song \"'/\\ä↭" - }, - "skip": "There is a limit of 200 free downloads / month for the test song" - }, - { - "name": "RedTube", - "url": "http://www.redtube.com/66418", - "file": "66418.mp4", - "md5": "7b8c22b5e7098a3e1c09709df1126d2d", - "info_dict":{ - "title":"Sucked on a toilet" - } - }, - { - "name": "Photobucket", - "url": "http://media.photobucket.com/user/rachaneronas/media/TiredofLinkBuildingTryBacklinkMyDomaincom_zpsc0c3b9fa.mp4.html?filters[term]=search&filters[primary]=videos&filters[secondary]=images&sort=1&o=0", - "file": "zpsc0c3b9fa.mp4", - "md5": "7dabfb92b0a31f6c16cebc0f8e60ff99", - "info_dict":{ - "title":"Tired of Link Building? Try BacklinkMyDomain.com!" - } - }, - { - "name": "Ina", - "url": "www.ina.fr/video/I12055569/francois-hollande-je-crois-que-c-est-clair-video.html", - "file": "I12055569.mp4", - "md5": "a667021bf2b41f8dc6049479d9bb38a3", - "info_dict":{ - "title":"François Hollande \"Je crois que c'est clair\"" - } - }, - { - "name": "Yahoo", - "url": "http://screen.yahoo.com/obama-celebrates-iraq-victory-27592561.html", - "file": "27592561.flv", - "md5": "c6179bed843512823fd284fa2e7f012d", - "info_dict": { - "title": "Obama Celebrates Iraq Victory" - }, - "skip": "Requires rtmpdump" - }, - { - "name": "Howcast", - "url": "http://www.howcast.com/videos/390161-How-to-Tie-a-Square-Knot-Properly", - "file": "390161.mp4", - "md5": "1d7ba54e2c9d7dc6935ef39e00529138", - "info_dict":{ - "title":"How to Tie a Square Knot Properly", - "description":"The square knot, also known as the reef knot, is one of the oldest, most basic knots to tie, and can be used in many different ways. Here's the proper way to tie a square knot." - } - }, - { - "name": "Vine", - "url": "https://vine.co/v/b9KOOWX7HUx", - "file": "b9KOOWX7HUx.mp4", - "md5": "2f36fed6235b16da96ce9b4dc890940d", - "info_dict":{ - "title": "Chicken.", - "uploader": "Jack Dorsey" - } - }, - { - "name": "Flickr", - "url": "http://www.flickr.com/photos/forestwander-nature-pictures/5645318632/in/photostream/", - "file": "5645318632.mp4", - "md5": "6fdc01adbc89d72fc9c4f15b4a4ba87b", - "info_dict":{ - "title": "Dark Hollow Waterfalls", - "uploader_id": "forestwander-nature-pictures", - "description": "Waterfalls in the Springtime at Dark Hollow Waterfalls. These are located just off of Skyline Drive in Virginia. They are only about 6/10 of a mile hike but it is a pretty steep hill and a good climb back up." - } - }, - { - "name": "Teamcoco", - "url": "http://teamcoco.com/video/louis-ck-interview-george-w-bush", - "file": "19705.mp4", - "md5": "27b6f7527da5acf534b15f21b032656e", - "info_dict":{ - "title": "Louis C.K. Interview Pt. 1 11/3/11", - "description": "Louis C.K. got starstruck by George W. Bush, so what? Part one." - } } ] diff --git a/youtube_dl/InfoExtractors.py b/youtube_dl/InfoExtractors.py deleted file mode 100755 index 7a882b4ae..000000000 --- a/youtube_dl/InfoExtractors.py +++ /dev/null @@ -1,4460 +0,0 @@ -#!/usr/bin/env python -# -*- coding: utf-8 -*- - -from __future__ import absolute_import - -import base64 -import datetime -import itertools -import netrc -import os -import re -import socket -import time -import email.utils -import xml.etree.ElementTree -import random -import math -import operator -import hashlib -import binascii -import urllib - -from .utils import * - - -class InfoExtractor(object): - """Information Extractor class. - - Information extractors are the classes that, given a URL, extract - information about the video (or videos) the URL refers to. This - information includes the real video URL, the video title, author and - others. The information is stored in a dictionary which is then - passed to the FileDownloader. The FileDownloader processes this - information possibly downloading the video to the file system, among - other possible outcomes. - - The dictionaries must include the following fields: - - id: Video identifier. - url: Final video URL. - title: Video title, unescaped. - ext: Video filename extension. - - The following fields are optional: - - format: The video format, defaults to ext (used for --get-format) - thumbnail: Full URL to a video thumbnail image. - description: One-line video description. - uploader: Full name of the video uploader. - upload_date: Video upload date (YYYYMMDD). - uploader_id: Nickname or id of the video uploader. - location: Physical location of the video. - player_url: SWF Player URL (used for rtmpdump). - subtitles: The subtitle file contents. - urlhandle: [internal] The urlHandle to be used to download the file, - like returned by urllib.request.urlopen - - The fields should all be Unicode strings. - - Subclasses of this one should re-define the _real_initialize() and - _real_extract() methods and define a _VALID_URL regexp. - Probably, they should also be added to the list of extractors. - - _real_extract() must return a *list* of information dictionaries as - described above. - - Finally, the _WORKING attribute should be set to False for broken IEs - in order to warn the users and skip the tests. - """ - - _ready = False - _downloader = None - _WORKING = True - - def __init__(self, downloader=None): - """Constructor. Receives an optional downloader.""" - self._ready = False - self.set_downloader(downloader) - - @classmethod - def suitable(cls, url): - """Receives a URL and returns True if suitable for this IE.""" - return re.match(cls._VALID_URL, url) is not None - - @classmethod - def working(cls): - """Getter method for _WORKING.""" - return cls._WORKING - - def initialize(self): - """Initializes an instance (authentication, etc).""" - if not self._ready: - self._real_initialize() - self._ready = True - - def extract(self, url): - """Extracts URL information and returns it in list of dicts.""" - self.initialize() - return self._real_extract(url) - - def set_downloader(self, downloader): - """Sets the downloader for this IE.""" - self._downloader = downloader - - def _real_initialize(self): - """Real initialization process. Redefine in subclasses.""" - pass - - def _real_extract(self, url): - """Real extraction process. Redefine in subclasses.""" - pass - - @property - def IE_NAME(self): - return type(self).__name__[:-2] - - def _request_webpage(self, url_or_request, video_id, note=None, errnote=None): - """ Returns the response handle """ - if note is None: - self.report_download_webpage(video_id) - elif note is not False: - self.to_screen(u'%s: %s' % (video_id, note)) - try: - return compat_urllib_request.urlopen(url_or_request) - except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: - if errnote is None: - errnote = u'Unable to download webpage' - raise ExtractorError(u'%s: %s' % (errnote, compat_str(err)), sys.exc_info()[2]) - - def _download_webpage_handle(self, url_or_request, video_id, note=None, errnote=None): - """ Returns a tuple (page content as string, URL handle) """ - urlh = self._request_webpage(url_or_request, video_id, note, errnote) - content_type = urlh.headers.get('Content-Type', '') - m = re.match(r'[a-zA-Z0-9_.-]+/[a-zA-Z0-9_.-]+\s*;\s*charset=(.+)', content_type) - if m: - encoding = m.group(1) - else: - encoding = 'utf-8' - webpage_bytes = urlh.read() - if self._downloader.params.get('dump_intermediate_pages', False): - try: - url = url_or_request.get_full_url() - except AttributeError: - url = url_or_request - self.to_screen(u'Dumping request to ' + url) - dump = base64.b64encode(webpage_bytes).decode('ascii') - self._downloader.to_screen(dump) - content = webpage_bytes.decode(encoding, 'replace') - return (content, urlh) - - def _download_webpage(self, url_or_request, video_id, note=None, errnote=None): - """ Returns the data of the page as a string """ - return self._download_webpage_handle(url_or_request, video_id, note, errnote)[0] - - def to_screen(self, msg): - """Print msg to screen, prefixing it with '[ie_name]'""" - self._downloader.to_screen(u'[%s] %s' % (self.IE_NAME, msg)) - - def report_extraction(self, id_or_name): - """Report information extraction.""" - self.to_screen(u'%s: Extracting information' % id_or_name) - - def report_download_webpage(self, video_id): - """Report webpage download.""" - self.to_screen(u'%s: Downloading webpage' % video_id) - - def report_age_confirmation(self): - """Report attempt to confirm age.""" - self.to_screen(u'Confirming age') - - #Methods for following #608 - #They set the correct value of the '_type' key - def video_result(self, video_info): - """Returns a video""" - video_info['_type'] = 'video' - return video_info - def url_result(self, url, ie=None): - """Returns a url that points to a page that should be processed""" - #TODO: ie should be the class used for getting the info - video_info = {'_type': 'url', - 'url': url, - 'ie_key': ie} - return video_info - def playlist_result(self, entries, playlist_id=None, playlist_title=None): - """Returns a playlist""" - video_info = {'_type': 'playlist', - 'entries': entries} - if playlist_id: - video_info['id'] = playlist_id - if playlist_title: - video_info['title'] = playlist_title - return video_info - -class SearchInfoExtractor(InfoExtractor): - """ - Base class for paged search queries extractors. - They accept urls in the format _SEARCH_KEY(|all|[0-9]):{query} - Instances should define _SEARCH_KEY and _MAX_RESULTS. - """ - - @classmethod - def _make_valid_url(cls): - return r'%s(?P|[1-9][0-9]*|all):(?P[\s\S]+)' % cls._SEARCH_KEY - - @classmethod - def suitable(cls, url): - return re.match(cls._make_valid_url(), url) is not None - - def _real_extract(self, query): - mobj = re.match(self._make_valid_url(), query) - if mobj is None: - raise ExtractorError(u'Invalid search query "%s"' % query) - - prefix = mobj.group('prefix') - query = mobj.group('query') - if prefix == '': - return self._get_n_results(query, 1) - elif prefix == 'all': - return self._get_n_results(query, self._MAX_RESULTS) - else: - n = int(prefix) - if n <= 0: - raise ExtractorError(u'invalid download number %s for query "%s"' % (n, query)) - elif n > self._MAX_RESULTS: - self._downloader.report_warning(u'%s returns max %i results (you requested %i)' % (self._SEARCH_KEY, self._MAX_RESULTS, n)) - n = self._MAX_RESULTS - return self._get_n_results(query, n) - - def _get_n_results(self, query, n): - """Get a specified number of results for a query""" - raise NotImplementedError("This method must be implemented by sublclasses") - - -class YoutubeIE(InfoExtractor): - """Information extractor for youtube.com.""" - - _VALID_URL = r"""^ - ( - (?:https?://)? # http(s):// (optional) - (?:youtu\.be/|(?:\w+\.)?youtube(?:-nocookie)?\.com/| - tube\.majestyc\.net/) # the various hostnames, with wildcard subdomains - (?:.*?\#/)? # handle anchor (#/) redirect urls - (?: # the various things that can precede the ID: - (?:(?:v|embed|e)/) # v/ or embed/ or e/ - |(?: # or the v= param in all its forms - (?:watch(?:_popup)?(?:\.php)?)? # preceding watch(_popup|.php) or nothing (like /?v=xxxx) - (?:\?|\#!?) # the params delimiter ? or # or #! - (?:.*?&)? # any other preceding param (like /?s=tuff&v=xxxx) - v= - ) - )? # optional -> youtube.com/xxxx is OK - )? # all until now is optional -> you can pass the naked ID - ([0-9A-Za-z_-]+) # here is it! the YouTube video ID - (?(1).+)? # if we found the ID, everything can follow - $""" - _LANG_URL = r'https://www.youtube.com/?hl=en&persist_hl=1&gl=US&persist_gl=1&opt_out_ackd=1' - _LOGIN_URL = 'https://accounts.google.com/ServiceLogin' - _AGE_URL = 'http://www.youtube.com/verify_age?next_url=/&gl=US&hl=en' - _NEXT_URL_RE = r'[\?&]next_url=([^&]+)' - _NETRC_MACHINE = 'youtube' - # Listed in order of quality - _available_formats = ['38', '37', '46', '22', '45', '35', '44', '34', '18', '43', '6', '5', '17', '13'] - _available_formats_prefer_free = ['38', '46', '37', '45', '22', '44', '35', '43', '34', '18', '6', '5', '17', '13'] - _video_extensions = { - '13': '3gp', - '17': 'mp4', - '18': 'mp4', - '22': 'mp4', - '37': 'mp4', - '38': 'video', # You actually don't know if this will be MOV, AVI or whatever - '43': 'webm', - '44': 'webm', - '45': 'webm', - '46': 'webm', - } - _video_dimensions = { - '5': '240x400', - '6': '???', - '13': '???', - '17': '144x176', - '18': '360x640', - '22': '720x1280', - '34': '360x640', - '35': '480x854', - '37': '1080x1920', - '38': '3072x4096', - '43': '360x640', - '44': '480x854', - '45': '720x1280', - '46': '1080x1920', - } - IE_NAME = u'youtube' - - @classmethod - def suitable(cls, url): - """Receives a URL and returns True if suitable for this IE.""" - if YoutubePlaylistIE.suitable(url): return False - return re.match(cls._VALID_URL, url, re.VERBOSE) is not None - - def report_lang(self): - """Report attempt to set language.""" - self.to_screen(u'Setting language') - - def report_login(self): - """Report attempt to log in.""" - self.to_screen(u'Logging in') - - def report_video_webpage_download(self, video_id): - """Report attempt to download video webpage.""" - self.to_screen(u'%s: Downloading video webpage' % video_id) - - def report_video_info_webpage_download(self, video_id): - """Report attempt to download video info webpage.""" - self.to_screen(u'%s: Downloading video info webpage' % video_id) - - def report_video_subtitles_download(self, video_id): - """Report attempt to download video info webpage.""" - self.to_screen(u'%s: Checking available subtitles' % video_id) - - def report_video_subtitles_request(self, video_id, sub_lang, format): - """Report attempt to download video info webpage.""" - self.to_screen(u'%s: Downloading video subtitles for %s.%s' % (video_id, sub_lang, format)) - - def report_video_subtitles_available(self, video_id, sub_lang_list): - """Report available subtitles.""" - sub_lang = ",".join(list(sub_lang_list.keys())) - self.to_screen(u'%s: Available subtitles for video: %s' % (video_id, sub_lang)) - - def report_information_extraction(self, video_id): - """Report attempt to extract video information.""" - self.to_screen(u'%s: Extracting video information' % video_id) - - def report_unavailable_format(self, video_id, format): - """Report extracted video URL.""" - self.to_screen(u'%s: Format %s not available' % (video_id, format)) - - def report_rtmp_download(self): - """Indicate the download will use the RTMP protocol.""" - self.to_screen(u'RTMP download detected') - - def _get_available_subtitles(self, video_id): - self.report_video_subtitles_download(video_id) - request = compat_urllib_request.Request('http://video.google.com/timedtext?hl=en&type=list&v=%s' % video_id) - try: - sub_list = compat_urllib_request.urlopen(request).read().decode('utf-8') - except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: - return (u'unable to download video subtitles: %s' % compat_str(err), None) - sub_lang_list = re.findall(r'name="([^"]*)"[^>]+lang_code="([\w\-]+)"', sub_list) - sub_lang_list = dict((l[1], l[0]) for l in sub_lang_list) - if not sub_lang_list: - return (u'video doesn\'t have subtitles', None) - return sub_lang_list - - def _list_available_subtitles(self, video_id): - sub_lang_list = self._get_available_subtitles(video_id) - self.report_video_subtitles_available(video_id, sub_lang_list) - - def _request_subtitle(self, sub_lang, sub_name, video_id, format): - """ - Return tuple: - (error_message, sub_lang, sub) - """ - self.report_video_subtitles_request(video_id, sub_lang, format) - params = compat_urllib_parse.urlencode({ - 'lang': sub_lang, - 'name': sub_name, - 'v': video_id, - 'fmt': format, - }) - url = 'http://www.youtube.com/api/timedtext?' + params - try: - sub = compat_urllib_request.urlopen(url).read().decode('utf-8') - except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: - return (u'unable to download video subtitles: %s' % compat_str(err), None, None) - if not sub: - return (u'Did not fetch video subtitles', None, None) - return (None, sub_lang, sub) - - def _extract_subtitle(self, video_id): - """ - Return a list with a tuple: - [(error_message, sub_lang, sub)] - """ - sub_lang_list = self._get_available_subtitles(video_id) - sub_format = self._downloader.params.get('subtitlesformat') - if isinstance(sub_lang_list,tuple): #There was some error, it didn't get the available subtitles - return [(sub_lang_list[0], None, None)] - if self._downloader.params.get('subtitleslang', False): - sub_lang = self._downloader.params.get('subtitleslang') - elif 'en' in sub_lang_list: - sub_lang = 'en' - else: - sub_lang = list(sub_lang_list.keys())[0] - if not sub_lang in sub_lang_list: - return [(u'no closed captions found in the specified language "%s"' % sub_lang, None, None)] - - subtitle = self._request_subtitle(sub_lang, sub_lang_list[sub_lang].encode('utf-8'), video_id, sub_format) - return [subtitle] - - def _extract_all_subtitles(self, video_id): - sub_lang_list = self._get_available_subtitles(video_id) - sub_format = self._downloader.params.get('subtitlesformat') - if isinstance(sub_lang_list,tuple): #There was some error, it didn't get the available subtitles - return [(sub_lang_list[0], None, None)] - subtitles = [] - for sub_lang in sub_lang_list: - subtitle = self._request_subtitle(sub_lang, sub_lang_list[sub_lang].encode('utf-8'), video_id, sub_format) - subtitles.append(subtitle) - return subtitles - - def _print_formats(self, formats): - print('Available formats:') - for x in formats: - print('%s\t:\t%s\t[%s]' %(x, self._video_extensions.get(x, 'flv'), self._video_dimensions.get(x, '???'))) - - def _real_initialize(self): - if self._downloader is None: - return - - username = None - password = None - downloader_params = self._downloader.params - - # Attempt to use provided username and password or .netrc data - if downloader_params.get('username', None) is not None: - username = downloader_params['username'] - password = downloader_params['password'] - elif downloader_params.get('usenetrc', False): - try: - info = netrc.netrc().authenticators(self._NETRC_MACHINE) - if info is not None: - username = info[0] - password = info[2] - else: - raise netrc.NetrcParseError('No authenticators for %s' % self._NETRC_MACHINE) - except (IOError, netrc.NetrcParseError) as err: - self._downloader.report_warning(u'parsing .netrc: %s' % compat_str(err)) - return - - # Set language - request = compat_urllib_request.Request(self._LANG_URL) - try: - self.report_lang() - compat_urllib_request.urlopen(request).read() - except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: - self._downloader.report_warning(u'unable to set language: %s' % compat_str(err)) - return - - # No authentication to be performed - if username is None: - return - - request = compat_urllib_request.Request(self._LOGIN_URL) - try: - login_page = compat_urllib_request.urlopen(request).read().decode('utf-8') - except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: - self._downloader.report_warning(u'unable to fetch login page: %s' % compat_str(err)) - return - - galx = None - dsh = None - match = re.search(re.compile(r']* id="gaia_loginform"', login_results) is not None: - self._downloader.report_warning(u'unable to log in: bad username or password') - return - except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: - self._downloader.report_warning(u'unable to log in: %s' % compat_str(err)) - return - - # Confirm age - age_form = { - 'next_url': '/', - 'action_confirm': 'Confirm', - } - request = compat_urllib_request.Request(self._AGE_URL, compat_urllib_parse.urlencode(age_form)) - try: - self.report_age_confirmation() - age_results = compat_urllib_request.urlopen(request).read().decode('utf-8') - except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: - raise ExtractorError(u'Unable to confirm age: %s' % compat_str(err)) - - def _extract_id(self, url): - mobj = re.match(self._VALID_URL, url, re.VERBOSE) - if mobj is None: - raise ExtractorError(u'Invalid URL: %s' % url) - video_id = mobj.group(2) - return video_id - - def _real_extract(self, url): - # Extract original video URL from URL with redirection, like age verification, using next_url parameter - mobj = re.search(self._NEXT_URL_RE, url) - if mobj: - url = 'https://www.youtube.com/' + compat_urllib_parse.unquote(mobj.group(1)).lstrip('/') - video_id = self._extract_id(url) - - # Get video webpage - self.report_video_webpage_download(video_id) - url = 'https://www.youtube.com/watch?v=%s&gl=US&hl=en&has_verified=1' % video_id - request = compat_urllib_request.Request(url) - try: - video_webpage_bytes = compat_urllib_request.urlopen(request).read() - except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: - raise ExtractorError(u'Unable to download video webpage: %s' % compat_str(err)) - - video_webpage = video_webpage_bytes.decode('utf-8', 'ignore') - - # Attempt to extract SWF player URL - mobj = re.search(r'swfConfig.*?"(http:\\/\\/.*?watch.*?-.*?\.swf)"', video_webpage) - if mobj is not None: - player_url = re.sub(r'\\(.)', r'\1', mobj.group(1)) - else: - player_url = None - - # Get video info - self.report_video_info_webpage_download(video_id) - for el_type in ['&el=embedded', '&el=detailpage', '&el=vevo', '']: - video_info_url = ('https://www.youtube.com/get_video_info?&video_id=%s%s&ps=default&eurl=&gl=US&hl=en' - % (video_id, el_type)) - video_info_webpage = self._download_webpage(video_info_url, video_id, - note=False, - errnote='unable to download video info webpage') - video_info = compat_parse_qs(video_info_webpage) - if 'token' in video_info: - break - if 'token' not in video_info: - if 'reason' in video_info: - raise ExtractorError(u'YouTube said: %s' % video_info['reason'][0]) - else: - raise ExtractorError(u'"token" parameter not in video info for unknown reason') - - # Check for "rental" videos - if 'ypc_video_rental_bar_text' in video_info and 'author' not in video_info: - raise ExtractorError(u'"rental" videos not supported') - - # Start extracting information - self.report_information_extraction(video_id) - - # uploader - if 'author' not in video_info: - raise ExtractorError(u'Unable to extract uploader name') - video_uploader = compat_urllib_parse.unquote_plus(video_info['author'][0]) - - # uploader_id - video_uploader_id = None - mobj = re.search(r'', video_webpage) - if mobj is not None: - video_uploader_id = mobj.group(1) - else: - self._downloader.report_warning(u'unable to extract uploader nickname') - - # title - if 'title' not in video_info: - raise ExtractorError(u'Unable to extract video title') - video_title = compat_urllib_parse.unquote_plus(video_info['title'][0]) - - # thumbnail image - if 'thumbnail_url' not in video_info: - self._downloader.report_warning(u'unable to extract video thumbnail') - video_thumbnail = '' - else: # don't panic if we can't find it - video_thumbnail = compat_urllib_parse.unquote_plus(video_info['thumbnail_url'][0]) - - # upload date - upload_date = None - mobj = re.search(r'id="eow-date.*?>(.*?)', video_webpage, re.DOTALL) - if mobj is not None: - upload_date = ' '.join(re.sub(r'[/,-]', r' ', mobj.group(1)).split()) - upload_date = unified_strdate(upload_date) - - # description - video_description = get_element_by_id("eow-description", video_webpage) - if video_description: - video_description = clean_html(video_description) - else: - fd_mobj = re.search(r'= 1: - url_map = {} - for url_data_str in video_info['url_encoded_fmt_stream_map'][0].split(','): - url_data = compat_parse_qs(url_data_str) - if 'itag' in url_data and 'url' in url_data: - url = url_data['url'][0] + '&signature=' + url_data['sig'][0] - if not 'ratebypass' in url: url += '&ratebypass=yes' - url_map[url_data['itag'][0]] = url - - format_limit = self._downloader.params.get('format_limit', None) - available_formats = self._available_formats_prefer_free if self._downloader.params.get('prefer_free_formats', False) else self._available_formats - if format_limit is not None and format_limit in available_formats: - format_list = available_formats[available_formats.index(format_limit):] - else: - format_list = available_formats - existing_formats = [x for x in format_list if x in url_map] - if len(existing_formats) == 0: - raise ExtractorError(u'no known formats available for video') - if self._downloader.params.get('listformats', None): - self._print_formats(existing_formats) - return - if req_format is None or req_format == 'best': - video_url_list = [(existing_formats[0], url_map[existing_formats[0]])] # Best quality - elif req_format == 'worst': - video_url_list = [(existing_formats[len(existing_formats)-1], url_map[existing_formats[len(existing_formats)-1]])] # worst quality - elif req_format in ('-1', 'all'): - video_url_list = [(f, url_map[f]) for f in existing_formats] # All formats - else: - # Specific formats. We pick the first in a slash-delimeted sequence. - # For example, if '1/2/3/4' is requested and '2' and '4' are available, we pick '2'. - req_formats = req_format.split('/') - video_url_list = None - for rf in req_formats: - if rf in url_map: - video_url_list = [(rf, url_map[rf])] - break - if video_url_list is None: - raise ExtractorError(u'requested format not available') - else: - raise ExtractorError(u'no conn or url_encoded_fmt_stream_map information found in video info') - - results = [] - for format_param, video_real_url in video_url_list: - # Extension - video_extension = self._video_extensions.get(format_param, 'flv') - - video_format = '{0} - {1}'.format(format_param if format_param else video_extension, - self._video_dimensions.get(format_param, '???')) - - results.append({ - 'id': video_id, - 'url': video_real_url, - 'uploader': video_uploader, - 'uploader_id': video_uploader_id, - 'upload_date': upload_date, - 'title': video_title, - 'ext': video_extension, - 'format': video_format, - 'thumbnail': video_thumbnail, - 'description': video_description, - 'player_url': player_url, - 'subtitles': video_subtitles, - 'duration': video_duration - }) - return results - - -class MetacafeIE(InfoExtractor): - """Information Extractor for metacafe.com.""" - - _VALID_URL = r'(?:http://)?(?:www\.)?metacafe\.com/watch/([^/]+)/([^/]+)/.*' - _DISCLAIMER = 'http://www.metacafe.com/family_filter/' - _FILTER_POST = 'http://www.metacafe.com/f/index.php?inputType=filter&controllerGroup=user' - IE_NAME = u'metacafe' - - def report_disclaimer(self): - """Report disclaimer retrieval.""" - self.to_screen(u'Retrieving disclaimer') - - def _real_initialize(self): - # Retrieve disclaimer - request = compat_urllib_request.Request(self._DISCLAIMER) - try: - self.report_disclaimer() - disclaimer = compat_urllib_request.urlopen(request).read() - except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: - raise ExtractorError(u'Unable to retrieve disclaimer: %s' % compat_str(err)) - - # Confirm age - disclaimer_form = { - 'filters': '0', - 'submit': "Continue - I'm over 18", - } - request = compat_urllib_request.Request(self._FILTER_POST, compat_urllib_parse.urlencode(disclaimer_form)) - try: - self.report_age_confirmation() - disclaimer = compat_urllib_request.urlopen(request).read() - except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: - raise ExtractorError(u'Unable to confirm age: %s' % compat_str(err)) - - def _real_extract(self, url): - # Extract id and simplified title from URL - mobj = re.match(self._VALID_URL, url) - if mobj is None: - raise ExtractorError(u'Invalid URL: %s' % url) - - video_id = mobj.group(1) - - # Check if video comes from YouTube - mobj2 = re.match(r'^yt-(.*)$', video_id) - if mobj2 is not None: - return [self.url_result('http://www.youtube.com/watch?v=%s' % mobj2.group(1), 'Youtube')] - - # Retrieve video webpage to extract further information - webpage = self._download_webpage('http://www.metacafe.com/watch/%s/' % video_id, video_id) - - # Extract URL, uploader and title from webpage - self.report_extraction(video_id) - mobj = re.search(r'(?m)&mediaURL=([^&]+)', webpage) - if mobj is not None: - mediaURL = compat_urllib_parse.unquote(mobj.group(1)) - video_extension = mediaURL[-3:] - - # Extract gdaKey if available - mobj = re.search(r'(?m)&gdaKey=(.*?)&', webpage) - if mobj is None: - video_url = mediaURL - else: - gdaKey = mobj.group(1) - video_url = '%s?__gda__=%s' % (mediaURL, gdaKey) - else: - mobj = re.search(r' name="flashvars" value="(.*?)"', webpage) - if mobj is None: - raise ExtractorError(u'Unable to extract media URL') - vardict = compat_parse_qs(mobj.group(1)) - if 'mediaData' not in vardict: - raise ExtractorError(u'Unable to extract media URL') - mobj = re.search(r'"mediaURL":"(?Phttp.*?)",(.*?)"key":"(?P.*?)"', vardict['mediaData'][0]) - if mobj is None: - raise ExtractorError(u'Unable to extract media URL') - mediaURL = mobj.group('mediaURL').replace('\\/', '/') - video_extension = mediaURL[-3:] - video_url = '%s?__gda__=%s' % (mediaURL, mobj.group('key')) - - mobj = re.search(r'(?im)(.*) - Video', webpage) - if mobj is None: - raise ExtractorError(u'Unable to extract title') - video_title = mobj.group(1).decode('utf-8') - - mobj = re.search(r'submitter=(.*?);', webpage) - if mobj is None: - raise ExtractorError(u'Unable to extract uploader nickname') - video_uploader = mobj.group(1) - - return [{ - 'id': video_id.decode('utf-8'), - 'url': video_url.decode('utf-8'), - 'uploader': video_uploader.decode('utf-8'), - 'upload_date': None, - 'title': video_title, - 'ext': video_extension.decode('utf-8'), - }] - -class DailymotionIE(InfoExtractor): - """Information Extractor for Dailymotion""" - - _VALID_URL = r'(?i)(?:https?://)?(?:www\.)?dailymotion\.[a-z]{2,3}/video/([^/]+)' - IE_NAME = u'dailymotion' - - def _real_extract(self, url): - # Extract id and simplified title from URL - mobj = re.match(self._VALID_URL, url) - if mobj is None: - raise ExtractorError(u'Invalid URL: %s' % url) - - video_id = mobj.group(1).split('_')[0].split('?')[0] - - video_extension = 'mp4' - - # Retrieve video webpage to extract further information - request = compat_urllib_request.Request(url) - request.add_header('Cookie', 'family_filter=off') - webpage = self._download_webpage(request, video_id) - - # Extract URL, uploader and title from webpage - self.report_extraction(video_id) - mobj = re.search(r'\s*var flashvars = (.*)', webpage) - if mobj is None: - raise ExtractorError(u'Unable to extract media URL') - flashvars = compat_urllib_parse.unquote(mobj.group(1)) - - for key in ['hd1080URL', 'hd720URL', 'hqURL', 'sdURL', 'ldURL', 'video_url']: - if key in flashvars: - max_quality = key - self.to_screen(u'Using %s' % key) - break - else: - raise ExtractorError(u'Unable to extract video URL') - - mobj = re.search(r'"' + max_quality + r'":"(.+?)"', flashvars) - if mobj is None: - raise ExtractorError(u'Unable to extract video URL') - - video_url = compat_urllib_parse.unquote(mobj.group(1)).replace('\\/', '/') - - # TODO: support choosing qualities - - mobj = re.search(r'', webpage) - if mobj is None: - raise ExtractorError(u'Unable to extract title') - video_title = unescapeHTML(mobj.group('title')) - - video_uploader = None - mobj = re.search(r'(?im)[^<]+?]+?>([^<]+?)', webpage) - if mobj is None: - # lookin for official user - mobj_official = re.search(r'', webpage) - if mobj_official is None: - self._downloader.report_warning(u'unable to extract uploader nickname') - else: - video_uploader = mobj_official.group(1) - else: - video_uploader = mobj.group(1) - - video_upload_date = None - mobj = re.search(r'
([0-9]{2})-([0-9]{2})-([0-9]{4})
', webpage) - if mobj is not None: - video_upload_date = mobj.group(3) + mobj.group(2) + mobj.group(1) - - return [{ - 'id': video_id, - 'url': video_url, - 'uploader': video_uploader, - 'upload_date': video_upload_date, - 'title': video_title, - 'ext': video_extension, - }] - - -class PhotobucketIE(InfoExtractor): - """Information extractor for photobucket.com.""" - - # TODO: the original _VALID_URL was: - # r'(?:http://)?(?:[a-z0-9]+\.)?photobucket\.com/.*[\?\&]current=(.*\.flv)' - # Check if it's necessary to keep the old extracion process - _VALID_URL = r'(?:http://)?(?:[a-z0-9]+\.)?photobucket\.com/.*(([\?\&]current=)|_)(?P.*)\.(?P(flv)|(mp4))' - IE_NAME = u'photobucket' - - def _real_extract(self, url): - # Extract id from URL - mobj = re.match(self._VALID_URL, url) - if mobj is None: - raise ExtractorError(u'Invalid URL: %s' % url) - - video_id = mobj.group('id') - - video_extension = mobj.group('ext') - - # Retrieve video webpage to extract further information - webpage = self._download_webpage(url, video_id) - - # Extract URL, uploader, and title from webpage - self.report_extraction(video_id) - # We try first by looking the javascript code: - mobj = re.search(r'Pb\.Data\.Shared\.put\(Pb\.Data\.Shared\.MEDIA, (?P.*?)\);', webpage) - if mobj is not None: - info = json.loads(mobj.group('json')) - return [{ - 'id': video_id, - 'url': info[u'downloadUrl'], - 'uploader': info[u'username'], - 'upload_date': datetime.date.fromtimestamp(info[u'creationDate']).strftime('%Y%m%d'), - 'title': info[u'title'], - 'ext': video_extension, - 'thumbnail': info[u'thumbUrl'], - }] - - # We try looking in other parts of the webpage - mobj = re.search(r'', webpage) - if mobj is None: - raise ExtractorError(u'Unable to extract media URL') - mediaURL = compat_urllib_parse.unquote(mobj.group(1)) - - video_url = mediaURL - - mobj = re.search(r'(.*) video by (.*) - Photobucket', webpage) - if mobj is None: - raise ExtractorError(u'Unable to extract title') - video_title = mobj.group(1).decode('utf-8') - - video_uploader = mobj.group(2).decode('utf-8') - - return [{ - 'id': video_id.decode('utf-8'), - 'url': video_url.decode('utf-8'), - 'uploader': video_uploader, - 'upload_date': None, - 'title': video_title, - 'ext': video_extension.decode('utf-8'), - }] - - -class YahooIE(InfoExtractor): - """Information extractor for screen.yahoo.com.""" - _VALID_URL = r'http://screen\.yahoo\.com/.*?-(?P\d*?)\.html' - - def _real_extract(self, url): - mobj = re.match(self._VALID_URL, url) - if mobj is None: - raise ExtractorError(u'Invalid URL: %s' % url) - video_id = mobj.group('id') - webpage = self._download_webpage(url, video_id) - m_id = re.search(r'YUI\.namespace\("Media"\)\.CONTENT_ID = "(?P.+?)";', webpage) - - if m_id is None: - # TODO: Check which url parameters are required - info_url = 'http://cosmos.bcst.yahoo.com/rest/v2/pops;lmsoverride=1;outputformat=mrss;cb=974419660;id=%s;rd=news.yahoo.com;datacontext=mdb;lg=KCa2IihxG3qE60vQ7HtyUy' % video_id - webpage = self._download_webpage(info_url, video_id, u'Downloading info webpage') - info_re = r'''<!\[CDATA\[(?P<title>.*?)\]\]>.* - .*?)\]\]>.* - .*?)\ .*\]\]>.* - https?://)?(?:(?:www|player)\.)?vimeo\.com/(?:(?:groups|album)/[^/]+/)?(?Pplay_redirect_hls\?clip_id=)?(?:videos?/)?(?P[0-9]+)' - IE_NAME = u'vimeo' - - def _real_extract(self, url, new_video=True): - # Extract ID from URL - mobj = re.match(self._VALID_URL, url) - if mobj is None: - raise ExtractorError(u'Invalid URL: %s' % url) - - video_id = mobj.group('id') - if not mobj.group('proto'): - url = 'https://' + url - if mobj.group('direct_link'): - url = 'https://vimeo.com/' + video_id - - # Retrieve video webpage to extract further information - request = compat_urllib_request.Request(url, None, std_headers) - webpage = self._download_webpage(request, video_id) - - # Now we begin extracting as much information as we can from what we - # retrieved. First we extract the information common to all extractors, - # and latter we extract those that are Vimeo specific. - self.report_extraction(video_id) - - # Extract the config JSON - try: - config = webpage.split(' = {config:')[1].split(',assets:')[0] - config = json.loads(config) - except: - if re.search('The creator of this video has not given you permission to embed it on this domain.', webpage): - raise ExtractorError(u'The author has restricted the access to this video, try with the "--referer" option') - else: - raise ExtractorError(u'Unable to extract info section') - - # Extract title - video_title = config["video"]["title"] - - # Extract uploader and uploader_id - video_uploader = config["video"]["owner"]["name"] - video_uploader_id = config["video"]["owner"]["url"].split('/')[-1] - - # Extract video thumbnail - video_thumbnail = config["video"]["thumbnail"] - - # Extract video description - video_description = get_element_by_attribute("itemprop", "description", webpage) - if video_description: video_description = clean_html(video_description) - else: video_description = u'' - - # Extract upload date - video_upload_date = None - mobj = re.search(r' 0: - video_quality = files[quality][0][2] - video_codec = files[quality][0][0] - video_extension = files[quality][0][1] - self.to_screen(u'%s: Downloading %s file at %s quality' % (video_id, video_codec.upper(), video_quality)) - break - else: - raise ExtractorError(u'No known codec found') - - video_url = "http://player.vimeo.com/play_redirect?clip_id=%s&sig=%s&time=%s&quality=%s&codecs=%s&type=moogaloop_local&embed_location=" \ - %(video_id, sig, timestamp, video_quality, video_codec.upper()) - - return [{ - 'id': video_id, - 'url': video_url, - 'uploader': video_uploader, - 'uploader_id': video_uploader_id, - 'upload_date': video_upload_date, - 'title': video_title, - 'ext': video_extension, - 'thumbnail': video_thumbnail, - 'description': video_description, - }] - - -class ArteTvIE(InfoExtractor): - """arte.tv information extractor.""" - - _VALID_URL = r'(?:http://)?videos\.arte\.tv/(?:fr|de)/videos/.*' - _LIVE_URL = r'index-[0-9]+\.html$' - - IE_NAME = u'arte.tv' - - def fetch_webpage(self, url): - request = compat_urllib_request.Request(url) - try: - self.report_download_webpage(url) - webpage = compat_urllib_request.urlopen(request).read() - except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: - raise ExtractorError(u'Unable to retrieve video webpage: %s' % compat_str(err)) - except ValueError as err: - raise ExtractorError(u'Invalid URL: %s' % url) - return webpage - - def grep_webpage(self, url, regex, regexFlags, matchTuples): - page = self.fetch_webpage(url) - mobj = re.search(regex, page, regexFlags) - info = {} - - if mobj is None: - raise ExtractorError(u'Invalid URL: %s' % url) - - for (i, key, err) in matchTuples: - if mobj.group(i) is None: - raise ExtractorError(err) - else: - info[key] = mobj.group(i) - - return info - - def extractLiveStream(self, url): - video_lang = url.split('/')[-4] - info = self.grep_webpage( - url, - r'src="(.*?/videothek_js.*?\.js)', - 0, - [ - (1, 'url', u'Invalid URL: %s' % url) - ] - ) - http_host = url.split('/')[2] - next_url = 'http://%s%s' % (http_host, compat_urllib_parse.unquote(info.get('url'))) - info = self.grep_webpage( - next_url, - r'(s_artestras_scst_geoFRDE_' + video_lang + '.*?)\'.*?' + - '(http://.*?\.swf).*?' + - '(rtmp://.*?)\'', - re.DOTALL, - [ - (1, 'path', u'could not extract video path: %s' % url), - (2, 'player', u'could not extract video player: %s' % url), - (3, 'url', u'could not extract video url: %s' % url) - ] - ) - video_url = u'%s/%s' % (info.get('url'), info.get('path')) - - def extractPlus7Stream(self, url): - video_lang = url.split('/')[-3] - info = self.grep_webpage( - url, - r'param name="movie".*?videorefFileUrl=(http[^\'"&]*)', - 0, - [ - (1, 'url', u'Invalid URL: %s' % url) - ] - ) - next_url = compat_urllib_parse.unquote(info.get('url')) - info = self.grep_webpage( - next_url, - r'""" - mobj = re.search(_title, webpage_src) - if mobj is not None: - title = mobj.group(1) - thumbnail = None - - results = [{ - 'id': video_id, - 'url' : video_url, - 'title' : title, - 'thumbnail' : thumbnail, - 'ext' : ext, - }] - return results - -class RBMARadioIE(InfoExtractor): - _VALID_URL = r'https?://(?:www\.)?rbmaradio\.com/shows/(?P[^/]+)$' - - def _real_extract(self, url): - m = re.match(self._VALID_URL, url) - video_id = m.group('videoID') - - webpage = self._download_webpage(url, video_id) - m = re.search(r'', webpage) - if not m: - raise ExtractorError(u'Cannot find metadata') - json_data = m.group(1) - - try: - data = json.loads(json_data) - except ValueError as e: - raise ExtractorError(u'Invalid JSON: ' + str(e)) - - video_url = data['akamai_url'] + '&cbr=256' - url_parts = compat_urllib_parse_urlparse(video_url) - video_ext = url_parts.path.rpartition('.')[2] - info = { - 'id': video_id, - 'url': video_url, - 'ext': video_ext, - 'title': data['title'], - 'description': data.get('teaser_text'), - 'location': data.get('country_of_origin'), - 'uploader': data.get('host', {}).get('name'), - 'uploader_id': data.get('host', {}).get('slug'), - 'thumbnail': data.get('image', {}).get('large_url_2x'), - 'duration': data.get('duration'), - } - return [info] - - -class YouPornIE(InfoExtractor): - """Information extractor for youporn.com.""" - _VALID_URL = r'^(?:https?://)?(?:\w+\.)?youporn\.com/watch/(?P[0-9]+)/(?P[^/]+)' - - def _print_formats(self, formats): - """Print all available formats""" - print(u'Available formats:') - print(u'ext\t\tformat') - print(u'---------------------------------') - for format in formats: - print(u'%s\t\t%s' % (format['ext'], format['format'])) - - def _specific(self, req_format, formats): - for x in formats: - if(x["format"]==req_format): - return x - return None - - def _real_extract(self, url): - mobj = re.match(self._VALID_URL, url) - if mobj is None: - raise ExtractorError(u'Invalid URL: %s' % url) - - video_id = mobj.group('videoid') - - req = compat_urllib_request.Request(url) - req.add_header('Cookie', 'age_verified=1') - webpage = self._download_webpage(req, video_id) - - # Get the video title - result = re.search(r'<h1.*?>(?P<title>.*)</h1>', webpage) - if result is None: - raise ExtractorError(u'Unable to extract video title') - video_title = result.group('title').strip() - - # Get the video date - result = re.search(r'Date:</label>(?P<date>.*) </li>', webpage) - if result is None: - self._downloader.report_warning(u'unable to extract video date') - upload_date = None - else: - upload_date = unified_strdate(result.group('date').strip()) - - # Get the video uploader - result = re.search(r'Submitted:</label>(?P<uploader>.*)</li>', webpage) - if result is None: - self._downloader.report_warning(u'unable to extract uploader') - video_uploader = None - else: - video_uploader = result.group('uploader').strip() - video_uploader = clean_html( video_uploader ) - - # Get all of the formats available - DOWNLOAD_LIST_RE = r'(?s)<ul class="downloadList">(?P<download_list>.*?)</ul>' - result = re.search(DOWNLOAD_LIST_RE, webpage) - if result is None: - raise ExtractorError(u'Unable to extract download list') - download_list_html = result.group('download_list').strip() - - # Get all of the links from the page - LINK_RE = r'(?s)<a href="(?P<url>[^"]+)">' - links = re.findall(LINK_RE, download_list_html) - if(len(links) == 0): - raise ExtractorError(u'ERROR: no known formats available for video') - - self.to_screen(u'Links found: %d' % len(links)) - - formats = [] - for link in links: - - # A link looks like this: - # http://cdn1.download.youporn.phncdn.com/201210/31/8004515/480p_370k_8004515/YouPorn%20-%20Nubile%20Films%20The%20Pillow%20Fight.mp4?nvb=20121113051249&nva=20121114051249&ir=1200&sr=1200&hash=014b882080310e95fb6a0 - # A path looks like this: - # /201210/31/8004515/480p_370k_8004515/YouPorn%20-%20Nubile%20Films%20The%20Pillow%20Fight.mp4 - video_url = unescapeHTML( link ) - path = compat_urllib_parse_urlparse( video_url ).path - extension = os.path.splitext( path )[1][1:] - format = path.split('/')[4].split('_')[:2] - size = format[0] - bitrate = format[1] - format = "-".join( format ) - title = u'%s-%s-%s' % (video_title, size, bitrate) - - formats.append({ - 'id': video_id, - 'url': video_url, - 'uploader': video_uploader, - 'upload_date': upload_date, - 'title': title, - 'ext': extension, - 'format': format, - 'thumbnail': None, - 'description': None, - 'player_url': None - }) - - if self._downloader.params.get('listformats', None): - self._print_formats(formats) - return - - req_format = self._downloader.params.get('format', None) - self.to_screen(u'Format: %s' % req_format) - - if req_format is None or req_format == 'best': - return [formats[0]] - elif req_format == 'worst': - return [formats[-1]] - elif req_format in ('-1', 'all'): - return formats - else: - format = self._specific( req_format, formats ) - if result is None: - raise ExtractorError(u'Requested format not available') - return [format] - - - -class PornotubeIE(InfoExtractor): - """Information extractor for pornotube.com.""" - _VALID_URL = r'^(?:https?://)?(?:\w+\.)?pornotube\.com(/c/(?P<channel>[0-9]+))?(/m/(?P<videoid>[0-9]+))(/(?P<title>.+))$' - - def _real_extract(self, url): - mobj = re.match(self._VALID_URL, url) - if mobj is None: - raise ExtractorError(u'Invalid URL: %s' % url) - - video_id = mobj.group('videoid') - video_title = mobj.group('title') - - # Get webpage content - webpage = self._download_webpage(url, video_id) - - # Get the video URL - VIDEO_URL_RE = r'url: "(?P<url>http://video[0-9].pornotube.com/.+\.flv)",' - result = re.search(VIDEO_URL_RE, webpage) - if result is None: - raise ExtractorError(u'Unable to extract video url') - video_url = compat_urllib_parse.unquote(result.group('url')) - - #Get the uploaded date - VIDEO_UPLOADED_RE = r'<div class="video_added_by">Added (?P<date>[0-9\/]+) by' - result = re.search(VIDEO_UPLOADED_RE, webpage) - if result is None: - raise ExtractorError(u'Unable to extract video title') - upload_date = unified_strdate(result.group('date')) - - info = {'id': video_id, - 'url': video_url, - 'uploader': None, - 'upload_date': upload_date, - 'title': video_title, - 'ext': 'flv', - 'format': 'flv'} - - return [info] - -class YouJizzIE(InfoExtractor): - """Information extractor for youjizz.com.""" - _VALID_URL = r'^(?:https?://)?(?:\w+\.)?youjizz\.com/videos/(?P<videoid>[^.]+).html$' - - def _real_extract(self, url): - mobj = re.match(self._VALID_URL, url) - if mobj is None: - raise ExtractorError(u'Invalid URL: %s' % url) - - video_id = mobj.group('videoid') - - # Get webpage content - webpage = self._download_webpage(url, video_id) - - # Get the video title - result = re.search(r'<title>(?P<title>.*)', webpage) - if result is None: - raise ExtractorError(u'ERROR: unable to extract video title') - video_title = result.group('title').strip() - - # Get the embed page - result = re.search(r'https?://www.youjizz.com/videos/embed/(?P[0-9]+)', webpage) - if result is None: - raise ExtractorError(u'ERROR: unable to extract embed page') - - embed_page_url = result.group(0).strip() - video_id = result.group('videoid') - - webpage = self._download_webpage(embed_page_url, video_id) - - # Get the video URL - result = re.search(r'so.addVariable\("file",encodeURIComponent\("(?P[^"]+)"\)\);', webpage) - if result is None: - raise ExtractorError(u'ERROR: unable to extract video url') - video_url = result.group('source') - - info = {'id': video_id, - 'url': video_url, - 'title': video_title, - 'ext': 'flv', - 'format': 'flv', - 'player_url': embed_page_url} - - return [info] - -class EightTracksIE(InfoExtractor): - IE_NAME = '8tracks' - _VALID_URL = r'https?://8tracks.com/(?P[^/]+)/(?P[^/#]+)(?:#.*)?$' - - def _real_extract(self, url): - mobj = re.match(self._VALID_URL, url) - if mobj is None: - raise ExtractorError(u'Invalid URL: %s' % url) - playlist_id = mobj.group('id') - - webpage = self._download_webpage(url, playlist_id) - - m = re.search(r"PAGE.mix = (.*?);\n", webpage, flags=re.DOTALL) - if not m: - raise ExtractorError(u'Cannot find trax information') - json_like = m.group(1) - data = json.loads(json_like) - - session = str(random.randint(0, 1000000000)) - mix_id = data['id'] - track_count = data['tracks_count'] - first_url = 'http://8tracks.com/sets/%s/play?player=sm&mix_id=%s&format=jsonh' % (session, mix_id) - next_url = first_url - res = [] - for i in itertools.count(): - api_json = self._download_webpage(next_url, playlist_id, - note=u'Downloading song information %s/%s' % (str(i+1), track_count), - errnote=u'Failed to download song information') - api_data = json.loads(api_json) - track_data = api_data[u'set']['track'] - info = { - 'id': track_data['id'], - 'url': track_data['track_file_stream_url'], - 'title': track_data['performer'] + u' - ' + track_data['name'], - 'raw_title': track_data['name'], - 'uploader_id': data['user']['login'], - 'ext': 'm4a', - } - res.append(info) - if api_data['set']['at_last_track']: - break - next_url = 'http://8tracks.com/sets/%s/next?player=sm&mix_id=%s&format=jsonh&track_id=%s' % (session, mix_id, track_data['id']) - return res - -class KeekIE(InfoExtractor): - _VALID_URL = r'http://(?:www\.)?keek\.com/(?:!|\w+/keeks/)(?P\w+)' - IE_NAME = u'keek' - - def _real_extract(self, url): - m = re.match(self._VALID_URL, url) - video_id = m.group('videoID') - video_url = u'http://cdn.keek.com/keek/video/%s' % video_id - thumbnail = u'http://cdn.keek.com/keek/thumbnail/%s/w100/h75' % video_id - webpage = self._download_webpage(url, video_id) - m = re.search(r'[\S\s]+?

(?P.+?)

', webpage) - uploader = clean_html(m.group('uploader')) - info = { - 'id': video_id, - 'url': video_url, - 'ext': 'mp4', - 'title': title, - 'thumbnail': thumbnail, - 'uploader': uploader - } - return [info] - -class TEDIE(InfoExtractor): - _VALID_URL=r'''http://www\.ted\.com/ - ( - ((?Pplaylists)/(?P\d+)) # We have a playlist - | - ((?Ptalks)) # We have a simple talk - ) - (/lang/(.*?))? # The url may contain the language - /(?P\w+) # Here goes the name and then ".html" - ''' - - @classmethod - def suitable(cls, url): - """Receives a URL and returns True if suitable for this IE.""" - return re.match(cls._VALID_URL, url, re.VERBOSE) is not None - - def _real_extract(self, url): - m=re.match(self._VALID_URL, url, re.VERBOSE) - if m.group('type_talk'): - return [self._talk_info(url)] - else : - playlist_id=m.group('playlist_id') - name=m.group('name') - self.to_screen(u'Getting info of playlist %s: "%s"' % (playlist_id,name)) - return [self._playlist_videos_info(url,name,playlist_id)] - - def _talk_video_link(self,mediaSlug): - '''Returns the video link for that mediaSlug''' - return 'http://download.ted.com/talks/%s.mp4' % mediaSlug - - def _playlist_videos_info(self,url,name,playlist_id=0): - '''Returns the videos of the playlist''' - video_RE=r''' - (?P.+?)

' - webpage=self._download_webpage(url, playlist_id, 'Downloading playlist webpage') - m_videos=re.finditer(video_RE,webpage,re.VERBOSE) - m_names=re.finditer(video_name_RE,webpage) - - playlist_RE = r'div class="headline">(\s*?)

(\s*?)(?P.*?)' - m_playlist = re.search(playlist_RE, webpage) - playlist_title = m_playlist.group('playlist_title') - - playlist_entries = [] - for m_video, m_name in zip(m_videos,m_names): - video_id=m_video.group('video_id') - talk_url='http://www.ted.com%s' % m_name.group('talk_url') - playlist_entries.append(self.url_result(talk_url, 'TED')) - return self.playlist_result(playlist_entries, playlist_id = playlist_id, playlist_title = playlist_title) - - def _talk_info(self, url, video_id=0): - """Return the video for the talk in the url""" - m=re.match(self._VALID_URL, url,re.VERBOSE) - videoName=m.group('name') - webpage=self._download_webpage(url, video_id, 'Downloading \"%s\" page' % videoName) - # If the url includes the language we get the title translated - title_RE=r'(?P.*)</span>' - title=re.search(title_RE, webpage).group('title') - info_RE=r'''<script\ type="text/javascript">var\ talkDetails\ =(.*?) - "id":(?P<videoID>[\d]+).*? - "mediaSlug":"(?P<mediaSlug>[\w\d]+?)"''' - thumb_RE=r'</span>[\s.]*</div>[\s.]*<img src="(?P<thumbnail>.*?)"' - thumb_match=re.search(thumb_RE,webpage) - info_match=re.search(info_RE,webpage,re.VERBOSE) - video_id=info_match.group('videoID') - mediaSlug=info_match.group('mediaSlug') - video_url=self._talk_video_link(mediaSlug) - info = { - 'id': video_id, - 'url': video_url, - 'ext': 'mp4', - 'title': title, - 'thumbnail': thumb_match.group('thumbnail') - } - return info - -class MySpassIE(InfoExtractor): - _VALID_URL = r'http://www.myspass.de/.*' - - def _real_extract(self, url): - META_DATA_URL_TEMPLATE = 'http://www.myspass.de/myspass/includes/apps/video/getvideometadataxml.php?id=%s' - - # video id is the last path element of the URL - # usually there is a trailing slash, so also try the second but last - url_path = compat_urllib_parse_urlparse(url).path - url_parent_path, video_id = os.path.split(url_path) - if not video_id: - _, video_id = os.path.split(url_parent_path) - - # get metadata - metadata_url = META_DATA_URL_TEMPLATE % video_id - metadata_text = self._download_webpage(metadata_url, video_id) - metadata = xml.etree.ElementTree.fromstring(metadata_text.encode('utf-8')) - - # extract values from metadata - url_flv_el = metadata.find('url_flv') - if url_flv_el is None: - raise ExtractorError(u'Unable to extract download url') - video_url = url_flv_el.text - extension = os.path.splitext(video_url)[1][1:] - title_el = metadata.find('title') - if title_el is None: - raise ExtractorError(u'Unable to extract title') - title = title_el.text - format_id_el = metadata.find('format_id') - if format_id_el is None: - format = ext - else: - format = format_id_el.text - description_el = metadata.find('description') - if description_el is not None: - description = description_el.text - else: - description = None - imagePreview_el = metadata.find('imagePreview') - if imagePreview_el is not None: - thumbnail = imagePreview_el.text - else: - thumbnail = None - info = { - 'id': video_id, - 'url': video_url, - 'title': title, - 'ext': extension, - 'format': format, - 'thumbnail': thumbnail, - 'description': description - } - return [info] - -class SpiegelIE(InfoExtractor): - _VALID_URL = r'https?://(?:www\.)?spiegel\.de/video/[^/]*-(?P<videoID>[0-9]+)(?:\.html)?(?:#.*)?$' - - def _real_extract(self, url): - m = re.match(self._VALID_URL, url) - video_id = m.group('videoID') - - webpage = self._download_webpage(url, video_id) - m = re.search(r'<div class="spVideoTitle">(.*?)</div>', webpage) - if not m: - raise ExtractorError(u'Cannot find title') - video_title = unescapeHTML(m.group(1)) - - xml_url = u'http://video2.spiegel.de/flash/' + video_id + u'.xml' - xml_code = self._download_webpage(xml_url, video_id, - note=u'Downloading XML', errnote=u'Failed to download XML') - - idoc = xml.etree.ElementTree.fromstring(xml_code) - last_type = idoc[-1] - filename = last_type.findall('./filename')[0].text - duration = float(last_type.findall('./duration')[0].text) - - video_url = 'http://video2.spiegel.de/flash/' + filename - video_ext = filename.rpartition('.')[2] - info = { - 'id': video_id, - 'url': video_url, - 'ext': video_ext, - 'title': video_title, - 'duration': duration, - } - return [info] - -class LiveLeakIE(InfoExtractor): - - _VALID_URL = r'^(?:http?://)?(?:\w+\.)?liveleak\.com/view\?(?:.*?)i=(?P<video_id>[\w_]+)(?:.*)' - IE_NAME = u'liveleak' - - def _real_extract(self, url): - mobj = re.match(self._VALID_URL, url) - if mobj is None: - raise ExtractorError(u'Invalid URL: %s' % url) - - video_id = mobj.group('video_id') - - webpage = self._download_webpage(url, video_id) - - m = re.search(r'file: "(.*?)",', webpage) - if not m: - raise ExtractorError(u'Unable to find video url') - video_url = m.group(1) - - m = re.search(r'<meta property="og:title" content="(?P<title>.*?)"', webpage) - if not m: - raise ExtractorError(u'Cannot find video title') - title = unescapeHTML(m.group('title')).replace('LiveLeak.com -', '').strip() - - m = re.search(r'<meta property="og:description" content="(?P<desc>.*?)"', webpage) - if m: - desc = unescapeHTML(m.group('desc')) - else: - desc = None - - m = re.search(r'By:.*?(\w+)</a>', webpage) - if m: - uploader = clean_html(m.group(1)) - else: - uploader = None - - info = { - 'id': video_id, - 'url': video_url, - 'ext': 'mp4', - 'title': title, - 'description': desc, - 'uploader': uploader - } - - return [info] - -class ARDIE(InfoExtractor): - _VALID_URL = r'^(?:https?://)?(?:(?:www\.)?ardmediathek\.de|mediathek\.daserste\.de)/(?:.*/)(?P<video_id>[^/\?]+)(?:\?.*)?' - _TITLE = r'<h1(?: class="boxTopHeadline")?>(?P<title>.*)</h1>' - _MEDIA_STREAM = r'mediaCollection\.addMediaStream\((?P<media_type>\d+), (?P<quality>\d+), "(?P<rtmp_url>[^"]*)", "(?P<video_url>[^"]*)", "[^"]*"\)' - - def _real_extract(self, url): - # determine video id from url - m = re.match(self._VALID_URL, url) - - numid = re.search(r'documentId=([0-9]+)', url) - if numid: - video_id = numid.group(1) - else: - video_id = m.group('video_id') - - # determine title and media streams from webpage - html = self._download_webpage(url, video_id) - title = re.search(self._TITLE, html).group('title') - streams = [m.groupdict() for m in re.finditer(self._MEDIA_STREAM, html)] - if not streams: - assert '"fsk"' in html - raise ExtractorError(u'This video is only available after 8:00 pm') - - # choose default media type and highest quality for now - stream = max([s for s in streams if int(s["media_type"]) == 0], - key=lambda s: int(s["quality"])) - - # there's two possibilities: RTMP stream or HTTP download - info = {'id': video_id, 'title': title, 'ext': 'mp4'} - if stream['rtmp_url']: - self.to_screen(u'RTMP download detected') - assert stream['video_url'].startswith('mp4:') - info["url"] = stream["rtmp_url"] - info["play_path"] = stream['video_url'] - else: - assert stream["video_url"].endswith('.mp4') - info["url"] = stream["video_url"] - return [info] - -class TumblrIE(InfoExtractor): - _VALID_URL = r'http://(?P<blog_name>.*?)\.tumblr\.com/((post)|(video))/(?P<id>\d*)/(.*?)' - - def _real_extract(self, url): - m_url = re.match(self._VALID_URL, url) - video_id = m_url.group('id') - blog = m_url.group('blog_name') - - url = 'http://%s.tumblr.com/post/%s/' % (blog, video_id) - webpage = self._download_webpage(url, video_id) - - re_video = r'src=\\x22(?P<video_url>http://%s\.tumblr\.com/video_file/%s/(.*?))\\x22 type=\\x22video/(?P<ext>.*?)\\x22' % (blog, video_id) - video = re.search(re_video, webpage) - if video is None: - self.to_screen("No video found") - return [] - video_url = video.group('video_url') - ext = video.group('ext') - - re_thumb = r'posters(.*?)\[\\x22(?P<thumb>.*?)\\x22' # We pick the first poster - thumb = re.search(re_thumb, webpage).group('thumb').replace('\\', '') - - # The only place where you can get a title, it's not complete, - # but searching in other places doesn't work for all videos - re_title = r'<title>(?P<title>.*?)' - title = unescapeHTML(re.search(re_title, webpage, re.DOTALL).group('title')) - - return [{'id': video_id, - 'url': video_url, - 'title': title, - 'thumbnail': thumb, - 'ext': ext - }] - -class BandcampIE(InfoExtractor): - _VALID_URL = r'http://.*?\.bandcamp\.com/track/(?P.*)' - - def _real_extract(self, url): - mobj = re.match(self._VALID_URL, url) - title = mobj.group('title') - webpage = self._download_webpage(url, title) - # We get the link to the free download page - m_download = re.search(r'freeDownloadPage: "(.*?)"', webpage) - if m_download is None: - raise ExtractorError(u'No free songs founded') - - download_link = m_download.group(1) - id = re.search(r'var TralbumData = {(.*?)id: (?P<id>\d*?)$', - webpage, re.MULTILINE|re.DOTALL).group('id') - - download_webpage = self._download_webpage(download_link, id, - 'Downloading free downloads page') - # We get the dictionary of the track from some javascrip code - info = re.search(r'items: (.*?),$', - download_webpage, re.MULTILINE).group(1) - info = json.loads(info)[0] - # We pick mp3-320 for now, until format selection can be easily implemented. - mp3_info = info[u'downloads'][u'mp3-320'] - # If we try to use this url it says the link has expired - initial_url = mp3_info[u'url'] - re_url = r'(?P<server>http://(.*?)\.bandcamp\.com)/download/track\?enc=mp3-320&fsig=(?P<fsig>.*?)&id=(?P<id>.*?)&ts=(?P<ts>.*)$' - m_url = re.match(re_url, initial_url) - #We build the url we will use to get the final track url - # This url is build in Bandcamp in the script download_bunde_*.js - request_url = '%s/statdownload/track?enc=mp3-320&fsig=%s&id=%s&ts=%s&.rand=665028774616&.vrs=1' % (m_url.group('server'), m_url.group('fsig'), id, m_url.group('ts')) - final_url_webpage = self._download_webpage(request_url, id, 'Requesting download url') - # If we could correctly generate the .rand field the url would be - #in the "download_url" key - final_url = re.search(r'"retry_url":"(.*?)"', final_url_webpage).group(1) - - track_info = {'id':id, - 'title' : info[u'title'], - 'ext' : 'mp3', - 'url' : final_url, - 'thumbnail' : info[u'thumb_url'], - 'uploader' : info[u'artist'] - } - - return [track_info] - -class RedTubeIE(InfoExtractor): - """Information Extractor for redtube""" - _VALID_URL = r'(?:http://)?(?:www\.)?redtube\.com/(?P<id>[0-9]+)' - - def _real_extract(self,url): - mobj = re.match(self._VALID_URL, url) - if mobj is None: - raise ExtractorError(u'Invalid URL: %s' % url) - - video_id = mobj.group('id') - video_extension = 'mp4' - webpage = self._download_webpage(url, video_id) - self.report_extraction(video_id) - mobj = re.search(r'<source src="'+'(.+)'+'" type="video/mp4">',webpage) - - if mobj is None: - raise ExtractorError(u'Unable to extract media URL') - - video_url = mobj.group(1) - mobj = re.search('<h1 class="videoTitle slidePanelMovable">(.+)</h1>',webpage) - if mobj is None: - raise ExtractorError(u'Unable to extract title') - video_title = mobj.group(1) - - return [{ - 'id': video_id, - 'url': video_url, - 'ext': video_extension, - 'title': video_title, - }] - -class InaIE(InfoExtractor): - """Information Extractor for Ina.fr""" - _VALID_URL = r'(?:http://)?(?:www\.)?ina\.fr/video/(?P<id>I[0-9]+)/.*' - - def _real_extract(self,url): - mobj = re.match(self._VALID_URL, url) - - video_id = mobj.group('id') - mrss_url='http://player.ina.fr/notices/%s.mrss' % video_id - video_extension = 'mp4' - webpage = self._download_webpage(mrss_url, video_id) - - mobj = re.search(r'<media:player url="(?P<mp4url>http://mp4.ina.fr/[^"]+\.mp4)', webpage) - if mobj is None: - raise ExtractorError(u'Unable to extract media URL') - video_url = mobj.group(1) - - mobj = re.search(r'<title><!\[CDATA\[(?P<titre>.*?)]]>', webpage) - if mobj is None: - raise ExtractorError(u'Unable to extract title') - video_title = mobj.group(1) - - return [{ - 'id': video_id, - 'url': video_url, - 'ext': video_extension, - 'title': video_title, - }] - -class HowcastIE(InfoExtractor): - """Information Extractor for Howcast.com""" - _VALID_URL = r'(?:https?://)?(?:www\.)?howcast\.com/videos/(?P\d+)' - - def _real_extract(self, url): - mobj = re.match(self._VALID_URL, url) - - video_id = mobj.group('id') - webpage_url = 'http://www.howcast.com/videos/' + video_id - webpage = self._download_webpage(webpage_url, video_id) - - self.report_extraction(video_id) - - mobj = re.search(r'\'?file\'?: "(http://mobile-media\.howcast\.com/[0-9]+\.mp4)"', webpage) - if mobj is None: - raise ExtractorError(u'Unable to extract video URL') - video_url = mobj.group(1) - - mobj = re.search(r'\w+)' - - def _real_extract(self, url): - - mobj = re.match(self._VALID_URL, url) - - video_id = mobj.group('id') - webpage_url = 'https://vine.co/v/' + video_id - webpage = self._download_webpage(webpage_url, video_id) - - self.report_extraction(video_id) - - mobj = re.search(r'.*?

(.+?)

', webpage, re.DOTALL) - if mobj is None: - raise ExtractorError(u'Unable to extract uploader') - uploader = mobj.group(1) - - return [{ - 'id': video_id, - 'url': video_url, - 'ext': 'mp4', - 'title': video_title, - 'thumbnail': thumbnail, - 'uploader': uploader, - }] - -class FlickrIE(InfoExtractor): - """Information Extractor for Flickr videos""" - _VALID_URL = r'(?:https?://)?(?:www\.)?flickr\.com/photos/(?P[\w\-_@]+)/(?P\d+).*' - - def _real_extract(self, url): - mobj = re.match(self._VALID_URL, url) - - video_id = mobj.group('id') - video_uploader_id = mobj.group('uploader_id') - webpage_url = 'http://www.flickr.com/photos/' + video_uploader_id + '/' + video_id - webpage = self._download_webpage(webpage_url, video_id) - - mobj = re.search(r"photo_secret: '(\w+)'", webpage) - if mobj is None: - raise ExtractorError(u'Unable to extract video secret') - secret = mobj.group(1) - - first_url = 'https://secure.flickr.com/apps/video/video_mtl_xml.gne?v=x&photo_id=' + video_id + '&secret=' + secret + '&bitrate=700&target=_self' - first_xml = self._download_webpage(first_url, video_id, 'Downloading first data webpage') - - mobj = re.search(r'(\d+-\d+)', first_xml) - if mobj is None: - raise ExtractorError(u'Unable to extract node_id') - node_id = mobj.group(1) - - second_url = 'https://secure.flickr.com/video_playlist.gne?node_id=' + node_id + '&tech=flash&mode=playlist&bitrate=700&secret=' + secret + '&rd=video.yahoo.com&noad=1' - second_xml = self._download_webpage(second_url, video_id, 'Downloading second data webpage') - - self.report_extraction(video_id) - - mobj = re.search(r'.*)' - - def _real_extract(self, url): - mobj = re.match(self._VALID_URL, url) - if mobj is None: - raise ExtractorError(u'Invalid URL: %s' % url) - url_title = mobj.group('url_title') - webpage = self._download_webpage(url, url_title) - - mobj = re.search(r'
(.*?)', data) - if mobj is None: - raise ExtractorError(u'Unable to extract video url') - video_url = mobj.group(1) - - return [{ - 'id': video_id, - 'url': video_url, - 'ext': 'mp4', - 'title': video_title, - 'thumbnail': thumbnail, - 'description': description, - }] - -def gen_extractors(): - """ Return a list of an instance of every supported extractor. - The order does matter; the first extractor matched is the one handling the URL. - """ - return [ - YoutubePlaylistIE(), - YoutubeChannelIE(), - YoutubeUserIE(), - YoutubeSearchIE(), - YoutubeIE(), - MetacafeIE(), - DailymotionIE(), - GoogleSearchIE(), - PhotobucketIE(), - YahooIE(), - YahooSearchIE(), - DepositFilesIE(), - FacebookIE(), - BlipTVIE(), - BlipTVUserIE(), - VimeoIE(), - MyVideoIE(), - ComedyCentralIE(), - EscapistIE(), - CollegeHumorIE(), - XVideosIE(), - SoundcloudSetIE(), - SoundcloudIE(), - InfoQIE(), - MixcloudIE(), - StanfordOpenClassroomIE(), - MTVIE(), - YoukuIE(), - XNXXIE(), - YouJizzIE(), - PornotubeIE(), - YouPornIE(), - GooglePlusIE(), - ArteTvIE(), - NBAIE(), - WorldStarHipHopIE(), - JustinTVIE(), - FunnyOrDieIE(), - SteamIE(), - UstreamIE(), - RBMARadioIE(), - EightTracksIE(), - KeekIE(), - TEDIE(), - MySpassIE(), - SpiegelIE(), - LiveLeakIE(), - ARDIE(), - TumblrIE(), - BandcampIE(), - RedTubeIE(), - InaIE(), - HowcastIE(), - VineIE(), - FlickrIE(), - TeamcocoIE(), - GenericIE() - ] - -def get_info_extractor(ie_name): - """Returns the info extractor class with the given ie_name""" - return globals()[ie_name+'IE'] diff --git a/youtube_dl/InfoExtractors/Generic.py b/youtube_dl/InfoExtractors/Generic.py new file mode 100644 index 000000000..ccb185d06 --- /dev/null +++ b/youtube_dl/InfoExtractors/Generic.py @@ -0,0 +1,145 @@ +#!/usr/bin/env python +# -*- coding: utf-8 -*- + +from __future__ import absolute_import + +import os +import re + +from ..utils import * +from .InfoExtractor import InfoExtractor + + +class GenericIE(InfoExtractor): + """Generic last-resort information extractor.""" + + _VALID_URL = r'.*' + IE_NAME = u'generic' + + def report_download_webpage(self, video_id): + """Report webpage download.""" + if not self._downloader.params.get('test', False): + self._downloader.report_warning(u'Falling back on generic information extractor.') + super(GenericIE, self).report_download_webpage(video_id) + + def report_following_redirect(self, new_url): + """Report information extraction.""" + self._downloader.to_screen(u'[redirect] Following redirect to %s' % new_url) + + def _test_redirect(self, url): + """Check if it is a redirect, like url shorteners, in case return the new url.""" + class HeadRequest(compat_urllib_request.Request): + def get_method(self): + return "HEAD" + + class HEADRedirectHandler(compat_urllib_request.HTTPRedirectHandler): + """ + Subclass the HTTPRedirectHandler to make it use our + HeadRequest also on the redirected URL + """ + def redirect_request(self, req, fp, code, msg, headers, newurl): + if code in (301, 302, 303, 307): + newurl = newurl.replace(' ', '%20') + newheaders = dict((k,v) for k,v in req.headers.items() + if k.lower() not in ("content-length", "content-type")) + return HeadRequest(newurl, + headers=newheaders, + origin_req_host=req.get_origin_req_host(), + unverifiable=True) + else: + raise compat_urllib_error.HTTPError(req.get_full_url(), code, msg, headers, fp) + + class HTTPMethodFallback(compat_urllib_request.BaseHandler): + """ + Fallback to GET if HEAD is not allowed (405 HTTP error) + """ + def http_error_405(self, req, fp, code, msg, headers): + fp.read() + fp.close() + + newheaders = dict((k,v) for k,v in req.headers.items() + if k.lower() not in ("content-length", "content-type")) + return self.parent.open(compat_urllib_request.Request(req.get_full_url(), + headers=newheaders, + origin_req_host=req.get_origin_req_host(), + unverifiable=True)) + + # Build our opener + opener = compat_urllib_request.OpenerDirector() + for handler in [compat_urllib_request.HTTPHandler, compat_urllib_request.HTTPDefaultErrorHandler, + HTTPMethodFallback, HEADRedirectHandler, + compat_urllib_request.HTTPErrorProcessor, compat_urllib_request.HTTPSHandler]: + opener.add_handler(handler()) + + response = opener.open(HeadRequest(url)) + if response is None: + raise ExtractorError(u'Invalid URL protocol') + new_url = response.geturl() + + if url == new_url: + return False + + self.report_following_redirect(new_url) + return new_url + + def _real_extract(self, url): + new_url = self._test_redirect(url) + if new_url: return [self.url_result(new_url)] + + video_id = url.split('/')[-1] + try: + webpage = self._download_webpage(url, video_id) + except ValueError as err: + # since this is the last-resort InfoExtractor, if + # this error is thrown, it'll be thrown here + raise ExtractorError(u'Invalid URL: %s' % url) + + self.report_extraction(video_id) + # Start with something easy: JW Player in SWFObject + mobj = re.search(r'flashvars: [\'"](?:.*&)?file=(http[^\'"&]*)', webpage) + if mobj is None: + # Broaden the search a little bit + mobj = re.search(r'[^A-Za-z0-9]?(?:file|source)=(http[^\'"&]*)', webpage) + if mobj is None: + # Broaden the search a little bit: JWPlayer JS loader + mobj = re.search(r'[^A-Za-z0-9]?file:\s*["\'](http[^\'"&]*)', webpage) + if mobj is None: + raise ExtractorError(u'Invalid URL: %s' % url) + + # It's possible that one of the regexes + # matched, but returned an empty group: + if mobj.group(1) is None: + raise ExtractorError(u'Invalid URL: %s' % url) + + video_url = compat_urllib_parse.unquote(mobj.group(1)) + video_id = os.path.basename(video_url) + + # here's a fun little line of code for you: + video_extension = os.path.splitext(video_id)[1][1:] + video_id = os.path.splitext(video_id)[0] + + # it's tempting to parse this further, but you would + # have to take into account all the variations like + # Video Title - Site Name + # Site Name | Video Title + # Video Title - Tagline | Site Name + # and so on and so forth; it's just not practical + mobj = re.search(r'(.*)', webpage) + if mobj is None: + raise ExtractorError(u'Unable to extract title') + video_title = mobj.group(1) + + # video uploader is domain name + mobj = re.match(r'(?:https?://)?([^/]*)/.*', url) + if mobj is None: + raise ExtractorError(u'Unable to extract title') + video_uploader = mobj.group(1) + + return [{ + 'id': video_id, + 'url': video_url, + 'uploader': video_uploader, + 'upload_date': None, + 'title': video_title, + 'ext': video_extension, + }] diff --git a/youtube_dl/InfoExtractors/InfoExtractor.py b/youtube_dl/InfoExtractors/InfoExtractor.py new file mode 100644 index 000000000..aac50c8c7 --- /dev/null +++ b/youtube_dl/InfoExtractors/InfoExtractor.py @@ -0,0 +1,218 @@ +#!/usr/bin/env python +# -*- coding: utf-8 -*- + +from __future__ import absolute_import + +import base64 +import re +import socket + +from ..utils import * + + +class InfoExtractor(object): + """Information Extractor class. + + Information extractors are the classes that, given a URL, extract + information about the video (or videos) the URL refers to. This + information includes the real video URL, the video title, author and + others. The information is stored in a dictionary which is then + passed to the FileDownloader. The FileDownloader processes this + information possibly downloading the video to the file system, among + other possible outcomes. + + The dictionaries must include the following fields: + + id: Video identifier. + url: Final video URL. + title: Video title, unescaped. + ext: Video filename extension. + + The following fields are optional: + + format: The video format, defaults to ext (used for --get-format) + thumbnail: Full URL to a video thumbnail image. + description: One-line video description. + uploader: Full name of the video uploader. + upload_date: Video upload date (YYYYMMDD). + uploader_id: Nickname or id of the video uploader. + location: Physical location of the video. + player_url: SWF Player URL (used for rtmpdump). + subtitles: The subtitle file contents. + urlhandle: [internal] The urlHandle to be used to download the file, + like returned by urllib.request.urlopen + + The fields should all be Unicode strings. + + Subclasses of this one should re-define the _real_initialize() and + _real_extract() methods and define a _VALID_URL regexp. + Probably, they should also be added to the list of extractors. + + _real_extract() must return a *list* of information dictionaries as + described above. + + Finally, the _WORKING attribute should be set to False for broken IEs + in order to warn the users and skip the tests. + """ + + _ready = False + _downloader = None + _WORKING = True + + def __init__(self, downloader=None): + """Constructor. Receives an optional downloader.""" + self._ready = False + self.set_downloader(downloader) + + @classmethod + def suitable(cls, url): + """Receives a URL and returns True if suitable for this IE.""" + return re.match(cls._VALID_URL, url) is not None + + @classmethod + def working(cls): + """Getter method for _WORKING.""" + return cls._WORKING + + def initialize(self): + """Initializes an instance (authentication, etc).""" + if not self._ready: + self._real_initialize() + self._ready = True + + def extract(self, url): + """Extracts URL information and returns it in list of dicts.""" + self.initialize() + return self._real_extract(url) + + def set_downloader(self, downloader): + """Sets the downloader for this IE.""" + self._downloader = downloader + + def _real_initialize(self): + """Real initialization process. Redefine in subclasses.""" + pass + + def _real_extract(self, url): + """Real extraction process. Redefine in subclasses.""" + pass + + @property + def IE_NAME(self): + return type(self).__name__[:-2] + + def _request_webpage(self, url_or_request, video_id, note=None, errnote=None): + """ Returns the response handle """ + if note is None: + self.report_download_webpage(video_id) + elif note is not False: + self.to_screen(u'%s: %s' % (video_id, note)) + try: + return compat_urllib_request.urlopen(url_or_request) + except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: + if errnote is None: + errnote = u'Unable to download webpage' + raise ExtractorError(u'%s: %s' % (errnote, compat_str(err)), sys.exc_info()[2]) + + def _download_webpage_handle(self, url_or_request, video_id, note=None, errnote=None): + """ Returns a tuple (page content as string, URL handle) """ + urlh = self._request_webpage(url_or_request, video_id, note, errnote) + content_type = urlh.headers.get('Content-Type', '') + m = re.match(r'[a-zA-Z0-9_.-]+/[a-zA-Z0-9_.-]+\s*;\s*charset=(.+)', content_type) + if m: + encoding = m.group(1) + else: + encoding = 'utf-8' + webpage_bytes = urlh.read() + if self._downloader.params.get('dump_intermediate_pages', False): + try: + url = url_or_request.get_full_url() + except AttributeError: + url = url_or_request + self.to_screen(u'Dumping request to ' + url) + dump = base64.b64encode(webpage_bytes).decode('ascii') + self._downloader.to_screen(dump) + content = webpage_bytes.decode(encoding, 'replace') + return (content, urlh) + + def _download_webpage(self, url_or_request, video_id, note=None, errnote=None): + """ Returns the data of the page as a string """ + return self._download_webpage_handle(url_or_request, video_id, note, errnote)[0] + + def to_screen(self, msg): + """Print msg to screen, prefixing it with '[ie_name]'""" + self._downloader.to_screen(u'[%s] %s' % (self.IE_NAME, msg)) + + def report_extraction(self, id_or_name): + """Report information extraction.""" + self.to_screen(u'%s: Extracting information' % id_or_name) + + def report_download_webpage(self, video_id): + """Report webpage download.""" + self.to_screen(u'%s: Downloading webpage' % video_id) + + def report_age_confirmation(self): + """Report attempt to confirm age.""" + self.to_screen(u'Confirming age') + + #Methods for following #608 + #They set the correct value of the '_type' key + def video_result(self, video_info): + """Returns a video""" + video_info['_type'] = 'video' + return video_info + def url_result(self, url, ie=None): + """Returns a url that points to a page that should be processed""" + #TODO: ie should be the class used for getting the info + video_info = {'_type': 'url', + 'url': url, + 'ie_key': ie} + return video_info + def playlist_result(self, entries, playlist_id=None, playlist_title=None): + """Returns a playlist""" + video_info = {'_type': 'playlist', + 'entries': entries} + if playlist_id: + video_info['id'] = playlist_id + if playlist_title: + video_info['title'] = playlist_title + return video_info + +class SearchInfoExtractor(InfoExtractor): + """ + Base class for paged search queries extractors. + They accept urls in the format _SEARCH_KEY(|all|[0-9]):{query} + Instances should define _SEARCH_KEY and _MAX_RESULTS. + """ + + @classmethod + def _make_valid_url(cls): + return r'%s(?P|[1-9][0-9]*|all):(?P[\s\S]+)' % cls._SEARCH_KEY + + @classmethod + def suitable(cls, url): + return re.match(cls._make_valid_url(), url) is not None + + def _real_extract(self, query): + mobj = re.match(self._make_valid_url(), query) + if mobj is None: + raise ExtractorError(u'Invalid search query "%s"' % query) + + prefix = mobj.group('prefix') + query = mobj.group('query') + if prefix == '': + return self._get_n_results(query, 1) + elif prefix == 'all': + return self._get_n_results(query, self._MAX_RESULTS) + else: + n = int(prefix) + if n <= 0: + raise ExtractorError(u'invalid download number %s for query "%s"' % (n, query)) + elif n > self._MAX_RESULTS: + self._downloader.report_warning(u'%s returns max %i results (you requested %i)' % (self._SEARCH_KEY, self._MAX_RESULTS, n)) + n = self._MAX_RESULTS + return self._get_n_results(query, n) + + def _get_n_results(self, query, n): + """Get a specified number of results for a query""" + raise NotImplementedError("This method must be implemented by sublclasses") diff --git a/youtube_dl/InfoExtractors/Youtube.py b/youtube_dl/InfoExtractors/Youtube.py new file mode 100644 index 000000000..09a062e4a --- /dev/null +++ b/youtube_dl/InfoExtractors/Youtube.py @@ -0,0 +1,727 @@ +#!/usr/bin/env python +# -*- coding: utf-8 -*- + +from __future__ import absolute_import + +import netrc +import re +import socket + +from ..utils import * +from .InfoExtractor import InfoExtractor +from .InfoExtractor import SearchInfoExtractor + + +class YoutubeIE(InfoExtractor): + """Information extractor for youtube.com.""" + + _VALID_URL = r"""^ + ( + (?:https?://)? # http(s):// (optional) + (?:youtu\.be/|(?:\w+\.)?youtube(?:-nocookie)?\.com/| + tube\.majestyc\.net/) # the various hostnames, with wildcard subdomains + (?:.*?\#/)? # handle anchor (#/) redirect urls + (?: # the various things that can precede the ID: + (?:(?:v|embed|e)/) # v/ or embed/ or e/ + |(?: # or the v= param in all its forms + (?:watch(?:_popup)?(?:\.php)?)? # preceding watch(_popup|.php) or nothing (like /?v=xxxx) + (?:\?|\#!?) # the params delimiter ? or # or #! + (?:.*?&)? # any other preceding param (like /?s=tuff&v=xxxx) + v= + ) + )? # optional -> youtube.com/xxxx is OK + )? # all until now is optional -> you can pass the naked ID + ([0-9A-Za-z_-]+) # here is it! the YouTube video ID + (?(1).+)? # if we found the ID, everything can follow + $""" + _LANG_URL = r'https://www.youtube.com/?hl=en&persist_hl=1&gl=US&persist_gl=1&opt_out_ackd=1' + _LOGIN_URL = 'https://accounts.google.com/ServiceLogin' + _AGE_URL = 'http://www.youtube.com/verify_age?next_url=/&gl=US&hl=en' + _NEXT_URL_RE = r'[\?&]next_url=([^&]+)' + _NETRC_MACHINE = 'youtube' + # Listed in order of quality + _available_formats = ['38', '37', '46', '22', '45', '35', '44', '34', '18', '43', '6', '5', '17', '13'] + _available_formats_prefer_free = ['38', '46', '37', '45', '22', '44', '35', '43', '34', '18', '6', '5', '17', '13'] + _video_extensions = { + '13': '3gp', + '17': 'mp4', + '18': 'mp4', + '22': 'mp4', + '37': 'mp4', + '38': 'video', # You actually don't know if this will be MOV, AVI or whatever + '43': 'webm', + '44': 'webm', + '45': 'webm', + '46': 'webm', + } + _video_dimensions = { + '5': '240x400', + '6': '???', + '13': '???', + '17': '144x176', + '18': '360x640', + '22': '720x1280', + '34': '360x640', + '35': '480x854', + '37': '1080x1920', + '38': '3072x4096', + '43': '360x640', + '44': '480x854', + '45': '720x1280', + '46': '1080x1920', + } + IE_NAME = u'youtube' + + @classmethod + def suitable(cls, url): + """Receives a URL and returns True if suitable for this IE.""" + if YoutubePlaylistIE.suitable(url): return False + return re.match(cls._VALID_URL, url, re.VERBOSE) is not None + + def report_lang(self): + """Report attempt to set language.""" + self.to_screen(u'Setting language') + + def report_login(self): + """Report attempt to log in.""" + self.to_screen(u'Logging in') + + def report_video_webpage_download(self, video_id): + """Report attempt to download video webpage.""" + self.to_screen(u'%s: Downloading video webpage' % video_id) + + def report_video_info_webpage_download(self, video_id): + """Report attempt to download video info webpage.""" + self.to_screen(u'%s: Downloading video info webpage' % video_id) + + def report_video_subtitles_download(self, video_id): + """Report attempt to download video info webpage.""" + self.to_screen(u'%s: Checking available subtitles' % video_id) + + def report_video_subtitles_request(self, video_id, sub_lang, format): + """Report attempt to download video info webpage.""" + self.to_screen(u'%s: Downloading video subtitles for %s.%s' % (video_id, sub_lang, format)) + + def report_video_subtitles_available(self, video_id, sub_lang_list): + """Report available subtitles.""" + sub_lang = ",".join(list(sub_lang_list.keys())) + self.to_screen(u'%s: Available subtitles for video: %s' % (video_id, sub_lang)) + + def report_information_extraction(self, video_id): + """Report attempt to extract video information.""" + self.to_screen(u'%s: Extracting video information' % video_id) + + def report_unavailable_format(self, video_id, format): + """Report extracted video URL.""" + self.to_screen(u'%s: Format %s not available' % (video_id, format)) + + def report_rtmp_download(self): + """Indicate the download will use the RTMP protocol.""" + self.to_screen(u'RTMP download detected') + + def _get_available_subtitles(self, video_id): + self.report_video_subtitles_download(video_id) + request = compat_urllib_request.Request('http://video.google.com/timedtext?hl=en&type=list&v=%s' % video_id) + try: + sub_list = compat_urllib_request.urlopen(request).read().decode('utf-8') + except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: + return (u'unable to download video subtitles: %s' % compat_str(err), None) + sub_lang_list = re.findall(r'name="([^"]*)"[^>]+lang_code="([\w\-]+)"', sub_list) + sub_lang_list = dict((l[1], l[0]) for l in sub_lang_list) + if not sub_lang_list: + return (u'video doesn\'t have subtitles', None) + return sub_lang_list + + def _list_available_subtitles(self, video_id): + sub_lang_list = self._get_available_subtitles(video_id) + self.report_video_subtitles_available(video_id, sub_lang_list) + + def _request_subtitle(self, sub_lang, sub_name, video_id, format): + """ + Return tuple: + (error_message, sub_lang, sub) + """ + self.report_video_subtitles_request(video_id, sub_lang, format) + params = compat_urllib_parse.urlencode({ + 'lang': sub_lang, + 'name': sub_name, + 'v': video_id, + 'fmt': format, + }) + url = 'http://www.youtube.com/api/timedtext?' + params + try: + sub = compat_urllib_request.urlopen(url).read().decode('utf-8') + except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: + return (u'unable to download video subtitles: %s' % compat_str(err), None, None) + if not sub: + return (u'Did not fetch video subtitles', None, None) + return (None, sub_lang, sub) + + def _extract_subtitle(self, video_id): + """ + Return a list with a tuple: + [(error_message, sub_lang, sub)] + """ + sub_lang_list = self._get_available_subtitles(video_id) + sub_format = self._downloader.params.get('subtitlesformat') + if isinstance(sub_lang_list,tuple): #There was some error, it didn't get the available subtitles + return [(sub_lang_list[0], None, None)] + if self._downloader.params.get('subtitleslang', False): + sub_lang = self._downloader.params.get('subtitleslang') + elif 'en' in sub_lang_list: + sub_lang = 'en' + else: + sub_lang = list(sub_lang_list.keys())[0] + if not sub_lang in sub_lang_list: + return [(u'no closed captions found in the specified language "%s"' % sub_lang, None, None)] + + subtitle = self._request_subtitle(sub_lang, sub_lang_list[sub_lang].encode('utf-8'), video_id, sub_format) + return [subtitle] + + def _extract_all_subtitles(self, video_id): + sub_lang_list = self._get_available_subtitles(video_id) + sub_format = self._downloader.params.get('subtitlesformat') + if isinstance(sub_lang_list,tuple): #There was some error, it didn't get the available subtitles + return [(sub_lang_list[0], None, None)] + subtitles = [] + for sub_lang in sub_lang_list: + subtitle = self._request_subtitle(sub_lang, sub_lang_list[sub_lang].encode('utf-8'), video_id, sub_format) + subtitles.append(subtitle) + return subtitles + + def _print_formats(self, formats): + print('Available formats:') + for x in formats: + print('%s\t:\t%s\t[%s]' %(x, self._video_extensions.get(x, 'flv'), self._video_dimensions.get(x, '???'))) + + def _real_initialize(self): + if self._downloader is None: + return + + username = None + password = None + downloader_params = self._downloader.params + + # Attempt to use provided username and password or .netrc data + if downloader_params.get('username', None) is not None: + username = downloader_params['username'] + password = downloader_params['password'] + elif downloader_params.get('usenetrc', False): + try: + info = netrc.netrc().authenticators(self._NETRC_MACHINE) + if info is not None: + username = info[0] + password = info[2] + else: + raise netrc.NetrcParseError('No authenticators for %s' % self._NETRC_MACHINE) + except (IOError, netrc.NetrcParseError) as err: + self._downloader.report_warning(u'parsing .netrc: %s' % compat_str(err)) + return + + # Set language + request = compat_urllib_request.Request(self._LANG_URL) + try: + self.report_lang() + compat_urllib_request.urlopen(request).read() + except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: + self._downloader.report_warning(u'unable to set language: %s' % compat_str(err)) + return + + # No authentication to be performed + if username is None: + return + + request = compat_urllib_request.Request(self._LOGIN_URL) + try: + login_page = compat_urllib_request.urlopen(request).read().decode('utf-8') + except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: + self._downloader.report_warning(u'unable to fetch login page: %s' % compat_str(err)) + return + + galx = None + dsh = None + match = re.search(re.compile(r']* id="gaia_loginform"', login_results) is not None: + self._downloader.report_warning(u'unable to log in: bad username or password') + return + except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: + self._downloader.report_warning(u'unable to log in: %s' % compat_str(err)) + return + + # Confirm age + age_form = { + 'next_url': '/', + 'action_confirm': 'Confirm', + } + request = compat_urllib_request.Request(self._AGE_URL, compat_urllib_parse.urlencode(age_form)) + try: + self.report_age_confirmation() + age_results = compat_urllib_request.urlopen(request).read().decode('utf-8') + except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: + raise ExtractorError(u'Unable to confirm age: %s' % compat_str(err)) + + def _extract_id(self, url): + mobj = re.match(self._VALID_URL, url, re.VERBOSE) + if mobj is None: + raise ExtractorError(u'Invalid URL: %s' % url) + video_id = mobj.group(2) + return video_id + + def _real_extract(self, url): + # Extract original video URL from URL with redirection, like age verification, using next_url parameter + mobj = re.search(self._NEXT_URL_RE, url) + if mobj: + url = 'https://www.youtube.com/' + compat_urllib_parse.unquote(mobj.group(1)).lstrip('/') + video_id = self._extract_id(url) + + # Get video webpage + self.report_video_webpage_download(video_id) + url = 'https://www.youtube.com/watch?v=%s&gl=US&hl=en&has_verified=1' % video_id + request = compat_urllib_request.Request(url) + try: + video_webpage_bytes = compat_urllib_request.urlopen(request).read() + except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: + raise ExtractorError(u'Unable to download video webpage: %s' % compat_str(err)) + + video_webpage = video_webpage_bytes.decode('utf-8', 'ignore') + + # Attempt to extract SWF player URL + mobj = re.search(r'swfConfig.*?"(http:\\/\\/.*?watch.*?-.*?\.swf)"', video_webpage) + if mobj is not None: + player_url = re.sub(r'\\(.)', r'\1', mobj.group(1)) + else: + player_url = None + + # Get video info + self.report_video_info_webpage_download(video_id) + for el_type in ['&el=embedded', '&el=detailpage', '&el=vevo', '']: + video_info_url = ('https://www.youtube.com/get_video_info?&video_id=%s%s&ps=default&eurl=&gl=US&hl=en' + % (video_id, el_type)) + video_info_webpage = self._download_webpage(video_info_url, video_id, + note=False, + errnote='unable to download video info webpage') + video_info = compat_parse_qs(video_info_webpage) + if 'token' in video_info: + break + if 'token' not in video_info: + if 'reason' in video_info: + raise ExtractorError(u'YouTube said: %s' % video_info['reason'][0]) + else: + raise ExtractorError(u'"token" parameter not in video info for unknown reason') + + # Check for "rental" videos + if 'ypc_video_rental_bar_text' in video_info and 'author' not in video_info: + raise ExtractorError(u'"rental" videos not supported') + + # Start extracting information + self.report_information_extraction(video_id) + + # uploader + if 'author' not in video_info: + raise ExtractorError(u'Unable to extract uploader name') + video_uploader = compat_urllib_parse.unquote_plus(video_info['author'][0]) + + # uploader_id + video_uploader_id = None + mobj = re.search(r'', video_webpage) + if mobj is not None: + video_uploader_id = mobj.group(1) + else: + self._downloader.report_warning(u'unable to extract uploader nickname') + + # title + if 'title' not in video_info: + raise ExtractorError(u'Unable to extract video title') + video_title = compat_urllib_parse.unquote_plus(video_info['title'][0]) + + # thumbnail image + if 'thumbnail_url' not in video_info: + self._downloader.report_warning(u'unable to extract video thumbnail') + video_thumbnail = '' + else: # don't panic if we can't find it + video_thumbnail = compat_urllib_parse.unquote_plus(video_info['thumbnail_url'][0]) + + # upload date + upload_date = None + mobj = re.search(r'id="eow-date.*?>(.*?)', video_webpage, re.DOTALL) + if mobj is not None: + upload_date = ' '.join(re.sub(r'[/,-]', r' ', mobj.group(1)).split()) + upload_date = unified_strdate(upload_date) + + # description + video_description = get_element_by_id("eow-description", video_webpage) + if video_description: + video_description = clean_html(video_description) + else: + fd_mobj = re.search(r'= 1: + url_map = {} + for url_data_str in video_info['url_encoded_fmt_stream_map'][0].split(','): + url_data = compat_parse_qs(url_data_str) + if 'itag' in url_data and 'url' in url_data: + url = url_data['url'][0] + '&signature=' + url_data['sig'][0] + if not 'ratebypass' in url: url += '&ratebypass=yes' + url_map[url_data['itag'][0]] = url + + format_limit = self._downloader.params.get('format_limit', None) + available_formats = self._available_formats_prefer_free if self._downloader.params.get('prefer_free_formats', False) else self._available_formats + if format_limit is not None and format_limit in available_formats: + format_list = available_formats[available_formats.index(format_limit):] + else: + format_list = available_formats + existing_formats = [x for x in format_list if x in url_map] + if len(existing_formats) == 0: + raise ExtractorError(u'no known formats available for video') + if self._downloader.params.get('listformats', None): + self._print_formats(existing_formats) + return + if req_format is None or req_format == 'best': + video_url_list = [(existing_formats[0], url_map[existing_formats[0]])] # Best quality + elif req_format == 'worst': + video_url_list = [(existing_formats[len(existing_formats)-1], url_map[existing_formats[len(existing_formats)-1]])] # worst quality + elif req_format in ('-1', 'all'): + video_url_list = [(f, url_map[f]) for f in existing_formats] # All formats + else: + # Specific formats. We pick the first in a slash-delimeted sequence. + # For example, if '1/2/3/4' is requested and '2' and '4' are available, we pick '2'. + req_formats = req_format.split('/') + video_url_list = None + for rf in req_formats: + if rf in url_map: + video_url_list = [(rf, url_map[rf])] + break + if video_url_list is None: + raise ExtractorError(u'requested format not available') + else: + raise ExtractorError(u'no conn or url_encoded_fmt_stream_map information found in video info') + + results = [] + for format_param, video_real_url in video_url_list: + # Extension + video_extension = self._video_extensions.get(format_param, 'flv') + + video_format = '{0} - {1}'.format(format_param if format_param else video_extension, + self._video_dimensions.get(format_param, '???')) + + results.append({ + 'id': video_id, + 'url': video_real_url, + 'uploader': video_uploader, + 'uploader_id': video_uploader_id, + 'upload_date': upload_date, + 'title': video_title, + 'ext': video_extension, + 'format': video_format, + 'thumbnail': video_thumbnail, + 'description': video_description, + 'player_url': player_url, + 'subtitles': video_subtitles, + 'duration': video_duration + }) + return results + + +class YoutubeSearchIE(SearchInfoExtractor): + """Information Extractor for YouTube search queries.""" + _API_URL = 'https://gdata.youtube.com/feeds/api/videos?q=%s&start-index=%i&max-results=50&v=2&alt=jsonc' + _MAX_RESULTS = 1000 + IE_NAME = u'youtube:search' + _SEARCH_KEY = 'ytsearch' + + def report_download_page(self, query, pagenum): + """Report attempt to download search page with given number.""" + query = query.decode(preferredencoding()) + self._downloader.to_screen(u'[youtube] query "%s": Downloading page %s' % (query, pagenum)) + + def _get_n_results(self, query, n): + """Get a specified number of results for a query""" + + video_ids = [] + pagenum = 0 + limit = n + + while (50 * pagenum) < limit: + self.report_download_page(query, pagenum+1) + result_url = self._API_URL % (compat_urllib_parse.quote_plus(query), (50*pagenum)+1) + request = compat_urllib_request.Request(result_url) + try: + data = compat_urllib_request.urlopen(request).read().decode('utf-8') + except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: + raise ExtractorError(u'Unable to download API page: %s' % compat_str(err)) + api_response = json.loads(data)['data'] + + if not 'items' in api_response: + raise ExtractorError(u'[youtube] No video results') + + new_ids = list(video['id'] for video in api_response['items']) + video_ids += new_ids + + limit = min(n, api_response['totalItems']) + pagenum += 1 + + if len(video_ids) > n: + video_ids = video_ids[:n] + videos = [self.url_result('http://www.youtube.com/watch?v=%s' % id, 'Youtube') for id in video_ids] + return self.playlist_result(videos, query) + + +class YoutubePlaylistIE(InfoExtractor): + """Information Extractor for YouTube playlists.""" + + _VALID_URL = r"""(?: + (?:https?://)? + (?:\w+\.)? + youtube\.com/ + (?: + (?:course|view_play_list|my_playlists|artist|playlist|watch) + \? (?:.*?&)*? (?:p|a|list)= + | p/ + ) + ((?:PL|EC|UU)?[0-9A-Za-z-_]{10,}) + .* + | + ((?:PL|EC|UU)[0-9A-Za-z-_]{10,}) + )""" + _TEMPLATE_URL = 'https://gdata.youtube.com/feeds/api/playlists/%s?max-results=%i&start-index=%i&v=2&alt=json' + _MAX_RESULTS = 50 + IE_NAME = u'youtube:playlist' + + @classmethod + def suitable(cls, url): + """Receives a URL and returns True if suitable for this IE.""" + return re.match(cls._VALID_URL, url, re.VERBOSE) is not None + + def _real_extract(self, url): + # Extract playlist id + mobj = re.match(self._VALID_URL, url, re.VERBOSE) + if mobj is None: + raise ExtractorError(u'Invalid URL: %s' % url) + + # Download playlist videos from API + playlist_id = mobj.group(1) or mobj.group(2) + page_num = 1 + videos = [] + + while True: + url = self._TEMPLATE_URL % (playlist_id, self._MAX_RESULTS, self._MAX_RESULTS * (page_num - 1) + 1) + page = self._download_webpage(url, playlist_id, u'Downloading page #%s' % page_num) + + try: + response = json.loads(page) + except ValueError as err: + raise ExtractorError(u'Invalid JSON in API response: ' + compat_str(err)) + + if 'feed' not in response: + raise ExtractorError(u'Got a malformed response from YouTube API') + playlist_title = response['feed']['title']['$t'] + if 'entry' not in response['feed']: + # Number of videos is a multiple of self._MAX_RESULTS + break + + videos += [ (entry['yt$position']['$t'], entry['content']['src']) + for entry in response['feed']['entry'] + if 'content' in entry ] + + if len(response['feed']['entry']) < self._MAX_RESULTS: + break + page_num += 1 + + videos = [v[1] for v in sorted(videos)] + + url_results = [self.url_result(url, 'Youtube') for url in videos] + return [self.playlist_result(url_results, playlist_id, playlist_title)] + + +class YoutubeChannelIE(InfoExtractor): + """Information Extractor for YouTube channels.""" + + _VALID_URL = r"^(?:https?://)?(?:youtu\.be|(?:\w+\.)?youtube(?:-nocookie)?\.com)/channel/([0-9A-Za-z_-]+)" + _TEMPLATE_URL = 'http://www.youtube.com/channel/%s/videos?sort=da&flow=list&view=0&page=%s&gl=US&hl=en' + _MORE_PAGES_INDICATOR = 'yt-uix-load-more' + _MORE_PAGES_URL = 'http://www.youtube.com/channel_ajax?action_load_more_videos=1&flow=list&paging=%s&view=0&sort=da&channel_id=%s' + IE_NAME = u'youtube:channel' + + def extract_videos_from_page(self, page): + ids_in_page = [] + for mobj in re.finditer(r'href="/watch\?v=([0-9A-Za-z_-]+)&?', page): + if mobj.group(1) not in ids_in_page: + ids_in_page.append(mobj.group(1)) + return ids_in_page + + def _real_extract(self, url): + # Extract channel id + mobj = re.match(self._VALID_URL, url) + if mobj is None: + raise ExtractorError(u'Invalid URL: %s' % url) + + # Download channel page + channel_id = mobj.group(1) + video_ids = [] + pagenum = 1 + + url = self._TEMPLATE_URL % (channel_id, pagenum) + page = self._download_webpage(url, channel_id, + u'Downloading page #%s' % pagenum) + + # Extract video identifiers + ids_in_page = self.extract_videos_from_page(page) + video_ids.extend(ids_in_page) + + # Download any subsequent channel pages using the json-based channel_ajax query + if self._MORE_PAGES_INDICATOR in page: + while True: + pagenum = pagenum + 1 + + url = self._MORE_PAGES_URL % (pagenum, channel_id) + page = self._download_webpage(url, channel_id, + u'Downloading page #%s' % pagenum) + + page = json.loads(page) + + ids_in_page = self.extract_videos_from_page(page['content_html']) + video_ids.extend(ids_in_page) + + if self._MORE_PAGES_INDICATOR not in page['load_more_widget_html']: + break + + self._downloader.to_screen(u'[youtube] Channel %s: Found %i videos' % (channel_id, len(video_ids))) + + urls = ['http://www.youtube.com/watch?v=%s' % id for id in video_ids] + url_entries = [self.url_result(url, 'Youtube') for url in urls] + return [self.playlist_result(url_entries, channel_id)] + + +class YoutubeUserIE(InfoExtractor): + """Information Extractor for YouTube users.""" + + _VALID_URL = r'(?:(?:(?:https?://)?(?:\w+\.)?youtube\.com/user/)|ytuser:)([A-Za-z0-9_-]+)' + _TEMPLATE_URL = 'http://gdata.youtube.com/feeds/api/users/%s' + _GDATA_PAGE_SIZE = 50 + _GDATA_URL = 'http://gdata.youtube.com/feeds/api/users/%s/uploads?max-results=%d&start-index=%d' + _VIDEO_INDICATOR = r'/watch\?v=(.+?)[\<&]' + IE_NAME = u'youtube:user' + + def _real_extract(self, url): + # Extract username + mobj = re.match(self._VALID_URL, url) + if mobj is None: + raise ExtractorError(u'Invalid URL: %s' % url) + + username = mobj.group(1) + + # Download video ids using YouTube Data API. Result size per + # query is limited (currently to 50 videos) so we need to query + # page by page until there are no video ids - it means we got + # all of them. + + video_ids = [] + pagenum = 0 + + while True: + start_index = pagenum * self._GDATA_PAGE_SIZE + 1 + + gdata_url = self._GDATA_URL % (username, self._GDATA_PAGE_SIZE, start_index) + page = self._download_webpage(gdata_url, username, + u'Downloading video ids from %d to %d' % (start_index, start_index + self._GDATA_PAGE_SIZE)) + + # Extract video identifiers + ids_in_page = [] + + for mobj in re.finditer(self._VIDEO_INDICATOR, page): + if mobj.group(1) not in ids_in_page: + ids_in_page.append(mobj.group(1)) + + video_ids.extend(ids_in_page) + + # A little optimization - if current page is not + # "full", ie. does not contain PAGE_SIZE video ids then + # we can assume that this page is the last one - there + # are no more ids on further pages - no need to query + # again. + + if len(ids_in_page) < self._GDATA_PAGE_SIZE: + break + + pagenum += 1 + + urls = ['http://www.youtube.com/watch?v=%s' % video_id for video_id in video_ids] + url_results = [self.url_result(url, 'Youtube') for url in urls] + return [self.playlist_result(url_results, playlist_title = username)] diff --git a/youtube_dl/InfoExtractors/__init__.py b/youtube_dl/InfoExtractors/__init__.py new file mode 100755 index 000000000..70b667818 --- /dev/null +++ b/youtube_dl/InfoExtractors/__init__.py @@ -0,0 +1,28 @@ +#!/usr/bin/env python +# -*- coding: utf-8 -*- + +from __future__ import absolute_import + + +IEs = [ + ('Youtube', ['YoutubePlaylistIE', 'YoutubeChannelIE', + 'YoutubeUserIE', 'YoutubeSearchIE', 'YoutubeIE']), + ('Generic', ['GenericIE']), + ] + + +IE_list = [] +for module, IE_names in IEs: + _mod = __import__('youtube_dl.InfoExtractors.' + module, + globals(), locals(), IE_names) + for IE_name in IE_names: + IE_list.append(getattr(_mod, IE_name)) + + +def gen_extractors(): + """ Return a list of an instance of every supported extractor """ + return [ IE() for IE in IE_list ] + +def get_info_extractor(IE_name): + """Returns the info extractor class with the given ie_name""" + return next(IE for IE in IE_list if IE.__name__ == IE_name + 'IE') From 1452a3a9593bb980844f0124777b4e42234bf882 Mon Sep 17 00:00:00 2001 From: Filippo Valsorda Date: Wed, 5 Jun 2013 12:18:14 +0200 Subject: [PATCH 2/5] edit __init__.py to handle suggestions --- youtube_dl/InfoExtractors/__init__.py | 10 ++++++---- 1 file changed, 6 insertions(+), 4 deletions(-) diff --git a/youtube_dl/InfoExtractors/__init__.py b/youtube_dl/InfoExtractors/__init__.py index 70b667818..c8317bf96 100755 --- a/youtube_dl/InfoExtractors/__init__.py +++ b/youtube_dl/InfoExtractors/__init__.py @@ -12,11 +12,13 @@ IEs = [ IE_list = [] +IE_dict = {} for module, IE_names in IEs: - _mod = __import__('youtube_dl.InfoExtractors.' + module, - globals(), locals(), IE_names) + _mod = __import__('youtube_dl.InfoExtractors.' + module, globals(), IE_names) for IE_name in IE_names: - IE_list.append(getattr(_mod, IE_name)) + IE = getattr(_mod, IE_name) + IE_list.append(IE) + IE_dict[IE_name] = IE def gen_extractors(): @@ -25,4 +27,4 @@ def gen_extractors(): def get_info_extractor(IE_name): """Returns the info extractor class with the given ie_name""" - return next(IE for IE in IE_list if IE.__name__ == IE_name + 'IE') + return IE_dict[IE_name] From aec82b2b2276433a98cb6e670a218813aed94425 Mon Sep 17 00:00:00 2001 From: Filippo Valsorda Date: Wed, 5 Jun 2013 12:22:23 +0200 Subject: [PATCH 3/5] still know Python? --- youtube_dl/InfoExtractors/__init__.py | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/youtube_dl/InfoExtractors/__init__.py b/youtube_dl/InfoExtractors/__init__.py index c8317bf96..0657d42f2 100755 --- a/youtube_dl/InfoExtractors/__init__.py +++ b/youtube_dl/InfoExtractors/__init__.py @@ -14,7 +14,8 @@ IEs = [ IE_list = [] IE_dict = {} for module, IE_names in IEs: - _mod = __import__('youtube_dl.InfoExtractors.' + module, globals(), IE_names) + _mod = __import__('youtube_dl.InfoExtractors.' + module, + globals(), fromlist=IE_names) for IE_name in IE_names: IE = getattr(_mod, IE_name) IE_list.append(IE) From 238d0b87baba730fee5ce642350b6cc7b4a0a937 Mon Sep 17 00:00:00 2001 From: Filippo Valsorda Date: Wed, 5 Jun 2013 12:26:14 +0200 Subject: [PATCH 4/5] Travis is making me lazy to run tests --- youtube_dl/InfoExtractors/__init__.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/youtube_dl/InfoExtractors/__init__.py b/youtube_dl/InfoExtractors/__init__.py index 0657d42f2..d84e37e8a 100755 --- a/youtube_dl/InfoExtractors/__init__.py +++ b/youtube_dl/InfoExtractors/__init__.py @@ -28,4 +28,4 @@ def gen_extractors(): def get_info_extractor(IE_name): """Returns the info extractor class with the given ie_name""" - return IE_dict[IE_name] + return IE_dict[IE_name + 'IE'] From 5a5f27e7a2f623ada4afde8b090b41d5db2db72c Mon Sep 17 00:00:00 2001 From: Filippo Valsorda Date: Wed, 5 Jun 2013 12:50:29 +0200 Subject: [PATCH 5/5] Accept simple strings in IEs --- youtube_dl/InfoExtractors/__init__.py | 17 ++++++++++++++--- 1 file changed, 14 insertions(+), 3 deletions(-) diff --git a/youtube_dl/InfoExtractors/__init__.py b/youtube_dl/InfoExtractors/__init__.py index d84e37e8a..1e0d49b10 100755 --- a/youtube_dl/InfoExtractors/__init__.py +++ b/youtube_dl/InfoExtractors/__init__.py @@ -3,19 +3,30 @@ from __future__ import absolute_import - +""" +List here the InfoExtractors. +The order matter! URLs will be handled by the first IE to match. +List either the name of the file, like 'Foo', containing 'FooIE' or +a tuple like (FILENAME, [IE_name, IE_name, ...]) +""" IEs = [ ('Youtube', ['YoutubePlaylistIE', 'YoutubeChannelIE', 'YoutubeUserIE', 'YoutubeSearchIE', 'YoutubeIE']), - ('Generic', ['GenericIE']), + 'Generic', ] IE_list = [] IE_dict = {} -for module, IE_names in IEs: +for entry in IEs: + if type(entry) == type(''): + module, IE_names = entry, [entry + 'IE'] + else: + module, IE_names = entry + _mod = __import__('youtube_dl.InfoExtractors.' + module, globals(), fromlist=IE_names) + for IE_name in IE_names: IE = getattr(_mod, IE_name) IE_list.append(IE)