track custom youtube-dl distribution

youtube_dl/extractor/__init__.py

@@ -0,0 +1,46 @@
+from __future__ import unicode_literals
+
+try:
+    from .lazy_extractors import *
+    from .lazy_extractors import _ALL_CLASSES
+    _LAZY_LOADER = True
+except ImportError:
+    _LAZY_LOADER = False
+    from .extractors import *
+
+    _ALL_CLASSES = [
+        klass
+        for name, klass in globals().items()
+        if name.endswith('IE') and name != 'GenericIE'
+    ]
+    #_ALL_CLASSES.append(GenericIE)
+
+
+def gen_extractor_classes():
+    """ Return a list of supported extractors.
+    The order does matter; the first extractor matched is the one handling the URL.
+    """
+    return _ALL_CLASSES
+
+
+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 [klass() for klass in gen_extractor_classes()]
+
+
+def list_extractors(age_limit):
+    """
+    Return a list of extractors that are suitable for the given age,
+    sorted by extractor ID.
+    """
+
+    return sorted(
+        filter(lambda ie: ie.is_suitable(age_limit), gen_extractors()),
+        key=lambda ie: ie.IE_NAME.lower())
+
+
+def get_info_extractor(ie_name):
+    """Returns the info extractor class with the given ie_name"""
+    return globals()[ie_name + 'IE']

youtube_dl/extractor/commonmistakes.py

@@ -0,0 +1,50 @@
+from __future__ import unicode_literals
+
+import sys
+
+from .common import InfoExtractor
+from ..utils import ExtractorError
+
+
+class CommonMistakesIE(InfoExtractor):
+    IE_DESC = False  # Do not list
+    _VALID_URL = r'''(?x)
+        (?:url|URL)$
+    '''
+
+    _TESTS = [{
+        'url': 'url',
+        'only_matching': True,
+    }, {
+        'url': 'URL',
+        'only_matching': True,
+    }]
+
+    def _real_extract(self, url):
+        msg = (
+            'You\'ve asked youtube-dl to download the URL "%s". '
+            'That doesn\'t make any sense. '
+            'Simply remove the parameter in your command or configuration.'
+        ) % url
+        if not self._downloader.params.get('verbose'):
+            msg += ' Add -v to the command line to see what arguments and configuration youtube-dl got.'
+        raise ExtractorError(msg, expected=True)
+
+
+class UnicodeBOMIE(InfoExtractor):
+        IE_DESC = False
+        _VALID_URL = r'(?P<bom>\ufeff)(?P<id>.*)$'
+
+        # Disable test for python 3.2 since BOM is broken in re in this version
+        # (see https://github.com/rg3/youtube-dl/issues/9751)
+        _TESTS = [] if (3, 0) < sys.version_info <= (3, 3) else [{
+            'url': '\ufeffhttp://www.youtube.com/watch?v=BaW_jenozKc',
+            'only_matching': True,
+        }]
+
+        def _real_extract(self, url):
+            real_url = self._match_id(url)
+            self.report_warning(
+                'Your URL starts with a Byte Order Mark (BOM). '
+                'Removing the BOM and looking for "%s" ...' % real_url)
+            return self.url_result(real_url)

youtube_dl/extractor/commonprotocols.py

@@ -0,0 +1,60 @@
+from __future__ import unicode_literals
+
+from .common import InfoExtractor
+from ..compat import (
+    compat_urlparse,
+)
+
+
+class RtmpIE(InfoExtractor):
+    IE_DESC = False  # Do not list
+    _VALID_URL = r'(?i)rtmp[est]?://.+'
+
+    _TESTS = [{
+        'url': 'rtmp://cp44293.edgefcs.net/ondemand?auth=daEcTdydfdqcsb8cZcDbAaCbhamacbbawaS-bw7dBb-bWG-GqpGFqCpNCnGoyL&aifp=v001&slist=public/unsecure/audio/2c97899446428e4301471a8cb72b4b97--audio--pmg-20110908-0900a_flv_aac_med_int.mp4',
+        'only_matching': True,
+    }, {
+        'url': 'rtmp://edge.live.hitbox.tv/live/dimak',
+        'only_matching': True,
+    }]
+
+    def _real_extract(self, url):
+        video_id = self._generic_id(url)
+        title = self._generic_title(url)
+        return {
+            'id': video_id,
+            'title': title,
+            'formats': [{
+                'url': url,
+                'ext': 'flv',
+                'format_id': compat_urlparse.urlparse(url).scheme,
+            }],
+        }
+
+
+class MmsIE(InfoExtractor):
+    IE_DESC = False  # Do not list
+    _VALID_URL = r'(?i)mms://.+'
+
+    _TEST = {
+        # Direct MMS link
+        'url': 'mms://kentro.kaist.ac.kr/200907/MilesReid(0709).wmv',
+        'info_dict': {
+            'id': 'MilesReid(0709)',
+            'ext': 'wmv',
+            'title': 'MilesReid(0709)',
+        },
+        'params': {
+            'skip_download': True,  # rtsp downloads, requiring mplayer or mpv
+        },
+    }
+
+    def _real_extract(self, url):
+        video_id = self._generic_id(url)
+        title = self._generic_title(url)
+
+        return {
+            'id': video_id,
+            'title': title,
+            'url': url,
+        }

youtube_dl/extractor/extractors.py

@@ -0,0 +1,31 @@
+# flake8: noqa
+from __future__ import unicode_literals
+
+
+from .commonmistakes import CommonMistakesIE, UnicodeBOMIE
+from .commonprotocols import (
+    MmsIE,
+    RtmpIE,
+)
+
+from .openload import OpenloadIE
+
+from .youtube import (
+    YoutubeIE,
+    YoutubeChannelIE,
+    YoutubeFavouritesIE,
+    YoutubeHistoryIE,
+    YoutubeLiveIE,
+    YoutubePlaylistIE,
+    YoutubePlaylistsIE,
+    YoutubeRecommendedIE,
+    YoutubeSearchDateIE,
+    YoutubeSearchIE,
+    YoutubeSearchURLIE,
+    YoutubeShowIE,
+    YoutubeSubscriptionsIE,
+    YoutubeTruncatedIDIE,
+    YoutubeTruncatedURLIE,
+    YoutubeUserIE,
+    YoutubeWatchLaterIE,
+)

youtube_dl/extractor/openload.py

@@ -0,0 +1,379 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import json
+import os
+import re
+import subprocess
+import tempfile
+
+from .common import InfoExtractor
+from ..compat import (
+    compat_urlparse,
+    compat_kwargs,
+)
+from ..utils import (
+    check_executable,
+    determine_ext,
+    encodeArgument,
+    ExtractorError,
+    get_element_by_id,
+    get_exe_version,
+    is_outdated_version,
+    std_headers,
+)
+
+
+def cookie_to_dict(cookie):
+    cookie_dict = {
+        'name': cookie.name,
+        'value': cookie.value,
+    }
+    if cookie.port_specified:
+        cookie_dict['port'] = cookie.port
+    if cookie.domain_specified:
+        cookie_dict['domain'] = cookie.domain
+    if cookie.path_specified:
+        cookie_dict['path'] = cookie.path
+    if cookie.expires is not None:
+        cookie_dict['expires'] = cookie.expires
+    if cookie.secure is not None:
+        cookie_dict['secure'] = cookie.secure
+    if cookie.discard is not None:
+        cookie_dict['discard'] = cookie.discard
+    try:
+        if (cookie.has_nonstandard_attr('httpOnly') or
+                cookie.has_nonstandard_attr('httponly') or
+                cookie.has_nonstandard_attr('HttpOnly')):
+            cookie_dict['httponly'] = True
+    except TypeError:
+        pass
+    return cookie_dict
+
+
+def cookie_jar_to_list(cookie_jar):
+    return [cookie_to_dict(cookie) for cookie in cookie_jar]
+
+
+class PhantomJSwrapper(object):
+    """PhantomJS wrapper class
+
+    This class is experimental.
+    """
+
+    _TEMPLATE = r'''
+        phantom.onError = function(msg, trace) {{
+          var msgStack = ['PHANTOM ERROR: ' + msg];
+          if(trace && trace.length) {{
+            msgStack.push('TRACE:');
+            trace.forEach(function(t) {{
+              msgStack.push(' -> ' + (t.file || t.sourceURL) + ': ' + t.line
+                + (t.function ? ' (in function ' + t.function +')' : ''));
+            }});
+          }}
+          console.error(msgStack.join('\n'));
+          phantom.exit(1);
+        }};
+        var page = require('webpage').create();
+        var fs = require('fs');
+        var read = {{ mode: 'r', charset: 'utf-8' }};
+        var write = {{ mode: 'w', charset: 'utf-8' }};
+        JSON.parse(fs.read("{cookies}", read)).forEach(function(x) {{
+          phantom.addCookie(x);
+        }});
+        page.settings.resourceTimeout = {timeout};
+        page.settings.userAgent = "{ua}";
+        page.onLoadStarted = function() {{
+          page.evaluate(function() {{
+            delete window._phantom;
+            delete window.callPhantom;
+          }});
+        }};
+        var saveAndExit = function() {{
+          fs.write("{html}", page.content, write);
+          fs.write("{cookies}", JSON.stringify(phantom.cookies), write);
+          phantom.exit();
+        }};
+        page.onLoadFinished = function(status) {{
+          if(page.url === "") {{
+            page.setContent(fs.read("{html}", read), "{url}");
+          }}
+          else {{
+            {jscode}
+          }}
+        }};
+        page.open("");
+    '''
+
+    _TMP_FILE_NAMES = ['script', 'html', 'cookies']
+
+    @staticmethod
+    def _version():
+        return get_exe_version('phantomjs', version_re=r'([0-9.]+)')
+
+    def __init__(self, extractor, required_version=None, timeout=10000):
+        self._TMP_FILES = {}
+
+        self.exe = check_executable('phantomjs', ['-v'])
+        if not self.exe:
+            raise ExtractorError('PhantomJS executable not found in PATH, '
+                                 'download it from http://phantomjs.org',
+                                 expected=True)
+
+        self.extractor = extractor
+
+        if required_version:
+            version = self._version()
+            if is_outdated_version(version, required_version):
+                self.extractor._downloader.report_warning(
+                    'Your copy of PhantomJS is outdated, update it to version '
+                    '%s or newer if you encounter any errors.' % required_version)
+
+        self.options = {
+            'timeout': timeout,
+        }
+        for name in self._TMP_FILE_NAMES:
+            tmp = tempfile.NamedTemporaryFile(delete=False)
+            tmp.close()
+            self._TMP_FILES[name] = tmp
+
+    def __del__(self):
+        for name in self._TMP_FILE_NAMES:
+            try:
+                os.remove(self._TMP_FILES[name].name)
+            except (IOError, OSError, KeyError):
+                pass
+
+    def _save_cookies(self, url):
+        cookies = cookie_jar_to_list(self.extractor._downloader.cookiejar)
+        for cookie in cookies:
+            if 'path' not in cookie:
+                cookie['path'] = '/'
+            if 'domain' not in cookie:
+                cookie['domain'] = compat_urlparse.urlparse(url).netloc
+        with open(self._TMP_FILES['cookies'].name, 'wb') as f:
+            f.write(json.dumps(cookies).encode('utf-8'))
+
+    def _load_cookies(self):
+        with open(self._TMP_FILES['cookies'].name, 'rb') as f:
+            cookies = json.loads(f.read().decode('utf-8'))
+        for cookie in cookies:
+            if cookie['httponly'] is True:
+                cookie['rest'] = {'httpOnly': None}
+            if 'expiry' in cookie:
+                cookie['expire_time'] = cookie['expiry']
+            self.extractor._set_cookie(**compat_kwargs(cookie))
+
+    def get(self, url, html=None, video_id=None, note=None, note2='Executing JS on webpage', headers={}, jscode='saveAndExit();'):
+        """
+        Downloads webpage (if needed) and executes JS
+
+        Params:
+            url: website url
+            html: optional, html code of website
+            video_id: video id
+            note: optional, displayed when downloading webpage
+            note2: optional, displayed when executing JS
+            headers: custom http headers
+            jscode: code to be executed when page is loaded
+
+        Returns tuple with:
+            * downloaded website (after JS execution)
+            * anything you print with `console.log` (but not inside `page.execute`!)
+
+        In most cases you don't need to add any `jscode`.
+        It is executed in `page.onLoadFinished`.
+        `saveAndExit();` is mandatory, use it instead of `phantom.exit()`
+        It is possible to wait for some element on the webpage, for example:
+            var check = function() {
+              var elementFound = page.evaluate(function() {
+                return document.querySelector('#b.done') !== null;
+              });
+              if(elementFound)
+                saveAndExit();
+              else
+                window.setTimeout(check, 500);
+            }
+
+            page.evaluate(function(){
+              document.querySelector('#a').click();
+            });
+            check();
+        """
+        if 'saveAndExit();' not in jscode:
+            raise ExtractorError('`saveAndExit();` not found in `jscode`')
+        if not html:
+            html = self.extractor._download_webpage(url, video_id, note=note, headers=headers)
+        with open(self._TMP_FILES['html'].name, 'wb') as f:
+            f.write(html.encode('utf-8'))
+
+        self._save_cookies(url)
+
+        replaces = self.options
+        replaces['url'] = url
+        user_agent = headers.get('User-Agent') or std_headers['User-Agent']
+        replaces['ua'] = user_agent.replace('"', '\\"')
+        replaces['jscode'] = jscode
+
+        for x in self._TMP_FILE_NAMES:
+            replaces[x] = self._TMP_FILES[x].name.replace('\\', '\\\\').replace('"', '\\"')
+
+        with open(self._TMP_FILES['script'].name, 'wb') as f:
+            f.write(self._TEMPLATE.format(**replaces).encode('utf-8'))
+
+        if video_id is None:
+            self.extractor.to_screen('%s' % (note2,))
+        else:
+            self.extractor.to_screen('%s: %s' % (video_id, note2))
+
+        p = subprocess.Popen([
+            self.exe, '--ssl-protocol=any',
+            self._TMP_FILES['script'].name
+        ], stdout=subprocess.PIPE, stderr=subprocess.PIPE)
+        out, err = p.communicate()
+        if p.returncode != 0:
+            raise ExtractorError(
+                'Executing JS failed\n:' + encodeArgument(err))
+        with open(self._TMP_FILES['html'].name, 'rb') as f:
+            html = f.read().decode('utf-8')
+
+        self._load_cookies()
+
+        return (html, encodeArgument(out))
+
+
+class OpenloadIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?(?:openload\.(?:co|io|link)|oload\.(?:tv|stream|site|xyz|win|download))/(?:f|embed)/(?P<id>[a-zA-Z0-9-_]+)'
+
+    _TESTS = [{
+        'url': 'https://openload.co/f/kUEfGclsU9o',
+        'md5': 'bf1c059b004ebc7a256f89408e65c36e',
+        'info_dict': {
+            'id': 'kUEfGclsU9o',
+            'ext': 'mp4',
+            'title': 'skyrim_no-audio_1080.mp4',
+            'thumbnail': r're:^https?://.*\.jpg$',
+        },
+    }, {
+        'url': 'https://openload.co/embed/rjC09fkPLYs',
+        'info_dict': {
+            'id': 'rjC09fkPLYs',
+            'ext': 'mp4',
+            'title': 'movie.mp4',
+            'thumbnail': r're:^https?://.*\.jpg$',
+            'subtitles': {
+                'en': [{
+                    'ext': 'vtt',
+                }],
+            },
+        },
+        'params': {
+            'skip_download': True,  # test subtitles only
+        },
+    }, {
+        'url': 'https://openload.co/embed/kUEfGclsU9o/skyrim_no-audio_1080.mp4',
+        'only_matching': True,
+    }, {
+        'url': 'https://openload.io/f/ZAn6oz-VZGE/',
+        'only_matching': True,
+    }, {
+        'url': 'https://openload.co/f/_-ztPaZtMhM/',
+        'only_matching': True,
+    }, {
+        # unavailable via https://openload.co/f/Sxz5sADo82g/, different layout
+        # for title and ext
+        'url': 'https://openload.co/embed/Sxz5sADo82g/',
+        'only_matching': True,
+    }, {
+        # unavailable via https://openload.co/embed/e-Ixz9ZR5L0/ but available
+        # via https://openload.co/f/e-Ixz9ZR5L0/
+        'url': 'https://openload.co/f/e-Ixz9ZR5L0/',
+        'only_matching': True,
+    }, {
+        'url': 'https://oload.tv/embed/KnG-kKZdcfY/',
+        'only_matching': True,
+    }, {
+        'url': 'http://www.openload.link/f/KnG-kKZdcfY',
+        'only_matching': True,
+    }, {
+        'url': 'https://oload.stream/f/KnG-kKZdcfY',
+        'only_matching': True,
+    }, {
+        'url': 'https://oload.xyz/f/WwRBpzW8Wtk',
+        'only_matching': True,
+    }, {
+        'url': 'https://oload.win/f/kUEfGclsU9o',
+        'only_matching': True,
+    }, {
+        'url': 'https://oload.download/f/kUEfGclsU9o',
+        'only_matching': True,
+    }, {
+        # Its title has not got its extension but url has it
+        'url': 'https://oload.download/f/N4Otkw39VCw/Tomb.Raider.2018.HDRip.XviD.AC3-EVO.avi.mp4',
+        'only_matching': True,
+    }]
+
+    _USER_AGENT = 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/56.0.2924.87 Safari/537.36'
+
+    @staticmethod
+    def _extract_urls(webpage):
+        return re.findall(
+            r'<iframe[^>]+src=["\']((?:https?://)?(?:openload\.(?:co|io)|oload\.tv)/embed/[a-zA-Z0-9-_]+)',
+            webpage)
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+        url_pattern = 'https://openload.co/%%s/%s/' % video_id
+        headers = {
+            'User-Agent': self._USER_AGENT,
+        }
+
+        for path in ('embed', 'f'):
+            page_url = url_pattern % path
+            last = path == 'f'
+            webpage = self._download_webpage(
+                page_url, video_id, 'Downloading %s webpage' % path,
+                headers=headers, fatal=last)
+            if not webpage:
+                continue
+            if 'File not found' in webpage or 'deleted by the owner' in webpage:
+                if not last:
+                    continue
+                raise ExtractorError('File not found', expected=True, video_id=video_id)
+            break
+
+        phantom = PhantomJSwrapper(self, required_version='2.0')
+        webpage, _ = phantom.get(page_url, html=webpage, video_id=video_id, headers=headers)
+
+        decoded_id = (get_element_by_id('streamurl', webpage) or
+                      get_element_by_id('streamuri', webpage) or
+                      get_element_by_id('streamurj', webpage) or
+                      self._search_regex(
+                          (r'>\s*([\w-]+~\d{10,}~\d+\.\d+\.0\.0~[\w-]+)\s*<',
+                           r'>\s*([\w~-]+~\d+\.\d+\.\d+\.\d+~[\w~-]+)',
+                           r'>\s*([\w-]+~\d{10,}~(?:[a-f\d]+:){2}:~[\w-]+)\s*<',
+                           r'>\s*([\w~-]+~[a-f0-9:]+~[\w~-]+)\s*<',
+                           r'>\s*([\w~-]+~[a-f0-9:]+~[\w~-]+)'), webpage,
+                          'stream URL'))
+
+        video_url = 'https://openload.co/stream/%s?mime=true' % decoded_id
+
+        title = self._og_search_title(webpage, default=None) or self._search_regex(
+            r'<span[^>]+class=["\']title["\'][^>]*>([^<]+)', webpage,
+            'title', default=None) or self._html_search_meta(
+            'description', webpage, 'title', fatal=True)
+
+        entries = self._parse_html5_media_entries(page_url, webpage, video_id)
+        entry = entries[0] if entries else {}
+        subtitles = entry.get('subtitles')
+
+        info_dict = {
+            'id': video_id,
+            'title': title,
+            'thumbnail': entry.get('thumbnail') or self._og_search_thumbnail(webpage, default=None),
+            'url': video_url,
+            'ext': determine_ext(title, None) or determine_ext(url, 'mp4'),
+            'subtitles': subtitles,
+            'http_headers': headers,
+        }
+        return info_dict