diff options
author | bashonly <bashonly@bashonly.com> | 2023-07-04 16:40:56 -0500 |
---|---|---|
committer | dirkf <fieldhouse@gmx.net> | 2023-07-18 10:50:46 +0100 |
commit | 21438a4194376c3a9b1e5c322c825d43a1b03d6e (patch) | |
tree | a5a50e6b9d99e4a4bda08604682addd275ce7207 /youtube_dl | |
parent | 8334ec961b802ad7ef8571b776c5fc727206dc9b (diff) | |
download | youtube-dl-21438a4194376c3a9b1e5c322c825d43a1b03d6e.tar.xz |
[downloader/external] Fix cookie support
Diffstat (limited to 'youtube_dl')
-rw-r--r-- | youtube_dl/downloader/common.py | 5 | ||||
-rw-r--r-- | youtube_dl/downloader/external.py | 124 |
2 files changed, 106 insertions, 23 deletions
diff --git a/youtube_dl/downloader/common.py b/youtube_dl/downloader/common.py index 08c98b336..afb4ee33d 100644 --- a/youtube_dl/downloader/common.py +++ b/youtube_dl/downloader/common.py @@ -13,9 +13,7 @@ from ..utils import ( error_to_compat_str, format_bytes, shell_quote, - T, timeconvert, - traverse_obj, ) @@ -379,9 +377,6 @@ class FileDownloader(object): else '%.2f' % sleep_interval)) time.sleep(sleep_interval) - info_dict['http_headers'] = dict(traverse_obj(info_dict, ( - 'http_headers', T(dict.items), lambda _, pair: pair[0].lower() != 'cookie'))) or None - return self.real_download(filename, info_dict) def real_download(self, filename, info_dict): diff --git a/youtube_dl/downloader/external.py b/youtube_dl/downloader/external.py index 1b6bd1fa2..7fc864e85 100644 --- a/youtube_dl/downloader/external.py +++ b/youtube_dl/downloader/external.py @@ -1,9 +1,10 @@ from __future__ import unicode_literals -import os.path +import os import re import subprocess import sys +import tempfile import time from .common import FileDownloader @@ -23,6 +24,8 @@ from ..utils import ( check_executable, is_outdated_version, process_communicate_or_kill, + T, + traverse_obj, ) @@ -30,6 +33,7 @@ class ExternalFD(FileDownloader): def real_download(self, filename, info_dict): self.report_destination(filename) tmpfilename = self.temp_name(filename) + self._cookies_tempfile = None try: started = time.time() @@ -42,6 +46,13 @@ class ExternalFD(FileDownloader): # should take place retval = 0 self.to_screen('[%s] Interrupted by user' % self.get_basename()) + finally: + if self._cookies_tempfile and os.path.isfile(self._cookies_tempfile): + try: + os.remove(self._cookies_tempfile) + except OSError: + self.report_warning( + 'Unable to delete temporary cookies file "{0}"'.format(self._cookies_tempfile)) if retval == 0: status = { @@ -97,6 +108,16 @@ class ExternalFD(FileDownloader): def _configuration_args(self, default=[]): return cli_configuration_args(self.params, 'external_downloader_args', default) + def _write_cookies(self): + if not self.ydl.cookiejar.filename: + tmp_cookies = tempfile.NamedTemporaryFile(suffix='.cookies', delete=False) + tmp_cookies.close() + self._cookies_tempfile = tmp_cookies.name + self.to_screen('[download] Writing temporary cookies file to "{0}"'.format(self._cookies_tempfile)) + # real_download resets _cookies_tempfile; if it's None, save() will write to cookiejar.filename + self.ydl.cookiejar.save(self._cookies_tempfile, ignore_discard=True, ignore_expires=True) + return self.ydl.cookiejar.filename or self._cookies_tempfile + def _call_downloader(self, tmpfilename, info_dict): """ Either overwrite this or implement _make_cmd """ cmd = [encodeArgument(a) for a in self._make_cmd(tmpfilename, info_dict)] @@ -110,13 +131,21 @@ class ExternalFD(FileDownloader): self.to_stderr(stderr.decode('utf-8', 'replace')) return p.returncode + @staticmethod + def _header_items(info_dict): + return traverse_obj( + info_dict, ('http_headers', T(dict.items), Ellipsis)) + class CurlFD(ExternalFD): AVAILABLE_OPT = '-V' def _make_cmd(self, tmpfilename, info_dict): - cmd = [self.exe, '--location', '-o', tmpfilename] - for key, val in info_dict['http_headers'].items(): + cmd = [self.exe, '--location', '-o', tmpfilename, '--compressed'] + cookie_header = self.ydl.cookiejar.get_cookie_header(info_dict['url']) + if cookie_header: + cmd += ['--cookie', cookie_header] + for key, val in self._header_items(info_dict): cmd += ['--header', '%s: %s' % (key, val)] cmd += self._bool_option('--continue-at', 'continuedl', '-', '0') cmd += self._valueless_option('--silent', 'noprogress') @@ -151,8 +180,11 @@ class AxelFD(ExternalFD): def _make_cmd(self, tmpfilename, info_dict): cmd = [self.exe, '-o', tmpfilename] - for key, val in info_dict['http_headers'].items(): + for key, val in self._header_items(info_dict): cmd += ['-H', '%s: %s' % (key, val)] + cookie_header = self.ydl.cookiejar.get_cookie_header(info_dict['url']) + if cookie_header: + cmd += ['-H', 'Cookie: {0}'.format(cookie_header), '--max-redirect=0'] cmd += self._configuration_args() cmd += ['--', info_dict['url']] return cmd @@ -162,8 +194,10 @@ class WgetFD(ExternalFD): AVAILABLE_OPT = '--version' def _make_cmd(self, tmpfilename, info_dict): - cmd = [self.exe, '-O', tmpfilename, '-nv', '--no-cookies'] - for key, val in info_dict['http_headers'].items(): + cmd = [self.exe, '-O', tmpfilename, '-nv', '--compression=auto'] + if self.ydl.cookiejar.get_cookie_header(info_dict['url']): + cmd += ['--load-cookies', self._write_cookies()] + for key, val in self._header_items(info_dict): cmd += ['--header', '%s: %s' % (key, val)] cmd += self._option('--limit-rate', 'ratelimit') retry = self._option('--tries', 'retries') @@ -182,21 +216,58 @@ class WgetFD(ExternalFD): class Aria2cFD(ExternalFD): AVAILABLE_OPT = '-v' + @staticmethod + def _aria2c_filename(fn): + return fn if os.path.isabs(fn) else os.path.join('.', fn) + def _make_cmd(self, tmpfilename, info_dict): - cmd = [self.exe, '-c'] - cmd += self._configuration_args([ - '--min-split-size', '1M', '--max-connection-per-server', '4']) - dn = os.path.dirname(tmpfilename) - if dn: - cmd += ['--dir', dn] - cmd += ['--out', os.path.basename(tmpfilename)] - for key, val in info_dict['http_headers'].items(): + cmd = [self.exe, '-c', + '--console-log-level=warn', '--summary-interval=0', '--download-result=hide', + '--http-accept-gzip=true', '--file-allocation=none', '-x16', '-j16', '-s16'] + if 'fragments' in info_dict: + cmd += ['--allow-overwrite=true', '--allow-piece-length-change=true'] + else: + cmd += ['--min-split-size', '1M'] + + if self.ydl.cookiejar.get_cookie_header(info_dict['url']): + cmd += ['--load-cookies={0}'.format(self._write_cookies())] + for key, val in self._header_items(info_dict): cmd += ['--header', '%s: %s' % (key, val)] + cmd += self._configuration_args(['--max-connection-per-server', '4']) + cmd += ['--out', os.path.basename(tmpfilename)] + cmd += self._option('--max-overall-download-limit', 'ratelimit') cmd += self._option('--interface', 'source_address') cmd += self._option('--all-proxy', 'proxy') cmd += self._bool_option('--check-certificate', 'nocheckcertificate', 'false', 'true', '=') cmd += self._bool_option('--remote-time', 'updatetime', 'true', 'false', '=') - cmd += ['--', info_dict['url']] + cmd += self._bool_option('--show-console-readout', 'noprogress', 'false', 'true', '=') + cmd += self._configuration_args() + + # aria2c strips out spaces from the beginning/end of filenames and paths. + # We work around this issue by adding a "./" to the beginning of the + # filename and relative path, and adding a "/" at the end of the path. + # See: https://github.com/yt-dlp/yt-dlp/issues/276 + # https://github.com/ytdl-org/youtube-dl/issues/20312 + # https://github.com/aria2/aria2/issues/1373 + dn = os.path.dirname(tmpfilename) + if dn: + cmd += ['--dir', self._aria2c_filename(dn) + os.path.sep] + if 'fragments' not in info_dict: + cmd += ['--out', self._aria2c_filename(os.path.basename(tmpfilename))] + cmd += ['--auto-file-renaming=false'] + if 'fragments' in info_dict: + cmd += ['--file-allocation=none', '--uri-selector=inorder'] + url_list_file = '%s.frag.urls' % (tmpfilename, ) + url_list = [] + for frag_index, fragment in enumerate(info_dict['fragments']): + fragment_filename = '%s-Frag%d' % (os.path.basename(tmpfilename), frag_index) + url_list.append('%s\n\tout=%s' % (fragment['url'], self._aria2c_filename(fragment_filename))) + stream, _ = self.sanitize_open(url_list_file, 'wb') + stream.write('\n'.join(url_list).encode()) + stream.close() + cmd += ['-i', self._aria2c_filename(url_list_file)] + else: + cmd += ['--', info_dict['url']] return cmd @@ -235,8 +306,10 @@ class Aria2pFD(ExternalFD): } options['dir'] = os.path.dirname(tmpfilename) or os.path.abspath('.') options['out'] = os.path.basename(tmpfilename) + if self.ydl.cookiejar.get_cookie_header(info_dict['url']): + options['load-cookies'] = self._write_cookies() options['header'] = [] - for key, val in info_dict['http_headers'].items(): + for key, val in self._header_items(info_dict): options['header'].append('{0}: {1}'.format(key, val)) download = aria2.add_uris([info_dict['url']], options) status = { @@ -265,8 +338,16 @@ class HttpieFD(ExternalFD): def _make_cmd(self, tmpfilename, info_dict): cmd = ['http', '--download', '--output', tmpfilename, info_dict['url']] - for key, val in info_dict['http_headers'].items(): + for key, val in self._header_items(info_dict): cmd += ['%s:%s' % (key, val)] + + # httpie 3.1.0+ removes the Cookie header on redirect, so this should be safe for now. [1] + # If we ever need cookie handling for redirects, we can export the cookiejar into a session. [2] + # 1: https://github.com/httpie/httpie/security/advisories/GHSA-9w4w-cpc8-h2fq + # 2: https://httpie.io/docs/cli/sessions + cookie_header = self.ydl.cookiejar.get_cookie_header(info_dict['url']) + if cookie_header: + cmd += ['Cookie:%s' % cookie_header] return cmd @@ -312,7 +393,14 @@ class FFmpegFD(ExternalFD): # if end_time: # args += ['-t', compat_str(end_time - start_time)] - if info_dict['http_headers'] and re.match(r'^https?://', url): + cookies = self.ydl.cookiejar.get_cookies_for_url(url) + if cookies: + args.extend(['-cookies', ''.join( + '{0}={1}; path={2}; domain={3};\r\n'.format( + cookie.name, cookie.value, cookie.path, cookie.domain) + for cookie in cookies)]) + + if info_dict.get('http_headers') and re.match(r'^https?://', url): # Trailing \r\n after each HTTP header is important to prevent warning from ffmpeg/avconv: # [http @ 00000000003d2fa0] No trailing CRLF found in HTTP header. headers = handle_youtubedl_headers(info_dict['http_headers']) |