如何解决ffmpeg在下载youtube视频时“清单太大”

TL; DR

我正在尝试以编程方式下载 YouTube视频的一部分。众所周知的程序不适用于某些视频，我想克服这种情况。

上下文

我正在尝试以编程方式下载 YouTube视频的一部分。如How to download portion of video with youtube-dl command?中所述，您可以通过以下命令来实现。

#Converts a human-readable URL to longer URLs for internal use.
~ $ youtube-dl --get-url 'https://www.youtube.com/watch?v=POrfo478HRI'
https://r1---sn-3pm76n76.googlevideo.com/videoplayback?expire=1597593009&ei=UQE5X7bXEYqO4QLs17KACQ&ip=10.100.238.99&id=o-AD0HfadoeimIbuf9t8Anru5X9V7cER08YyOz4iKZOCHL&itag=299&aitags=133%2C134%2C135%2C136%2C137%2C160%2C242%2C243%2C244%2C247%2C248%2C278%2C298%2C299%2C302%2C303&source=youtube&requiressl=yes&mh=NV&mm=31%2C26&mn=sn-3pm76n76%2Csn-oguesnze&ms=au%2Conr&mv=m&mvi=1&pl=17&initcwndbps=908750&vprv=1&mime=video%2Fmp4&gir=yes&clen=9205054989&dur=17931.000&lmt=1597347724488304&mt=1597571322&fvip=1&keepalive=yes&fexp=23883098&c=WEB&txp=7316222&sparams=expire%2Cei%2Cip%2Cid%2Caitags%2Csource%2Crequiressl%2Cvprv%2Cmime%2Cgir%2Cclen%2Cdur%2Clmt&sig=AOq0QJ8wRQIgYDqTm0CnLfG0kxsYNrAdtAUYB7alnowiaBArU8R5bBwCIQDeCRDJHFO_PWcnbeFaJvip80deuboN4Pi1x3eRhJBxlQ%3D%3D&lsparams=mh%2Cmm%2Cmn%2Cms%2Cmv%2Cmvi%2Cpl%2Cinitcwndbps&lsig=AG3C_xAwRgIhAKTkyTIGVkbWWVx4lqFz77rOdG1FzkM5XuTnYOqtHT4oAiEA9u1R47jPz-mjj4DRXpqqqpUbfaFMW36KbG_pR2jovFM%3D&ratebypass=yes
https://r1---sn-3pm76n76.googlevideo.com/videoplayback?expire=1597593009&ei=UQE5X7bXEYqO4QLs17KACQ&ip=10.100.238.99&id=o-AD0HfadoeimIbuf9t8Anru5X9V7cER08YyOz4iKZOCHL&itag=251&source=youtube&requiressl=yes&mh=NV&mm=31%2C26&mn=sn-3pm76n76%2Csn-oguesnze&ms=au%2Conr&mv=m&mvi=1&pl=17&initcwndbps=908750&vprv=1&mime=audio%2Fwebm&gir=yes&clen=268945126&dur=17931.001&lmt=1597348111385641&mt=1597571322&fvip=1&keepalive=yes&fexp=23883098&c=WEB&txp=7311222&sparams=expire%2Cei%2Cip%2Cid%2Citag%2Csource%2Crequiressl%2Cvprv%2Cmime%2Cgir%2Cclen%2Cdur%2Clmt&sig=AOq0QJ8wRgIhAK8XGu3vHucHwqERZ7TQniuvKEd-NTbMkwZbu8EXJ5E3AiEAiS7OYFOcsIrD2xp-AJNzucj1H9ZKMlmkCl_sU7__dZo%3D&lsparams=mh%2Cmm%2Cmn%2Cms%2Cmv%2Cmvi%2Cpl%2Cinitcwndbps&lsig=AG3C_xAwRgIhAKTkyTIGVkbWWVx4lqFz77rOdG1FzkM5XuTnYOqtHT4oAiEA9u1R47jPz-mjj4DRXpqqqpUbfaFMW36KbG_pR2jovFM%3D&ratebypass=yes

#Stores the result to a variable.
~ $ url_video_part='https://r1---sn-3pm76n76.googlevideo.com/videoplayback?expire=1597593009&ei=UQE5X7bXEYqO4QLs17KACQ&ip=10.100.238.99&id=o-AD0HfadoeimIbuf9t8Anru5X9V7cER08YyOz4iKZOCHL&itag=299&aitags=133%2C134%2C135%2C136%2C137%2C160%2C242%2C243%2C244%2C247%2C248%2C278%2C298%2C299%2C302%2C303&source=youtube&requiressl=yes&mh=NV&mm=31%2C26&mn=sn-3pm76n76%2Csn-oguesnze&ms=au%2Conr&mv=m&mvi=1&pl=17&initcwndbps=908750&vprv=1&mime=video%2Fmp4&gir=yes&clen=9205054989&dur=17931.000&lmt=1597347724488304&mt=1597571322&fvip=1&keepalive=yes&fexp=23883098&c=WEB&txp=7316222&sparams=expire%2Cei%2Cip%2Cid%2Caitags%2Csource%2Crequiressl%2Cvprv%2Cmime%2Cgir%2Cclen%2Cdur%2Clmt&sig=AOq0QJ8wRQIgYDqTm0CnLfG0kxsYNrAdtAUYB7alnowiaBArU8R5bBwCIQDeCRDJHFO_PWcnbeFaJvip80deuboN4Pi1x3eRhJBxlQ%3D%3D&lsparams=mh%2Cmm%2Cmn%2Cms%2Cmv%2Cmvi%2Cpl%2Cinitcwndbps&lsig=AG3C_xAwRgIhAKTkyTIGVkbWWVx4lqFz77rOdG1FzkM5XuTnYOqtHT4oAiEA9u1R47jPz-mjj4DRXpqqqpUbfaFMW36KbG_pR2jovFM%3D&ratebypass=yes'

#Stores the result to a variable.
~ $ url_audio_part='https://r1---sn-3pm76n76.googlevideo.com/videoplayback?expire=1597593009&ei=UQE5X7bXEYqO4QLs17KACQ&ip=10.100.238.99&id=o-AD0HfadoeimIbuf9t8Anru5X9V7cER08YyOz4iKZOCHL&itag=251&source=youtube&requiressl=yes&mh=NV&mm=31%2C26&mn=sn-3pm76n76%2Csn-oguesnze&ms=au%2Conr&mv=m&mvi=1&pl=17&initcwndbps=908750&vprv=1&mime=audio%2Fwebm&gir=yes&clen=268945126&dur=17931.001&lmt=1597348111385641&mt=1597571322&fvip=1&keepalive=yes&fexp=23883098&c=WEB&txp=7311222&sparams=expire%2Cei%2Cip%2Cid%2Citag%2Csource%2Crequiressl%2Cvprv%2Cmime%2Cgir%2Cclen%2Cdur%2Clmt&sig=AOq0QJ8wRgIhAK8XGu3vHucHwqERZ7TQniuvKEd-NTbMkwZbu8EXJ5E3AiEAiS7OYFOcsIrD2xp-AJNzucj1H9ZKMlmkCl_sU7__dZo%3D&lsparams=mh%2Cmm%2Cmn%2Cms%2Cmv%2Cmvi%2Cpl%2Cinitcwndbps&lsig=AG3C_xAwRgIhAKTkyTIGVkbWWVx4lqFz77rOdG1FzkM5XuTnYOqtHT4oAiEA9u1R47jPz-mjj4DRXpqqqpUbfaFMW36KbG_pR2jovFM%3D&ratebypass=yes'

#Downloads the video for one minute from 12:45.
~ $ ffmpeg    -ss 12:45 -i "${url_video_part}"    -ss 12:45 -i "${url_audio_part}"    -t 1:00 output.mkv
frame= 3600 fps= 49 q=-1.0 Lsize=   27047kB time=00:00:59.99 bitrate=3692.9kbits/s speed=0.813x

这确实很好。

问题

对于某些视频^†1，内部URL的格式为https://manifest.googlevideo.com/...，而不是上面示例中的https://RANDOM-STRING.googlevideo.com/...。

~ $ youtube-dl --get-url 'https://www.youtube.com/watch?v=KRMfnvLODZQ'
https://manifest.googlevideo.com/api/manifest/dash/expire/1597593866/ei/qgQ5X5maN8-ilQSu15W4CQ/ip/10.100.238.99/id/KRMfnvLODZQ.1/source/yt_live_broadcast/requiressl/yes/hfr/all/as/fmp4_audio_clear%2Cwebm_audio_clear%2Cwebm2_audio_clear%2Cfmp4_sd_hd_clear%2Cwebm2_sd_hd_clear/force_finished/1/vprv/1/keepalive/yes/fexp/23883098/itag/0/playlist_type/DVR/sparams/expire%2Cei%2Cip%2Cid%2Csource%2Crequiressl%2Chfr%2Cas%2Cforce_finished%2Cvprv%2Citag%2Cplaylist_type/sig/AOq0QJ8wRQIhANYKQvKTN8BRmhp7--4tQp9aqRp_qIGjh472BTYLyJ62AiAautFF86xGDHOe7mYlYaZ1W86EM4k2e6UiJS9Q91dMog%3D%3D
https://manifest.googlevideo.com/api/manifest/dash/expire/1597593866/ei/qgQ5X5maN8-ilQSu15W4CQ/ip/10.100.238.99/id/KRMfnvLODZQ.1/source/yt_live_broadcast/requiressl/yes/hfr/all/as/fmp4_audio_clear%2Cwebm_audio_clear%2Cwebm2_audio_clear%2Cfmp4_sd_hd_clear%2Cwebm2_sd_hd_clear/force_finished/1/vprv/1/keepalive/yes/fexp/23883098/itag/0/playlist_type/DVR/sparams/expire%2Cei%2Cip%2Cid%2Csource%2Crequiressl%2Chfr%2Cas%2Cforce_finished%2Cvprv%2Citag%2Cplaylist_type/sig/AOq0QJ8wRQIhANYKQvKTN8BRmhp7--4tQp9aqRp_qIGjh472BTYLyJ62AiAautFF86xGDHOe7mYlYaZ1W86EM4k2e6UiJS9Q91dMog%3D%3D

下载此类视频始终会失败。

~ $ ffmpeg    -ss 12:45 -i "${url_video_part}"    -ss 12:45 -i "${url_audio_part}"    -t 1:00 output.mkv
[dash @ 0x55fa6a1a4400] Manifest too large: 3776722
https://manifest.googlevideo.com/...: Invalid data found when processing input

请注意，仅使用youtube-dl即可下载全部此类视频。

_{†1：据我们测试，最近（例如昨天）上传的视频具有这样的URL，而当我们在一天左右的时间后重试相同的youtube-dl --get-url命令时，我们会得到“无辜的” URL，例如在第一个示例中。}

问题

如何下载部分视频

通过为ffmpeg指定其他选项，
通过某种方式强制检索https://RANDOM-STRING.googlevideo.com/...形式的内部URL，
还是我根本不知道的事情？

我当前的解决方法是等待一两天，直到内部URL更改为止，这比下载完整的视频然后进行裁剪要慢得多。

环境

~ $ lsb_release -d
Description:    Arch Linux

~ $ ffmpeg -version
ffmpeg version n4.3.1 Copyright (c) 2000-2020 the FFmpeg developers
built with gcc 10.1.0 (GCC)
configuration: --prefix=/usr --disable-debug --disable-static --disable-stripping --enable-avisynth --enable-fontconfig --enable-gmp --enable-gnutls --enable-gpl --enable-ladspa --enable-libaom --enable-libass --enable-libbluray --enable-libdav1d --enable-libdrm --enable-libfreetype --enable-libfribidi --enable-libgsm --enable-libiec61883 --enable-libjack --enable-libmfx --enable-libmodplug --enable-libmp3lame --enable-libopencore_amrnb --enable-libopencore_amrwb --enable-libopenjpeg --enable-libopus --enable-libpulse --enable-librav1e --enable-libsoxr --enable-libspeex --enable-libsrt --enable-libssh --enable-libtheora --enable-libv4l2 --enable-libvidstab --enable-libvmaf --enable-libvorbis --enable-libvpx --enable-libwebp --enable-libx264 --enable-libx265 --enable-libxcb --enable-libxml2 --enable-libxvid --enable-nvdec --enable-nvenc --enable-omx --enable-shared --enable-version3
libavutil      56. 51.100 / 56. 51.100
libavcodec     58. 91.100 / 58. 91.100
libavformat    58. 45.100 / 58. 45.100
libavdevice    58. 10.100 / 58. 10.100
libavfilter     7. 85.100 /  7. 85.100
libswscale      5.  7.100 /  5.  7.100
libswresample   3.  7.100 /  3.  7.100
libpostproc    55.  7.100 / 55.  7.100

~ $ youtube-dl --version
2020.07.28

解决方法

有两种解决方法。

1。只需等待一两天。

如OP中所述，您可以等待几天，以将内部URL从https://manifest.googlevideo.com/api/manifest/dash/...更改为https://r1---sn-3pm76n76.googlevideo.com/...。^†1。

据我测试，这至少需要一整天的时间（7-10小时）。

2。使用HLS代替DASH。

2.1解决方法

通过以下步骤，您可以立即下载视频。

url='https://www.youtube.com/watch?v=pkP-hgGJdvY'

_{（我选择此URL作为示例，因为youtube-dl --get-url "${url}"在编写此答案时返回https://manifest.googlevideo.com/api/...。）}

curl "${url}"
从curl的输出中，提取格式为internal_url的URL https://manifest.googlevideo.com/api/manifest/hls_variant/...。请注意，此URL包含/hls_variant/而不是/dash/。
ffmpeg -i "${internal_url}"，可以选择是否下载视频。

第2步和第3步的一个衬里

curl --silent "${url}" | grep -o 'https:\\/\\/[^"]\+googlevideo.com\\/[^"]\+' | grep 'hls_variant' | sed 's/\\//g'

2.2已知问题

即使采用上述方法，对于长时间（2个多小时）的实时存档（最近已完成实时流式传输（例如，几个小时或一天前）），也只能将整个视频的最后2个小时完全下载 ^†2，即使您可以在浏览器或官方应用上播放视频也没有问题。此问题并非特定于ffmpeg，实际上youtube-dl的GitHub存储库有一个相关问题：Youtube: Only downloading last 2 hours of 3 hour video. (with hack as a workaround) #26330。（顺便说一句，此问题的操作包括解决此问题的方法，由于未进行测试，因此在此不写。）

3结论

考虑†1和†2，当目标视频的YouTube服务器端处理尚未完全完成时，似乎会出现此问题。尽管我根本不知道该怎么做，但是可能应该有一些优雅的方法来解决最后两个小时的问题。

我能够使用youtube-dl --youtube-skip-dash-manifest跳过清单文件并获得最终URL。

为了跟进这个问题，我遇到了这个问题，一些挖掘发现 ffmpeg merged a patch to fix this 早在 9 月份。虽然截至 2021 年 2 月 6 日它还没有发布，但它应该在 4.3.1（可能是 4.3.2 或 4.4）之后的任何版本中。

有用的是，它还解释了问题是如何产生的 - 对于未知情况的“软限制”被重新解释为硬限制，而 9 月的补丁通过完全移除限制修复了误解。

ffmpeg在下载youtube视频时“清单太大”