Filtering out videos based on video format, resolution, or duration is currently not possible without downloading the full video file and doing the filtering in the postprocessor/output stage.

I recently learned that you can use ffprobe to figure out the metadata of a video without actually downloading the full thing. That's how your browser also knows how long a video (for the video timeline) is without downloading it completely.

Input #0, mov,mp4,m4a,3gp,3g2,mj2, from 'https://c5.coomer.su/data/xy.mp4':
  Metadata:
    major_brand     : isom
    minor_version   : 512
    compatible_brands: isomiso2avc1mp41
    encoder         : Lavf58.29.100
  Duration: 00:08:23.59, start: 0.000000, bitrate: 1212 kb/s
  Stream #0:0[0x1](und): Video: h264 (Main) (avc1 / 0x31637661), yuv420p(tv, bt709, progressive), 1280x720 [SAR 1:1 DAR 16:9], 1013 kb/s, 25 fps, 25 tbr, 12800 tbn (default)
    Metadata:
      handler_name    : Core Media Video
      vendor_id       : [0][0][0][0]
  Stream #0:1[0x2](eng): Audio: aac (LC) (mp4a / 0x6134706D), 48000 Hz, stereo, fltp, 192 kb/s (default)
    Metadata:
      handler_name    : Core Media Audio
      vendor_id       : [0][0][0][0]

This gives us: Duration, Resolution, Framerate, Video-Ratio, Bitrate, Video and Audio format, Audio channel (stereo) and much more.

Similar to extractor.*.image-filter we could have a extractor.*.video-filter that does ffprobe in the background and provides all these fields for the filter string.

What do you think?