Marware/better-ffmpeg-progress

Run an FFmpeg command & see percentage progress, ETA, etc.

Better FFmpeg Progress

Runs an FFmpeg command and uses tqdm to show a progress bar.

Example:

39%|███████████████████████████████████████████ | 23.581/60.226 [00:19<00:34, 1.07s/s]

Where:

• 39% is the percentage progress.
• 23.581 seconds of the input file have been processed.
• 60.226 is the duration of the input file in seconds.
• 00:19 is the time elapsed since the FFmpeg process started.
• 00:34 is the estimated time required for the FFmpeg process to complete.
• 1.07 shows how many seconds of the input file are processed per second.

Installation:

pip3 install better-ffmpeg-progress --upgrade

Usage:

Create an instance of the FfmpegProcess class and supply a list of arguments like you would to subprocess.run():

from better_ffmpeg_progress import FfmpegProcess
# Pass a list of FFmpeg arguments, like you would if using subprocess.run()
process = FfmpegProcess(["ffmpeg", "-i", "input.mp4", "-c:a", "libmp3lame", "output.mp3"])
# Use the run method to run the FFmpeg command.
process.run()

The run method takes the following optional arguments:

• progress_handler

  • You can create a function if you would like to do something with the following values:

    • Percentage progress. [float]
    • Speed, e.g. 22.3x which means that 22.3 seconds of the input are processed every second. [string]
    • ETA in seconds. [float]
    • Estimated output filesize in bytes. [float]
      • Note: This is not accurate. Please take the value with a grain of salt.

    The function will receive the aforementioned metrics as arguments, about two times per second.

    Here's an example of a progress handler that you can create:

    def handle_progress_info(percentage, speed, eta, estimated_filesize):
        print(f"The FFmpeg process is {percentage}% complete. ETA is {eta} seconds.")
        print(f"Estimated Output Filesize: {estimated_filesize / 1_000_000} MB")

    Then you simply set the value of the progress_handler argument to the name of your function, like so:

    process.run(progress_handler=handle_progress_info)

• ffmpeg_output_file

  • The ffmpeg_output_file argument allows you define where you want the output of FFmpeg to be saved. By default, this is saved in a folder named "ffmpeg_output", with the filename [<input_filename>].txt, but you can change this using the ffmpeg_output_file argument.

Here's an example where both the progress_handler and ffmpeg_output_file parameters are used:

process.run(progress_handler=handle_progress_info, ffmpeg_output_file="ffmpeg_log.txt")

Changelog:

[19th September 2022]

• Add the ability to specify process_complete_handler, a function to run when the FFmpeg process is complete. E.g.

  process.run(progress_handler=None, ffmpeg_output_file=None, process_complete_handler=my_function)

• Add 0.001 to tqdm's total parameter to prevent the chance of getting TqdmWarning: clamping frac to range [0, 1]