#video-processing #editing #ffmpeg #concatenate #save #input-file

app trimmeroni

Cut and concatenate video clips without reencoding

4 releases

0.2.3 Jul 14, 2024
0.2.2 Feb 3, 2024
0.2.1 Nov 28, 2023
0.1.0 May 9, 2023

#31 in Video

29 downloads per month

MIT license

15KB
347 lines

trimmeroni

This little tool aims to make it easier to cut and concatenate together any number of video clips, from any number of input files, into one output file, without reencoding. Compared to editing the clips using video editing or transcoding software, this preserves original video quality, and saves processing time.

It uses the ffmpeg program under the hood to copy each requested video segment from the source files into a temporary directory, and then concatenate them (insert one after another) into the destination file.

The only requirement is that you have ffmpeg installed on your system, accessible from PATH.

Usage

Examples

From "cool_gameplay.mkv", grab the segment between 08:34 and 11:02, and save it to "360_noscope.mp4"

trimmeroni -i 'cool_gameplay.mkv @ 08:34 - 11:02' "360_noscope.mp4"

From "camera_footage.mp4", use the segments 04:20 - 04:59, 06:44 - 09:00, and from 14:40 to the end, concatenate them, and save the result to "check_this_out.mp4"

trimmeroni -i 'camera_footage.mp4 @ 04:20 - 04:59, 06:44 - 09:00, 14:40 - ' "check_this_out.mp4"

Take 38:08 to 45:35, and 1:02:20 to 1:17:52 from "Day1.mp4", and from the beginning to 29:26 from "Day2.mp4", concatenate, and save the result to "Weekend bike adventure.mp4"

trimmeroni -i 'Day1.mp4 @ 38:08 - 45:35, 1:02:20 - 1:17:52' -i 'Day2.mp4 @ - 29:26' "Weekend bike adventure.mp4"

Details

Usage: trimmeroni [OPTIONS] <OUTPUT_NAME>

Arguments:
  <OUTPUT_NAME>

Options:
  -i, --input-clip <INPUT_CLIP_SPECS>
  -v, --verbose                        Print additional information while working
  -h, --help                           Print help information
  -V, --version                        Print version information

<INPUT_CLIP_SPECS> specifies an input file and the timecodes of clips to copy from it. It should be formatted as follows:

INPUT_NAME @ TIMECODE - TIMECODE[, TIMECODE - TIMECODE, ...]

where

  • INPUT_NAME is the input file name; if the file name contains "@", use another "@" to escape it ("@@"); there is no need to escape spaces;
  • each TIMECODE can be any seek position supported by ffmpeg (e.g. in h:mm:ss format), or empty, meaning either the start or end of the entire video file;
  • "TIMECODE - TIMECODE," can be repeated any number of times in order to grab more than one clip from the input file.

You can use -i many times to use clips from many source files.

Remember to quote your <INPUT_CLIP_SPECS>, so that trimmeroni receives it as a single string.

The start of each clip will be snapped to the nearest keyframe before the specified timecode. For recordings from OBS Studio, for example, this may mean that the clip will include a segment from up to 2 seconds earlier.

Use -v or --verbose to get additional info from trimmeroni and ffmpeg.

Installing

If you have cargo, you can install this program from crates.io:

cargo install trimmeroni

Building

This project uses Rust's stable toolchain, 2021 edition.

To build locally from source:

git clone https://github.com/outfrost/trimmeroni.git
cd trimmeroni
cargo build

You can then run the debug binary like so:

cargo run -- -i 'file.mkv @ 00:02 - 01:12' "out.mp4"

or install it to your cargo binaries with:

cargo install --path .

Contributing

Issues and pull requests welcome.

By opening an issue or pull request for this repository, you acknowledge and agree that the contributions included in your issue or pull request may be published, used, copied, modified, repurposed, and reused without limitation, under the terms of the MIT License (included in LICENSE), regardless of whether or not the pull request is ever merged, and whether or not the issue is ever resolved.

If you would like your name to be included in the list of contributors in the copyright message, please edit the LICENSE file as part of your pull request.

Dependencies

~5–15MB
~199K SLoC