Programmatically controlling LosslessCut / API / CLI

[mifi]

Suggested ways to programmatically control LosslessCut. Feel free to add suggestions in comments below.

command line arguments

• open file or multiple files on launch
• seek to specified time value after file opened on launch Possibility to pass start time via command line #1204
• …

API / actions / macros

I could imagine some API that allows automating actions in LosslessCut, internally and externally. This could be tied to custom keyboard mappings #254

• HTTP API
• allow executing keyboard actions from command line
• seek to specified time value Programmatically moving time cursor through HTTP API #2087

Run a custom command from segments

See also afterExport event #974 #1347

Ability to press a button in losslesscut and have it run a custom command line (or more commands, e.g. one cmd per segment) and with ability to inject information about the current segments into this command. I think this requires some coding in one way or another by the user.

I think such a «hack» can currently be achieved by creating a custom bash script. Losslesscut will auto save a project .llc file with json5 containing information about each segment. You could create a bash script that you trigger by a keyboard shortcut (outside losslesscut). Then this script will then read the .llc file and run the desired commands. The workflow would be:

1. Edit segments in losslesscut
2. once you’re done, trigger your bash script

Scripting / automation of actions

from #2135 (reply in thread):

Maybe add scripting feature for existing commands, so we can make our own automation scripts in javascript.
Then we could assign them to keys or add them to menus, right click commands, icons.
This could massively expand capabilities and add a ton of flexibility to UI.

This it could also ease your development of hard coding everything.

For example, this is the best feature of BLENDER with its console and scripting feature or REAPER DAW, they are basically a giant command library with a UI where everything can be scripted and automated. The users then created an entire ecosystem with thousands of scripts and add-ons.

[mifi]

this was closed by accident

[mifi]

there's now a new http api

[7f5d3nbctkx8gcxwek9e7z3him6f44a7]

Feature request: Immediate execution of detectSilentScenes with parameters via API

Similar to the recently implemented goToTimecodeDirect action (ref: #2087), I'd like to propose adding the ability to execute detectSilentScenes with specified parameters via the HTTP API.

Current implementation:
curl -X POST http://localhost:8080/api/action/detectSilentScenes
This opens the silent detection dialog in the GUI, requiring manual input.

Proposed implementation:
curl -X POST http://localhost:8080/api/action/detectSilentScenes --json "{\"noise\": \"-50\", \"duration\": \"1.0\"}"

This would allow:

1. Specifying parameters (noise threshold and duration) directly in the API call.
2. Immediate execution of the silent detection without manual intervention.

Benefits:

• Enables full automation of silent detection via CLI or scripts.
• Consistent with other API actions like goToTimecodeDirect.
• Improves workflow efficiency for users who frequently use silent detection with specific parameters.

I believe this enhancement would significantly improve the programmatic control of LosslessCut, especially for automated workflows and batch processing scenarios.