Capture pane output

A pane runs asynchronously: sending a command does not mean its output is already on screen. Capture repeatedly until the expected line appears, with a deadline so a failed command cannot leave the program waiting forever.

This complete program creates a private tmux server, captures its output, and cleans up. Follow the setup and run instructions below. You need tmux and a Unix environment; no existing tmux session is required.

Read what’s on screen¶

The program sends printf with a leading newline, then waits for the complete line libtmux capture ready. The newline keeps a late shell prompt off that line. Matching the whole line avoids mistaking the echoed command for its output.

use std::error::Error;
use std::time::Duration;

use libtmux::{NewSessionOptions, Server};
use tokio::time::{sleep, timeout};

#[tokio::main]
async fn main() -> Result<(), Box<dyn Error>> {
    let directory = tempfile::tempdir()?;
    let server = Server::builder()
        .socket_path(directory.path().join("tmux.sock"))
        .config_file("/dev/null")
        .build()?;

    let captured = timeout(Duration::from_secs(5), async {
        let session = server
            .new_session(NewSessionOptions::new("capture").command("env ENV=/dev/null sh"))
            .await?;
        let panes = session.panes().await?;
        let pane = panes.first().ok_or("the session has no pane")?;
        pane.send_line("printf '\\nlibtmux capture ready\\n'").await?;
        loop {
            let lines = pane.capture().await?;
            if let Some(line) = lines
                .iter()
                .find(|line| line.as_bytes() == b"libtmux capture ready")
            {
                println!("{}", line.to_string_lossy());
                return Ok::<_, Box<dyn Error>>(());
            }
            sleep(Duration::from_millis(20)).await;
        }
    })
    .await;

    let killed = server.kill().await;
    let closed = server.shutdown().await;
    captured??;
    killed?;
    closed?;
    Ok(())
}

Wait for output or completion¶

The program above checks the captured screen for up to five seconds. The short pause between checks limits polling; the observed output determines when the loop finishes. A tmux capture is a view of the screen and scrollback, so it can miss output that has already scrolled away. Use a stream or a completion signal for long-running commands when that distinction matters.

Capturing output covers capture options, while Sending keys explains why sending and waiting are separate operations.

Setup and run¶

Use an empty directory. The commands pin the library revision used to verify the program.

Save the program as src/main.rs and this file as Cargo.toml. Use the repository’s Rust 1.97.1 toolchain for this pinned revision.

[package]
name = "capture-example"
version = "0.1.0"
edition = "2024"

[dependencies]
libtmux = { path = "libtmux/crates/libtmux" }
tempfile = "3.27.0"
tokio = { version = "1.53.1", features = ["macros", "rt-multi-thread", "time"] }

$ git clone https://github.com/libtmux/libtmux-rs libtmux &&
  git -C libtmux checkout d4e08b4eaab62ef4eeedab79b47973ae9a1de310 &&
  cargo +1.97.1 run

Where this comes from¶

This complete program was run against the library revision pinned above. The displayed code is checked against the bytes from that run.