hyperhive/hive-c0re/src/dashboard/lifecycle_ops.rs

//! Container lifecycle endpoints for the dashboard.
//!
//! Rebuild / restart / start / stop (hard + graceful) / update-all all
//! submit DAGs to the job queue (`job_queue::submit`), so each shows a
//! visible queued→running transient on the dashboard — a direct
//! sub-second start/stop only flashed the badge. Start/stop also
//! persist the agent's `wanted` power intent before submitting; the
//! DAG's `Reconcile` converges to it. Destroy delegates to
//! `actions::destroy` (optionally purging).

use axum::{
    extract::{Form, Path as AxumPath, Query, State},
    http::StatusCode,
    response::{IntoResponse, Response},
};
use serde::Deserialize;

/// Query params for `post_kill`. `?graceful=1` routes to the graceful-stop
/// orchestration (quiesce the harness, flush `/state`, then container stop)
/// instead of an immediate hard stop. Defaults false → today's hard kill.
#[derive(Deserialize)]
pub(super) struct KillParams {
    #[serde(default)]
    graceful: bool,
}

use super::{AppState, error_response, guard_agent_name, strip_container_prefix};
use crate::job_queue::{Source, submit};
use crate::{actions, lifecycle};

pub(super) async fn post_rebuild(
    State(state): State<AppState>,
    AxumPath(name): AxumPath<String>,
) -> Response {
    let logical = strip_container_prefix(&name);
    if let Some(reject) = guard_agent_name(&state, &logical).await {
        return reject;
    }
    submit::rebuild(
        &state.coord,
        &logical,
        Source::Manual,
        "manual via dashboard ↻ R3BU1LD button".to_owned(),
    );
    (StatusCode::OK, "ok").into_response()
}

pub(super) async fn post_kill(
    State(state): State<AppState>,
    AxumPath(name): AxumPath<String>,
    Query(params): Query<KillParams>,
) -> Response {
    let logical = strip_container_prefix(&name);
    if let Some(reject) = guard_agent_name(&state, &logical).await {
        return reject;
    }
    if params.graceful {
        // Graceful stop: submit the quiesce DAG (signal the harness →
        // one stop-checkpoint turn → drain → container stop, with a
        // timeout fallback to a hard stop). The agent's lifecycle
        // lease keeps it from racing an in-flight rebuild for the same
        // agent, and per-node progress surfaces on the queue snapshot.
        submit::graceful_stop(
            &state.coord,
            &logical,
            Source::Manual,
            "manual via dashboard graceful stop".to_owned(),
        );
        return (StatusCode::OK, "ok").into_response();
    }
    // Manager is stoppable from the dashboard like any other
    // agent. The host's dashboard server keeps running (it's
    // hive-c0re, not the manager container), per-agent approvals
    // submitted by other sub-agents still process through the
    // host-side approval queue without the manager up, and
    // operator-driven meta-input updates work from the dashboard
    // either way. The MCP-surface self-kill guard in
    // `socket_server.rs::ManagerRequest::Kill` stays in place: a
    // manager calling Kill on its own container is self-suicide
    // mid-call, not a legitimate operator action.
    submit::stop(
        &state.coord,
        &logical,
        Source::Manual,
        "manual via dashboard stop".to_owned(),
    );
    (StatusCode::OK, "ok").into_response()
}

pub(super) async fn post_restart(
    State(state): State<AppState>,
    AxumPath(name): AxumPath<String>,
) -> Response {
    let logical = strip_container_prefix(&name);
    if let Some(reject) = guard_agent_name(&state, &logical).await {
        return reject;
    }
    submit::restart(
        &state.coord,
        &logical,
        Source::Manual,
        "manual via dashboard ↺ R3START button".to_owned(),
    );
    (StatusCode::OK, "ok").into_response()
}

pub(super) async fn post_start(
    State(state): State<AppState>,
    AxumPath(name): AxumPath<String>,
) -> Response {
    let logical = strip_container_prefix(&name);
    if let Some(reject) = guard_agent_name(&state, &logical).await {
        return reject;
    }
    submit::start(
        &state.coord,
        &logical,
        Source::Manual,
        "manual via dashboard start".to_owned(),
    );
    (StatusCode::OK, "ok").into_response()
}

pub(super) async fn post_update_all(State(state): State<AppState>) -> Response {
    let containers = lifecycle::list().await.unwrap_or_default();
    for container in containers {
        let Some(logical) = container
            .strip_prefix(lifecycle::AGENT_PREFIX)
            .map(str::to_owned)
        else {
            continue;
        };
        submit::rebuild(
            &state.coord,
            &logical,
            Source::Manual,
            "manual via dashboard 🌀 UPDATE ALL".to_owned(),
        );
    }
    (StatusCode::OK, "ok").into_response()
}

#[derive(Deserialize, Default)]
pub(super) struct DestroyForm {
    #[serde(default)]
    purge: Option<String>,
}

pub(super) async fn post_destroy(
    State(state): State<AppState>,
    AxumPath(name): AxumPath<String>,
    Form(form): Form<DestroyForm>,
) -> Response {
    if let Some(reject) = guard_agent_name(&state, &name).await {
        return reject;
    }
    // Checkbox semantics: any non-empty value (axum sends "on") = purge.
    let purge = form.purge.as_deref().is_some_and(|v| !v.is_empty());
    // `actions::destroy` rescans the container list on success, so the
    // `ContainerRemoved` event lands before we return 200. The matching
    // form carries `data-no-refresh`.
    match actions::destroy(&state.coord, &name, purge).await {
        Ok(()) => (StatusCode::OK, "ok").into_response(),
        Err(e) => error_response(&format!("destroy {name} failed: {e:#}")),
    }
}