platform/src/servers/plugins/install.ts

import { recordPluginInstall, removePluginInstall, setPluginEnabled } from 'officerdb';
import { PLATFORM_DIR } from '../data-path';
import { deleteProcess, processStatus, startProcess, stopProcess } from '../app-store/pm2';
import { addPluginToEcosystem, pluginProcessName, removePluginFromEcosystem } from './ecosystem';
import { discoverPlugins } from './discover';
import { refreshPluginMounts, snapshotPlugins } from './mount';
import { generatePluginSchemas, pushSchema } from './schema';
import { installDependencies, manualInstallHint, reportDependencies, stillMissing } from './os-deps';
import type { DiscoveredPlugin } from './manifest';

// The install runner: the four verbs, each as a short ordered list of effects.
//
// ── Order is the whole design ──
//
// Every verb does its work in the order that leaves the system coherent if it stops halfway, because it
// can. Bringing something UP goes outside-in (make it possible, then start it, then record it, then
// expose it); taking something DOWN goes inside-out (stop exposing it, stop it, then forget it). The
// worst intermediate state is then "recorded but not running", which the UI can show and a retry fixes —
// never "running but forgotten", which nothing can see and nothing will clean up.
//
// ── What each verb touches ──
//
//              host deps   ecosystem   PM2     row         mounts    tables
//   install    install     add         start   upsert      rebuild   (see below)
//   uninstall  untouched   remove      delete  delete      rebuild   untouched
//   enable     untouched   —           start   enabled=t   rebuild   untouched
//   disable    untouched   —           stop    enabled=f   rebuild   untouched
//
// Host dependencies are installed and never removed. A package is a machine-wide resource that other
// things may have started depending on the moment it appeared, so uninstalling a plugin has no business
// deciding that `ffmpeg` should go — the same reasoning as tables, one level down.
//
// Nothing here drops a table, ever. Uninstall means "stop running this", and for a plugin holding a
// user's data the two are unrecoverably different — see `plugin_installs` schema.
//
// The schema push happens on install and NEVER on uninstall. `drizzle-kit push` drops what it cannot
// see, so the generated barrel follows the plugin DIRECTORIES rather than the install table — uninstall
// leaves both the barrel entry and every row alone, and only deleting a plugin's source can lose data.
// See plugins/schema.ts.

/**
 * Reported as each step completes, for the streaming endpoint.
 *
 * The runner does not know or care whether anyone is listening — it calls this and carries on, so the
 * non-streaming path is the same code with no callback rather than a second implementation.
 */
export type OnStep = (step: string) => void | Promise<void>;

/**
 * A beat between steps. Deliberate, and not decoration — do not remove it as a pointless sleep.
 *
 * Some of this work is genuinely slow (`pm2 start` measures ~770ms) and some is not: writing a row and
 * rebuilding the router are effectively instant. Without a pause the fast steps arrive in a single frame,
 * so the log jumps from empty to finished and the reader cannot tell "it worked" from "nothing happened".
 * The interval IS the feedback — it is what makes each step something you saw occur rather than something
 * you found already done.
 *
 * Small enough that a script does not care, long enough that a person can follow it. Only applied when
 * something is listening: `onStep` absent means no beat, so the JSON path runs at full speed.
 */
const STEP_BEAT_MS = 220;

export type PluginActionResult = {
  ok: boolean;
  appName: string;
  /** What actually happened, in order. Returned so the UI can show a real account rather than a spinner. */
  steps: string[];
  /** Present when a step failed. The plugin is left in the last coherent state above. */
  error?: string;
};

/** PM2 is only involved when the plugin actually has a sidecar. Most won't. */
const hasSidecar = (plugin: DiscoveredPlugin) => !!plugin.sidecar;

/** Record a step, tell whoever is listening, and pause so the next one does not land in the same frame. */
async function step(steps: string[], onStep: OnStep | undefined, text: string): Promise<void> {
  steps.push(text);
  await onStep?.(text);
  if (onStep) await Bun.sleep(STEP_BEAT_MS);
}

async function findPlugin(appName: string): Promise<DiscoveredPlugin | null> {
  const { states } = await snapshotPlugins();
  return states.find((s) => s.plugin.appName === appName)?.plugin ?? null;
}

/**
 * Install, or upgrade one already installed.
 *
 * Idempotent by construction: the ecosystem entry is replaced rather than appended, the row is an upsert,
 * and the mount is a rebuild. Re-running after a failure resumes rather than duplicating.
 *
 * `enabled` is deliberately untouched on the upgrade path — re-installing something the owner had
 * switched off must not switch it back on.
 */
export async function installPlugin(appName: string, onStep?: OnStep): Promise<PluginActionResult> {
  const plugin = await findPlugin(appName);
  if (!plugin) return { ok: false, appName, steps: [], error: `No plugin directory named "${appName}"` };

  const steps: string[] = [];
  try {
    // HOST BINARIES FIRST, before a table is created or a row is written.
    //
    // Music is why. Without `ffprobe` its indexer does not fail — it writes a complete library where
    // every track is a bare filename with no cover, and reports success. So a plugin whose dependencies
    // cannot be satisfied must not install at all: the alternative is a plugin that runs, answers 200,
    // and quietly produces nothing, which is far harder to diagnose than a refusal.
    //
    // First because it is the only step that can refuse for a reason the owner can act on, and refusing
    // before any effect means there is nothing to undo. Everything below this line changes the machine.
    const deps = reportDependencies(plugin);
    if (deps.missing.length) {
      if (!deps.canInstall) {
        const hint = manualInstallHint(deps.manager, deps.packagesToInstall);
        return {
          ok: false,
          appName,
          steps,
          error:
            `${appName} needs ${deps.missing.map((d) => d.binary).join(', ')}, which ${deps.missing.length === 1 ? 'is' : 'are'} not installed. ` +
            `${deps.blockedReason ?? ''}${hint ? ` Run: ${hint}` : ''}`.trim(),
        };
      }

      await step(steps, onStep, `dependencies: installing ${deps.packagesToInstall.join(', ')} with ${deps.manager}`);
      const installed = await installDependencies(deps.manager!, deps.packagesToInstall);
      if (!installed.ok) {
        return {
          ok: false,
          appName,
          steps,
          error: `dependency install failed: ${installed.error ?? installed.output}`,
        };
      }

      // Observed, not inferred. A package manager exiting 0 having installed something that does not
      // provide the binary the plugin spawns is exactly the failure this whole check exists to catch,
      // and trusting the exit code would reproduce it one layer up.
      const absent = stillMissing(plugin);
      if (absent.length) {
        return {
          ok: false,
          appName,
          steps,
          error: `installed ${deps.packagesToInstall.join(', ')}, but ${absent.join(', ')} still ${absent.length === 1 ? 'is' : 'are'} not on PATH`,
        };
      }
      await step(steps, onStep, `dependencies: ${deps.missing.map((d) => d.binary).join(', ')} now on PATH`);
    } else if (deps.dependencies.length) {
      // Said out loud rather than skipped silently: "it was already there" and "we never looked" are
      // different facts, and only one of them is reassuring.
      await step(steps, onStep, `dependencies: ${deps.dependencies.map((d) => d.binary).join(', ')} already present`);
    }

    if (plugin.schema) {
      // The barrel first, then the push. It is generated from the DIRECTORIES on disk rather than from
      // what is installed — see schema.ts for why that difference is the whole safety property.
      const { plugins } = await discoverPlugins();
      generatePluginSchemas(plugins);

      const pushed = await pushSchema();
      if (!pushed.ok) {
        return { ok: false, appName, steps, error: `schema push failed: ${pushed.output}` };
      }
      await step(steps, onStep, `schema: applied in ${pushed.ms}ms`);
    }

    await recordPluginInstall(appName, plugin.manifest.version);
    await step(steps, onStep, `recorded at ${plugin.manifest.version}`);

    // MOUNT BEFORE STARTING THE SIDECAR, and the order is not cosmetic.
    //
    // `createSidecarProxy` learns its sidecar's port from a one-shot event (`<name>:server`), and it
    // subscribes when the plugin's router module is first imported — which happens here, at mount. Start
    // the process first and it announces its port to nobody: the sidecar is online, the routes are
    // mounted, and every request answers `503 sidecar not available` until something makes it reconnect.
    //
    // Found installing offscale, whose sidecar binds its own HTTP server. `example` never caught it
    // because it has no listener to announce.
    const { mounted } = await refreshPluginMounts();
    await step(
      steps,
      onStep,
      mounted.length ? `mounted: ${mounted.join(', ')}` : 'mounted: nothing (no api/router.ts)',
    );

    if (hasSidecar(plugin)) {
      addPluginToEcosystem(plugin);
      await step(steps, onStep, `ecosystem: ${pluginProcessName(appName)} added`);

      const started = await startProcess(pluginProcessName(appName), PLATFORM_DIR);
      if (!started.ok) {
        // The entry stays. A sidecar that will not start is a plugin to retry or debug, and removing the
        // entry would take away the thing that makes `pm2 logs officer-<name>` work.
        return { ok: false, appName, steps, error: `sidecar failed to start: ${started.error}` };
      }
      await step(steps, onStep, 'sidecar: started');
    }

    return { ok: true, appName, steps };
  } catch (err) {
    return { ok: false, appName, steps, error: err instanceof Error ? err.message : String(err) };
  }
}

/**
 * Uninstall: stop answering, stop running, forget.
 *
 * Keeps every table and row the plugin owns, and keeps the directory. Reinstalling is therefore a restore
 * rather than a fresh start, which is the whole reason not to drop anything here.
 */
export async function uninstallPlugin(appName: string, onStep?: OnStep): Promise<PluginActionResult> {
  const steps: string[] = [];
  const plugin = await findPlugin(appName);

  // Row first: the mount rebuild below reads it, and a failure after this point leaves the plugin
  // unmounted and stopped rather than half-visible.
  const removed = await removePluginInstall(appName);
  if (!removed) return { ok: false, appName, steps, error: `"${appName}" is not installed` };
  await step(steps, onStep, 'install record removed');

  const { mounted } = await refreshPluginMounts();
  await step(steps, onStep, `unmounted (now: ${mounted.join(', ') || 'no plugin routes'})`);

  if (plugin && hasSidecar(plugin)) {
    await stopProcess(pluginProcessName(appName), PLATFORM_DIR);
    await deleteProcess(pluginProcessName(appName), PLATFORM_DIR);
    removePluginFromEcosystem(appName);
    await step(steps, onStep, 'sidecar: stopped, deleted, ecosystem entry removed');
  }

  await step(steps, onStep, 'tables and data: untouched');
  return { ok: true, appName, steps };
}

/** Enable: mount and run again. Disable: the reversible middle — unmount and stop, keep everything. */
export async function setPluginRunning(
  appName: string,
  enabled: boolean,
  onStep?: OnStep,
): Promise<PluginActionResult> {
  const steps: string[] = [];
  const row = await setPluginEnabled(appName, enabled);
  if (!row) return { ok: false, appName, steps, error: `"${appName}" is not installed` };
  await step(steps, onStep, enabled ? 'enabled' : 'disabled');

  const plugin = await findPlugin(appName);

  // Enabling mounts BEFORE starting, for the same reason install does: the proxy has to be listening
  // before the sidecar announces its port. Disabling is the mirror — stop answering, then stop the
  // process — so neither direction leaves a mounted route in front of a sidecar that cannot be reached.
  if (enabled) {
    const { mounted } = await refreshPluginMounts();
    await step(steps, onStep, `mounts: ${mounted.join(', ') || 'no plugin routes'}`);
  }

  if (plugin && hasSidecar(plugin)) {
    const name = pluginProcessName(appName);
    const result = enabled ? await startProcess(name, PLATFORM_DIR) : await stopProcess(name, PLATFORM_DIR);
    await step(steps, onStep, result.ok ? `sidecar: ${enabled ? 'started' : 'stopped'}` : `sidecar: ${result.error}`);
  }

  if (!enabled) {
    const { mounted } = await refreshPluginMounts();
    await step(steps, onStep, `mounts: ${mounted.join(', ') || 'no plugin routes'}`);
  }
  return { ok: true, appName, steps };
}

/** Whether a plugin's sidecar is actually up, for the UI. `null` when it has none or PM2 has not seen it. */
export async function pluginProcessStatus(plugin: DiscoveredPlugin): Promise<string | null> {
  if (!hasSidecar(plugin)) return null;
  return processStatus(pluginProcessName(plugin.appName), PLATFORM_DIR);
}