feat(node): persistent index cache, visible scan progress, adaptive reconcile, and delta sync

Indexer performance work, in four parts:

- Persistent (path, size, mtime) -> hash cache (indexer/cache.py) so a node
  restart no longer re-hashes every file — measured at 23 minutes for a
  114 GB library on a slow disk before this, near-instant after. Hashing
  is deliberately kept sequential (max_workers=1): it was never actually
  concurrent despite the pool size, and two interleaved reads seek-thrash
  a spinning disk instead of going faster.

- Byte-based scan progress (IndexProgress), surfaced via the loopback
  index-status route, the handshake ack, and a periodic INDEX_PROGRESS
  push to connected peers — drives a progress bar in the Create Group
  wizard and "add a directory" in Settings, and an animated presence dot.
  Guaranteed to settle back to idle via try/finally and a final push on
  the scanning->false transition.

- The reconcile backstop's directory walks now run in the executor
  instead of blocking the daemon's event loop; its interval defaults to
  10 min (was 60s) with adaptive backoff to 2h when nothing changes,
  reset on a real change or a peer connecting, and is now a per-group
  operator setting (signed op + group Settings UI).

- INDEX_DELTA wired up (protocol support existed, nothing called it):
  _on_index_change now sends additions/deletions instead of rebuilding
  the full entries list, coalesced over a short window so a burst of
  file events produces one push, and the hub swarm registration for
  public groups only (re-)registers newly added hashes.

Also fixes several bugs found while testing the above against real
libraries (a 114 GB and a 100+ GB group on a USB HDD):

- /api/reload blocked until the reload — including a brand-new group's
  full initial scan — finished, which the Electron bridge's fixed 30s
  call timeout turned into a hard failure on any real library. The route
  now fires the reload without waiting (ops.start_reload), matching
  add_root/remove_root's existing pattern; the wizard's own step order
  was fixed to wait for the group to actually appear hosted before the
  steps that need it (extra roots, GEK), with retries for the residual
  race between that and the daemon's own bookkeeping.
- transport.js's hand-rolled msgpack codec had no case for uint64/int64
  (0xcf/0xd3) and crashed decoding any message containing one — hit by
  IndexProgress.scanned_bytes/total_bytes for any group over ~4.3 GB.
  Verified against real msgpack-encoded bytes from the Python side.
- chat_hist_resp, and this change's own index_progress and
  set_scan_settings_ack pushes, were not routed by message type and
  could be handed to an unrelated pending request by the transport's
  "oldest pending" fallback, stalling it until its own 30s timeout and
  corrupting whatever received the wrong reply in its place.

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_016SF6RKNBKg9qejmoMJ9ybA

1 files changed, 110 insertions, 1 deletions

diff --git a/packages/meshbay-hub/src/meshbay_hub/static/group-settings.js b/packages/meshbay-hub/src/meshbay_hub/static/group-settings.js
index a14fd21..0b64a03 100644
--- a/packages/meshbay-hub/src/meshbay_hub/static/group-settings.js
+++ b/packages/meshbay-hub/src/meshbay_hub/static/group-settings.js
@@ -22,6 +22,7 @@ function GroupSettingsPanel({ groupId, group, token, transportRef, gekRef,
                              isNodeAdmin, userId, operatorPaired, connected,
                              memberUpload, onMemberUpload,
                              enabledApps, onEnabledApps,
+                             scanSettings, onScanSettings,
                              onPaired, onLeft }) {
   const [members, setMembers] = useState([]);
   const [adminId, setAdminId] = useState('');
@@ -36,6 +37,10 @@ function GroupSettingsPanel({ groupId, group, token, transportRef, gekRef,
   const [nodeGroupName, setNodeGroupName] = useState('');
   const [nodeBusy, setNodeBusy] = useState(false);
   const [nodeMsg, setNodeMsg] = useState('');
+  // Bytes-based indexing progress while a newly added directory is being
+  // scanned — same source as the Create Group wizard's step, see
+  // platform.watchIndexProgress.
+  const [nodeIndexProgress, setNodeIndexProgress] = useState(null);
 
   const loadNodeInfo = useCallback(async () => {
     if (!platform.node.available) return;
@@ -194,6 +199,51 @@ function GroupSettingsPanel({ groupId, group, token, transportRef, gekRef,
     }
   }, [transportRef, onEnabledApps, activeApps]);
 
+  const [scanBusy, setScanBusy] = useState(false);
+  const [scanMsg, setScanMsg] = useState('');
+  const [reconcileMinutes, setReconcileMinutes] = useState(
+    scanSettings ? Math.round(scanSettings.reconcile_interval_secs / 60) : 10);
+  const [debounceSeconds, setDebounceSeconds] = useState(
+    scanSettings ? Math.round(scanSettings.debounce_secs) : 2);
+  // The node is the source of truth; once it has answered, the fields track
+  // it rather than whatever this browser guessed before connecting.
+  useEffect(() => {
+    if (!scanSettings) return;
+    setReconcileMinutes(Math.round(scanSettings.reconcile_interval_secs / 60));
+    setDebounceSeconds(Math.round(scanSettings.debounce_secs));
+  }, [scanSettings]);
+
+  /**
+   * How often the reconciliation backstop runs, and how long a changed file
+   * is left alone before being hashed. Same shape as toggleApp: signed, and
+   * the fields do not claim success until the node has confirmed it.
+   */
+  const saveScanSettings = useCallback(async () => {
+    const transport = transportRef && transportRef.current;
+    setScanMsg('');
+    setScanBusy(true);
+    try {
+      if (!transport || !transport.connected) {
+        throw new Error('Not connected to the node');
+      }
+      const sk = transport.sessionKeys && transport.sessionKeys.skEdB64;
+      const signFn = (sk && window.MeshBayKeys)
+        ? (transcript) => window.MeshBayKeys.signBytes(sk, transcript)
+        : null;
+      await transport.setScanSettings(reconcileMinutes * 60, debounceSeconds, signFn);
+      const applied = {
+        reconcile_interval_secs: reconcileMinutes * 60,
+        debounce_secs: debounceSeconds,
+      };
+      if (onScanSettings) onScanSettings(applied);
+      setScanMsg(t('settings_node.scan_saved'));
+    } catch (err) {
+      setScanMsg(err.message);
+    } finally {
+      setScanBusy(false);
+    }
+  }, [transportRef, onScanSettings, reconcileMinutes, debounceSeconds]);
+
   const [removing, setRemoving] = useState('');
 
   /**
@@ -382,6 +432,37 @@ function GroupSettingsPanel({ groupId, group, token, transportRef, gekRef,
         </div>
       `}
 
+      ${/* How hard the node works watching its own disk — indexer.py
+            DirectoryIndexer. A performance knob, not a permission: it
+            changes nothing about who can see or do what. */
+        isNodeAdmin && connected && html`
+        <div class="settings-section">
+          <h3 class="settings-heading">${t('settings_node.scan_title')}</h3>
+          <p class="settings-hint">${t('settings_node.scan_hint')}</p>
+          <div class="settings-row">
+            <label class="settings-label">
+              ${t('settings_node.scan_reconcile_label')}
+              <input type="number" min="1" max="1440" step="1"
+                value=${reconcileMinutes} disabled=${scanBusy}
+                onInput=${e => setReconcileMinutes(Number(e.target.value))} />
+            </label>
+          </div>
+          <div class="settings-row">
+            <label class="settings-label">
+              ${t('settings_node.scan_debounce_label')}
+              <input type="number" min="0" max="300" step="1"
+                value=${debounceSeconds} disabled=${scanBusy}
+                onInput=${e => setDebounceSeconds(Number(e.target.value))} />
+            </label>
+          </div>
+          <button class="btn btn-small btn-secondary" style="margin-top:8px"
+            disabled=${scanBusy} onClick=${saveScanSettings}>
+            ${scanBusy ? t('settings_node.scan_saving') : t('settings_node.scan_save')}
+          </button>
+          ${scanMsg && html`<p class="settings-hint">${scanMsg}</p>`}
+        </div>
+      `}
+
       ${/* Roots management (Electron-only, when node is local) */
         nodeDetected && nodeRoots.length > 0 && html`
         <div class="settings-section">
@@ -425,12 +506,18 @@ function GroupSettingsPanel({ groupId, group, token, transportRef, gekRef,
               onClick=${async () => {
                 const chosen = await platform.rootPicker.choose();
                 if (!chosen) return;
-                setNodeBusy(true); setNodeMsg('');
+                setNodeBusy(true); setNodeMsg(''); setNodeIndexProgress(null);
                 try {
                   await platform.node.call('POST',
                     '/api/groups/' + groupId + '/roots',
                     { path: chosen.path, name: chosen.name });
                   await platform.node.call('POST', '/api/reload');
+                  // The root is already scanning in the background on the
+                  // node regardless of whether anyone watches this — see
+                  // the "closing the client" test in test_hot_reload_*.py.
+                  // This is only about not leaving the operator staring at
+                  // an unchanged screen while it happens.
+                  await platform.watchIndexProgress(groupId, setNodeIndexProgress);
                   setNodeMsg(t('node.root_added'));
                   await loadNodeInfo();
                 } catch (err) { setNodeMsg(platform.bridgeMessage(err)); }
@@ -438,6 +525,28 @@ function GroupSettingsPanel({ groupId, group, token, transportRef, gekRef,
               }}>
               <${Icon} name="folder-plus" /> ${t('node.add_root')}
             </button>
+            ${nodeIndexProgress && nodeIndexProgress.scanning && html`
+              <div class="index-progress" style="margin-top:8px">
+                <div class="index-progress-bar">
+                  <div class="index-progress-fill" style="width:${
+                    nodeIndexProgress.total_bytes
+                      ? Math.min(100, Math.round(
+                          100 * nodeIndexProgress.scanned_bytes / nodeIndexProgress.total_bytes))
+                      : 0}%"></div>
+                </div>
+                <div class="index-progress-label">${t('wizard.indexing_progress', {
+                  pct: nodeIndexProgress.total_bytes
+                    ? Math.min(100, Math.round(
+                        100 * nodeIndexProgress.scanned_bytes / nodeIndexProgress.total_bytes))
+                    : 0,
+                })}</div>
+                ${nodeIndexProgress.current_dir && html`
+                  <div class="index-progress-dir">
+                    ${t('wizard.indexing_current_dir', { dir: nodeIndexProgress.current_dir })}
+                  </div>
+                `}
+              </div>
+            `}
           </div>
         </div>
       `}