feat: the first pages, and what a browser walk found behind them

Phase 4. `/rust` is the server list and the module's landing page (D12);
`/rust/servers/:id` is one server with four tabs — feed, leaderboard, who is
on, wipes (D13). Everything selectable lives in the URL, so any view of the
page is a link. The feed and the presence list poll every twenty seconds while
the tab is visible and not at all when it is not (D14); the leaderboard and the
wipe list load once. `site.footer.status` is filled with a live server and
player count (D15).

Nothing on these pages calls a game server. Every field comes from this
module's own tables, which is what the phase criterion is about: the site
renders the last thing each server said while every server is off.

Walking that criterion in a browser against a live rig found four defects, two
of them already shipped in phase 3:

  * An unreachable refresh called `putState` — the whole-row write — with two
    fields, so a host that rebooted lost its hostname, map, size, seed and wipe
    id. The list then read "Offline" with nothing beside it, which is not "here
    is what we know" but "we have never heard of it". `markUnreachable` now
    moves three columns and mentions no others.
  * "Last reported" read `updated_at`, which a FAILED poll writes too — so an
    offline server claimed it had reported just now, every thirty seconds, for
    as long as it stayed down. `last_seen_at` is the new column, moved only by a
    frame that arrived.
  * Feed rows showed a bare time of day, so three events from six weeks ago all
    read as this afternoon once the feed was filtered to a past wipe.
  * `/rust/servers/typo` rendered core's ErrorState under its own heading and
    read "No such server / Something went wrong", sending a reader who mistyped
    a URL looking for an outage.

Also: a detail route (`GET …/servers/:id`), because it is the only route under
that path that can say a server does not exist — the other four answer an empty
list for an id nobody configured, and each of those is a good answer to its own
question.

`useAsync` cannot poll: it blanks its data on every dependency change, so a
twenty-second refresh built on it would clear the killfeed and re-fill it four
times a minute. `hooks/usePolled.js` is the module's own, invisible when it
succeeds and keeping the rows when it fails.

The client test fake was *nearly* core — it prefixed routes without stripping
the trailing separator, so the first module to register an index route failed
the nav check for a link that works in a browser. It now copies core's line
character for character.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_016wDDVXWMDz82WqE1i969r4

server/model/servers/servers.db.js

@@ -80,11 +80,56 @@ async function listState() {
     `SELECT server_id AS serverId, reachable, online, players, max_players AS maxPlayers,
             hostname, level, seed, world_size AS worldSize, boot_id AS bootId,
             save_created_at AS saveCreatedAt, wipe_id AS wipeId, protocol,
-            updated_at AS updatedAt
+            last_seen_at AS lastSeenAt, updated_at AS updatedAt
        FROM ${STATE}`,
   )
 }
 
+/** One server's observed state, or `null`. The single-row twin of `listState`. */
+async function getState(serverId) {
+  const rows = await core.query(
+    `SELECT server_id AS serverId, reachable, online, players, max_players AS maxPlayers,
+            hostname, level, seed, world_size AS worldSize, boot_id AS bootId,
+            save_created_at AS saveCreatedAt, wipe_id AS wipeId, protocol,
+            last_seen_at AS lastSeenAt, updated_at AS updatedAt
+       FROM ${STATE}
+      WHERE server_id = ?`,
+    [serverId],
+  )
+  return rows[0] || null
+}
+
+/**
+ * Mark a server unreachable **without forgetting what it last said**.
+ *
+ * `putState` replaces the row whole, which is right when a sidecar answered: the
+ * frame it answered with is the complete truth about that server. It is wrong
+ * when nothing answered. A refresh that cannot reach a sidecar knows exactly one
+ * new fact — that it could not reach it — and writing the whole row from that
+ * one fact sets `hostname`, `level`, `seed`, `world_size` and `wipe_id` to NULL.
+ *
+ * The site's whole premise is that it renders the last thing each server said
+ * while every server is off. A row blanked the first time a game host reboots
+ * cannot do that: the page loses the map, the size, the seed and the wipe, and
+ * what it shows is not "offline, here is what we know" but "offline, and we have
+ * never heard of it". It is invisible in every test that stubs a reachable
+ * sidecar, and it shows up as a page that was complete an hour ago.
+ *
+ * So: three columns move, and the description stays where it is.
+ */
+async function markUnreachable(serverId, reachable = false) {
+  await core.query(
+    `INSERT INTO ${STATE} (server_id, reachable, online, players, updated_at)
+     VALUES (?, ?, 0, 0, CURRENT_TIMESTAMP)
+     ON DUPLICATE KEY UPDATE
+       reachable = VALUES(reachable),
+       online = 0,
+       players = 0,
+       updated_at = CURRENT_TIMESTAMP`,
+    [serverId, reachable ? 1 : 0],
+  )
+}
+
 /**
  * Replace one server's observed state.
  *
@@ -100,15 +145,20 @@ async function putState(state) {
   await core.query(
     `INSERT INTO ${STATE}
        (server_id, reachable, online, players, max_players, hostname, level, seed,
-        world_size, boot_id, save_created_at, wipe_id, protocol, raw, updated_at)
-     VALUES (?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, CURRENT_TIMESTAMP)
+        world_size, boot_id, save_created_at, wipe_id, protocol, raw, last_seen_at, updated_at)
+     VALUES (?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, CURRENT_TIMESTAMP, CURRENT_TIMESTAMP)
      ON DUPLICATE KEY UPDATE
        reachable = VALUES(reachable), online = VALUES(online), players = VALUES(players),
        max_players = VALUES(max_players), hostname = VALUES(hostname), level = VALUES(level),
        seed = VALUES(seed), world_size = VALUES(world_size), boot_id = VALUES(boot_id),
        save_created_at = VALUES(save_created_at), wipe_id = VALUES(wipe_id),
        protocol = VALUES(protocol),
-       raw = VALUES(raw), updated_at = CURRENT_TIMESTAMP`,
+       raw = VALUES(raw),
+       -- Only a frame moves this; an unreachable write leaves it alone, which is
+       -- what lets a page say how long a server has been down rather than how
+       -- recently we failed to reach it.
+       last_seen_at = CURRENT_TIMESTAMP,
+       updated_at = CURRENT_TIMESTAMP`,
     [
       state.serverId,
       state.reachable ? 1 : 0,
@@ -136,5 +186,7 @@ module.exports = {
   upsertServer,
   deleteServer,
   listState,
+  getState,
+  markUnreachable,
   putState,
 }

server/model/servers/servers.model.js

@@ -73,6 +73,7 @@ async function listPublic(now = Date.now()) {
 
 function shapePublic(row, state, now) {
   const updatedAt = state && state.updatedAt ? new Date(state.updatedAt) : null
+  const lastSeenAt = state && state.lastSeenAt ? new Date(state.lastSeenAt) : null
   const stale = !updatedAt || now - updatedAt.getTime() > STALE_AFTER_MS
 
   return {
@@ -87,11 +88,43 @@ function shapePublic(row, state, now) {
     level: (state && state.level) || null,
     worldSize: state && state.worldSize != null ? Number(state.worldSize) : null,
     seed: state && state.seed != null ? Number(state.seed) : null,
+    // The CURRENT wipe, from the state row rather than from the newest row in
+    // `rust_wipes`. The two usually agree and the state row is the one that is
+    // right when they do not: a wipe list is derived from events that have been
+    // ingested, so a server that has just wiped and said nothing since has a new
+    // wipe id here and no row there at all.
+    wipeId: (state && state.wipeId) || null,
+    wipedAt: (state && state.saveCreatedAt) || null,
+    // Two timestamps, because they are two facts. `lastSeenAt` is when a frame
+    // last arrived and is what a page means by "last reported"; `updatedAt` is
+    // when this module last wrote the row, and is what `stale` is computed from.
+    // Reading the second as the first is what made an offline server claim it had
+    // reported just now, on every failed poll, for as long as it stayed down.
+    lastSeenAt: lastSeenAt ? lastSeenAt.toISOString() : null,
     updatedAt: updatedAt ? updatedAt.toISOString() : null,
     stale,
   }
 }
 
+/**
+ * One enabled server, or `null`.
+ *
+ * It exists because `/rust/servers/:id` is a page and a page needs to be able to
+ * 404. A detail view built by fetching the list and finding the row in it cannot
+ * tell "no such server" from "a server that has said nothing" — both are an
+ * absence — and renders an empty page under a heading for a server that does not
+ * exist. Filtering happens here, where `enabled = 0` and "never configured" are
+ * the same answer on purpose: a disabled server is not a 403, it is not there.
+ */
+async function getPublic(id, now = Date.now()) {
+  if (!id) return null
+
+  const row = await db.getServer(id)
+  if (!row || !row.enabled) return null
+
+  return shapePublic(row, await db.getState(row.id), now)
+}
+
 /**
  * The admin view: configuration plus reachability, and **no token**.
  *
@@ -133,6 +166,7 @@ module.exports = {
   withToken,
   listForPolling,
   listPublic,
+  getPublic,
   listForAdmin,
   shapePublic,
   encryptToken,