@nebulr-group/bridge-svelte 0.8.1 → 0.8.2

package/dist/core/bridge-runtime.js

@@ -303,8 +303,16 @@ export function startBridgeRuntime(options = {}) {
         // TBP-660 — the TOKEN refresh above is skipped for a self-induced
         // reconnect; the STATE catch-up is not. Losing pushes is a property of
         // the socket swap, whoever caused it, and the catch-up cannot loop.
-        if (_connectedOnce)
-            requestCatchUp();
+        //
+        // TBP-686 — and it runs on the FIRST connect too. It used to be gated on
+        // `_connectedOnce`, which read as "only repair a reconnect" — but the
+        // first connect is exactly where the snapshot goes missing: the server
+        // publishes it fire-and-forget during authorize, before the subscription
+        // is live, so it loses the race and there is nothing to replay. The result
+        // was a signed-in page with a null workspace id, name and branding for the
+        // whole session. `requestCatchUp` already returns early when signed out
+        // and coalesces concurrent calls, so this costs one request per session.
+        requestCatchUp();
         _connectedOnce = true;
         for (const fn of _onOpenSubs) {
             try {

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@nebulr-group/bridge-svelte",
-  "version": "0.8.1",
+  "version": "0.8.2",
   "description": "Bridge Svelte library, This library helps you to add bridge authentication and feature flags, and payments to your svelte application.",
   "author": "Iman Pouya",
   "license": "MIT",
@@ -79,7 +79,7 @@
     "typescript": "^6.0.0",
     "vite": "^6.2.6",
     "vitest": "^4.1.4",
-    "@nebulr-group/bridge-auth-core": "0.7.1"
+    "@nebulr-group/bridge-auth-core": "0.7.2"
   },
   "keywords": [
     "svelte",