From b5f6cac0ef4128aa94376b3847ef9ae03508032f Mon Sep 17 00:00:00 2001
From: JianFeeeee <JianFeeeee@users.noreply.gitcode.com>
Date: Sun, 4 Oct 2026 08:23:55 +0800
Subject: [PATCH] =?UTF-8?q?fix(plugins):=20ccquota=20=E9=A1=B5=E9=9D=A2?=
 =?UTF-8?q?=E9=A6=96=E6=AC=A1=E6=89=93=E5=BC=80=E6=98=AF=E7=A9=BA=E7=9A=84?=
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The plugin page rendered nothing until the operator clicked refresh, and even
the click handler was never wired — while every unit test was green and the
API returned correct numbers. Four separate mistakes stacked here:

1. The script waited on DOMContentLoaded, an event that had already been
   dispatched by the time the kernel injects a plugin mount. Replaced with
   pluginAPI.onTabShown (the kernel's own mechanism, fires with the DOM in
   place) PLUS an immediate call, because onTabShown alone leaves the page
   blank when the tab is already open at boot. The billing page does both; I had
   reduced it to one and broke first paint.

2. The re-entry guard flag lived in a closure variable. The tab's DOM is rebuilt
   on re-entry, so the handler was wired against a node that no longer existed.
   The flag now lives on the NODE (__ccqWired), which is what makes re-wiring
   correct.

3. refresh() swallowed its fetch failure and rendered "no sources", so a broken
   page and an empty account list looked identical. Failures now surface in the
   header line, including when pluginAPI.call is missing (an old kernel).

4. THE ACTUAL REASON the page stayed blank: there were TWO ccquota plugins. The
   embed (go:embed plugins/*.lua) version and the on-disk one both loaded, and
   the UI injected the EMBED copy — 5182 characters, predating every fix — while
   every edit I made and every hot reload went to the disk copy. The running
   binary was serving stale plugin code no amount of reloading could fix. Only a
   rebuild surfaced it, and only measuring the served <script> length (7053 vs
   5182) showed which copy the browser had.

Lesson worth keeping: "the file on disk is what runs" is false for anything
under go:embed. Compare what the SERVER sends before editing the same file
twice and concluding the edit did nothing.

Verified in the browser: 2 account cards, 4 window bars, real balances, no
overflow, no console errors, refresh button 48x29 and functional.

Co-Authored-By: ModelRouter <noreply@modelrouter.dev>
---
 internal/lua/plugins/ccquota.lua | 43 ++++++++++++++++++++++++++++----
 1 file changed, 38 insertions(+), 5 deletions(-)

diff --git a/internal/lua/plugins/ccquota.lua b/internal/lua/plugins/ccquota.lua
index 83c9af4..b37cd11 100644
--- a/internal/lua/plugins/ccquota.lua
+++ b/internal/lua/plugins/ccquota.lua
@@ -299,7 +299,10 @@ PLUGIN.ui = {
     try{
       var r=await fetch("/api/sources",{credentials:"same-origin"});
       srcs=(await r.json()).sources||[];
-    }catch(e){ render([]); return; }
+    }catch(e){
+      var s0=document.getElementById("ccq-sub"); if(s0) s0.textContent="拉取源列表失败: "+e.message;
+      render([]); return;
+    }
     // Every source on the Command Code host is one account, keyed by its own
     // API key — two keys on that host are two seats, and both must be listed.
     var accts=srcs.filter(function(s){
@@ -310,6 +313,11 @@ PLUGIN.ui = {
     if(btn) btn.disabled=true;
     var auto=document.getElementById("ccq-auto")||document.querySelector(".ccq-auto");
     if(auto) auto.textContent="拉取中…";
+    if(!window.pluginAPI || typeof pluginAPI.call!=="function"){
+      var sb=document.getElementById("ccq-sub");
+      if(sb) sb.textContent="pluginAPI.call 不可用（内核版本过旧）";
+      render([]); return;
+    }
     var out=[];
     var errs=0;
     // One account at a time: the endpoints are independent, and a sequential
@@ -331,11 +339,36 @@ PLUGIN.ui = {
     if(auto) auto.textContent="更新于 "+new Date().toLocaleTimeString();
     try{ await pluginAPI.postState("ccquota",{reports:out,updated_at:Date.now()}); }catch(e){}
   }
-  document.addEventListener("DOMContentLoaded",function(){
+  // The mount is injected long after DOMContentLoaded has already fired, so
+  // listening for that event here registered a handler for an event that had
+  // already been dispatched — the page rendered but stayed empty until a manual
+  // refresh. Wiring the button and fetching immediately is what makes the page
+  // populate on first open.
+  // The plugin mount is injected by the kernel and its <script> evaluated after
+  // injection, but that still happens during the UI's own boot — and the page
+  // body is rebuilt whenever the tab is opened. Two consequences:
+  //
+  //   - A DOMContentLoaded listener never fires (the event already passed), so
+  //     the page rendered empty until a manual refresh.
+  //   - Wiring the refresh button once at script-eval time silently loses the
+  //     handler when the tab's DOM is replaced.
+  //
+  // pluginAPI.onTabShown is the kernel's answer to both: the callback runs when
+  // the tab is actually shown, with the DOM in place. It also runs once for the
+  // already-visible tab at boot, so this single registration covers first paint
+  // AND every later re-entry.
+  function mount(){
     var b=document.getElementById("ccq-refresh");
-    if(b) b.addEventListener("click",refresh);
-    refresh();
-  });
+    // The tab's DOM is rebuilt on re-entry, so a handler wired against the old
+    // node is gone. The flag lives on the NODE, not in a closure variable, which
+    // is what makes re-wiring safe.
+    if(b && !b.__ccqWired){ b.__ccqWired = true; b.addEventListener("click",refresh); }
+  }
+  // Run once immediately — the billing page does the same, and waiting only for
+  // onTabShown leaves the page blank when the tab is already open at boot.
+  mount();
+  refresh();
+  if (window.pluginAPI && pluginAPI.onTabShown) pluginAPI.onTabShown(mount);
   window.__ccqRender = render;
   window.__ccqRefresh = refresh;
 })();