fix(logs): preserve standalone detail token labels

changelog.d/features/9620-cache-read-write-logs.md

@@ -0,0 +1,2 @@
+- Show cache-read and cache-write token counts in request log rows and details when providers
+  report them.

src/shared/components/RequestLoggerV2.tsx

@@ -95,6 +95,7 @@ const RequestLoggerV2 = forwardRef<RequestLoggerV2Handle, { initialSelectedId?:
   (props, ref) => {
     const { initialSelectedId } = props as any;
     const t = useTranslations("requestLogger");
+    const tCache = useTranslations("cache");
     const { emailsVisible } = useEmailPrivacyStore();
 
     // Get translated status filters
@@ -1514,6 +1515,30 @@ const RequestLoggerV2 = forwardRef<RequestLoggerV2Handle, { initialSelectedId?:
                                 <span className="text-emerald-700 dark:text-emerald-400">
                                   {log.tokens?.out?.toLocaleString() || 0}
                                 </span>
+                                {log.tokens?.cacheRead != null && log.tokens.cacheRead > 0 && (
+                                  <>
+                                    <span className="mx-1 text-border">|</span>
+                                    <span className="text-text-muted">CR:</span>{" "}
+                                    <span
+                                      className="text-sky-700 dark:text-sky-400"
+                                      title={tCache("cachedTokensCol")}
+                                    >
+                                      {log.tokens.cacheRead.toLocaleString()}
+                                    </span>
+                                  </>
+                                )}
+                                {log.tokens?.cacheWrite != null && log.tokens.cacheWrite > 0 && (
+                                  <>
+                                    <span className="mx-1 text-border">|</span>
+                                    <span className="text-text-muted">CW:</span>{" "}
+                                    <span
+                                      className="text-amber-700 dark:text-amber-400"
+                                      title={tCache("cacheCreation")}
+                                    >
+                                      {log.tokens.cacheWrite.toLocaleString()}
+                                    </span>
+                                  </>
+                                )}
                                 {log.tokens?.compressed != null && log.tokens.compressed > 0 && (
                                   <>
                                     <span className="mx-1 text-border">|</span>

tests/unit/ui/request-logger-cache-tokens.test.tsx

@@ -0,0 +1,158 @@
+// @vitest-environment jsdom
+import React, { act } from "react";
+import { createRoot, type Root } from "react-dom/client";
+import { afterEach, beforeEach, describe, expect, it, vi } from "vitest";
+
+vi.mock("next-intl", () => ({
+  useTranslations: (namespace?: string) => (key: string) =>
+    namespace === "cache"
+      ? ({ cachedTokensCol: "Cache Read", cacheCreation: "Cache Write" }[key] ?? key)
+      : key,
+}));
+
+vi.mock("next/navigation", () => ({
+  useRouter: () => ({ push: vi.fn(), replace: vi.fn(), refresh: vi.fn() }),
+}));
+
+vi.mock("@/store/emailPrivacyStore", () => ({
+  default: () => ({ emailsVisible: true }),
+}));
+
+const RequestLoggerV2 = (await import("@/shared/components/RequestLoggerV2")).default;
+const RequestLoggerDetail = (await import("@/shared/components/RequestLoggerDetail")).default;
+
+let container: HTMLElement;
+let root: Root;
+
+const populatedLog = {
+  id: "log-cache",
+  status: 200,
+  method: "POST",
+  path: "/v1/chat/completions",
+  model: "gpt-cache",
+  provider: "openai",
+  timestamp: "2026-08-10T12:00:00.000Z",
+  duration: 1_000,
+  tokens: {
+    in: 1_000,
+    out: 250,
+    cacheRead: 800,
+    cacheWrite: 120,
+    reasoning: 50,
+    compressed: 20,
+  },
+};
+
+const noop = () => {};
+
+async function render(component: React.ReactNode) {
+  await act(async () => {
+    root.render(component);
+  });
+}
+
+beforeEach(() => {
+  container = document.createElement("div");
+  document.body.appendChild(container);
+  root = createRoot(container);
+});
+
+afterEach(async () => {
+  await act(async () => {
+    root.unmount();
+  });
+  container.remove();
+  vi.unstubAllGlobals();
+});
+
+describe("request log cache token metrics (#9620)", () => {
+  it("renders cache read/write beside the existing row token metrics", async () => {
+    const emptyCacheLog = {
+      ...populatedLog,
+      id: "log-no-cache",
+      model: "gpt-no-cache",
+      timestamp: "2026-08-10T11:59:00.000Z",
+      tokens: { ...populatedLog.tokens, cacheRead: null, cacheWrite: 0 },
+    };
+    vi.stubGlobal(
+      "fetch",
+      vi.fn(async (input: RequestInfo | URL) => {
+        const url = String(input);
+        if (url.startsWith("/api/usage/call-logs")) {
+          return Response.json([populatedLog, emptyCacheLog]);
+        }
+        if (url.startsWith("/api/provider-nodes")) return Response.json({ nodes: [] });
+        if (url.startsWith("/api/logs/detail")) return Response.json({ enabled: false });
+        return Response.json({});
+      })
+    );
+
+    await render(<RequestLoggerV2 />);
+    await act(async () => {
+      await Promise.resolve();
+    });
+
+    const row = Array.from(container.querySelectorAll("tbody tr")).find((candidate) =>
+      candidate.textContent?.includes("gpt-cache")
+    );
+    expect(row?.textContent).toContain("TI: 1,000");
+    expect(row?.textContent).toContain("TO: 250");
+    expect(row?.textContent).toContain("CR: 800");
+    expect(row?.textContent).toContain("CW: 120");
+    expect(row?.textContent).toContain("↓20");
+
+    const emptyRow = Array.from(container.querySelectorAll("tbody tr")).find((candidate) =>
+      candidate.textContent?.includes("gpt-no-cache")
+    );
+    expect(emptyRow?.textContent).toContain("TI: 1,000");
+    expect(emptyRow?.textContent).toContain("TO: 250");
+    expect(emptyRow?.textContent).not.toContain("CR:");
+    expect(emptyRow?.textContent).not.toContain("CW:");
+  });
+
+  it("distinguishes cache read from cache write in the detail view", async () => {
+    await render(
+      <RequestLoggerDetail
+        log={populatedLog}
+        detail={populatedLog}
+        loading={false}
+        debugEnabled={false}
+        onClose={noop}
+        onCopy={async () => true}
+      />
+    );
+
+    const inputGroup = container.querySelector('[data-testid="token-group-input"]');
+    const outputGroup = container.querySelector('[data-testid="token-group-output"]');
+    expect(inputGroup?.textContent).toContain("Total In: 1,000");
+    expect(inputGroup?.textContent).toContain("Cache Read: 800");
+    expect(inputGroup?.textContent).toContain("Cache Write: 120");
+    expect(inputGroup?.textContent).toContain("Compressed:");
+    expect(outputGroup?.textContent).toContain("Total Out: 250");
+    expect(outputGroup?.textContent).toContain("Reasoning: 50");
+  });
+
+  it("handles historical null and zero cache values without inventing usage", async () => {
+    const emptyCacheLog = {
+      ...populatedLog,
+      id: "log-no-cache",
+      tokens: { ...populatedLog.tokens, cacheRead: null, cacheWrite: 0 },
+    };
+
+    await render(
+      <RequestLoggerDetail
+        log={emptyCacheLog}
+        detail={emptyCacheLog}
+        loading={false}
+        debugEnabled={false}
+        onClose={noop}
+        onCopy={async () => true}
+      />
+    );
+
+    const inputGroup = container.querySelector('[data-testid="token-group-input"]');
+    expect(inputGroup?.textContent).toContain("Cache Read: N/A");
+    expect(inputGroup?.textContent).toContain("Cache Write: 0");
+    expect(inputGroup?.textContent).toContain("Total In: 1,000");
+  });
+});