fix(cli): capture writes inline SVG logos whole and parakeet takes long audio (#5662)

* fix(cli): capture writes inline SVG logos whole and parakeet takes long audio

* fix(cli): keep the decode timeout comment under the line limit

packages/cli/src/capture/assetDownloader.test.ts

@@ -351,6 +351,7 @@ describe("drop counts — why a referenced asset is not in the capture", () => {
       expect(css).toContain("assets/fonts/font-0.woff2");
       expect(drops).toEqual({
         "size-floor": 0,
+        "size-ceiling": 0,
         "budget-exhausted": 0,
         "cap-reached": 0,
         unavailable: 0,
@@ -421,6 +422,32 @@ describe("drop counts — why a referenced asset is not in the capture", () => {
       expect(drops["cap-reached"]).toBe(4);
     });
   });
+
+  it("writes an inline SVG logo over 10 KB whole, and skips one over the size cap uncut", async () => {
+    await withTempDir(async (dir) => {
+      const shapes = (n: number) =>
+        Array.from({ length: n }, (_, i) => `<path d="M${i} 0h1v10h-1z"/>`).join("");
+      const logo = `<svg viewBox="0 0 3000 10">${shapes(3000)}</svg>`;
+      const huge = `<svg viewBox="0 0 1 10">${shapes(1)}<desc>${"x".repeat(3 * 1024 * 1024)}</desc></svg>`;
+      expect(logo.length).toBeGreaterThan(10_000);
+      const { assets, drops } = await downloadAssets(
+        {
+          svgs: [
+            { outerHTML: logo, isLogo: true },
+            { outerHTML: huge, isLogo: false },
+          ],
+          sections: [],
+          ogImage: "",
+        } as unknown as DesignTokens,
+        dir,
+      );
+      expect(assets).toHaveLength(1);
+      const written = readFileSync(join(dir, assets[0]!.localPath), "utf-8");
+      expect(written).toBe(toStandaloneSvg(logo));
+      expect(written.endsWith("</svg>")).toBe(true);
+      expect(drops["size-ceiling"]).toBe(1);
+    });
+  });
 });
 
 describe("asset fetches present the same identity as the page navigation", () => {

packages/cli/src/capture/assetDownloader.ts

@@ -40,6 +40,8 @@ interface DownloadBudgetOptions {
 export type AssetDropReason =
   /** Fetched, then judged too small to be a real asset rather than a spacer or tracking pixel. */
   | "size-floor"
+  /** Larger than any one asset may be; skipped whole, never cut. */
+  | "size-ceiling"
   /** The post-navigation clock ran out before this one was reached. */
   | "budget-exhausted"
   /** A per-run or per-family limit was already met. */
@@ -51,7 +53,13 @@ export type AssetDropCounts = Record<AssetDropReason, number>;
 
 /** A tally with every reason at zero — the shape a caller merges into. */
 export function noDrops(): AssetDropCounts {
-  return { "size-floor": 0, "budget-exhausted": 0, "cap-reached": 0, unavailable: 0 };
+  return {
+    "size-floor": 0,
+    "size-ceiling": 0,
+    "budget-exhausted": 0,
+    "cap-reached": 0,
+    unavailable: 0,
+  };
 }
 
 /** Sum two tallies. Used to fold the font pass and the asset pass into one capture-wide count. */
@@ -326,6 +334,7 @@ export async function downloadAssets(
   ensureCaptureDirSync(outputDir, join(outputDir, "assets", "svgs"));
   const usedSvgNames = new Set<string>();
   const MAX_INLINE_SVGS = 30;
+  const MAX_INLINE_SVG_BYTES = 2 * 1024 * 1024;
   drops["cap-reached"] += Math.max(0, tokens.svgs.length - MAX_INLINE_SVGS);
   for (let i = 0; i < tokens.svgs.length && i < MAX_INLINE_SVGS; i++) {
     const svg = tokens.svgs[i]!;
@@ -335,6 +344,10 @@ export async function downloadAssets(
     }
     // Hash the bytes that actually land on disk, so the filename still can't drift from content.
     const svgFile = toStandaloneSvg(svg.outerHTML);
+    if (Buffer.byteLength(svgFile, "utf-8") > MAX_INLINE_SVG_BYTES) {
+      drops["size-ceiling"]++;
+      continue;
+    }
     const slug = svgContentHashSlug(svgFile, !!svg.isLogo);
     let finalSlug = slug;
     let suffix = 2;

packages/cli/src/capture/tokenExtractor.test.ts

@@ -0,0 +1,37 @@
+import { describe, expect, it } from "vitest";
+import { runInNewContext } from "node:vm";
+import { Window } from "happy-dom";
+import type { Page } from "puppeteer-core";
+import { extractTokens } from "./tokenExtractor.js";
+
+async function tokensOf(body: string) {
+  const window = new Window();
+  window.document.body.innerHTML = body;
+  const evaluate: Page["evaluate"] = async (script, ..._args) => {
+    if (typeof script !== "string") throw new Error("Expected a page expression");
+    return runInNewContext(script, {
+      window,
+      document: window.document,
+      location: window.location,
+      getComputedStyle: window.getComputedStyle.bind(window),
+    });
+  };
+  try {
+    return await extractTokens({ evaluate } as Page);
+  } finally {
+    await window.happyDOM.close();
+  }
+}
+
+describe("capture tokens", () => {
+  it("keeps an inline SVG logo over 10 KB whole", async () => {
+    const shapes = Array.from({ length: 3000 }, (_, i) => `<path d="M${i} 0h1v10h-1z"></path>`);
+    const logo = `<svg viewBox="0 0 3000 10" width="300">${shapes.join("")}</svg>`;
+    expect(logo.length).toBeGreaterThan(10_000);
+    const tokens = await tokensOf(`<header class="logo"><a href="/">${logo}</a></header>`);
+    expect(tokens.svgs).toHaveLength(1);
+    expect(tokens.svgs[0]!.outerHTML.length).toBe(logo.length);
+    expect(tokens.svgs[0]!.outerHTML.endsWith("</svg>")).toBe(true);
+    expect(tokens.svgs[0]!.isLogo).toBe(true);
+  });
+});

packages/cli/src/capture/tokenExtractor.ts

@@ -386,7 +386,7 @@ const EXTRACT_SCRIPT = `(() => {
       viewBox: svg.getAttribute("viewBox") || undefined,
       width: Math.round(rect.width),
       height: Math.round(rect.height),
-      outerHTML: svg.outerHTML.slice(0, 10000),
+      outerHTML: svg.outerHTML,
       isLogo: isLogo
     };
   }).filter(Boolean).slice(0, 50);

packages/cli/src/whisper/sherpa.test.ts

@@ -14,7 +14,9 @@ import { basename, dirname, join } from "node:path";
 import { pathToFileURL } from "node:url";
 import { afterEach, beforeEach, describe, expect, it, vi } from "vitest";
 import type { DownloadOptions } from "../cloud/download.js";
+import { execFileSync } from "node:child_process";
 import {
+  decodeTimeoutMs,
   ensureParakeetModel,
   installSherpaRuntime,
   SHERPA_RUNTIME_DIR,
@@ -470,3 +472,16 @@ describe("sherpaUnsupportedReason", () => {
     );
   });
 });
+
+describe("decodeTimeoutMs", () => {
+  // 1851.245125 s of audio once gave 3702490.25, which execFile refused as out of range.
+  it.each([1851.245125, 10_000_000])(
+    "gives execFile a timeout it accepts for %s s of audio",
+    (seconds) => {
+      const timeout = decodeTimeoutMs(seconds);
+      expect(Number.isInteger(timeout)).toBe(true);
+      expect(timeout).toBeGreaterThanOrEqual(Math.min(seconds * 2000, 2 ** 31 - 1));
+      expect(() => execFileSync(process.execPath, ["-e", ""], { timeout })).not.toThrow();
+    },
+  );
+});

packages/cli/src/whisper/sherpa.ts

@@ -82,8 +82,9 @@ const PARAKEET_MODEL_FILES: readonly ModelFile[] = [
   },
 ];
 
-/** At least 30 minutes, and twice the audio length for slow CPUs. */
-const decodeTimeoutMs = (audioSeconds: number) => Math.max(1_800_000, audioSeconds * 2000);
+/** At least 30 minutes, twice the audio length for slow CPUs; whole ms, within Node's timer limit. */
+export const decodeTimeoutMs = (audioSeconds: number) =>
+  Math.min(2 ** 31 - 1, Math.max(1_800_000, Math.ceil(audioSeconds * 2000)));
 
 /** The platforms sherpa-onnx-node publishes a native package for. */
 const SUPPORTED_TARGETS = [