@lincoln504/pi-research 1.6.5 → 1.6.7

package/src/web-research/pdf-extract-worker.ts

@@ -0,0 +1,42 @@
+/**
+ * PDF extraction worker entry — bundled by scripts/build.cjs into
+ * pdf-extract-worker.mjs and spawned by pdf-extraction.ts.
+ *
+ * Deliberately dumb: receive {id, bytes}, run the SAME parsePdfCore the
+ * in-process path runs (one code path, no behavioral drift), post the
+ * classified result back. All metrics/logging stay parent-side so both paths
+ * emit identically.
+ */
+
+import { parentPort } from 'node:worker_threads';
+import { parsePdfCore, PdfExtractError } from './pdf-extraction.ts';
+
+interface WorkerRequest {
+  id: number;
+  bytes: Uint8Array;
+}
+
+if (parentPort !== null) {
+  const port = parentPort;
+  port.on('message', (msg: WorkerRequest) => {
+    void (async () => {
+      try {
+        const result = await parsePdfCore(msg.bytes);
+        port.postMessage({ id: msg.id, ok: true, markdown: result.markdown, pageCount: result.pageCount });
+      } catch (e) {
+        if (e instanceof PdfExtractError) {
+          // Worker-side structured clone drops Error causes — the kind and the
+          // message (the load-bearing classification strings) always survive.
+          port.postMessage({ id: msg.id, ok: false, kind: e.kind, message: e.message });
+        } else {
+          port.postMessage({
+            id: msg.id,
+            ok: false,
+            kind: 'extraction_failed',
+            message: e instanceof Error ? e.message : String(e),
+          });
+        }
+      }
+    })();
+  });
+}

package/src/web-research/pdf-extraction.ts

@@ -0,0 +1,392 @@
+/**
+ * PDF extraction — off the main thread when possible, identical when not.
+ *
+ * Relocated verbatim from web-scraper.ts (which previously ran the whole
+ * parse inline) to fix its own recorded TODO(hardening, medium): the
+ * pdf-oxide-wasm parse is irreducibly synchronous (wasm-bindgen, no async
+ * API), so a large/complex PDF blocks the MAIN process's event loop for
+ * seconds — starving every concurrent scrape, the TUI render loop, and
+ * timers.
+ *
+ * The parse now runs in a dedicated worker_threads Worker (NOT the browser
+ * pool: cluster IPC would JSON-serialize the bytes — the exact hazard behind
+ * the bufferB64 round-trip — a pool worker eagerly boots Firefox before any
+ * task, and poolifier never settles in-flight execute() promises on worker
+ * death). postMessage transfers the byte buffer zero-copy after a defensive
+ * slice(); a wedged worker is terminate()d on timeout — the only way to
+ * interrupt a sync wasm call — and the parse retried once in-process.
+ *
+ * Behavioral contract (byte-identical to the web-scraper original):
+ *   - error STRINGS and their classification kinds are load-bearing:
+ *     isBenignScrapeFailure distinguishes 'Could not extract content from
+ *     PDF (…)' (benign, per-URL) from 'PDF extraction unavailable
+ *     (pdf-oxide-wasm failed to load: …)' (infrastructure, NOT benign);
+ *   - metrics: scrape_pdf_conversion_ms / scrape_pdf_conversions_total{status,
+ *     pages} on success, scrape_pdf_errors_total{error_type} per failure kind;
+ *   - log levels: warn (size), error (native module), debug (extraction);
+ *   - WasmPdfDocument.free() runs on every exit path (native memory is not
+ *     GC-reclaimed).
+ *
+ * The worker bundle (pdf-extract-worker.mjs) is built by scripts/build.cjs
+ * next to this file; when absent (unbuilt dev tree, exotic install) the
+ * in-process path runs unchanged — bounded by MAX_PDF_SIZE and the scrape
+ * task's deadline, exactly like before. PI_RESEARCH_PDF_WORKER=off disables
+ * the worker path outright (also the vitest default, so the module-level
+ * vi.mock('pdf-oxide-wasm') seams keep reaching the parse).
+ */
+
+import { existsSync } from 'node:fs';
+import { dirname, join } from 'node:path';
+import { fileURLToPath } from 'node:url';
+import { Worker } from 'node:worker_threads';
+
+// ESM-safe module directory. Bare __dirname is undefined in the bundled
+// dist/cli.mjs (esbuild format: 'esm') and in jiti-loaded ESM — before this
+// fix the packaged CLI threw ReferenceError on the first PDF and every
+// extraction failed. Same pattern as worker-pool-manager.ts.
+const MODULE_DIR = dirname(fileURLToPath(import.meta.url));
+import { logger } from '../logger.ts';
+import { metrics } from '../utils/metrics.ts';
+import { errorTracker } from '../utils/error-tracker.ts';
+import { MAX_PDF_SIZE } from './scraper-types.ts';
+
+// ============================================================================
+// Classified errors — the classification contract survives the worker hop.
+// ============================================================================
+
+export type PdfExtractErrorKind = 'size_exceeded' | 'native_module_unavailable' | 'extraction_failed';
+
+export class PdfExtractError extends Error {
+  readonly kind: PdfExtractErrorKind;
+  constructor(kind: PdfExtractErrorKind, message: string, options?: { cause?: unknown }) {
+    super(message, options);
+    this.name = 'PdfExtractError';
+    this.kind = kind;
+  }
+}
+
+// ============================================================================
+// In-process parse core (also the worker's payload — one code path).
+// ============================================================================
+
+interface PdfModule {
+  WasmPdfDocument: new (bytes: Uint8Array) => {
+    pageCount(): number;
+    toMarkdownAll(): string;
+    toMarkdown(page: number): string;
+    free(): void;
+  };
+}
+
+let pdfModulePromise: Promise<PdfModule> | null = null;
+
+/** Cached dynamic import — vitest's vi.mock('pdf-oxide-wasm') seam keeps working. */
+function loadPdfModule(): Promise<PdfModule> {
+  if (pdfModulePromise === null) {
+    pdfModulePromise = import('pdf-oxide-wasm') as Promise<PdfModule>;
+  }
+  return pdfModulePromise;
+}
+
+export interface PdfParseResult {
+  markdown: string;
+  pageCount: number;
+}
+
+/**
+ * The parse itself: size guard, module load (caught SEPARATELY from parse —
+ * an infrastructure fault is not a per-PDF outcome), doc.free() on every
+ * exit, toMarkdownAll with per-page fallback. Throws PdfExtractError with
+ * the historical message strings; no metrics/logging here so the caller can
+ * emit them once regardless of which context did the work.
+ */
+export async function parsePdfCore(bytes: Uint8Array): Promise<PdfParseResult> {
+  if (bytes.length > MAX_PDF_SIZE) {
+    const sizeMB = Math.round(bytes.length / 1024 / 1024);
+    // THROW, never an in-band "*Error: ...*" string: a returned banner flowed
+    // into validateContent as if it were page content (see git history).
+    throw new PdfExtractError('size_exceeded', `PDF too large (${sizeMB}MB, max 100MB)`);
+  }
+
+  let WasmPdfDocument: PdfModule['WasmPdfDocument'];
+  try {
+    ({ WasmPdfDocument } = await loadPdfModule());
+  } catch (e) {
+    const msg = e instanceof Error ? e.message : String(e);
+    // Deliberately does NOT start with 'Could not extract content from PDF' —
+    // isBenignScrapeFailure must NOT classify this as routine.
+    throw new PdfExtractError(
+      'native_module_unavailable',
+      `PDF extraction unavailable (pdf-oxide-wasm failed to load: ${msg})`,
+      { cause: e },
+    );
+  }
+
+  try {
+    const doc = new WasmPdfDocument(bytes);
+    // free() is native/WASM-side memory, not GC-reclaimed — must run on every
+    // exit, including when toMarkdownAll() AND the per-page fallback throw.
+    try {
+      const pageCount = doc.pageCount();
+      let markdown = `# PDF Document\n\n**Pages:** ${pageCount}\n\n`;
+      try {
+        markdown += doc.toMarkdownAll();
+      } catch {
+        for (let i = 0; i < pageCount; i++) {
+          markdown += `## Page ${i + 1}\n\n${doc.toMarkdown(i)}\n\n`;
+        }
+      }
+      return { markdown, pageCount };
+    } finally {
+      doc.free();
+    }
+  } catch (e) {
+    const msg = e instanceof Error ? e.message : String(e);
+    // See size branch: a failure must FAIL the scrape, not travel in-band.
+    throw new PdfExtractError('extraction_failed', `Could not extract content from PDF (${msg})`, { cause: e });
+  }
+}
+
+// ============================================================================
+// Worker transport.
+// ============================================================================
+
+const WORKER_FILENAME = 'pdf-extract-worker.mjs';
+/** Sync wasm cannot be cancelled; terminate() after this and retry in-process. */
+const PDF_WORKER_TIMEOUT_MS = 30_000;
+
+interface PendingRequest {
+  resolve: (r: PdfParseResult) => void;
+  reject: (e: Error) => void;
+  timer: ReturnType<typeof setTimeout>;
+  /** The Worker instance serving this request — timeout teardown must target
+   * THIS worker, never whatever is current when the timer fires. */
+  worker: Worker;
+}
+
+let pdfWorker: Worker | null = null;
+let nextRequestId = 1;
+const pendingRequests = new Map<number, PendingRequest>();
+let warnedWorkerUnavailable = false;
+
+type WorkerInvoke = (bytes: Uint8Array) => Promise<PdfParseResult>;
+
+function resolveWorkerPath(): string | null {
+  const candidate = join(MODULE_DIR, WORKER_FILENAME);
+  return existsSync(candidate) ? candidate : null;
+}
+
+/** True when the worker bundle is present next to this module and the off
+ *  kill-switch is not set. Exported for the regression test that pins the
+ *  ESM-safe resolution (bare __dirname would throw here in dist/cli.mjs). */
+export function pdfWorkerEnabled(): boolean {
+  if (process.env['PI_RESEARCH_PDF_WORKER'] === 'off') return false;
+  return resolveWorkerPath() !== null;
+}
+
+function failAllPending(reason: string): void {
+  for (const [id, pending] of pendingRequests) {
+    clearTimeout(pending.timer);
+    pendingRequests.delete(id);
+    pending.reject(new Error(`PDF worker ${reason}`));
+  }
+}
+
+function getOrCreatePdfWorker(): Worker | null {
+  if (pdfWorker !== null) return pdfWorker;
+  const workerPath = resolveWorkerPath();
+  if (workerPath === null) {
+    if (!warnedWorkerUnavailable) {
+      warnedWorkerUnavailable = true;
+      logger.warn(
+        `[Scrapers] PDF worker bundle absent (${WORKER_FILENAME}) — PDF extraction stays on the main thread this session. ` +
+          `Run the build step (npm install / prepare) to regenerate it.`,
+      );
+    }
+    return null;
+  }
+  let worker: Worker;
+  try {
+    worker = new Worker(workerPath);
+  } catch (e) {
+    const msg = e instanceof Error ? e.message : String(e);
+    if (!warnedWorkerUnavailable) {
+      warnedWorkerUnavailable = true;
+      logger.warn(`[Scrapers] PDF worker failed to spawn (${msg}) — falling back to in-process extraction.`);
+    }
+    return null;
+  }
+  pdfWorker = worker;
+  // An idle worker must never keep the process alive at shutdown; there is no
+  // browser process to reap and no cleanup ordering to respect (unlike the
+  // browser pool), so unref replaces dispose wiring entirely.
+  try {
+    worker.unref();
+  } catch {
+    // unref() on a terminating worker can throw on some Node versions — the
+    // exit handler below already clears our reference.
+  }
+  worker.on('message', (msg: { id: number; ok: boolean; markdown?: string; pageCount?: number; kind?: PdfExtractErrorKind; message?: string }) => {
+    const pending = pendingRequests.get(msg.id);
+    if (pending === undefined) return;
+    pendingRequests.delete(msg.id);
+    clearTimeout(pending.timer);
+    if (msg.ok) {
+      pending.resolve({ markdown: msg.markdown ?? '', pageCount: msg.pageCount ?? 0 });
+    } else {
+      pending.reject(new PdfExtractError(msg.kind ?? 'extraction_failed', msg.message ?? 'PDF worker failed'));
+    }
+  });
+  // Crash/exit handlers act only when the dying worker is the CURRENT one: a
+  // terminate()-then-respawn sequence can leave the old worker's exit event
+  // arriving after a fresh worker was created — without this identity check a
+  // stale exit would null the global and fail the NEW worker's requests.
+  worker.on('error', (e) => {
+    if (pdfWorker === worker) {
+      pdfWorker = null;
+      failAllPending(`crashed: ${e instanceof Error ? e.message : String(e)}`);
+    }
+  });
+  worker.on('exit', (code) => {
+    if (pdfWorker === worker) {
+      pdfWorker = null;
+      failAllPending(`exited (code ${code})`);
+    }
+  });
+  return worker;
+}
+
+function extractViaWorkerImpl(bytes: Uint8Array): Promise<PdfParseResult> {
+  const worker = getOrCreatePdfWorker();
+  if (worker === null) {
+    // Absent/spawn-failed worker is NOT a transport failure of a started
+    // parse: report it as a plain rejection so the caller falls back cleanly.
+    return Promise.reject(new Error('PDF worker unavailable'));
+  }
+  const id = nextRequestId++;
+  // Defensive copy: the transfer neuters the buffer it moves, and while both
+  // current call sites never touch pdfBytes after the await, a future caller
+  // must not be able to hand itself a detached ArrayBuffer by accident. A
+  // one-shot copy of ≤100MB is tens of milliseconds — correctness first.
+  const transferable = bytes.slice();
+  return new Promise<PdfParseResult>((resolve, reject) => {
+    const timer = setTimeout(() => {
+      // The worker instance serving THIS request — terminate that one, never
+      // whatever happens to be current when the timer fires.
+      const doomed = pendingRequests.get(id)?.worker ?? null;
+      // Drop THIS request first so failAllPending below (which rejects every
+      // REMAINING pending request and clears their timers) cannot double-fire
+      // on it — siblings fail fast instead of hanging out their own 30s.
+      pendingRequests.delete(id);
+      // A sync wasm parse wedged past the deadline can only be stopped by
+      // terminating its thread. The NEXT PDF spawns a fresh worker.
+      pdfWorker = null;
+      failAllPending(`timed out after ${PDF_WORKER_TIMEOUT_MS}ms`);
+      doomed?.terminate().catch(() => undefined);
+      reject(new Error(`PDF worker timed out after ${PDF_WORKER_TIMEOUT_MS}ms`));
+    }, PDF_WORKER_TIMEOUT_MS);
+    pendingRequests.set(id, { resolve, reject, timer, worker });
+    try {
+      worker.postMessage({ id, bytes: transferable }, [transferable.buffer]);
+    } catch (e) {
+      // Synchronous transport failure: this request falls back in-process.
+      // Clear the armed timer and the entry, else the timer fires later and
+      // terminates whatever worker is CURRENT (an unrelated in-flight parse).
+      pendingRequests.delete(id);
+      clearTimeout(timer);
+      reject(new Error(`PDF worker transport failed: ${e instanceof Error ? e.message : String(e)}`));
+    }
+  });
+}
+
+/** Test seam: the orchestration below accepts an injected transport. */
+const workerInvoker: WorkerInvoke = extractViaWorkerImpl;
+
+export function disposePdfWorkerPool(): void {
+  pdfWorker?.terminate().catch(() => undefined);
+  pdfWorker = null;
+  failAllPending('disposed');
+}
+
+// ============================================================================
+// Orchestration — the public surface web-scraper used to define inline.
+// ============================================================================
+
+function emitFailure(e: PdfExtractError): void {
+  if (e.kind === 'size_exceeded') {
+    logger.warn(`[Scrapers] ${e.message}, skipping extraction`);
+    metrics.increment('scrape_pdf_errors_total', 1, { error_type: 'size_exceeded' });
+    return;
+  }
+  if (e.kind === 'native_module_unavailable') {
+    logger.error(`[Scrapers] pdf-oxide-wasm failed to load — PDF extraction is unavailable (not just for this URL): ${e.cause instanceof Error ? e.cause.message : e.message}`);
+    metrics.increment('scrape_pdf_errors_total', 1, { error_type: 'native_module_unavailable' });
+    errorTracker.trackError(e.cause instanceof Error ? e.cause : e.message, {
+      component: 'scrapers',
+      operation: 'pdf-extract',
+      contentType: 'pdf',
+      errorType: 'native_module_unavailable',
+    });
+    return;
+  }
+  // debug, not error: a malformed/encrypted/scanned-image-only PDF is an
+  // expected, non-actionable per-URL outcome (see isBenignScrapeFailure).
+  logger.debug(`[Scrapers] PDF extraction failed: ${e.cause instanceof Error ? e.cause.message : e.message}`);
+  metrics.increment('scrape_pdf_errors_total', 1, { error_type: 'extraction_failed' });
+  errorTracker.trackError(e.cause instanceof Error ? e.cause : e.message, {
+    component: 'scrapers',
+    operation: 'pdf-extract',
+    contentType: 'pdf',
+    errorType: 'extraction_failed',
+  });
+}
+
+export async function extractPdfToMarkdown(
+  bytes: Uint8Array,
+  opts: { invokeWorker?: WorkerInvoke } = {},
+): Promise<string> {
+  const start = Date.now();
+  const invoke = opts.invokeWorker ?? workerInvoker;
+
+  let result: PdfParseResult;
+  if (pdfWorkerEnabled()) {
+    try {
+      result = await invoke(bytes);
+    } catch (e) {
+      // A CLASSIFIED error is a real answer from inside the worker (size,
+      // module, parse): report it, do not redo the work in-process.
+      if (e instanceof PdfExtractError) {
+        emitFailure(e);
+        throw e;
+      }
+      // Transport failure (spawn/timeout/crash): one bounded in-process
+      // retry — today's behavior, still capped by MAX_PDF_SIZE and the
+      // scrape deadline.
+      const msg = e instanceof Error ? e.message : String(e);
+      logger.warn(`[Scrapers] PDF worker path failed (${msg}); retrying extraction in-process`);
+      try {
+        result = await parsePdfCore(bytes);
+      } catch (e2) {
+        if (e2 instanceof PdfExtractError) {
+          emitFailure(e2);
+          throw e2;
+        }
+        throw e2;
+      }
+    }
+  } else {
+    try {
+      result = await parsePdfCore(bytes);
+    } catch (e) {
+      if (e instanceof PdfExtractError) {
+        emitFailure(e);
+        throw e;
+      }
+      throw e;
+    }
+  }
+
+  metrics.observe('scrape_pdf_conversion_ms', Date.now() - start);
+  metrics.increment('scrape_pdf_conversions_total', 1, { status: 'success', pages: String(result.pageCount) });
+  return result.markdown;
+}

package/src/web-research/web-scraper.ts

@@ -30,6 +30,9 @@ import { readBodyCapped, BodyTooLargeError } from '../utils/http-body.ts';
 import { getServiceContainer } from '../core/service-registry.ts';
 import type { ServiceContainer } from '../core/service-registry.ts';
 import { cacheScrapedContent } from '../utils/shared-links.ts';
+// PDF extraction now lives in pdf-extraction.ts (worker-thread offload with an
+// identical in-process fallback); same export name and contract as before.
+import { extractPdfToMarkdown } from './pdf-extraction.ts';
 
 /** Backoff before the fetch layer's single transient retry. */
 const FETCH_RETRY_DELAY_MS = 250;
@@ -75,117 +78,6 @@ async function convertToMarkdown(html: string): Promise<string> {
   return converter(html);
 }
 
-// TODO(hardening, medium): this synchronous WASM parse runs on the MAIN
-// process's event loop — a large/complex PDF can block it for seconds,
-// starving every concurrent scrape, the TUI render loop, and timers. The
-// correct fix is a worker_threads-based extraction (mirroring the browser
-// pool's CPU isolation pattern), but it is deliberately NOT done in this
-// pass: pdf-oxide-wasm is dynamically imported here and unit tests mock the
-// module via vitest, which cannot reach into a real worker thread — moving
-// the parse requires a companion test-seam redesign first. Bounded today by
-// MAX_PDF_SIZE (100MB) and the scrape task's own deadline.
-async function extractPdfToMarkdown(bytes: Uint8Array): Promise<string> {
-  if (bytes.length > MAX_PDF_SIZE) {
-    const sizeMB = Math.round(bytes.length / 1024 / 1024);
-    logger.warn(`[Scrapers] PDF too large (${sizeMB}MB, max 100MB), skipping extraction`);
-    metrics.increment('scrape_pdf_errors_total', 1, { error_type: 'size_exceeded' });
-    // THROW, never an in-band "*Error: ...*" string: a returned banner flowed
-    // into validateContent as if it were page content, and a verbose one
-    // (≥50 words / ≥200 chars) cleared the stub gate — the failure banner was
-    // then cached and cited as a successful scrape.
-    throw new Error(`PDF too large (${sizeMB}MB, max 100MB)`);
-  }
-
-  const pdfExtractionStart = Date.now();
-
-  // The native module load is caught SEPARATELY from the parse/extraction
-  // logic below: a failure here (missing prebuilt binary for this platform,
-  // corrupted install) is an INFRASTRUCTURE fault — PDF extraction is broken
-  // for every URL, not just this one — not the routine, expected outcome of
-  // a malformed/encrypted/scanned-image-only PDF. Sharing one catch with the
-  // parse logic used to log both at `debug` and lump both under
-  // 'extraction_failed', so a broken install produced an endless stream of
-  // silent per-PDF "failures" instead of ever surfacing the real cause.
-  let WasmPdfDocument: (new (bytes: Uint8Array) => {
-    pageCount(): number;
-    toMarkdownAll(): string;
-    toMarkdown(page: number): string;
-    free(): void;
-  });
-  try {
-    ({ WasmPdfDocument } = await import('pdf-oxide-wasm'));
-  } catch (e) {
-    const msg = e instanceof Error ? e.message : String(e);
-    logger.error(`[Scrapers] pdf-oxide-wasm failed to load — PDF extraction is unavailable (not just for this URL): ${msg}`);
-    metrics.increment('scrape_pdf_errors_total', 1, { error_type: 'native_module_unavailable' });
-    errorTracker.trackError(e instanceof Error ? e : String(e), {
-      component: 'scrapers',
-      operation: 'pdf-extract',
-      contentType: 'pdf',
-      errorType: 'native_module_unavailable',
-    });
-    // Deliberately does NOT start with 'Could not extract content from PDF' —
-    // isBenignScrapeFailure must NOT classify this as the routine per-PDF
-    // outcome it is for the parse-failure branch below.
-    throw new Error(`PDF extraction unavailable (pdf-oxide-wasm failed to load: ${msg})`, { cause: e });
-  }
-
-  try {
-    const doc = new WasmPdfDocument(bytes);
-    // free() is native/WASM-side memory, not GC-reclaimed — it must run on
-    // every exit, including the case toMarkdownAll() AND the per-page fallback
-    // both throw (a malformed/encrypted PDF the codebase already treats as
-    // routine on the open web, so this path fires regularly). It used to sit
-    // after the fallback loop, reachable only when nothing below it threw —
-    // the one case that needed the fallback landing squarely on the one case
-    // that skipped the free(), leaking one WasmPdfDocument per occurrence in a
-    // browser-server process that stays warm across many research sessions.
-    try {
-      const pageCount = doc.pageCount();
-
-      let markdown = `# PDF Document\n\n**Pages:** ${pageCount}\n\n`;
-
-      try {
-        markdown += doc.toMarkdownAll();
-      } catch {
-        for (let i = 0; i < pageCount; i++) {
-          markdown += `## Page ${i + 1}\n\n${doc.toMarkdown(i)}\n\n`;
-        }
-      }
-
-      const pdfExtractionDuration = Date.now() - pdfExtractionStart;
-      metrics.observe('scrape_pdf_conversion_ms', pdfExtractionDuration);
-      metrics.increment('scrape_pdf_conversions_total', 1, { status: 'success', pages: String(pageCount) });
-      return markdown;
-    } finally {
-      doc.free();
-    }
-  } catch (e) {
-    const msg = e instanceof Error ? e.message : String(e);
-    // debug, not error: a malformed/encrypted/scanned-image-only PDF is an
-    // expected, non-actionable per-URL outcome, not a genuine fault — the same
-    // reason isBenignScrapeFailure (scraper-utils.ts) keeps stub/blocked/HTTP-error
-    // content out of ERROR. This is called from both the fetch layer (where a
-    // browser-layer fallback always follows — the fetch layer's own catch at line
-    // ~529 already logs ANY of its failures at debug for that reason) and the
-    // browser layer (the last resort — isBenignScrapeFailure now recognizes this
-    // message so that terminal call site's own classification stays consistent
-    // too). Logging ERROR here, before either caller gets a chance to classify,
-    // pre-empted both of those and always alarmed regardless.
-    logger.debug(`[Scrapers] PDF extraction failed: ${msg}`);
-    metrics.increment('scrape_pdf_errors_total', 1, { error_type: 'extraction_failed' });
-    errorTracker.trackError(e instanceof Error ? e : String(e), {
-      component: 'scrapers',
-      operation: 'pdf-extract',
-      contentType: 'pdf',
-      errorType: 'extraction_failed',
-    });
-    // See the size branch above: an extraction failure must FAIL the scrape,
-    // not travel in-band as pseudo-content.
-    throw new Error(`Could not extract content from PDF (${msg})`, { cause: e });
-  }
-}
-
 /**
  * Decode an HTML response body using its declared charset. response.text() blindly UTF-8-decodes,
  * corrupting non-UTF-8 pages (Shift_JIS/GBK/EUC-KR/ISO-8859-1). Prefer the Content-Type charset,