Function tryExtractPdf

src/pdf.js:145–165

(base64Data)

 * @returns {{ text: string, pageCount: number } | null}
 */
export function tryExtractPdf(base64Data) {
  try {
    const buf = Buffer.from(base64Data, 'base64');
    if (buf.length < 5 || buf.subarray(0, 5).toString() !== '%PDF-') return null;

    const text = extractPdfText(buf);
    if (!text.trim()) {
      log.warn('PDF has no extractable text layer (scanned/image-only PDF)');
      return { text: '', pageCount: 0 };
    }

    const pageCount = (buf.toString('latin1').match(/\/Type\s*\/Page\b/g) || []).length;
    return { text, pageCount };
  } catch (e) {
    log.warn(`PDF extraction failed: ${e.message}`);
    if (/exceeds safety limit|maxOutputLength|too large|Buffer larger/i.test(e.message) || e.code === 'ERR_BUFFER_TOO_LARGE') {
      return { text: 'PDF 内容无法提取', pageCount: 0 };
    }
    return null;
  }
}