@dragon708/docmind-node 1.0.0 → 1.1.0

package/dist/index.d.ts

@@ -1,5 +1,5 @@
 import { DetectFileKindInput, NamedInput, AnalysisResult } from '@dragon708/docmind-shared';
-export { AnalysisAnalyzer, AnalysisResult, DetectFileKindInput, DocxAnalysisCoreResult, FileKind, FileKindMetadata, GenericAnalysisResult, ImageAnalysisCoreResult, PdfAnalysisCoreResult, TextAnalysisResult } from '@dragon708/docmind-shared';
+export { AnalysisAnalyzer, AnalysisResult, DetectFileKindInput, DocxAnalysisCoreResult, FileKind, FileKindMetadata, GenericAnalysisResult, ImageAnalysisCoreResult, PdfAnalysisCoreResult, TextAnalysisResult, detectFileKind } from '@dragon708/docmind-shared';
 import { OcrOptions } from '@dragon708/docmind-ocr';
 import { PdfAnalyzeOptions } from '@dragon708/docmind-pdf';
 
@@ -31,4 +31,25 @@ declare function resolveNodeAnalyzeInput(input: NodeAnalyzeInput): Promise<Detec
  */
 declare function analyzeFile(input: NodeAnalyzeInput, options?: NodeAnalyzeOptions): Promise<AnalysisResult>;
 
-export { type NodeAnalyzeInput, type NodeAnalyzeOptions, analyzeFile, bufferToInput, readFileToInput, resolveNodeAnalyzeInput };
+/**
+ * Text only: PDF → `extractTextFromPdf` (capa de texto, sin OCR); DOCX → `extractTextFromDocx`;
+ * imagen → `ocr`; texto → `analyzeText`.
+ */
+declare function extractText(input: NodeAnalyzeInput, options?: NodeAnalyzeOptions): Promise<AnalysisResult>;
+/**
+ * Metadatos: PDF → `extractPdfMetadata`; resto con mejor esfuerzo o aviso.
+ * El resultado sigue siendo `AnalysisResult` (PDF rellena `metadata` en forma `PdfAnalysisCoreResult`).
+ */
+declare function extractMetadata(input: NodeAnalyzeInput, options?: NodeAnalyzeOptions): Promise<AnalysisResult>;
+/**
+ * HTML: DOCX → `convertDocxToHtml`; PDF/texto → `<pre>` a partir de texto extraído;
+ * imágenes → vacío con aviso.
+ */
+declare function convertToHtml(input: NodeAnalyzeInput, options?: NodeAnalyzeOptions): Promise<AnalysisResult>;
+/**
+ * OCR: PDF → `analyzePdf` con `ocr: "force"`; imagen → `ocr`; DOCX → texto estructurado con aviso
+ * (sin OCR); texto → `analyzeText`.
+ */
+declare function runOcr(input: NodeAnalyzeInput, options?: NodeAnalyzeOptions): Promise<AnalysisResult>;
+
+export { type NodeAnalyzeInput, type NodeAnalyzeOptions, analyzeFile, bufferToInput, convertToHtml, extractMetadata, extractText, readFileToInput, resolveNodeAnalyzeInput, runOcr };

package/dist/index.js

@@ -1,7 +1,8 @@
 import { assertValidAnalyzeFileInput, detectFileKind, notImplementedResult, UNKNOWN_FORMAT_WARNING, analyzeText, toUint8Array, isNamedInput, isBinaryInput, isBlob, isFile } from '@dragon708/docmind-shared';
-import { analyzeDocx } from '@dragon708/docmind-docx';
+export { detectFileKind } from '@dragon708/docmind-shared';
+import { extractTextFromDocx, convertDocxToHtml, analyzeDocx } from '@dragon708/docmind-docx';
 import { ocr } from '@dragon708/docmind-ocr';
-import { analyzePdf } from '@dragon708/docmind-pdf';
+import { extractTextFromPdf, extractPdfMetadata, analyzePdf } from '@dragon708/docmind-pdf';
 import { readFile } from 'fs/promises';
 import { basename } from 'path';
 import { fileURLToPath } from 'url';
@@ -168,7 +169,383 @@ async function analyzeFile(input, options) {
       return notImplementedResult(fileKind, "none", [UNKNOWN_FORMAT_WARNING]);
   }
 }
+function throwIfAborted(signal) {
+  if (signal?.aborted) {
+    const err = new Error("The operation was aborted");
+    err.name = "AbortError";
+    throw err;
+  }
+}
+async function prepare(input) {
+  const resolved = await resolveNodeAnalyzeInput(input);
+  assertValidAnalyzeFileInput(resolved);
+  return resolved;
+}
+function escapeHtmlMinimal(s) {
+  return s.replace(/&/g, "&amp;").replace(/</g, "&lt;").replace(/>/g, "&gt;").replace(/"/g, "&quot;");
+}
+var DOCX_METADATA_STUB = "Structured document metadata for DOCX is not exposed as a separate API; use extractText or analyzeFile.";
+var IMAGE_METADATA_NOTE = "Raster images have no document metadata bundle in this API.";
+async function extractText(input, options) {
+  throwIfAborted(options?.signal);
+  const resolved = await prepare(input);
+  const kind = detectFileKind(resolved);
+  const signal = options?.signal;
+  switch (kind) {
+    case "pdf": {
+      const data = await bytesFromDetectInput(resolved);
+      if (data.byteLength === 0) {
+        return {
+          fileKind: "pdf",
+          analyzer: "pdf",
+          status: "ok",
+          kind: "pdf",
+          text: "",
+          pages: 0,
+          metadata: { info: {} },
+          warnings: ["No document bytes were provided for analysis."],
+          needsOCR: false,
+          ocrUsed: false
+        };
+      }
+      const r = await extractTextFromPdf(data);
+      return {
+        fileKind: "pdf",
+        analyzer: "pdf",
+        status: "ok",
+        kind: "pdf",
+        text: r.text,
+        pages: r.pages,
+        metadata: { info: {} },
+        warnings: r.warnings,
+        needsOCR: false,
+        ocrUsed: false
+      };
+    }
+    case "docx": {
+      const data = await bytesFromDetectInput(resolved);
+      if (data.byteLength === 0) {
+        return {
+          fileKind: "docx",
+          analyzer: "docx",
+          status: "ok",
+          kind: "docx",
+          text: "",
+          html: "",
+          warnings: ["No document bytes were provided for analysis."]
+        };
+      }
+      const r = await extractTextFromDocx(data);
+      return {
+        fileKind: "docx",
+        analyzer: "docx",
+        status: "ok",
+        kind: "docx",
+        text: r.text,
+        html: "",
+        warnings: r.warnings
+      };
+    }
+    case "image": {
+      const data = await bytesFromDetectInput(resolved);
+      if (data.byteLength === 0) {
+        return {
+          fileKind: "image",
+          analyzer: "image",
+          status: "ok",
+          kind: "image",
+          text: "",
+          confidence: 0,
+          ocrUsed: true,
+          warnings: ["No image bytes were provided for analysis."]
+        };
+      }
+      const ocrOpts = {
+        ...options?.ocr ?? {},
+        signal: options?.ocr?.signal ?? signal
+      };
+      const r = await ocr(data, ocrOpts);
+      return {
+        fileKind: "image",
+        analyzer: "image",
+        status: "ok",
+        kind: "image",
+        text: r.text,
+        confidence: r.confidence,
+        ocrUsed: r.ocrUsed,
+        warnings: []
+      };
+    }
+    case "text":
+      return analyzeText(resolved, { signal });
+    default:
+      return notImplementedResult(kind, "none", [UNKNOWN_FORMAT_WARNING]);
+  }
+}
+async function extractMetadata(input, options) {
+  throwIfAborted(options?.signal);
+  const resolved = await prepare(input);
+  const kind = detectFileKind(resolved);
+  const signal = options?.signal;
+  switch (kind) {
+    case "pdf": {
+      const data = await bytesFromDetectInput(resolved);
+      if (data.byteLength === 0) {
+        return {
+          fileKind: "pdf",
+          analyzer: "pdf",
+          status: "ok",
+          kind: "pdf",
+          text: "",
+          pages: 0,
+          metadata: { info: {} },
+          warnings: ["No document bytes were provided for analysis."],
+          needsOCR: false,
+          ocrUsed: false
+        };
+      }
+      const r = await extractPdfMetadata(data);
+      return {
+        fileKind: "pdf",
+        analyzer: "pdf",
+        status: "ok",
+        kind: "pdf",
+        text: "",
+        pages: 0,
+        metadata: r.metadata,
+        warnings: r.warnings,
+        needsOCR: false,
+        ocrUsed: false
+      };
+    }
+    case "docx":
+      return {
+        fileKind: "docx",
+        analyzer: "docx",
+        status: "ok",
+        kind: "docx",
+        text: "",
+        html: "",
+        warnings: [DOCX_METADATA_STUB]
+      };
+    case "image":
+      return {
+        fileKind: "image",
+        analyzer: "image",
+        status: "ok",
+        kind: "image",
+        text: "",
+        confidence: 0,
+        ocrUsed: true,
+        warnings: [IMAGE_METADATA_NOTE]
+      };
+    case "text":
+      return analyzeText(resolved, { signal });
+    default:
+      return notImplementedResult(kind, "none", [UNKNOWN_FORMAT_WARNING]);
+  }
+}
+async function convertToHtml(input, options) {
+  throwIfAborted(options?.signal);
+  const resolved = await prepare(input);
+  const kind = detectFileKind(resolved);
+  const signal = options?.signal;
+  switch (kind) {
+    case "docx": {
+      const data = await bytesFromDetectInput(resolved);
+      if (data.byteLength === 0) {
+        return {
+          fileKind: "docx",
+          analyzer: "docx",
+          status: "ok",
+          kind: "docx",
+          text: "",
+          html: "",
+          warnings: ["No document bytes were provided for analysis."]
+        };
+      }
+      const [textPart, htmlPart] = await Promise.all([
+        extractTextFromDocx(data),
+        convertDocxToHtml(data)
+      ]);
+      return {
+        fileKind: "docx",
+        analyzer: "docx",
+        status: "ok",
+        kind: "docx",
+        text: textPart.text,
+        html: htmlPart.html,
+        warnings: [...textPart.warnings, ...htmlPart.warnings]
+      };
+    }
+    case "pdf": {
+      const data = await bytesFromDetectInput(resolved);
+      if (data.byteLength === 0) {
+        return {
+          fileKind: "pdf",
+          analyzer: "pdf",
+          status: "ok",
+          kind: "pdf",
+          text: "",
+          pages: 0,
+          metadata: { info: {} },
+          warnings: ["No document bytes were provided for analysis."],
+          needsOCR: false,
+          ocrUsed: false
+        };
+      }
+      const r = await extractTextFromPdf(data);
+      const html = `<pre>${escapeHtmlMinimal(r.text)}</pre>`;
+      return {
+        fileKind: "pdf",
+        analyzer: "pdf",
+        status: "ok",
+        kind: "pdf",
+        text: r.text,
+        pages: r.pages,
+        metadata: { info: {} },
+        html,
+        warnings: [
+          ...r.warnings,
+          "PDF HTML is a plain-text preview wrapped in <pre> (not a visual layout)."
+        ],
+        needsOCR: false,
+        ocrUsed: false
+      };
+    }
+    case "text": {
+      const t = await analyzeText(resolved, { signal });
+      const html = `<pre>${escapeHtmlMinimal(t.text)}</pre>`;
+      return {
+        ...t,
+        html,
+        warnings: [
+          ...t.warnings,
+          "HTML for plain text is a <pre> wrapper around decoded UTF-8 content."
+        ]
+      };
+    }
+    case "image":
+      return {
+        fileKind: "image",
+        analyzer: "image",
+        status: "ok",
+        kind: "image",
+        text: "",
+        confidence: 0,
+        ocrUsed: true,
+        warnings: ["No HTML representation for raster images; use extractText / runOcr."]
+      };
+    default:
+      return notImplementedResult(kind, "none", [UNKNOWN_FORMAT_WARNING]);
+  }
+}
+async function runOcr(input, options) {
+  throwIfAborted(options?.signal);
+  const resolved = await prepare(input);
+  const kind = detectFileKind(resolved);
+  const signal = options?.signal;
+  const lang = options?.ocr?.langs ?? options?.pdf?.ocrLangs;
+  switch (kind) {
+    case "pdf": {
+      const data = await bytesFromDetectInput(resolved);
+      if (data.byteLength === 0) {
+        return {
+          fileKind: "pdf",
+          analyzer: "pdf",
+          status: "ok",
+          kind: "pdf",
+          text: "",
+          pages: 0,
+          metadata: { info: {} },
+          warnings: ["No document bytes were provided for analysis."],
+          needsOCR: false,
+          ocrUsed: false
+        };
+      }
+      const r = await analyzePdf(data, {
+        ...options?.pdf,
+        ocr: "force",
+        ocrLangs: lang ?? options?.pdf?.ocrLangs,
+        signal: options?.pdf?.signal ?? signal
+      });
+      return {
+        fileKind: "pdf",
+        analyzer: "pdf",
+        status: "ok",
+        kind: "pdf",
+        text: r.text,
+        pages: r.pages,
+        metadata: r.metadata,
+        warnings: [...r.warnings],
+        needsOCR: r.needsOCR,
+        ocrUsed: r.ocrUsed
+      };
+    }
+    case "image": {
+      const data = await bytesFromDetectInput(resolved);
+      if (data.byteLength === 0) {
+        return {
+          fileKind: "image",
+          analyzer: "image",
+          status: "ok",
+          kind: "image",
+          text: "",
+          confidence: 0,
+          ocrUsed: true,
+          warnings: ["No image bytes were provided for analysis."]
+        };
+      }
+      const ocrOpts = {
+        ...options?.ocr ?? {},
+        signal: options?.ocr?.signal ?? signal
+      };
+      const r = await ocr(data, ocrOpts);
+      return {
+        fileKind: "image",
+        analyzer: "image",
+        status: "ok",
+        kind: "image",
+        text: r.text,
+        confidence: r.confidence,
+        ocrUsed: r.ocrUsed,
+        warnings: []
+      };
+    }
+    case "docx": {
+      const data = await bytesFromDetectInput(resolved);
+      if (data.byteLength === 0) {
+        return {
+          fileKind: "docx",
+          analyzer: "docx",
+          status: "ok",
+          kind: "docx",
+          text: "",
+          html: "",
+          warnings: ["No document bytes were provided for analysis."]
+        };
+      }
+      const r = await analyzeDocx(data);
+      return {
+        fileKind: "docx",
+        analyzer: "docx",
+        status: "ok",
+        kind: "docx",
+        text: r.text,
+        html: r.html,
+        warnings: [
+          ...r.warnings,
+          "OCR does not apply to DOCX; returned structured text/HTML extract."
+        ]
+      };
+    }
+    case "text":
+      return analyzeText(resolved, { signal });
+    default:
+      return notImplementedResult(kind, "none", [UNKNOWN_FORMAT_WARNING]);
+  }
+}
 
-export { analyzeFile, bufferToInput, readFileToInput, resolveNodeAnalyzeInput };
+export { analyzeFile, bufferToInput, convertToHtml, extractMetadata, extractText, readFileToInput, resolveNodeAnalyzeInput, runOcr };
 //# sourceMappingURL=index.js.map
 //# sourceMappingURL=index.js.map

package/dist/index.js.map

@@ -1 +1 @@
-{"version":3,"sources":["../src/inputBytes.ts","../src/analyzers/docx.ts","../src/analyzers/image.ts","../src/analyzers/pdf.ts","../src/resolveNodeInput.ts","../src/analyze.ts"],"names":["extractDocx","runPdf"],"mappings":";;;;;;;;;AASO,SAAS,kBAAkB,KAAA,EAAoD;AACpF,EAAA,OAAO,YAAA,CAAa,KAAK,CAAA,IAAK,aAAA,CAAc,KAAK,KAAK,MAAA,CAAO,KAAK,CAAA,IAAK,MAAA,CAAO,KAAK,CAAA;AACrF;AAGA,eAAsB,qBAAqB,KAAA,EAAiD;AAC1F,EAAA,IAAI,CAAC,iBAAA,CAAkB,KAAK,CAAA,EAAG;AAC7B,IAAA,OAAO,IAAI,WAAW,CAAC,CAAA;AAAA,EACzB;AACA,EAAA,OAAO,aAAa,KAAK,CAAA;AAC3B;;;ACZA,eAAsB,kBAAA,CACpB,OACA,MAAA,EACyB;AACzB,EAAA,IAAI,QAAQ,OAAA,EAAS;AACnB,IAAA,MAAM,GAAA,GAAM,IAAI,KAAA,CAAM,2BAA2B,CAAA;AACjD,IAAA,GAAA,CAAI,IAAA,GAAO,YAAA;AACX,IAAA,MAAM,GAAA;AAAA,EACR;AAEA,EAAA,MAAM,IAAA,GAAO,MAAM,oBAAA,CAAqB,KAAK,CAAA;AAC7C,EAAA,IAAI,IAAA,CAAK,eAAe,CAAA,EAAG;AACzB,IAAA,OAAO;AAAA,MACL,QAAA,EAAU,MAAA;AAAA,MACV,QAAA,EAAU,MAAA;AAAA,MACV,MAAA,EAAQ,IAAA;AAAA,MACR,IAAA,EAAM,MAAA;AAAA,MACN,IAAA,EAAM,EAAA;AAAA,MACN,IAAA,EAAM,EAAA;AAAA,MACN,QAAA,EAAU,CAAC,+CAA+C;AAAA,KAC5D;AAAA,EACF;AAEA,EAAA,MAAM,CAAA,GAAI,MAAMA,WAAA,CAAY,IAAI,CAAA;AAChC,EAAA,OAAO;AAAA,IACL,QAAA,EAAU,MAAA;AAAA,IACV,QAAA,EAAU,MAAA;AAAA,IACV,MAAA,EAAQ,IAAA;AAAA,IACR,IAAA,EAAM,MAAA;AAAA,IACN,MAAM,CAAA,CAAE,IAAA;AAAA,IACR,MAAM,CAAA,CAAE,IAAA;AAAA,IACR,QAAA,EAAU,CAAC,GAAG,CAAA,CAAE,QAAQ;AAAA,GAC1B;AACF;AChCA,eAAsB,mBAAA,CACpB,OACA,OAAA,EACyB;AACzB,EAAA,IAAI,OAAA,EAAS,QAAQ,OAAA,EAAS;AAC5B,IAAA,MAAM,GAAA,GAAM,IAAI,KAAA,CAAM,2BAA2B,CAAA;AACjD,IAAA,GAAA,CAAI,IAAA,GAAO,YAAA;AACX,IAAA,MAAM,GAAA;AAAA,EACR;AAEA,EAAA,MAAM,IAAA,GAAO,MAAM,oBAAA,CAAqB,KAAK,CAAA;AAC7C,EAAA,IAAI,IAAA,CAAK,eAAe,CAAA,EAAG;AACzB,IAAA,OAAO;AAAA,MACL,QAAA,EAAU,OAAA;AAAA,MACV,QAAA,EAAU,OAAA;AAAA,MACV,MAAA,EAAQ,IAAA;AAAA,MACR,IAAA,EAAM,OAAA;AAAA,MACN,IAAA,EAAM,EAAA;AAAA,MACN,UAAA,EAAY,CAAA;AAAA,MACZ,OAAA,EAAS,IAAA;AAAA,MACT,QAAA,EAAU,CAAC,4CAA4C;AAAA,KACzD;AAAA,EACF;AAEA,EAAA,MAAM,OAAA,GAAU;AAAA,IACd,GAAI,OAAA,EAAS,GAAA,IAAO,EAAC;AAAA,IACrB,MAAA,EAAQ,OAAA,EAAS,GAAA,EAAK,MAAA,IAAU,OAAA,EAAS;AAAA,GAC3C;AAEA,EAAA,MAAM,CAAA,GAAI,MAAM,GAAA,CAAI,IAAA,EAAM,OAAO,CAAA;AACjC,EAAA,OAAO;AAAA,IACL,QAAA,EAAU,OAAA;AAAA,IACV,QAAA,EAAU,OAAA;AAAA,IACV,MAAA,EAAQ,IAAA;AAAA,IACR,IAAA,EAAM,OAAA;AAAA,IACN,MAAM,CAAA,CAAE,IAAA;AAAA,IACR,YAAY,CAAA,CAAE,UAAA;AAAA,IACd,SAAS,CAAA,CAAE,OAAA;AAAA,IACX,UAAU;AAAC,GACb;AACF;AClCA,eAAsB,iBAAA,CACpB,OACA,OAAA,EACyB;AACzB,EAAA,IAAI,OAAA,EAAS,QAAQ,OAAA,EAAS;AAC5B,IAAA,MAAM,GAAA,GAAM,IAAI,KAAA,CAAM,2BAA2B,CAAA;AACjD,IAAA,GAAA,CAAI,IAAA,GAAO,YAAA;AACX,IAAA,MAAM,GAAA;AAAA,EACR;AAEA,EAAA,MAAM,IAAA,GAAO,MAAM,oBAAA,CAAqB,KAAK,CAAA;AAC7C,EAAA,IAAI,IAAA,CAAK,eAAe,CAAA,EAAG;AACzB,IAAA,OAAO;AAAA,MACL,QAAA,EAAU,KAAA;AAAA,MACV,QAAA,EAAU,KAAA;AAAA,MACV,MAAA,EAAQ,IAAA;AAAA,MACR,IAAA,EAAM,KAAA;AAAA,MACN,IAAA,EAAM,EAAA;AAAA,MACN,KAAA,EAAO,CAAA;AAAA,MACP,QAAA,EAAU,EAAE,IAAA,EAAM,EAAC,EAAE;AAAA,MACrB,QAAA,EAAU,CAAC,+CAA+C,CAAA;AAAA,MAC1D,QAAA,EAAU,KAAA;AAAA,MACV,OAAA,EAAS;AAAA,KACX;AAAA,EACF;AAEA,EAAA,MAAM,UAAU,OAAA,EAAS,GAAA;AACzB,EAAA,MAAM,OAAA,GAA6B;AAAA,IACjC,GAAG,OAAA;AAAA,IACH,GAAA,EAAK,SAAS,GAAA,IAAO,MAAA;AAAA,IACrB,QAAA,EAAU,OAAA,EAAS,QAAA,IAAY,OAAA,EAAS,GAAA,EAAK,KAAA;AAAA,IAC7C,MAAA,EAAQ,OAAA,EAAS,MAAA,IAAU,OAAA,EAAS;AAAA,GACtC;AAEA,EAAA,MAAM,CAAA,GAAI,MAAMC,UAAA,CAAO,IAAA,EAAM,OAAO,CAAA;AACpC,EAAA,OAAO;AAAA,IACL,QAAA,EAAU,KAAA;AAAA,IACV,QAAA,EAAU,KAAA;AAAA,IACV,MAAA,EAAQ,IAAA;AAAA,IACR,IAAA,EAAM,KAAA;AAAA,IACN,MAAM,CAAA,CAAE,IAAA;AAAA,IACR,OAAO,CAAA,CAAE,KAAA;AAAA,IACT,UAAU,CAAA,CAAE,QAAA;AAAA,IACZ,QAAA,EAAU,CAAC,GAAG,CAAA,CAAE,QAAQ,CAAA;AAAA,IACxB,UAAU,CAAA,CAAE,QAAA;AAAA,IACZ,SAAS,CAAA,CAAE;AAAA,GACb;AACF;AClDA,SAAS,aAAa,SAAA,EAAiC;AACrD,EAAA,OAAO,SAAA,YAAqB,GAAA,GAAM,aAAA,CAAc,SAAS,CAAA,GAAI,SAAA;AAC/D;AAKA,eAAsB,gBAAgB,IAAA,EAAiD;AACrF,EAAA,MAAM,MAAA,GAAS,aAAa,IAAI,CAAA;AAChC,EAAA,MAAM,IAAA,GAAO,MAAM,QAAA,CAAS,MAAM,CAAA;AAClC,EAAA,OAAO;AAAA,IACL,IAAA;AAAA,IACA,IAAA,EAAM,SAAS,MAAM;AAAA,GACvB;AACF;AAGO,SAAS,aAAA,CAAc,QAAgB,IAAA,EAAmC;AAC/E,EAAA,OAAO,IAAA,KAAS,SAAY,EAAE,IAAA,EAAM,QAAQ,IAAA,EAAK,GAAI,EAAE,IAAA,EAAM,MAAA,EAAO;AACtE;AAKA,eAAsB,wBAAwB,KAAA,EAAuD;AACnG,EAAA,IAAI,OAAO,KAAA,KAAU,QAAA,IAAY,KAAA,YAAiB,GAAA,EAAK;AACrD,IAAA,OAAO,gBAAgB,KAAK,CAAA;AAAA,EAC9B;AACA,EAAA,OAAO,KAAA;AACT;;;ACvBA,eAAsB,WAAA,CACpB,OACA,OAAA,EACyB;AACzB,EAAA,IAAI,OAAA,EAAS,QAAQ,OAAA,EAAS;AAC5B,IAAA,MAAM,GAAA,GAAM,IAAI,KAAA,CAAM,2BAA2B,CAAA;AACjD,IAAA,GAAA,CAAI,IAAA,GAAO,YAAA;AACX,IAAA,MAAM,GAAA;AAAA,EACR;AAEA,EAAA,MAAM,QAAA,GAAW,MAAM,uBAAA,CAAwB,KAAK,CAAA;AACpD,EAAA,2BAAA,CAA4B,QAAQ,CAAA;AAEpC,EAAA,MAAM,QAAA,GAAW,eAAe,QAAQ,CAAA;AAExC,EAAA,QAAQ,QAAA;AAAU,IAChB,KAAK,KAAA;AACH,MAAA,OAAO,iBAAA,CAAkB,UAAiC,OAAO,CAAA;AAAA,IACnE,KAAK,MAAA;AACH,MAAA,OAAO,kBAAA,CAAmB,QAAA,EAAiC,OAAA,EAAS,MAAM,CAAA;AAAA,IAC5E,KAAK,OAAA;AACH,MAAA,OAAO,mBAAA,CAAoB,UAAiC,OAAO,CAAA;AAAA,IACrE,KAAK,MAAA;AACH,MAAA,OAAO,YAAY,QAAA,EAAiC,EAAE,MAAA,EAAQ,OAAA,EAAS,QAAQ,CAAA;AAAA,IACjF;AACE,MAAA,OAAO,oBAAA,CAAqB,QAAA,EAAU,MAAA,EAAQ,CAAC,sBAAsB,CAAC,CAAA;AAAA;AAE5E","file":"index.js","sourcesContent":["import type { DetectFileKindInput, FileLikeInput } from \"@dragon708/docmind-shared\";\nimport {\n  isBinaryInput,\n  isBlob,\n  isFile,\n  isNamedInput,\n  toUint8Array,\n} from \"@dragon708/docmind-shared\";\n\nexport function isByteBackedInput(input: DetectFileKindInput): input is FileLikeInput {\n  return isNamedInput(input) || isBinaryInput(input) || isBlob(input) || isFile(input);\n}\n\n/** Resolves bytes when the input carries a body; otherwise an empty `Uint8Array`. */\nexport async function bytesFromDetectInput(input: DetectFileKindInput): Promise<Uint8Array> {\n  if (!isByteBackedInput(input)) {\n    return new Uint8Array(0);\n  }\n  return toUint8Array(input);\n}\n","import { analyzeDocx as extractDocx } from \"@dragon708/docmind-docx\";\nimport type { AnalysisResult, DetectFileKindInput } from \"@dragon708/docmind-shared\";\nimport { bytesFromDetectInput } from \"../inputBytes.js\";\n\n/**\n * DOCX → `@dragon708/docmind-docx`.\n */\nexport async function analyzeDocxForNode(\n  input: DetectFileKindInput,\n  signal?: AbortSignal,\n): Promise<AnalysisResult> {\n  if (signal?.aborted) {\n    const err = new Error(\"The operation was aborted\");\n    err.name = \"AbortError\";\n    throw err;\n  }\n\n  const data = await bytesFromDetectInput(input);\n  if (data.byteLength === 0) {\n    return {\n      fileKind: \"docx\",\n      analyzer: \"docx\",\n      status: \"ok\",\n      kind: \"docx\",\n      text: \"\",\n      html: \"\",\n      warnings: [\"No document bytes were provided for analysis.\"],\n    };\n  }\n\n  const r = await extractDocx(data);\n  return {\n    fileKind: \"docx\",\n    analyzer: \"docx\",\n    status: \"ok\",\n    kind: \"docx\",\n    text: r.text,\n    html: r.html,\n    warnings: [...r.warnings],\n  };\n}\n","import { ocr } from \"@dragon708/docmind-ocr\";\nimport type { AnalysisResult, DetectFileKindInput } from \"@dragon708/docmind-shared\";\nimport type { NodeAnalyzeOptions } from \"../nodeAnalyzeOptions.js\";\nimport { bytesFromDetectInput } from \"../inputBytes.js\";\n\n/**\n * Image → `@dragon708/docmind-ocr`.\n */\nexport async function analyzeImageForNode(\n  input: DetectFileKindInput,\n  options?: NodeAnalyzeOptions,\n): Promise<AnalysisResult> {\n  if (options?.signal?.aborted) {\n    const err = new Error(\"The operation was aborted\");\n    err.name = \"AbortError\";\n    throw err;\n  }\n\n  const data = await bytesFromDetectInput(input);\n  if (data.byteLength === 0) {\n    return {\n      fileKind: \"image\",\n      analyzer: \"image\",\n      status: \"ok\",\n      kind: \"image\",\n      text: \"\",\n      confidence: 0,\n      ocrUsed: true,\n      warnings: [\"No image bytes were provided for analysis.\"],\n    };\n  }\n\n  const ocrOpts = {\n    ...(options?.ocr ?? {}),\n    signal: options?.ocr?.signal ?? options?.signal,\n  };\n\n  const r = await ocr(data, ocrOpts);\n  return {\n    fileKind: \"image\",\n    analyzer: \"image\",\n    status: \"ok\",\n    kind: \"image\",\n    text: r.text,\n    confidence: r.confidence,\n    ocrUsed: r.ocrUsed,\n    warnings: [],\n  };\n}\n","import { analyzePdf as runPdf } from \"@dragon708/docmind-pdf\";\nimport type { PdfAnalyzeOptions } from \"@dragon708/docmind-pdf\";\nimport type { AnalysisResult, DetectFileKindInput } from \"@dragon708/docmind-shared\";\nimport type { NodeAnalyzeOptions } from \"../nodeAnalyzeOptions.js\";\nimport { bytesFromDetectInput } from \"../inputBytes.js\";\n\n/**\n * PDF → `@dragon708/docmind-pdf` (Node / pdf-parse + OCR).\n *\n * Unlike `analyzePdf` from `@dragon708/docmind-pdf` (OCR off unless set), `analyzeFile` defaults\n * to `pdf.ocr: \"auto\"`: when the PDF has\n * pages but almost no extractable text (typical scan), the raster OCR pipeline runs. Pass\n * `pdf: { ocr: \"off\" }` to skip OCR for speed.\n */\nexport async function analyzePdfForNode(\n  input: DetectFileKindInput,\n  options?: NodeAnalyzeOptions,\n): Promise<AnalysisResult> {\n  if (options?.signal?.aborted) {\n    const err = new Error(\"The operation was aborted\");\n    err.name = \"AbortError\";\n    throw err;\n  }\n\n  const data = await bytesFromDetectInput(input);\n  if (data.byteLength === 0) {\n    return {\n      fileKind: \"pdf\",\n      analyzer: \"pdf\",\n      status: \"ok\",\n      kind: \"pdf\",\n      text: \"\",\n      pages: 0,\n      metadata: { info: {} },\n      warnings: [\"No document bytes were provided for analysis.\"],\n      needsOCR: false,\n      ocrUsed: false,\n    };\n  }\n\n  const userPdf = options?.pdf;\n  const pdfOpts: PdfAnalyzeOptions = {\n    ...userPdf,\n    ocr: userPdf?.ocr ?? \"auto\",\n    ocrLangs: userPdf?.ocrLangs ?? options?.ocr?.langs,\n    signal: userPdf?.signal ?? options?.signal,\n  };\n\n  const r = await runPdf(data, pdfOpts);\n  return {\n    fileKind: \"pdf\",\n    analyzer: \"pdf\",\n    status: \"ok\",\n    kind: \"pdf\",\n    text: r.text,\n    pages: r.pages,\n    metadata: r.metadata,\n    warnings: [...r.warnings],\n    needsOCR: r.needsOCR,\n    ocrUsed: r.ocrUsed,\n  };\n}\n","import type { DetectFileKindInput, NamedInput } from \"@dragon708/docmind-shared\";\nimport { readFile } from \"node:fs/promises\";\nimport { basename } from \"node:path\";\nimport { fileURLToPath } from \"node:url\";\n\n/**\n * Inputs accepted by {@link analyzeFile} in this package.\n * Paths and `file:` URLs are read with `fs`; other values pass through as {@link DetectFileKindInput}.\n */\nexport type NodeAnalyzeInput = string | URL | DetectFileKindInput;\n\nfunction toPathString(pathOrUrl: string | URL): string {\n  return pathOrUrl instanceof URL ? fileURLToPath(pathOrUrl) : pathOrUrl;\n}\n\n/**\n * Reads a file from disk into a {@link NamedInput} (binary `Buffer`, basename as `name` for hints).\n */\nexport async function readFileToInput(path: string | URL): Promise<NamedInput<Buffer>> {\n  const fsPath = toPathString(path);\n  const data = await readFile(fsPath);\n  return {\n    data,\n    name: basename(fsPath),\n  };\n}\n\n/** Wraps a `Buffer` as a named payload when you already know the filename. */\nexport function bufferToInput(buffer: Buffer, name?: string): NamedInput<Buffer> {\n  return name !== undefined ? { data: buffer, name } : { data: buffer };\n}\n\n/**\n * Resolves paths / `file:` URLs to a {@link DetectFileKindInput}; leaves other values untouched.\n */\nexport async function resolveNodeAnalyzeInput(input: NodeAnalyzeInput): Promise<DetectFileKindInput> {\n  if (typeof input === \"string\" || input instanceof URL) {\n    return readFileToInput(input);\n  }\n  return input;\n}\n","import type { AnalysisResult, DetectFileKindInput } from \"@dragon708/docmind-shared\";\nimport {\n  analyzeText,\n  assertValidAnalyzeFileInput,\n  detectFileKind,\n  notImplementedResult,\n  UNKNOWN_FORMAT_WARNING,\n} from \"@dragon708/docmind-shared\";\nimport { analyzeDocxForNode } from \"./analyzers/docx.js\";\nimport { analyzeImageForNode } from \"./analyzers/image.js\";\nimport { analyzePdfForNode } from \"./analyzers/pdf.js\";\nimport type { NodeAnalyzeOptions } from \"./nodeAnalyzeOptions.js\";\nimport { resolveNodeAnalyzeInput, type NodeAnalyzeInput } from \"./resolveNodeInput.js\";\n\n/**\n * Node router: PDF, DOCX, images (OCR), and text. Paths and `file:` URLs are read via `fs`.\n */\nexport async function analyzeFile(\n  input: NodeAnalyzeInput,\n  options?: NodeAnalyzeOptions,\n): Promise<AnalysisResult> {\n  if (options?.signal?.aborted) {\n    const err = new Error(\"The operation was aborted\");\n    err.name = \"AbortError\";\n    throw err;\n  }\n\n  const resolved = await resolveNodeAnalyzeInput(input);\n  assertValidAnalyzeFileInput(resolved);\n\n  const fileKind = detectFileKind(resolved);\n\n  switch (fileKind) {\n    case \"pdf\":\n      return analyzePdfForNode(resolved as DetectFileKindInput, options);\n    case \"docx\":\n      return analyzeDocxForNode(resolved as DetectFileKindInput, options?.signal);\n    case \"image\":\n      return analyzeImageForNode(resolved as DetectFileKindInput, options);\n    case \"text\":\n      return analyzeText(resolved as DetectFileKindInput, { signal: options?.signal });\n    default:\n      return notImplementedResult(fileKind, \"none\", [UNKNOWN_FORMAT_WARNING]);\n  }\n}\n"]}
+{"version":3,"sources":["../src/inputBytes.ts","../src/analyzers/docx.ts","../src/analyzers/image.ts","../src/analyzers/pdf.ts","../src/resolveNodeInput.ts","../src/analyze.ts","../src/publicActions.ts"],"names":["extractDocx","runPdf","assertValidAnalyzeFileInput","detectFileKind","ocr","analyzeText","notImplementedResult","UNKNOWN_FORMAT_WARNING"],"mappings":";;;;;;;;;;AASO,SAAS,kBAAkB,KAAA,EAAoD;AACpF,EAAA,OAAO,YAAA,CAAa,KAAK,CAAA,IAAK,aAAA,CAAc,KAAK,KAAK,MAAA,CAAO,KAAK,CAAA,IAAK,MAAA,CAAO,KAAK,CAAA;AACrF;AAGA,eAAsB,qBAAqB,KAAA,EAAiD;AAC1F,EAAA,IAAI,CAAC,iBAAA,CAAkB,KAAK,CAAA,EAAG;AAC7B,IAAA,OAAO,IAAI,WAAW,CAAC,CAAA;AAAA,EACzB;AACA,EAAA,OAAO,aAAa,KAAK,CAAA;AAC3B;;;ACZA,eAAsB,kBAAA,CACpB,OACA,MAAA,EACyB;AACzB,EAAA,IAAI,QAAQ,OAAA,EAAS;AACnB,IAAA,MAAM,GAAA,GAAM,IAAI,KAAA,CAAM,2BAA2B,CAAA;AACjD,IAAA,GAAA,CAAI,IAAA,GAAO,YAAA;AACX,IAAA,MAAM,GAAA;AAAA,EACR;AAEA,EAAA,MAAM,IAAA,GAAO,MAAM,oBAAA,CAAqB,KAAK,CAAA;AAC7C,EAAA,IAAI,IAAA,CAAK,eAAe,CAAA,EAAG;AACzB,IAAA,OAAO;AAAA,MACL,QAAA,EAAU,MAAA;AAAA,MACV,QAAA,EAAU,MAAA;AAAA,MACV,MAAA,EAAQ,IAAA;AAAA,MACR,IAAA,EAAM,MAAA;AAAA,MACN,IAAA,EAAM,EAAA;AAAA,MACN,IAAA,EAAM,EAAA;AAAA,MACN,QAAA,EAAU,CAAC,+CAA+C;AAAA,KAC5D;AAAA,EACF;AAEA,EAAA,MAAM,CAAA,GAAI,MAAMA,WAAA,CAAY,IAAI,CAAA;AAChC,EAAA,OAAO;AAAA,IACL,QAAA,EAAU,MAAA;AAAA,IACV,QAAA,EAAU,MAAA;AAAA,IACV,MAAA,EAAQ,IAAA;AAAA,IACR,IAAA,EAAM,MAAA;AAAA,IACN,MAAM,CAAA,CAAE,IAAA;AAAA,IACR,MAAM,CAAA,CAAE,IAAA;AAAA,IACR,QAAA,EAAU,CAAC,GAAG,CAAA,CAAE,QAAQ;AAAA,GAC1B;AACF;AChCA,eAAsB,mBAAA,CACpB,OACA,OAAA,EACyB;AACzB,EAAA,IAAI,OAAA,EAAS,QAAQ,OAAA,EAAS;AAC5B,IAAA,MAAM,GAAA,GAAM,IAAI,KAAA,CAAM,2BAA2B,CAAA;AACjD,IAAA,GAAA,CAAI,IAAA,GAAO,YAAA;AACX,IAAA,MAAM,GAAA;AAAA,EACR;AAEA,EAAA,MAAM,IAAA,GAAO,MAAM,oBAAA,CAAqB,KAAK,CAAA;AAC7C,EAAA,IAAI,IAAA,CAAK,eAAe,CAAA,EAAG;AACzB,IAAA,OAAO;AAAA,MACL,QAAA,EAAU,OAAA;AAAA,MACV,QAAA,EAAU,OAAA;AAAA,MACV,MAAA,EAAQ,IAAA;AAAA,MACR,IAAA,EAAM,OAAA;AAAA,MACN,IAAA,EAAM,EAAA;AAAA,MACN,UAAA,EAAY,CAAA;AAAA,MACZ,OAAA,EAAS,IAAA;AAAA,MACT,QAAA,EAAU,CAAC,4CAA4C;AAAA,KACzD;AAAA,EACF;AAEA,EAAA,MAAM,OAAA,GAAU;AAAA,IACd,GAAI,OAAA,EAAS,GAAA,IAAO,EAAC;AAAA,IACrB,MAAA,EAAQ,OAAA,EAAS,GAAA,EAAK,MAAA,IAAU,OAAA,EAAS;AAAA,GAC3C;AAEA,EAAA,MAAM,CAAA,GAAI,MAAM,GAAA,CAAI,IAAA,EAAM,OAAO,CAAA;AACjC,EAAA,OAAO;AAAA,IACL,QAAA,EAAU,OAAA;AAAA,IACV,QAAA,EAAU,OAAA;AAAA,IACV,MAAA,EAAQ,IAAA;AAAA,IACR,IAAA,EAAM,OAAA;AAAA,IACN,MAAM,CAAA,CAAE,IAAA;AAAA,IACR,YAAY,CAAA,CAAE,UAAA;AAAA,IACd,SAAS,CAAA,CAAE,OAAA;AAAA,IACX,UAAU;AAAC,GACb;AACF;AClCA,eAAsB,iBAAA,CACpB,OACA,OAAA,EACyB;AACzB,EAAA,IAAI,OAAA,EAAS,QAAQ,OAAA,EAAS;AAC5B,IAAA,MAAM,GAAA,GAAM,IAAI,KAAA,CAAM,2BAA2B,CAAA;AACjD,IAAA,GAAA,CAAI,IAAA,GAAO,YAAA;AACX,IAAA,MAAM,GAAA;AAAA,EACR;AAEA,EAAA,MAAM,IAAA,GAAO,MAAM,oBAAA,CAAqB,KAAK,CAAA;AAC7C,EAAA,IAAI,IAAA,CAAK,eAAe,CAAA,EAAG;AACzB,IAAA,OAAO;AAAA,MACL,QAAA,EAAU,KAAA;AAAA,MACV,QAAA,EAAU,KAAA;AAAA,MACV,MAAA,EAAQ,IAAA;AAAA,MACR,IAAA,EAAM,KAAA;AAAA,MACN,IAAA,EAAM,EAAA;AAAA,MACN,KAAA,EAAO,CAAA;AAAA,MACP,QAAA,EAAU,EAAE,IAAA,EAAM,EAAC,EAAE;AAAA,MACrB,QAAA,EAAU,CAAC,+CAA+C,CAAA;AAAA,MAC1D,QAAA,EAAU,KAAA;AAAA,MACV,OAAA,EAAS;AAAA,KACX;AAAA,EACF;AAEA,EAAA,MAAM,UAAU,OAAA,EAAS,GAAA;AACzB,EAAA,MAAM,OAAA,GAA6B;AAAA,IACjC,GAAG,OAAA;AAAA,IACH,GAAA,EAAK,SAAS,GAAA,IAAO,MAAA;AAAA,IACrB,QAAA,EAAU,OAAA,EAAS,QAAA,IAAY,OAAA,EAAS,GAAA,EAAK,KAAA;AAAA,IAC7C,MAAA,EAAQ,OAAA,EAAS,MAAA,IAAU,OAAA,EAAS;AAAA,GACtC;AAEA,EAAA,MAAM,CAAA,GAAI,MAAMC,UAAA,CAAO,IAAA,EAAM,OAAO,CAAA;AACpC,EAAA,OAAO;AAAA,IACL,QAAA,EAAU,KAAA;AAAA,IACV,QAAA,EAAU,KAAA;AAAA,IACV,MAAA,EAAQ,IAAA;AAAA,IACR,IAAA,EAAM,KAAA;AAAA,IACN,MAAM,CAAA,CAAE,IAAA;AAAA,IACR,OAAO,CAAA,CAAE,KAAA;AAAA,IACT,UAAU,CAAA,CAAE,QAAA;AAAA,IACZ,QAAA,EAAU,CAAC,GAAG,CAAA,CAAE,QAAQ,CAAA;AAAA,IACxB,UAAU,CAAA,CAAE,QAAA;AAAA,IACZ,SAAS,CAAA,CAAE;AAAA,GACb;AACF;AClDA,SAAS,aAAa,SAAA,EAAiC;AACrD,EAAA,OAAO,SAAA,YAAqB,GAAA,GAAM,aAAA,CAAc,SAAS,CAAA,GAAI,SAAA;AAC/D;AAKA,eAAsB,gBAAgB,IAAA,EAAiD;AACrF,EAAA,MAAM,MAAA,GAAS,aAAa,IAAI,CAAA;AAChC,EAAA,MAAM,IAAA,GAAO,MAAM,QAAA,CAAS,MAAM,CAAA;AAClC,EAAA,OAAO;AAAA,IACL,IAAA;AAAA,IACA,IAAA,EAAM,SAAS,MAAM;AAAA,GACvB;AACF;AAGO,SAAS,aAAA,CAAc,QAAgB,IAAA,EAAmC;AAC/E,EAAA,OAAO,IAAA,KAAS,SAAY,EAAE,IAAA,EAAM,QAAQ,IAAA,EAAK,GAAI,EAAE,IAAA,EAAM,MAAA,EAAO;AACtE;AAKA,eAAsB,wBAAwB,KAAA,EAAuD;AACnG,EAAA,IAAI,OAAO,KAAA,KAAU,QAAA,IAAY,KAAA,YAAiB,GAAA,EAAK;AACrD,IAAA,OAAO,gBAAgB,KAAK,CAAA;AAAA,EAC9B;AACA,EAAA,OAAO,KAAA;AACT;;;ACvBA,eAAsB,WAAA,CACpB,OACA,OAAA,EACyB;AACzB,EAAA,IAAI,OAAA,EAAS,QAAQ,OAAA,EAAS;AAC5B,IAAA,MAAM,GAAA,GAAM,IAAI,KAAA,CAAM,2BAA2B,CAAA;AACjD,IAAA,GAAA,CAAI,IAAA,GAAO,YAAA;AACX,IAAA,MAAM,GAAA;AAAA,EACR;AAEA,EAAA,MAAM,QAAA,GAAW,MAAM,uBAAA,CAAwB,KAAK,CAAA;AACpD,EAAA,2BAAA,CAA4B,QAAQ,CAAA;AAEpC,EAAA,MAAM,QAAA,GAAW,eAAe,QAAQ,CAAA;AAExC,EAAA,QAAQ,QAAA;AAAU,IAChB,KAAK,KAAA;AACH,MAAA,OAAO,iBAAA,CAAkB,UAAiC,OAAO,CAAA;AAAA,IACnE,KAAK,MAAA;AACH,MAAA,OAAO,kBAAA,CAAmB,QAAA,EAAiC,OAAA,EAAS,MAAM,CAAA;AAAA,IAC5E,KAAK,OAAA;AACH,MAAA,OAAO,mBAAA,CAAoB,UAAiC,OAAO,CAAA;AAAA,IACrE,KAAK,MAAA;AACH,MAAA,OAAO,YAAY,QAAA,EAAiC,EAAE,MAAA,EAAQ,OAAA,EAAS,QAAQ,CAAA;AAAA,IACjF;AACE,MAAA,OAAO,oBAAA,CAAqB,QAAA,EAAU,MAAA,EAAQ,CAAC,sBAAsB,CAAC,CAAA;AAAA;AAE5E;ACrBA,SAAS,eAAe,MAAA,EAA4B;AAClD,EAAA,IAAI,QAAQ,OAAA,EAAS;AACnB,IAAA,MAAM,GAAA,GAAM,IAAI,KAAA,CAAM,2BAA2B,CAAA;AACjD,IAAA,GAAA,CAAI,IAAA,GAAO,YAAA;AACX,IAAA,MAAM,GAAA;AAAA,EACR;AACF;AAEA,eAAe,QAAQ,KAAA,EAAuD;AAC5E,EAAA,MAAM,QAAA,GAAW,MAAM,uBAAA,CAAwB,KAAK,CAAA;AACpD,EAAAC,4BAA4B,QAAQ,CAAA;AACpC,EAAA,OAAO,QAAA;AACT;AAEA,SAAS,kBAAkB,CAAA,EAAmB;AAC5C,EAAA,OAAO,CAAA,CACJ,OAAA,CAAQ,IAAA,EAAM,OAAO,EACrB,OAAA,CAAQ,IAAA,EAAM,MAAM,CAAA,CACpB,QAAQ,IAAA,EAAM,MAAM,CAAA,CACpB,OAAA,CAAQ,MAAM,QAAQ,CAAA;AAC3B;AAEA,IAAM,kBAAA,GACJ,yGAAA;AAEF,IAAM,mBAAA,GACJ,6DAAA;AAMF,eAAsB,WAAA,CACpB,OACA,OAAA,EACyB;AACzB,EAAA,cAAA,CAAe,SAAS,MAAM,CAAA;AAC9B,EAAA,MAAM,QAAA,GAAW,MAAM,OAAA,CAAQ,KAAK,CAAA;AACpC,EAAA,MAAM,IAAA,GAAOC,eAAe,QAAQ,CAAA;AACpC,EAAA,MAAM,SAAS,OAAA,EAAS,MAAA;AAExB,EAAA,QAAQ,IAAA;AAAM,IACZ,KAAK,KAAA,EAAO;AACV,MAAA,MAAM,IAAA,GAAO,MAAM,oBAAA,CAAqB,QAAQ,CAAA;AAChD,MAAA,IAAI,IAAA,CAAK,eAAe,CAAA,EAAG;AACzB,QAAA,OAAO;AAAA,UACL,QAAA,EAAU,KAAA;AAAA,UACV,QAAA,EAAU,KAAA;AAAA,UACV,MAAA,EAAQ,IAAA;AAAA,UACR,IAAA,EAAM,KAAA;AAAA,UACN,IAAA,EAAM,EAAA;AAAA,UACN,KAAA,EAAO,CAAA;AAAA,UACP,QAAA,EAAU,EAAE,IAAA,EAAM,EAAC,EAAE;AAAA,UACrB,QAAA,EAAU,CAAC,+CAA+C,CAAA;AAAA,UAC1D,QAAA,EAAU,KAAA;AAAA,UACV,OAAA,EAAS;AAAA,SACX;AAAA,MACF;AACA,MAAA,MAAM,CAAA,GAAI,MAAM,kBAAA,CAAmB,IAAI,CAAA;AACvC,MAAA,OAAO;AAAA,QACL,QAAA,EAAU,KAAA;AAAA,QACV,QAAA,EAAU,KAAA;AAAA,QACV,MAAA,EAAQ,IAAA;AAAA,QACR,IAAA,EAAM,KAAA;AAAA,QACN,MAAM,CAAA,CAAE,IAAA;AAAA,QACR,OAAO,CAAA,CAAE,KAAA;AAAA,QACT,QAAA,EAAU,EAAE,IAAA,EAAM,EAAC,EAAE;AAAA,QACrB,UAAU,CAAA,CAAE,QAAA;AAAA,QACZ,QAAA,EAAU,KAAA;AAAA,QACV,OAAA,EAAS;AAAA,OACX;AAAA,IACF;AAAA,IACA,KAAK,MAAA,EAAQ;AACX,MAAA,MAAM,IAAA,GAAO,MAAM,oBAAA,CAAqB,QAAQ,CAAA;AAChD,MAAA,IAAI,IAAA,CAAK,eAAe,CAAA,EAAG;AACzB,QAAA,OAAO;AAAA,UACL,QAAA,EAAU,MAAA;AAAA,UACV,QAAA,EAAU,MAAA;AAAA,UACV,MAAA,EAAQ,IAAA;AAAA,UACR,IAAA,EAAM,MAAA;AAAA,UACN,IAAA,EAAM,EAAA;AAAA,UACN,IAAA,EAAM,EAAA;AAAA,UACN,QAAA,EAAU,CAAC,+CAA+C;AAAA,SAC5D;AAAA,MACF;AACA,MAAA,MAAM,CAAA,GAAI,MAAM,mBAAA,CAAoB,IAAI,CAAA;AACxC,MAAA,OAAO;AAAA,QACL,QAAA,EAAU,MAAA;AAAA,QACV,QAAA,EAAU,MAAA;AAAA,QACV,MAAA,EAAQ,IAAA;AAAA,QACR,IAAA,EAAM,MAAA;AAAA,QACN,MAAM,CAAA,CAAE,IAAA;AAAA,QACR,IAAA,EAAM,EAAA;AAAA,QACN,UAAU,CAAA,CAAE;AAAA,OACd;AAAA,IACF;AAAA,IACA,KAAK,OAAA,EAAS;AACZ,MAAA,MAAM,IAAA,GAAO,MAAM,oBAAA,CAAqB,QAAQ,CAAA;AAChD,MAAA,IAAI,IAAA,CAAK,eAAe,CAAA,EAAG;AACzB,QAAA,OAAO;AAAA,UACL,QAAA,EAAU,OAAA;AAAA,UACV,QAAA,EAAU,OAAA;AAAA,UACV,MAAA,EAAQ,IAAA;AAAA,UACR,IAAA,EAAM,OAAA;AAAA,UACN,IAAA,EAAM,EAAA;AAAA,UACN,UAAA,EAAY,CAAA;AAAA,UACZ,OAAA,EAAS,IAAA;AAAA,UACT,QAAA,EAAU,CAAC,4CAA4C;AAAA,SACzD;AAAA,MACF;AACA,MAAA,MAAM,OAAA,GAAU;AAAA,QACd,GAAI,OAAA,EAAS,GAAA,IAAO,EAAC;AAAA,QACrB,MAAA,EAAQ,OAAA,EAAS,GAAA,EAAK,MAAA,IAAU;AAAA,OAClC;AACA,MAAA,MAAM,CAAA,GAAI,MAAMC,GAAAA,CAAI,IAAA,EAAM,OAAO,CAAA;AACjC,MAAA,OAAO;AAAA,QACL,QAAA,EAAU,OAAA;AAAA,QACV,QAAA,EAAU,OAAA;AAAA,QACV,MAAA,EAAQ,IAAA;AAAA,QACR,IAAA,EAAM,OAAA;AAAA,QACN,MAAM,CAAA,CAAE,IAAA;AAAA,QACR,YAAY,CAAA,CAAE,UAAA;AAAA,QACd,SAAS,CAAA,CAAE,OAAA;AAAA,QACX,UAAU;AAAC,OACb;AAAA,IACF;AAAA,IACA,KAAK,MAAA;AACH,MAAA,OAAOC,WAAAA,CAAY,QAAA,EAAiC,EAAE,MAAA,EAAQ,CAAA;AAAA,IAChE;AACE,MAAA,OAAOC,oBAAAA,CAAqB,IAAA,EAAM,MAAA,EAAQ,CAACC,sBAAsB,CAAC,CAAA;AAAA;AAExE;AAMA,eAAsB,eAAA,CACpB,OACA,OAAA,EACyB;AACzB,EAAA,cAAA,CAAe,SAAS,MAAM,CAAA;AAC9B,EAAA,MAAM,QAAA,GAAW,MAAM,OAAA,CAAQ,KAAK,CAAA;AACpC,EAAA,MAAM,IAAA,GAAOJ,eAAe,QAAQ,CAAA;AACpC,EAAA,MAAM,SAAS,OAAA,EAAS,MAAA;AAExB,EAAA,QAAQ,IAAA;AAAM,IACZ,KAAK,KAAA,EAAO;AACV,MAAA,MAAM,IAAA,GAAO,MAAM,oBAAA,CAAqB,QAAQ,CAAA;AAChD,MAAA,IAAI,IAAA,CAAK,eAAe,CAAA,EAAG;AACzB,QAAA,OAAO;AAAA,UACL,QAAA,EAAU,KAAA;AAAA,UACV,QAAA,EAAU,KAAA;AAAA,UACV,MAAA,EAAQ,IAAA;AAAA,UACR,IAAA,EAAM,KAAA;AAAA,UACN,IAAA,EAAM,EAAA;AAAA,UACN,KAAA,EAAO,CAAA;AAAA,UACP,QAAA,EAAU,EAAE,IAAA,EAAM,EAAC,EAAE;AAAA,UACrB,QAAA,EAAU,CAAC,+CAA+C,CAAA;AAAA,UAC1D,QAAA,EAAU,KAAA;AAAA,UACV,OAAA,EAAS;AAAA,SACX;AAAA,MACF;AACA,MAAA,MAAM,CAAA,GAAI,MAAM,kBAAA,CAAmB,IAAI,CAAA;AACvC,MAAA,OAAO;AAAA,QACL,QAAA,EAAU,KAAA;AAAA,QACV,QAAA,EAAU,KAAA;AAAA,QACV,MAAA,EAAQ,IAAA;AAAA,QACR,IAAA,EAAM,KAAA;AAAA,QACN,IAAA,EAAM,EAAA;AAAA,QACN,KAAA,EAAO,CAAA;AAAA,QACP,UAAU,CAAA,CAAE,QAAA;AAAA,QACZ,UAAU,CAAA,CAAE,QAAA;AAAA,QACZ,QAAA,EAAU,KAAA;AAAA,QACV,OAAA,EAAS;AAAA,OACX;AAAA,IACF;AAAA,IACA,KAAK,MAAA;AACH,MAAA,OAAO;AAAA,QACL,QAAA,EAAU,MAAA;AAAA,QACV,QAAA,EAAU,MAAA;AAAA,QACV,MAAA,EAAQ,IAAA;AAAA,QACR,IAAA,EAAM,MAAA;AAAA,QACN,IAAA,EAAM,EAAA;AAAA,QACN,IAAA,EAAM,EAAA;AAAA,QACN,QAAA,EAAU,CAAC,kBAAkB;AAAA,OAC/B;AAAA,IACF,KAAK,OAAA;AACH,MAAA,OAAO;AAAA,QACL,QAAA,EAAU,OAAA;AAAA,QACV,QAAA,EAAU,OAAA;AAAA,QACV,MAAA,EAAQ,IAAA;AAAA,QACR,IAAA,EAAM,OAAA;AAAA,QACN,IAAA,EAAM,EAAA;AAAA,QACN,UAAA,EAAY,CAAA;AAAA,QACZ,OAAA,EAAS,IAAA;AAAA,QACT,QAAA,EAAU,CAAC,mBAAmB;AAAA,OAChC;AAAA,IACF,KAAK,MAAA;AACH,MAAA,OAAOE,WAAAA,CAAY,QAAA,EAAiC,EAAE,MAAA,EAAQ,CAAA;AAAA,IAChE;AACE,MAAA,OAAOC,oBAAAA,CAAqB,IAAA,EAAM,MAAA,EAAQ,CAACC,sBAAsB,CAAC,CAAA;AAAA;AAExE;AAMA,eAAsB,aAAA,CACpB,OACA,OAAA,EACyB;AACzB,EAAA,cAAA,CAAe,SAAS,MAAM,CAAA;AAC9B,EAAA,MAAM,QAAA,GAAW,MAAM,OAAA,CAAQ,KAAK,CAAA;AACpC,EAAA,MAAM,IAAA,GAAOJ,eAAe,QAAQ,CAAA;AACpC,EAAA,MAAM,SAAS,OAAA,EAAS,MAAA;AAExB,EAAA,QAAQ,IAAA;AAAM,IACZ,KAAK,MAAA,EAAQ;AACX,MAAA,MAAM,IAAA,GAAO,MAAM,oBAAA,CAAqB,QAAQ,CAAA;AAChD,MAAA,IAAI,IAAA,CAAK,eAAe,CAAA,EAAG;AACzB,QAAA,OAAO;AAAA,UACL,QAAA,EAAU,MAAA;AAAA,UACV,QAAA,EAAU,MAAA;AAAA,UACV,MAAA,EAAQ,IAAA;AAAA,UACR,IAAA,EAAM,MAAA;AAAA,UACN,IAAA,EAAM,EAAA;AAAA,UACN,IAAA,EAAM,EAAA;AAAA,UACN,QAAA,EAAU,CAAC,+CAA+C;AAAA,SAC5D;AAAA,MACF;AACA,MAAA,MAAM,CAAC,QAAA,EAAU,QAAQ,CAAA,GAAI,MAAM,QAAQ,GAAA,CAAI;AAAA,QAC7C,oBAAoB,IAAI,CAAA;AAAA,QACxB,kBAAkB,IAAI;AAAA,OACvB,CAAA;AACD,MAAA,OAAO;AAAA,QACL,QAAA,EAAU,MAAA;AAAA,QACV,QAAA,EAAU,MAAA;AAAA,QACV,MAAA,EAAQ,IAAA;AAAA,QACR,IAAA,EAAM,MAAA;AAAA,QACN,MAAM,QAAA,CAAS,IAAA;AAAA,QACf,MAAM,QAAA,CAAS,IAAA;AAAA,QACf,UAAU,CAAC,GAAG,SAAS,QAAA,EAAU,GAAG,SAAS,QAAQ;AAAA,OACvD;AAAA,IACF;AAAA,IACA,KAAK,KAAA,EAAO;AACV,MAAA,MAAM,IAAA,GAAO,MAAM,oBAAA,CAAqB,QAAQ,CAAA;AAChD,MAAA,IAAI,IAAA,CAAK,eAAe,CAAA,EAAG;AACzB,QAAA,OAAO;AAAA,UACL,QAAA,EAAU,KAAA;AAAA,UACV,QAAA,EAAU,KAAA;AAAA,UACV,MAAA,EAAQ,IAAA;AAAA,UACR,IAAA,EAAM,KAAA;AAAA,UACN,IAAA,EAAM,EAAA;AAAA,UACN,KAAA,EAAO,CAAA;AAAA,UACP,QAAA,EAAU,EAAE,IAAA,EAAM,EAAC,EAAE;AAAA,UACrB,QAAA,EAAU,CAAC,+CAA+C,CAAA;AAAA,UAC1D,QAAA,EAAU,KAAA;AAAA,UACV,OAAA,EAAS;AAAA,SACX;AAAA,MACF;AACA,MAAA,MAAM,CAAA,GAAI,MAAM,kBAAA,CAAmB,IAAI,CAAA;AACvC,MAAA,MAAM,IAAA,GAAO,CAAA,KAAA,EAAQ,iBAAA,CAAkB,CAAA,CAAE,IAAI,CAAC,CAAA,MAAA,CAAA;AAC9C,MAAA,OAAO;AAAA,QACL,QAAA,EAAU,KAAA;AAAA,QACV,QAAA,EAAU,KAAA;AAAA,QACV,MAAA,EAAQ,IAAA;AAAA,QACR,IAAA,EAAM,KAAA;AAAA,QACN,MAAM,CAAA,CAAE,IAAA;AAAA,QACR,OAAO,CAAA,CAAE,KAAA;AAAA,QACT,QAAA,EAAU,EAAE,IAAA,EAAM,EAAC,EAAE;AAAA,QACrB,IAAA;AAAA,QACA,QAAA,EAAU;AAAA,UACR,GAAG,CAAA,CAAE,QAAA;AAAA,UACL;AAAA,SACF;AAAA,QACA,QAAA,EAAU,KAAA;AAAA,QACV,OAAA,EAAS;AAAA,OACX;AAAA,IACF;AAAA,IACA,KAAK,MAAA,EAAQ;AACX,MAAA,MAAM,IAAI,MAAME,WAAAA,CAAY,QAAA,EAAiC,EAAE,QAAQ,CAAA;AACvE,MAAA,MAAM,IAAA,GAAO,CAAA,KAAA,EAAQ,iBAAA,CAAkB,CAAA,CAAE,IAAI,CAAC,CAAA,MAAA,CAAA;AAC9C,MAAA,OAAO;AAAA,QACL,GAAG,CAAA;AAAA,QACH,IAAA;AAAA,QACA,QAAA,EAAU;AAAA,UACR,GAAG,CAAA,CAAE,QAAA;AAAA,UACL;AAAA;AACF,OACF;AAAA,IACF;AAAA,IACA,KAAK,OAAA;AACH,MAAA,OAAO;AAAA,QACL,QAAA,EAAU,OAAA;AAAA,QACV,QAAA,EAAU,OAAA;AAAA,QACV,MAAA,EAAQ,IAAA;AAAA,QACR,IAAA,EAAM,OAAA;AAAA,QACN,IAAA,EAAM,EAAA;AAAA,QACN,UAAA,EAAY,CAAA;AAAA,QACZ,OAAA,EAAS,IAAA;AAAA,QACT,QAAA,EAAU,CAAC,qEAAqE;AAAA,OAClF;AAAA,IACF;AACE,MAAA,OAAOC,oBAAAA,CAAqB,IAAA,EAAM,MAAA,EAAQ,CAACC,sBAAsB,CAAC,CAAA;AAAA;AAExE;AAMA,eAAsB,MAAA,CACpB,OACA,OAAA,EACyB;AACzB,EAAA,cAAA,CAAe,SAAS,MAAM,CAAA;AAC9B,EAAA,MAAM,QAAA,GAAW,MAAM,OAAA,CAAQ,KAAK,CAAA;AACpC,EAAA,MAAM,IAAA,GAAOJ,eAAe,QAAQ,CAAA;AACpC,EAAA,MAAM,SAAS,OAAA,EAAS,MAAA;AACxB,EAAA,MAAM,IAAA,GAAO,OAAA,EAAS,GAAA,EAAK,KAAA,IAAS,SAAS,GAAA,EAAK,QAAA;AAElD,EAAA,QAAQ,IAAA;AAAM,IACZ,KAAK,KAAA,EAAO;AACV,MAAA,MAAM,IAAA,GAAO,MAAM,oBAAA,CAAqB,QAAQ,CAAA;AAChD,MAAA,IAAI,IAAA,CAAK,eAAe,CAAA,EAAG;AACzB,QAAA,OAAO;AAAA,UACL,QAAA,EAAU,KAAA;AAAA,UACV,QAAA,EAAU,KAAA;AAAA,UACV,MAAA,EAAQ,IAAA;AAAA,UACR,IAAA,EAAM,KAAA;AAAA,UACN,IAAA,EAAM,EAAA;AAAA,UACN,KAAA,EAAO,CAAA;AAAA,UACP,QAAA,EAAU,EAAE,IAAA,EAAM,EAAC,EAAE;AAAA,UACrB,QAAA,EAAU,CAAC,+CAA+C,CAAA;AAAA,UAC1D,QAAA,EAAU,KAAA;AAAA,UACV,OAAA,EAAS;AAAA,SACX;AAAA,MACF;AACA,MAAA,MAAM,CAAA,GAAI,MAAM,UAAA,CAAW,IAAA,EAAM;AAAA,QAC/B,GAAG,OAAA,EAAS,GAAA;AAAA,QACZ,GAAA,EAAK,OAAA;AAAA,QACL,QAAA,EAAU,IAAA,IAAQ,OAAA,EAAS,GAAA,EAAK,QAAA;AAAA,QAChC,MAAA,EAAQ,OAAA,EAAS,GAAA,EAAK,MAAA,IAAU;AAAA,OACjC,CAAA;AACD,MAAA,OAAO;AAAA,QACL,QAAA,EAAU,KAAA;AAAA,QACV,QAAA,EAAU,KAAA;AAAA,QACV,MAAA,EAAQ,IAAA;AAAA,QACR,IAAA,EAAM,KAAA;AAAA,QACN,MAAM,CAAA,CAAE,IAAA;AAAA,QACR,OAAO,CAAA,CAAE,KAAA;AAAA,QACT,UAAU,CAAA,CAAE,QAAA;AAAA,QACZ,QAAA,EAAU,CAAC,GAAG,CAAA,CAAE,QAAQ,CAAA;AAAA,QACxB,UAAU,CAAA,CAAE,QAAA;AAAA,QACZ,SAAS,CAAA,CAAE;AAAA,OACb;AAAA,IACF;AAAA,IACA,KAAK,OAAA,EAAS;AACZ,MAAA,MAAM,IAAA,GAAO,MAAM,oBAAA,CAAqB,QAAQ,CAAA;AAChD,MAAA,IAAI,IAAA,CAAK,eAAe,CAAA,EAAG;AACzB,QAAA,OAAO;AAAA,UACL,QAAA,EAAU,OAAA;AAAA,UACV,QAAA,EAAU,OAAA;AAAA,UACV,MAAA,EAAQ,IAAA;AAAA,UACR,IAAA,EAAM,OAAA;AAAA,UACN,IAAA,EAAM,EAAA;AAAA,UACN,UAAA,EAAY,CAAA;AAAA,UACZ,OAAA,EAAS,IAAA;AAAA,UACT,QAAA,EAAU,CAAC,4CAA4C;AAAA,SACzD;AAAA,MACF;AACA,MAAA,MAAM,OAAA,GAAU;AAAA,QACd,GAAI,OAAA,EAAS,GAAA,IAAO,EAAC;AAAA,QACrB,MAAA,EAAQ,OAAA,EAAS,GAAA,EAAK,MAAA,IAAU;AAAA,OAClC;AACA,MAAA,MAAM,CAAA,GAAI,MAAMC,GAAAA,CAAI,IAAA,EAAM,OAAO,CAAA;AACjC,MAAA,OAAO;AAAA,QACL,QAAA,EAAU,OAAA;AAAA,QACV,QAAA,EAAU,OAAA;AAAA,QACV,MAAA,EAAQ,IAAA;AAAA,QACR,IAAA,EAAM,OAAA;AAAA,QACN,MAAM,CAAA,CAAE,IAAA;AAAA,QACR,YAAY,CAAA,CAAE,UAAA;AAAA,QACd,SAAS,CAAA,CAAE,OAAA;AAAA,QACX,UAAU;AAAC,OACb;AAAA,IACF;AAAA,IACA,KAAK,MAAA,EAAQ;AACX,MAAA,MAAM,IAAA,GAAO,MAAM,oBAAA,CAAqB,QAAQ,CAAA;AAChD,MAAA,IAAI,IAAA,CAAK,eAAe,CAAA,EAAG;AACzB,QAAA,OAAO;AAAA,UACL,QAAA,EAAU,MAAA;AAAA,UACV,QAAA,EAAU,MAAA;AAAA,UACV,MAAA,EAAQ,IAAA;AAAA,UACR,IAAA,EAAM,MAAA;AAAA,UACN,IAAA,EAAM,EAAA;AAAA,UACN,IAAA,EAAM,EAAA;AAAA,UACN,QAAA,EAAU,CAAC,+CAA+C;AAAA,SAC5D;AAAA,MACF;AACA,MAAA,MAAM,CAAA,GAAI,MAAM,WAAA,CAAY,IAAI,CAAA;AAChC,MAAA,OAAO;AAAA,QACL,QAAA,EAAU,MAAA;AAAA,QACV,QAAA,EAAU,MAAA;AAAA,QACV,MAAA,EAAQ,IAAA;AAAA,QACR,IAAA,EAAM,MAAA;AAAA,QACN,MAAM,CAAA,CAAE,IAAA;AAAA,QACR,MAAM,CAAA,CAAE,IAAA;AAAA,QACR,QAAA,EAAU;AAAA,UACR,GAAG,CAAA,CAAE,QAAA;AAAA,UACL;AAAA;AACF,OACF;AAAA,IACF;AAAA,IACA,KAAK,MAAA;AACH,MAAA,OAAOC,WAAAA,CAAY,QAAA,EAAiC,EAAE,MAAA,EAAQ,CAAA;AAAA,IAChE;AACE,MAAA,OAAOC,oBAAAA,CAAqB,IAAA,EAAM,MAAA,EAAQ,CAACC,sBAAsB,CAAC,CAAA;AAAA;AAExE","file":"index.js","sourcesContent":["import type { DetectFileKindInput, FileLikeInput } from \"@dragon708/docmind-shared\";\nimport {\n  isBinaryInput,\n  isBlob,\n  isFile,\n  isNamedInput,\n  toUint8Array,\n} from \"@dragon708/docmind-shared\";\n\nexport function isByteBackedInput(input: DetectFileKindInput): input is FileLikeInput {\n  return isNamedInput(input) || isBinaryInput(input) || isBlob(input) || isFile(input);\n}\n\n/** Resolves bytes when the input carries a body; otherwise an empty `Uint8Array`. */\nexport async function bytesFromDetectInput(input: DetectFileKindInput): Promise<Uint8Array> {\n  if (!isByteBackedInput(input)) {\n    return new Uint8Array(0);\n  }\n  return toUint8Array(input);\n}\n","import { analyzeDocx as extractDocx } from \"@dragon708/docmind-docx\";\nimport type { AnalysisResult, DetectFileKindInput } from \"@dragon708/docmind-shared\";\nimport { bytesFromDetectInput } from \"../inputBytes.js\";\n\n/**\n * DOCX → `@dragon708/docmind-docx`.\n */\nexport async function analyzeDocxForNode(\n  input: DetectFileKindInput,\n  signal?: AbortSignal,\n): Promise<AnalysisResult> {\n  if (signal?.aborted) {\n    const err = new Error(\"The operation was aborted\");\n    err.name = \"AbortError\";\n    throw err;\n  }\n\n  const data = await bytesFromDetectInput(input);\n  if (data.byteLength === 0) {\n    return {\n      fileKind: \"docx\",\n      analyzer: \"docx\",\n      status: \"ok\",\n      kind: \"docx\",\n      text: \"\",\n      html: \"\",\n      warnings: [\"No document bytes were provided for analysis.\"],\n    };\n  }\n\n  const r = await extractDocx(data);\n  return {\n    fileKind: \"docx\",\n    analyzer: \"docx\",\n    status: \"ok\",\n    kind: \"docx\",\n    text: r.text,\n    html: r.html,\n    warnings: [...r.warnings],\n  };\n}\n","import { ocr } from \"@dragon708/docmind-ocr\";\nimport type { AnalysisResult, DetectFileKindInput } from \"@dragon708/docmind-shared\";\nimport type { NodeAnalyzeOptions } from \"../nodeAnalyzeOptions.js\";\nimport { bytesFromDetectInput } from \"../inputBytes.js\";\n\n/**\n * Image → `@dragon708/docmind-ocr`.\n */\nexport async function analyzeImageForNode(\n  input: DetectFileKindInput,\n  options?: NodeAnalyzeOptions,\n): Promise<AnalysisResult> {\n  if (options?.signal?.aborted) {\n    const err = new Error(\"The operation was aborted\");\n    err.name = \"AbortError\";\n    throw err;\n  }\n\n  const data = await bytesFromDetectInput(input);\n  if (data.byteLength === 0) {\n    return {\n      fileKind: \"image\",\n      analyzer: \"image\",\n      status: \"ok\",\n      kind: \"image\",\n      text: \"\",\n      confidence: 0,\n      ocrUsed: true,\n      warnings: [\"No image bytes were provided for analysis.\"],\n    };\n  }\n\n  const ocrOpts = {\n    ...(options?.ocr ?? {}),\n    signal: options?.ocr?.signal ?? options?.signal,\n  };\n\n  const r = await ocr(data, ocrOpts);\n  return {\n    fileKind: \"image\",\n    analyzer: \"image\",\n    status: \"ok\",\n    kind: \"image\",\n    text: r.text,\n    confidence: r.confidence,\n    ocrUsed: r.ocrUsed,\n    warnings: [],\n  };\n}\n","import { analyzePdf as runPdf } from \"@dragon708/docmind-pdf\";\nimport type { PdfAnalyzeOptions } from \"@dragon708/docmind-pdf\";\nimport type { AnalysisResult, DetectFileKindInput } from \"@dragon708/docmind-shared\";\nimport type { NodeAnalyzeOptions } from \"../nodeAnalyzeOptions.js\";\nimport { bytesFromDetectInput } from \"../inputBytes.js\";\n\n/**\n * PDF → `@dragon708/docmind-pdf` (Node / pdf-parse + OCR).\n *\n * Unlike `analyzePdf` from `@dragon708/docmind-pdf` (OCR off unless set), `analyzeFile` defaults\n * to `pdf.ocr: \"auto\"`: when the PDF has\n * pages but almost no extractable text (typical scan), the raster OCR pipeline runs. Pass\n * `pdf: { ocr: \"off\" }` to skip OCR for speed.\n */\nexport async function analyzePdfForNode(\n  input: DetectFileKindInput,\n  options?: NodeAnalyzeOptions,\n): Promise<AnalysisResult> {\n  if (options?.signal?.aborted) {\n    const err = new Error(\"The operation was aborted\");\n    err.name = \"AbortError\";\n    throw err;\n  }\n\n  const data = await bytesFromDetectInput(input);\n  if (data.byteLength === 0) {\n    return {\n      fileKind: \"pdf\",\n      analyzer: \"pdf\",\n      status: \"ok\",\n      kind: \"pdf\",\n      text: \"\",\n      pages: 0,\n      metadata: { info: {} },\n      warnings: [\"No document bytes were provided for analysis.\"],\n      needsOCR: false,\n      ocrUsed: false,\n    };\n  }\n\n  const userPdf = options?.pdf;\n  const pdfOpts: PdfAnalyzeOptions = {\n    ...userPdf,\n    ocr: userPdf?.ocr ?? \"auto\",\n    ocrLangs: userPdf?.ocrLangs ?? options?.ocr?.langs,\n    signal: userPdf?.signal ?? options?.signal,\n  };\n\n  const r = await runPdf(data, pdfOpts);\n  return {\n    fileKind: \"pdf\",\n    analyzer: \"pdf\",\n    status: \"ok\",\n    kind: \"pdf\",\n    text: r.text,\n    pages: r.pages,\n    metadata: r.metadata,\n    warnings: [...r.warnings],\n    needsOCR: r.needsOCR,\n    ocrUsed: r.ocrUsed,\n  };\n}\n","import type { DetectFileKindInput, NamedInput } from \"@dragon708/docmind-shared\";\nimport { readFile } from \"node:fs/promises\";\nimport { basename } from \"node:path\";\nimport { fileURLToPath } from \"node:url\";\n\n/**\n * Inputs accepted by {@link analyzeFile} in this package.\n * Paths and `file:` URLs are read with `fs`; other values pass through as {@link DetectFileKindInput}.\n */\nexport type NodeAnalyzeInput = string | URL | DetectFileKindInput;\n\nfunction toPathString(pathOrUrl: string | URL): string {\n  return pathOrUrl instanceof URL ? fileURLToPath(pathOrUrl) : pathOrUrl;\n}\n\n/**\n * Reads a file from disk into a {@link NamedInput} (binary `Buffer`, basename as `name` for hints).\n */\nexport async function readFileToInput(path: string | URL): Promise<NamedInput<Buffer>> {\n  const fsPath = toPathString(path);\n  const data = await readFile(fsPath);\n  return {\n    data,\n    name: basename(fsPath),\n  };\n}\n\n/** Wraps a `Buffer` as a named payload when you already know the filename. */\nexport function bufferToInput(buffer: Buffer, name?: string): NamedInput<Buffer> {\n  return name !== undefined ? { data: buffer, name } : { data: buffer };\n}\n\n/**\n * Resolves paths / `file:` URLs to a {@link DetectFileKindInput}; leaves other values untouched.\n */\nexport async function resolveNodeAnalyzeInput(input: NodeAnalyzeInput): Promise<DetectFileKindInput> {\n  if (typeof input === \"string\" || input instanceof URL) {\n    return readFileToInput(input);\n  }\n  return input;\n}\n","import type { AnalysisResult, DetectFileKindInput } from \"@dragon708/docmind-shared\";\nimport {\n  analyzeText,\n  assertValidAnalyzeFileInput,\n  detectFileKind,\n  notImplementedResult,\n  UNKNOWN_FORMAT_WARNING,\n} from \"@dragon708/docmind-shared\";\nimport { analyzeDocxForNode } from \"./analyzers/docx.js\";\nimport { analyzeImageForNode } from \"./analyzers/image.js\";\nimport { analyzePdfForNode } from \"./analyzers/pdf.js\";\nimport type { NodeAnalyzeOptions } from \"./nodeAnalyzeOptions.js\";\nimport { resolveNodeAnalyzeInput, type NodeAnalyzeInput } from \"./resolveNodeInput.js\";\n\n/**\n * Node router: PDF, DOCX, images (OCR), and text. Paths and `file:` URLs are read via `fs`.\n */\nexport async function analyzeFile(\n  input: NodeAnalyzeInput,\n  options?: NodeAnalyzeOptions,\n): Promise<AnalysisResult> {\n  if (options?.signal?.aborted) {\n    const err = new Error(\"The operation was aborted\");\n    err.name = \"AbortError\";\n    throw err;\n  }\n\n  const resolved = await resolveNodeAnalyzeInput(input);\n  assertValidAnalyzeFileInput(resolved);\n\n  const fileKind = detectFileKind(resolved);\n\n  switch (fileKind) {\n    case \"pdf\":\n      return analyzePdfForNode(resolved as DetectFileKindInput, options);\n    case \"docx\":\n      return analyzeDocxForNode(resolved as DetectFileKindInput, options?.signal);\n    case \"image\":\n      return analyzeImageForNode(resolved as DetectFileKindInput, options);\n    case \"text\":\n      return analyzeText(resolved as DetectFileKindInput, { signal: options?.signal });\n    default:\n      return notImplementedResult(fileKind, \"none\", [UNKNOWN_FORMAT_WARNING]);\n  }\n}\n","import type { AnalysisResult, DetectFileKindInput } from \"@dragon708/docmind-shared\";\nimport {\n  analyzeText,\n  assertValidAnalyzeFileInput,\n  detectFileKind,\n  notImplementedResult,\n  UNKNOWN_FORMAT_WARNING,\n} from \"@dragon708/docmind-shared\";\nimport {\n  analyzePdf,\n  extractPdfMetadata,\n  extractTextFromPdf,\n} from \"@dragon708/docmind-pdf\";\nimport {\n  analyzeDocx,\n  convertDocxToHtml,\n  extractTextFromDocx,\n} from \"@dragon708/docmind-docx\";\nimport { ocr } from \"@dragon708/docmind-ocr\";\nimport { bytesFromDetectInput } from \"./inputBytes.js\";\nimport type { NodeAnalyzeOptions } from \"./nodeAnalyzeOptions.js\";\nimport { resolveNodeAnalyzeInput, type NodeAnalyzeInput } from \"./resolveNodeInput.js\";\n\nfunction throwIfAborted(signal?: AbortSignal): void {\n  if (signal?.aborted) {\n    const err = new Error(\"The operation was aborted\");\n    err.name = \"AbortError\";\n    throw err;\n  }\n}\n\nasync function prepare(input: NodeAnalyzeInput): Promise<DetectFileKindInput> {\n  const resolved = await resolveNodeAnalyzeInput(input);\n  assertValidAnalyzeFileInput(resolved);\n  return resolved;\n}\n\nfunction escapeHtmlMinimal(s: string): string {\n  return s\n    .replace(/&/g, \"&amp;\")\n    .replace(/</g, \"&lt;\")\n    .replace(/>/g, \"&gt;\")\n    .replace(/\"/g, \"&quot;\");\n}\n\nconst DOCX_METADATA_STUB =\n  \"Structured document metadata for DOCX is not exposed as a separate API; use extractText or analyzeFile.\";\n\nconst IMAGE_METADATA_NOTE =\n  \"Raster images have no document metadata bundle in this API.\";\n\n/**\n * Text only: PDF → `extractTextFromPdf` (capa de texto, sin OCR); DOCX → `extractTextFromDocx`;\n * imagen → `ocr`; texto → `analyzeText`.\n */\nexport async function extractText(\n  input: NodeAnalyzeInput,\n  options?: NodeAnalyzeOptions,\n): Promise<AnalysisResult> {\n  throwIfAborted(options?.signal);\n  const resolved = await prepare(input);\n  const kind = detectFileKind(resolved);\n  const signal = options?.signal;\n\n  switch (kind) {\n    case \"pdf\": {\n      const data = await bytesFromDetectInput(resolved);\n      if (data.byteLength === 0) {\n        return {\n          fileKind: \"pdf\",\n          analyzer: \"pdf\",\n          status: \"ok\",\n          kind: \"pdf\",\n          text: \"\",\n          pages: 0,\n          metadata: { info: {} },\n          warnings: [\"No document bytes were provided for analysis.\"],\n          needsOCR: false,\n          ocrUsed: false,\n        };\n      }\n      const r = await extractTextFromPdf(data);\n      return {\n        fileKind: \"pdf\",\n        analyzer: \"pdf\",\n        status: \"ok\",\n        kind: \"pdf\",\n        text: r.text,\n        pages: r.pages,\n        metadata: { info: {} },\n        warnings: r.warnings,\n        needsOCR: false,\n        ocrUsed: false,\n      };\n    }\n    case \"docx\": {\n      const data = await bytesFromDetectInput(resolved);\n      if (data.byteLength === 0) {\n        return {\n          fileKind: \"docx\",\n          analyzer: \"docx\",\n          status: \"ok\",\n          kind: \"docx\",\n          text: \"\",\n          html: \"\",\n          warnings: [\"No document bytes were provided for analysis.\"],\n        };\n      }\n      const r = await extractTextFromDocx(data);\n      return {\n        fileKind: \"docx\",\n        analyzer: \"docx\",\n        status: \"ok\",\n        kind: \"docx\",\n        text: r.text,\n        html: \"\",\n        warnings: r.warnings,\n      };\n    }\n    case \"image\": {\n      const data = await bytesFromDetectInput(resolved);\n      if (data.byteLength === 0) {\n        return {\n          fileKind: \"image\",\n          analyzer: \"image\",\n          status: \"ok\",\n          kind: \"image\",\n          text: \"\",\n          confidence: 0,\n          ocrUsed: true,\n          warnings: [\"No image bytes were provided for analysis.\"],\n        };\n      }\n      const ocrOpts = {\n        ...(options?.ocr ?? {}),\n        signal: options?.ocr?.signal ?? signal,\n      };\n      const r = await ocr(data, ocrOpts);\n      return {\n        fileKind: \"image\",\n        analyzer: \"image\",\n        status: \"ok\",\n        kind: \"image\",\n        text: r.text,\n        confidence: r.confidence,\n        ocrUsed: r.ocrUsed,\n        warnings: [],\n      };\n    }\n    case \"text\":\n      return analyzeText(resolved as DetectFileKindInput, { signal });\n    default:\n      return notImplementedResult(kind, \"none\", [UNKNOWN_FORMAT_WARNING]);\n  }\n}\n\n/**\n * Metadatos: PDF → `extractPdfMetadata`; resto con mejor esfuerzo o aviso.\n * El resultado sigue siendo `AnalysisResult` (PDF rellena `metadata` en forma `PdfAnalysisCoreResult`).\n */\nexport async function extractMetadata(\n  input: NodeAnalyzeInput,\n  options?: NodeAnalyzeOptions,\n): Promise<AnalysisResult> {\n  throwIfAborted(options?.signal);\n  const resolved = await prepare(input);\n  const kind = detectFileKind(resolved);\n  const signal = options?.signal;\n\n  switch (kind) {\n    case \"pdf\": {\n      const data = await bytesFromDetectInput(resolved);\n      if (data.byteLength === 0) {\n        return {\n          fileKind: \"pdf\",\n          analyzer: \"pdf\",\n          status: \"ok\",\n          kind: \"pdf\",\n          text: \"\",\n          pages: 0,\n          metadata: { info: {} },\n          warnings: [\"No document bytes were provided for analysis.\"],\n          needsOCR: false,\n          ocrUsed: false,\n        };\n      }\n      const r = await extractPdfMetadata(data);\n      return {\n        fileKind: \"pdf\",\n        analyzer: \"pdf\",\n        status: \"ok\",\n        kind: \"pdf\",\n        text: \"\",\n        pages: 0,\n        metadata: r.metadata,\n        warnings: r.warnings,\n        needsOCR: false,\n        ocrUsed: false,\n      };\n    }\n    case \"docx\":\n      return {\n        fileKind: \"docx\",\n        analyzer: \"docx\",\n        status: \"ok\",\n        kind: \"docx\",\n        text: \"\",\n        html: \"\",\n        warnings: [DOCX_METADATA_STUB],\n      };\n    case \"image\":\n      return {\n        fileKind: \"image\",\n        analyzer: \"image\",\n        status: \"ok\",\n        kind: \"image\",\n        text: \"\",\n        confidence: 0,\n        ocrUsed: true,\n        warnings: [IMAGE_METADATA_NOTE],\n      };\n    case \"text\":\n      return analyzeText(resolved as DetectFileKindInput, { signal });\n    default:\n      return notImplementedResult(kind, \"none\", [UNKNOWN_FORMAT_WARNING]);\n  }\n}\n\n/**\n * HTML: DOCX → `convertDocxToHtml`; PDF/texto → `<pre>` a partir de texto extraído;\n * imágenes → vacío con aviso.\n */\nexport async function convertToHtml(\n  input: NodeAnalyzeInput,\n  options?: NodeAnalyzeOptions,\n): Promise<AnalysisResult> {\n  throwIfAborted(options?.signal);\n  const resolved = await prepare(input);\n  const kind = detectFileKind(resolved);\n  const signal = options?.signal;\n\n  switch (kind) {\n    case \"docx\": {\n      const data = await bytesFromDetectInput(resolved);\n      if (data.byteLength === 0) {\n        return {\n          fileKind: \"docx\",\n          analyzer: \"docx\",\n          status: \"ok\",\n          kind: \"docx\",\n          text: \"\",\n          html: \"\",\n          warnings: [\"No document bytes were provided for analysis.\"],\n        };\n      }\n      const [textPart, htmlPart] = await Promise.all([\n        extractTextFromDocx(data),\n        convertDocxToHtml(data),\n      ]);\n      return {\n        fileKind: \"docx\",\n        analyzer: \"docx\",\n        status: \"ok\",\n        kind: \"docx\",\n        text: textPart.text,\n        html: htmlPart.html,\n        warnings: [...textPart.warnings, ...htmlPart.warnings],\n      };\n    }\n    case \"pdf\": {\n      const data = await bytesFromDetectInput(resolved);\n      if (data.byteLength === 0) {\n        return {\n          fileKind: \"pdf\",\n          analyzer: \"pdf\",\n          status: \"ok\",\n          kind: \"pdf\",\n          text: \"\",\n          pages: 0,\n          metadata: { info: {} },\n          warnings: [\"No document bytes were provided for analysis.\"],\n          needsOCR: false,\n          ocrUsed: false,\n        };\n      }\n      const r = await extractTextFromPdf(data);\n      const html = `<pre>${escapeHtmlMinimal(r.text)}</pre>`;\n      return {\n        fileKind: \"pdf\",\n        analyzer: \"pdf\",\n        status: \"ok\",\n        kind: \"pdf\",\n        text: r.text,\n        pages: r.pages,\n        metadata: { info: {} },\n        html,\n        warnings: [\n          ...r.warnings,\n          \"PDF HTML is a plain-text preview wrapped in <pre> (not a visual layout).\",\n        ],\n        needsOCR: false,\n        ocrUsed: false,\n      } as AnalysisResult;\n    }\n    case \"text\": {\n      const t = await analyzeText(resolved as DetectFileKindInput, { signal });\n      const html = `<pre>${escapeHtmlMinimal(t.text)}</pre>`;\n      return {\n        ...t,\n        html,\n        warnings: [\n          ...t.warnings,\n          \"HTML for plain text is a <pre> wrapper around decoded UTF-8 content.\",\n        ],\n      } as AnalysisResult;\n    }\n    case \"image\":\n      return {\n        fileKind: \"image\",\n        analyzer: \"image\",\n        status: \"ok\",\n        kind: \"image\",\n        text: \"\",\n        confidence: 0,\n        ocrUsed: true,\n        warnings: [\"No HTML representation for raster images; use extractText / runOcr.\"],\n      };\n    default:\n      return notImplementedResult(kind, \"none\", [UNKNOWN_FORMAT_WARNING]);\n  }\n}\n\n/**\n * OCR: PDF → `analyzePdf` con `ocr: \"force\"`; imagen → `ocr`; DOCX → texto estructurado con aviso\n * (sin OCR); texto → `analyzeText`.\n */\nexport async function runOcr(\n  input: NodeAnalyzeInput,\n  options?: NodeAnalyzeOptions,\n): Promise<AnalysisResult> {\n  throwIfAborted(options?.signal);\n  const resolved = await prepare(input);\n  const kind = detectFileKind(resolved);\n  const signal = options?.signal;\n  const lang = options?.ocr?.langs ?? options?.pdf?.ocrLangs;\n\n  switch (kind) {\n    case \"pdf\": {\n      const data = await bytesFromDetectInput(resolved);\n      if (data.byteLength === 0) {\n        return {\n          fileKind: \"pdf\",\n          analyzer: \"pdf\",\n          status: \"ok\",\n          kind: \"pdf\",\n          text: \"\",\n          pages: 0,\n          metadata: { info: {} },\n          warnings: [\"No document bytes were provided for analysis.\"],\n          needsOCR: false,\n          ocrUsed: false,\n        };\n      }\n      const r = await analyzePdf(data, {\n        ...options?.pdf,\n        ocr: \"force\",\n        ocrLangs: lang ?? options?.pdf?.ocrLangs,\n        signal: options?.pdf?.signal ?? signal,\n      });\n      return {\n        fileKind: \"pdf\",\n        analyzer: \"pdf\",\n        status: \"ok\",\n        kind: \"pdf\",\n        text: r.text,\n        pages: r.pages,\n        metadata: r.metadata,\n        warnings: [...r.warnings],\n        needsOCR: r.needsOCR,\n        ocrUsed: r.ocrUsed,\n      };\n    }\n    case \"image\": {\n      const data = await bytesFromDetectInput(resolved);\n      if (data.byteLength === 0) {\n        return {\n          fileKind: \"image\",\n          analyzer: \"image\",\n          status: \"ok\",\n          kind: \"image\",\n          text: \"\",\n          confidence: 0,\n          ocrUsed: true,\n          warnings: [\"No image bytes were provided for analysis.\"],\n        };\n      }\n      const ocrOpts = {\n        ...(options?.ocr ?? {}),\n        signal: options?.ocr?.signal ?? signal,\n      };\n      const r = await ocr(data, ocrOpts);\n      return {\n        fileKind: \"image\",\n        analyzer: \"image\",\n        status: \"ok\",\n        kind: \"image\",\n        text: r.text,\n        confidence: r.confidence,\n        ocrUsed: r.ocrUsed,\n        warnings: [],\n      };\n    }\n    case \"docx\": {\n      const data = await bytesFromDetectInput(resolved);\n      if (data.byteLength === 0) {\n        return {\n          fileKind: \"docx\",\n          analyzer: \"docx\",\n          status: \"ok\",\n          kind: \"docx\",\n          text: \"\",\n          html: \"\",\n          warnings: [\"No document bytes were provided for analysis.\"],\n        };\n      }\n      const r = await analyzeDocx(data);\n      return {\n        fileKind: \"docx\",\n        analyzer: \"docx\",\n        status: \"ok\",\n        kind: \"docx\",\n        text: r.text,\n        html: r.html,\n        warnings: [\n          ...r.warnings,\n          \"OCR does not apply to DOCX; returned structured text/HTML extract.\",\n        ],\n      };\n    }\n    case \"text\":\n      return analyzeText(resolved as DetectFileKindInput, { signal });\n    default:\n      return notImplementedResult(kind, \"none\", [UNKNOWN_FORMAT_WARNING]);\n  }\n}\n"]}

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@dragon708/docmind-node",
-  "version": "1.0.0",
+  "version": "1.1.0",
   "description": "Node.js DocMind entry: PDF, DOCX, OCR, text, and fs helpers.",
   "type": "module",
   "main": "./dist/index.js",