@dragon708/docmind-browser 0.1.0-alpha.1 → 1.1.0

package/dist/index.d.ts

@@ -1,14 +1,43 @@
-import { AnalyzeOptions, AnalysisResult } from '@dragon708/docmind-core';
-export { AnalysisAnalyzer, AnalysisResult, AnalyzeOptions, DetectFileKindInput, FileKind, FileKindMetadata, GenericAnalysisResult, TextAnalysisResult } from '@dragon708/docmind-core';
+import { AnalysisResult } 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';
 
+/** Options for {@link analyzeFile} in the browser entry (no PDF pipeline). */
+interface BrowserAnalyzeOptions {
+    readonly signal?: AbortSignal;
+    readonly ocr?: OcrOptions;
+}
+
+/** PDF is not processed in the browser; use `@dragon708/docmind-node` on the server. */
+declare const BROWSER_PDF_UNSUPPORTED_WARNING = "PDF text extraction is not available in the browser runtime; use @dragon708/docmind-node on the server.";
 /**
  * Inputs supported by the browser entry (DOM types only — no `fs`, no Node `Buffer` in the public surface).
- * For richer hints (`NamedInput`, etc.), import `analyzeFile` from the `core` package directly.
  */
 type BrowserAnalyzeInput = File | Blob | ArrayBuffer;
 /**
- * Thin wrapper around `@dragon708/docmind-core`’s `analyzeFile` for browser runtimes.
+ * Browser-only router: DOCX, images (OCR), and text. PDF yields `not_implemented` with a clear warning.
+ */
+declare function analyzeFile(input: BrowserAnalyzeInput, options?: BrowserAnalyzeOptions): Promise<AnalysisResult>;
+
+/**
+ * Text only: DOCX → `extractTextFromDocx`; imagen → `ocr`; texto → `analyzeText`.
+ * PDF no está soportado en el navegador (mismo aviso que `analyzeFile`).
+ */
+declare function extractText(input: BrowserAnalyzeInput, options?: BrowserAnalyzeOptions): Promise<AnalysisResult>;
+/**
+ * Metadatos: en el navegador no hay pipeline PDF ni metadatos DOCX dedicados;
+ * DOCX/imagen con avisos; texto → `analyzeText`.
+ */
+declare function extractMetadata(input: BrowserAnalyzeInput, options?: BrowserAnalyzeOptions): Promise<AnalysisResult>;
+/**
+ * HTML: DOCX → `extractTextFromDocx` + `convertDocxToHtml`; texto → `<pre>`;
+ * PDF/imagen no aplican en browser como HTML rico.
+ */
+declare function convertToHtml(input: BrowserAnalyzeInput, options?: BrowserAnalyzeOptions): Promise<AnalysisResult>;
+/**
+ * OCR: imagen → `ocr`; DOCX → `analyzeDocx` con aviso (sin OCR); texto → `analyzeText`.
+ * PDF no soportado en browser.
  */
-declare function analyzeFile(input: BrowserAnalyzeInput, options?: AnalyzeOptions): Promise<AnalysisResult>;
+declare function runOcr(input: BrowserAnalyzeInput, options?: BrowserAnalyzeOptions): Promise<AnalysisResult>;
 
-export { type BrowserAnalyzeInput, analyzeFile };
+export { BROWSER_PDF_UNSUPPORTED_WARNING, type BrowserAnalyzeInput, type BrowserAnalyzeOptions, analyzeFile, convertToHtml, extractMetadata, extractText, runOcr };

package/dist/index.js

@@ -1,6 +1,76 @@
-import { analyzeFile as analyzeFile$1, InvalidInputError } from '@dragon708/docmind-core';
+import { assertValidAnalyzeFileInput, detectFileKind, notImplementedResult, UNKNOWN_FORMAT_WARNING, analyzeText, toUint8Array, InvalidInputError } from '@dragon708/docmind-shared';
+export { detectFileKind } from '@dragon708/docmind-shared';
+import { extractTextFromDocx, convertDocxToHtml, analyzeDocx } from '@dragon708/docmind-docx';
+import { ocr } from '@dragon708/docmind-ocr';
 
 // src/analyzeFile.ts
+async function analyzeDocxForBrowser(input, signal) {
+  if (signal?.aborted) {
+    const err = new Error("The operation was aborted");
+    err.name = "AbortError";
+    throw err;
+  }
+  const data = await toUint8Array(input);
+  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]
+  };
+}
+async function analyzeImageForBrowser(input, options) {
+  if (options?.signal?.aborted) {
+    const err = new Error("The operation was aborted");
+    err.name = "AbortError";
+    throw err;
+  }
+  const data = await toUint8Array(input);
+  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 ?? options?.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: []
+  };
+}
+
+// src/analyzeFile.ts
+var BROWSER_PDF_UNSUPPORTED_WARNING = "PDF text extraction is not available in the browser runtime; use @dragon708/docmind-node on the server.";
 function assertBrowserInput(input) {
   const ok = input instanceof File || input instanceof Blob || input instanceof ArrayBuffer;
   if (!ok) {
@@ -8,10 +78,292 @@ function assertBrowserInput(input) {
   }
 }
 async function analyzeFile(input, options) {
+  if (options?.signal?.aborted) {
+    const err = new Error("The operation was aborted");
+    err.name = "AbortError";
+    throw err;
+  }
   assertBrowserInput(input);
-  return analyzeFile$1(input, options);
+  assertValidAnalyzeFileInput(input);
+  const fileKind = detectFileKind(input);
+  const bytesInput = input;
+  switch (fileKind) {
+    case "pdf":
+      return notImplementedResult("pdf", "pdf", [BROWSER_PDF_UNSUPPORTED_WARNING]);
+    case "docx":
+      return analyzeDocxForBrowser(bytesInput, options?.signal);
+    case "image":
+      return analyzeImageForBrowser(bytesInput, options);
+    case "text":
+      return analyzeText(bytesInput, { signal: options?.signal });
+    default:
+      return notImplementedResult(fileKind, "none", [UNKNOWN_FORMAT_WARNING]);
+  }
+}
+function assertBrowserInput2(input) {
+  const ok = input instanceof File || input instanceof Blob || input instanceof ArrayBuffer;
+  if (!ok) {
+    throw new InvalidInputError("Expected a File, Blob, or ArrayBuffer.");
+  }
+}
+function throwIfAborted(signal) {
+  if (signal?.aborted) {
+    const err = new Error("The operation was aborted");
+    err.name = "AbortError";
+    throw err;
+  }
+}
+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 kindOf(input) {
+  assertBrowserInput2(input);
+  assertValidAnalyzeFileInput(input);
+  return input;
+}
+async function extractText(input, options) {
+  throwIfAborted(options?.signal);
+  const resolved = await kindOf(input);
+  const kind = detectFileKind(resolved);
+  const bytesInput = input;
+  const signal = options?.signal;
+  switch (kind) {
+    case "pdf":
+      return notImplementedResult("pdf", "pdf", [BROWSER_PDF_UNSUPPORTED_WARNING]);
+    case "docx": {
+      const data = await toUint8Array(bytesInput);
+      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 toUint8Array(bytesInput);
+      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(bytesInput, { signal });
+    default:
+      return notImplementedResult(kind, "none", [UNKNOWN_FORMAT_WARNING]);
+  }
+}
+async function extractMetadata(input, options) {
+  throwIfAborted(options?.signal);
+  const resolved = await kindOf(input);
+  const kind = detectFileKind(resolved);
+  const bytesInput = input;
+  const signal = options?.signal;
+  switch (kind) {
+    case "pdf":
+      return notImplementedResult("pdf", "pdf", [BROWSER_PDF_UNSUPPORTED_WARNING]);
+    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(bytesInput, { signal });
+    default:
+      return notImplementedResult(kind, "none", [UNKNOWN_FORMAT_WARNING]);
+  }
+}
+async function convertToHtml(input, options) {
+  throwIfAborted(options?.signal);
+  const resolved = await kindOf(input);
+  const kind = detectFileKind(resolved);
+  const bytesInput = input;
+  const signal = options?.signal;
+  switch (kind) {
+    case "pdf":
+      return notImplementedResult("pdf", "pdf", [BROWSER_PDF_UNSUPPORTED_WARNING]);
+    case "docx": {
+      const data = await toUint8Array(bytesInput);
+      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 "text": {
+      const t = await analyzeText(bytesInput, { 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 kindOf(input);
+  const kind = detectFileKind(resolved);
+  const bytesInput = input;
+  const signal = options?.signal;
+  switch (kind) {
+    case "pdf":
+      return notImplementedResult("pdf", "pdf", [BROWSER_PDF_UNSUPPORTED_WARNING]);
+    case "image": {
+      const data = await toUint8Array(bytesInput);
+      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 toUint8Array(bytesInput);
+      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(bytesInput, { signal });
+    default:
+      return notImplementedResult(kind, "none", [UNKNOWN_FORMAT_WARNING]);
+  }
 }
 
-export { analyzeFile };
+export { BROWSER_PDF_UNSUPPORTED_WARNING, analyzeFile, convertToHtml, extractMetadata, extractText, runOcr };
 //# sourceMappingURL=index.js.map
 //# sourceMappingURL=index.js.map

package/dist/index.js.map

@@ -1 +1 @@
-{"version":3,"sources":["../src/analyzeFile.ts"],"names":["analyzeWithCore"],"mappings":";;;AASA,SAAS,mBAAmB,KAAA,EAAsD;AAChF,EAAA,MAAM,EAAA,GACJ,KAAA,YAAiB,IAAA,IACjB,KAAA,YAAiB,QACjB,KAAA,YAAiB,WAAA;AACnB,EAAA,IAAI,CAAC,EAAA,EAAI;AACP,IAAA,MAAM,IAAI,kBAAkB,wCAAwC,CAAA;AAAA,EACtE;AACF;AAKA,eAAsB,WAAA,CACpB,OACA,OAAA,EACyB;AACzB,EAAA,kBAAA,CAAmB,KAAK,CAAA;AACxB,EAAA,OAAOA,aAAA,CAAgB,OAAO,OAAO,CAAA;AACvC","file":"index.js","sourcesContent":["import { analyzeFile as analyzeWithCore, InvalidInputError } from \"@dragon708/docmind-core\";\r\nimport type { AnalyzeOptions, AnalysisResult } from \"@dragon708/docmind-core\";\r\n\r\n/**\r\n * Inputs supported by the browser entry (DOM types only — no `fs`, no Node `Buffer` in the public surface).\r\n * For richer hints (`NamedInput`, etc.), import `analyzeFile` from the `core` package directly.\r\n */\r\nexport type BrowserAnalyzeInput = File | Blob | ArrayBuffer;\r\n\r\nfunction assertBrowserInput(input: unknown): asserts input is BrowserAnalyzeInput {\r\n  const ok =\r\n    input instanceof File ||\r\n    input instanceof Blob ||\r\n    input instanceof ArrayBuffer;\r\n  if (!ok) {\r\n    throw new InvalidInputError(\"Expected a File, Blob, or ArrayBuffer.\");\r\n  }\r\n}\r\n\r\n/**\r\n * Thin wrapper around `@dragon708/docmind-core`’s `analyzeFile` for browser runtimes.\r\n */\r\nexport async function analyzeFile(\r\n  input: BrowserAnalyzeInput,\r\n  options?: AnalyzeOptions,\r\n): Promise<AnalysisResult> {\r\n  assertBrowserInput(input);\r\n  return analyzeWithCore(input, options);\r\n}\r\n"]}
+{"version":3,"sources":["../src/analyzers/docx.ts","../src/analyzers/image.ts","../src/analyzeFile.ts","../src/publicActions.ts"],"names":["extractDocx","toUint8Array","assertBrowserInput","InvalidInputError","assertValidAnalyzeFileInput","detectFileKind","notImplementedResult","ocr","analyzeText","UNKNOWN_FORMAT_WARNING"],"mappings":";;;;;;AAOA,eAAsB,qBAAA,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,YAAA,CAAa,KAAK,CAAA;AACrC,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,sBAAA,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,MAAMC,YAAAA,CAAa,KAAK,CAAA;AACrC,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;;;ACjCO,IAAM,+BAAA,GACX;AAOF,SAAS,mBAAmB,KAAA,EAAsD;AAChF,EAAA,MAAM,EAAA,GACJ,KAAA,YAAiB,IAAA,IACjB,KAAA,YAAiB,QACjB,KAAA,YAAiB,WAAA;AACnB,EAAA,IAAI,CAAC,EAAA,EAAI;AACP,IAAA,MAAM,IAAI,kBAAkB,wCAAwC,CAAA;AAAA,EACtE;AACF;AAKA,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,kBAAA,CAAmB,KAAK,CAAA;AACxB,EAAA,2BAAA,CAA4B,KAAK,CAAA;AAEjC,EAAA,MAAM,QAAA,GAAW,eAAe,KAA4B,CAAA;AAE5D,EAAA,MAAM,UAAA,GAAa,KAAA;AAEnB,EAAA,QAAQ,QAAA;AAAU,IAChB,KAAK,KAAA;AACH,MAAA,OAAO,oBAAA,CAAqB,KAAA,EAAO,KAAA,EAAO,CAAC,+BAA+B,CAAC,CAAA;AAAA,IAC7E,KAAK,MAAA;AACH,MAAA,OAAO,qBAAA,CAAsB,UAAA,EAAY,OAAA,EAAS,MAAM,CAAA;AAAA,IAC1D,KAAK,OAAA;AACH,MAAA,OAAO,sBAAA,CAAuB,YAAY,OAAO,CAAA;AAAA,IACnD,KAAK,MAAA;AACH,MAAA,OAAO,YAAY,UAAA,EAAY,EAAE,MAAA,EAAQ,OAAA,EAAS,QAAQ,CAAA;AAAA,IAC5D;AACE,MAAA,OAAO,oBAAA,CAAqB,QAAA,EAAU,MAAA,EAAQ,CAAC,sBAAsB,CAAC,CAAA;AAAA;AAE5E;AC7CA,SAASC,oBAAmB,KAAA,EAAsD;AAChF,EAAA,MAAM,EAAA,GACJ,KAAA,YAAiB,IAAA,IACjB,KAAA,YAAiB,QACjB,KAAA,YAAiB,WAAA;AACnB,EAAA,IAAI,CAAC,EAAA,EAAI;AACP,IAAA,MAAM,IAAIC,kBAAkB,wCAAwC,CAAA;AAAA,EACtE;AACF;AAEA,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,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;AAEF,eAAe,OAAO,KAAA,EAA0D;AAC9E,EAAAD,oBAAmB,KAAK,CAAA;AACxB,EAAAE,4BAA4B,KAAK,CAAA;AACjC,EAAA,OAAO,KAAA;AACT;AAMA,eAAsB,WAAA,CACpB,OACA,OAAA,EACyB;AACzB,EAAA,cAAA,CAAe,SAAS,MAAM,CAAA;AAC9B,EAAA,MAAM,QAAA,GAAW,MAAM,MAAA,CAAO,KAAK,CAAA;AACnC,EAAA,MAAM,IAAA,GAAOC,eAAe,QAAQ,CAAA;AACpC,EAAA,MAAM,UAAA,GAAa,KAAA;AACnB,EAAA,MAAM,SAAS,OAAA,EAAS,MAAA;AAExB,EAAA,QAAQ,IAAA;AAAM,IACZ,KAAK,KAAA;AACH,MAAA,OAAOC,oBAAAA,CAAqB,KAAA,EAAO,KAAA,EAAO,CAAC,+BAA+B,CAAC,CAAA;AAAA,IAC7E,KAAK,MAAA,EAAQ;AACX,MAAA,MAAM,IAAA,GAAO,MAAML,YAAAA,CAAa,UAAU,CAAA;AAC1C,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,MAAMA,YAAAA,CAAa,UAAU,CAAA;AAC1C,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,MAAMM,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,UAAA,EAAY,EAAE,MAAA,EAAQ,CAAA;AAAA,IAC3C;AACE,MAAA,OAAOF,oBAAAA,CAAqB,IAAA,EAAM,MAAA,EAAQ,CAACG,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,MAAA,CAAO,KAAK,CAAA;AACnC,EAAA,MAAM,IAAA,GAAOJ,eAAe,QAAQ,CAAA;AACpC,EAAA,MAAM,UAAA,GAAa,KAAA;AACnB,EAAA,MAAM,SAAS,OAAA,EAAS,MAAA;AAExB,EAAA,QAAQ,IAAA;AAAM,IACZ,KAAK,KAAA;AACH,MAAA,OAAOC,oBAAAA,CAAqB,KAAA,EAAO,KAAA,EAAO,CAAC,+BAA+B,CAAC,CAAA;AAAA,IAC7E,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,UAAA,EAAY,EAAE,MAAA,EAAQ,CAAA;AAAA,IAC3C;AACE,MAAA,OAAOF,oBAAAA,CAAqB,IAAA,EAAM,MAAA,EAAQ,CAACG,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,MAAA,CAAO,KAAK,CAAA;AACnC,EAAA,MAAM,IAAA,GAAOJ,eAAe,QAAQ,CAAA;AACpC,EAAA,MAAM,UAAA,GAAa,KAAA;AACnB,EAAA,MAAM,SAAS,OAAA,EAAS,MAAA;AAExB,EAAA,QAAQ,IAAA;AAAM,IACZ,KAAK,KAAA;AACH,MAAA,OAAOC,oBAAAA,CAAqB,KAAA,EAAO,KAAA,EAAO,CAAC,+BAA+B,CAAC,CAAA;AAAA,IAC7E,KAAK,MAAA,EAAQ;AACX,MAAA,MAAM,IAAA,GAAO,MAAML,YAAAA,CAAa,UAAU,CAAA;AAC1C,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,MAAA,EAAQ;AACX,MAAA,MAAM,IAAI,MAAMO,WAAAA,CAAY,UAAA,EAAY,EAAE,QAAQ,CAAA;AAClD,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,OAAOF,oBAAAA,CAAqB,IAAA,EAAM,MAAA,EAAQ,CAACG,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,MAAA,CAAO,KAAK,CAAA;AACnC,EAAA,MAAM,IAAA,GAAOJ,eAAe,QAAQ,CAAA;AACpC,EAAA,MAAM,UAAA,GAAa,KAAA;AACnB,EAAA,MAAM,SAAS,OAAA,EAAS,MAAA;AAExB,EAAA,QAAQ,IAAA;AAAM,IACZ,KAAK,KAAA;AACH,MAAA,OAAOC,oBAAAA,CAAqB,KAAA,EAAO,KAAA,EAAO,CAAC,+BAA+B,CAAC,CAAA;AAAA,IAC7E,KAAK,OAAA,EAAS;AACZ,MAAA,MAAM,IAAA,GAAO,MAAML,YAAAA,CAAa,UAAU,CAAA;AAC1C,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,MAAMM,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,MAAMN,YAAAA,CAAa,UAAU,CAAA;AAC1C,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,OAAOO,WAAAA,CAAY,UAAA,EAAY,EAAE,MAAA,EAAQ,CAAA;AAAA,IAC3C;AACE,MAAA,OAAOF,oBAAAA,CAAqB,IAAA,EAAM,MAAA,EAAQ,CAACG,sBAAsB,CAAC,CAAA;AAAA;AAExE","file":"index.js","sourcesContent":["import { analyzeDocx as extractDocx } from \"@dragon708/docmind-docx\";\nimport type { AnalysisResult, FileLikeInput } from \"@dragon708/docmind-shared\";\nimport { toUint8Array } from \"@dragon708/docmind-shared\";\n\n/**\n * DOCX → `@dragon708/docmind-docx` (browser-safe: Mammoth + JSZip).\n */\nexport async function analyzeDocxForBrowser(\n  input: FileLikeInput,\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 toUint8Array(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, FileLikeInput } from \"@dragon708/docmind-shared\";\nimport { toUint8Array } from \"@dragon708/docmind-shared\";\nimport type { BrowserAnalyzeOptions } from \"../browserAnalyzeOptions.js\";\n\n/**\n * Image → `@dragon708/docmind-ocr` (Tesseract in WASM / browser).\n */\nexport async function analyzeImageForBrowser(\n  input: FileLikeInput,\n  options?: BrowserAnalyzeOptions,\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 toUint8Array(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 type { AnalysisResult, FileLikeInput } from \"@dragon708/docmind-shared\";\r\nimport {\r\n  analyzeText,\r\n  assertValidAnalyzeFileInput,\r\n  detectFileKind,\r\n  InvalidInputError,\r\n  notImplementedResult,\r\n  UNKNOWN_FORMAT_WARNING,\r\n} from \"@dragon708/docmind-shared\";\r\nimport type { DetectFileKindInput } from \"@dragon708/docmind-shared\";\r\nimport type { BrowserAnalyzeOptions } from \"./browserAnalyzeOptions.js\";\r\nimport { analyzeDocxForBrowser } from \"./analyzers/docx.js\";\r\nimport { analyzeImageForBrowser } from \"./analyzers/image.js\";\r\n\r\n/** PDF is not processed in the browser; use `@dragon708/docmind-node` on the server. */\r\nexport const BROWSER_PDF_UNSUPPORTED_WARNING =\r\n  \"PDF text extraction is not available in the browser runtime; use @dragon708/docmind-node on the server.\";\r\n\r\n/**\r\n * Inputs supported by the browser entry (DOM types only — no `fs`, no Node `Buffer` in the public surface).\r\n */\r\nexport type BrowserAnalyzeInput = File | Blob | ArrayBuffer;\r\n\r\nfunction assertBrowserInput(input: unknown): asserts input is BrowserAnalyzeInput {\r\n  const ok =\r\n    input instanceof File ||\r\n    input instanceof Blob ||\r\n    input instanceof ArrayBuffer;\r\n  if (!ok) {\r\n    throw new InvalidInputError(\"Expected a File, Blob, or ArrayBuffer.\");\r\n  }\r\n}\r\n\r\n/**\r\n * Browser-only router: DOCX, images (OCR), and text. PDF yields `not_implemented` with a clear warning.\r\n */\r\nexport async function analyzeFile(\r\n  input: BrowserAnalyzeInput,\r\n  options?: BrowserAnalyzeOptions,\r\n): Promise<AnalysisResult> {\r\n  if (options?.signal?.aborted) {\r\n    const err = new Error(\"The operation was aborted\");\r\n    err.name = \"AbortError\";\r\n    throw err;\r\n  }\r\n\r\n  assertBrowserInput(input);\r\n  assertValidAnalyzeFileInput(input);\r\n\r\n  const fileKind = detectFileKind(input as DetectFileKindInput);\r\n\r\n  const bytesInput = input as FileLikeInput;\r\n\r\n  switch (fileKind) {\r\n    case \"pdf\":\r\n      return notImplementedResult(\"pdf\", \"pdf\", [BROWSER_PDF_UNSUPPORTED_WARNING]);\r\n    case \"docx\":\r\n      return analyzeDocxForBrowser(bytesInput, options?.signal);\r\n    case \"image\":\r\n      return analyzeImageForBrowser(bytesInput, options);\r\n    case \"text\":\r\n      return analyzeText(bytesInput, { signal: options?.signal });\r\n    default:\r\n      return notImplementedResult(fileKind, \"none\", [UNKNOWN_FORMAT_WARNING]);\r\n  }\r\n}\r\n","import type { AnalysisResult, DetectFileKindInput, FileLikeInput } from \"@dragon708/docmind-shared\";\nimport {\n  analyzeText,\n  assertValidAnalyzeFileInput,\n  detectFileKind,\n  InvalidInputError,\n  notImplementedResult,\n  UNKNOWN_FORMAT_WARNING,\n  toUint8Array,\n} from \"@dragon708/docmind-shared\";\nimport {\n  analyzeDocx,\n  convertDocxToHtml,\n  extractTextFromDocx,\n} from \"@dragon708/docmind-docx\";\nimport { ocr } from \"@dragon708/docmind-ocr\";\nimport type { BrowserAnalyzeOptions } from \"./browserAnalyzeOptions.js\";\nimport { BROWSER_PDF_UNSUPPORTED_WARNING } from \"./analyzeFile.js\";\nimport type { BrowserAnalyzeInput } from \"./analyzeFile.js\";\n\nfunction assertBrowserInput(input: unknown): asserts input is BrowserAnalyzeInput {\n  const ok =\n    input instanceof File ||\n    input instanceof Blob ||\n    input instanceof ArrayBuffer;\n  if (!ok) {\n    throw new InvalidInputError(\"Expected a File, Blob, or ArrayBuffer.\");\n  }\n}\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\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\nasync function kindOf(input: BrowserAnalyzeInput): Promise<DetectFileKindInput> {\n  assertBrowserInput(input);\n  assertValidAnalyzeFileInput(input);\n  return input as DetectFileKindInput;\n}\n\n/**\n * Text only: DOCX → `extractTextFromDocx`; imagen → `ocr`; texto → `analyzeText`.\n * PDF no está soportado en el navegador (mismo aviso que `analyzeFile`).\n */\nexport async function extractText(\n  input: BrowserAnalyzeInput,\n  options?: BrowserAnalyzeOptions,\n): Promise<AnalysisResult> {\n  throwIfAborted(options?.signal);\n  const resolved = await kindOf(input);\n  const kind = detectFileKind(resolved);\n  const bytesInput = input as FileLikeInput;\n  const signal = options?.signal;\n\n  switch (kind) {\n    case \"pdf\":\n      return notImplementedResult(\"pdf\", \"pdf\", [BROWSER_PDF_UNSUPPORTED_WARNING]);\n    case \"docx\": {\n      const data = await toUint8Array(bytesInput);\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 toUint8Array(bytesInput);\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(bytesInput, { signal });\n    default:\n      return notImplementedResult(kind, \"none\", [UNKNOWN_FORMAT_WARNING]);\n  }\n}\n\n/**\n * Metadatos: en el navegador no hay pipeline PDF ni metadatos DOCX dedicados;\n * DOCX/imagen con avisos; texto → `analyzeText`.\n */\nexport async function extractMetadata(\n  input: BrowserAnalyzeInput,\n  options?: BrowserAnalyzeOptions,\n): Promise<AnalysisResult> {\n  throwIfAborted(options?.signal);\n  const resolved = await kindOf(input);\n  const kind = detectFileKind(resolved);\n  const bytesInput = input as FileLikeInput;\n  const signal = options?.signal;\n\n  switch (kind) {\n    case \"pdf\":\n      return notImplementedResult(\"pdf\", \"pdf\", [BROWSER_PDF_UNSUPPORTED_WARNING]);\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(bytesInput, { signal });\n    default:\n      return notImplementedResult(kind, \"none\", [UNKNOWN_FORMAT_WARNING]);\n  }\n}\n\n/**\n * HTML: DOCX → `extractTextFromDocx` + `convertDocxToHtml`; texto → `<pre>`;\n * PDF/imagen no aplican en browser como HTML rico.\n */\nexport async function convertToHtml(\n  input: BrowserAnalyzeInput,\n  options?: BrowserAnalyzeOptions,\n): Promise<AnalysisResult> {\n  throwIfAborted(options?.signal);\n  const resolved = await kindOf(input);\n  const kind = detectFileKind(resolved);\n  const bytesInput = input as FileLikeInput;\n  const signal = options?.signal;\n\n  switch (kind) {\n    case \"pdf\":\n      return notImplementedResult(\"pdf\", \"pdf\", [BROWSER_PDF_UNSUPPORTED_WARNING]);\n    case \"docx\": {\n      const data = await toUint8Array(bytesInput);\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 \"text\": {\n      const t = await analyzeText(bytesInput, { 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: imagen → `ocr`; DOCX → `analyzeDocx` con aviso (sin OCR); texto → `analyzeText`.\n * PDF no soportado en browser.\n */\nexport async function runOcr(\n  input: BrowserAnalyzeInput,\n  options?: BrowserAnalyzeOptions,\n): Promise<AnalysisResult> {\n  throwIfAborted(options?.signal);\n  const resolved = await kindOf(input);\n  const kind = detectFileKind(resolved);\n  const bytesInput = input as FileLikeInput;\n  const signal = options?.signal;\n\n  switch (kind) {\n    case \"pdf\":\n      return notImplementedResult(\"pdf\", \"pdf\", [BROWSER_PDF_UNSUPPORTED_WARNING]);\n    case \"image\": {\n      const data = await toUint8Array(bytesInput);\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 toUint8Array(bytesInput);\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(bytesInput, { signal });\n    default:\n      return notImplementedResult(kind, \"none\", [UNKNOWN_FORMAT_WARNING]);\n  }\n}\n"]}

package/package.json

@@ -1,7 +1,7 @@
 {
   "name": "@dragon708/docmind-browser",
-  "version": "0.1.0-alpha.1",
-  "description": "Browser-safe DocMind entry (File, Blob, ArrayBuffer) delegating to core.",
+  "version": "1.1.0",
+  "description": "Browser DocMind entry: DOCX, OCR, text (no PDF; use @dragon708/docmind-node on the server).",
   "type": "module",
   "sideEffects": false,
   "main": "./dist/index.js",
@@ -32,11 +32,14 @@
   ],
   "license": "MIT",
   "dependencies": {
-    "@dragon708/docmind-core": "^0.1.0-alpha.1"
+    "@dragon708/docmind-docx": "^1.0.0",
+    "@dragon708/docmind-ocr": "^1.0.0",
+    "@dragon708/docmind-shared": "^1.0.0"
   },
   "devDependencies": {
     "@types/node": "^20.19.37",
     "tsup": "^8.5.1",
-    "typescript": "^5.9.3"
+    "typescript": "^5.9.3",
+    "vitest": "^1.6.1"
   }
 }