@dragon708/docmind-browser 0.1.0-alpha.1 → 1.0.0

package/dist/index.d.ts

@@ -1,14 +1,22 @@
-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 } 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?: AnalyzeOptions): Promise<AnalysisResult>;
+declare function analyzeFile(input: BrowserAnalyzeInput, options?: BrowserAnalyzeOptions): Promise<AnalysisResult>;
 
-export { type BrowserAnalyzeInput, analyzeFile };
+export { BROWSER_PDF_UNSUPPORTED_WARNING, type BrowserAnalyzeInput, type BrowserAnalyzeOptions, analyzeFile };

package/dist/index.js

@@ -1,6 +1,75 @@
-import { analyzeFile as analyzeFile$1, InvalidInputError } from '@dragon708/docmind-core';
+import { assertValidAnalyzeFileInput, detectFileKind, notImplementedResult, UNKNOWN_FORMAT_WARNING, analyzeText, InvalidInputError, toUint8Array } from '@dragon708/docmind-shared';
+import { 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 +77,29 @@ 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]);
+  }
 }
 
-export { analyzeFile };
+export { BROWSER_PDF_UNSUPPORTED_WARNING, analyzeFile };
 //# 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"],"names":["extractDocx","toUint8Array"],"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","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"]}

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": "1.0.0",
+  "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"
   }
 }