feat: add @docen/image

package.json

@@ -16,10 +16,7 @@
     "type": "git",
     "url": "git+https://github.com/docenjs/docen.git"
   },
-  "keywords": [
-    "pandoc",
-    "mammoth"
-  ],
+  "keywords": ["pandoc", "mammoth"],
   "author": {
     "name": "Demo Macro",
     "email": "[email protected]",

packages/csv/package.json

@@ -4,9 +4,7 @@
   "description": "Programmed implementation of csv format, powered by Demo Macro.",
   "main": "dist/index.mjs",
   "types": "dist/index.d.ts",
-  "files": [
-    "dist"
-  ],
+  "files": ["dist"],
   "exports": {
     ".": {
       "import": "./dist/index.mjs",
@@ -20,12 +18,7 @@
     "type": "git",
     "url": "git+https://github.com/docenjs/docen.git"
   },
-  "keywords": [
-    "pandoc",
-    "papaparse",
-    "csv",
-    "json"
-  ],
+  "keywords": ["pandoc", "papaparse", "csv", "json"],
   "author": {
     "name": "Demo Macro",
     "email": "[email protected]",

packages/csv/src/json.ts

@@ -19,7 +19,7 @@ export function convertCSVToJSON(source: string, options?: csvConvertOptions) {
     // const regex = /("[^"]*"|[^,]+)/g;
     const regex = new RegExp(
       `(${quotechar}[^${quotechar}]*${quotechar}|[^${delimiter}]+)`,
-      "g",
+      "g"
     );
 
     return row.match(regex) ?? [];
@@ -40,7 +40,7 @@ export function convertCSVToJSON(source: string, options?: csvConvertOptions) {
   } else if (options?.header) {
     header = Array.from(
       { length: maxColumns },
-      (_, i) => `${options?.header}${i + 1}`,
+      (_, i) => `${options?.header}${i + 1}`
     );
   }
 
@@ -52,7 +52,7 @@ export function convertCSVToJSON(source: string, options?: csvConvertOptions) {
 
   if (header.length < maxColumns) {
     header = header.concat(
-      Array.from({ length: maxColumns - header.length }, (_, i) => `${i + 1}`),
+      Array.from({ length: maxColumns - header.length }, (_, i) => `${i + 1}`)
     );
   }
 
@@ -80,7 +80,7 @@ export function convertCSVToJSON(source: string, options?: csvConvertOptions) {
 export function convertJSONToCSV(
   // biome-ignore lint/suspicious/noExplicitAny: <explanation>
   source: Record<string, any>[],
-  options?: csvConvertOptions,
+  options?: csvConvertOptions
 ) {
   const delimiter = options?.delimiter ?? ",";
 
@@ -103,7 +103,7 @@ export function convertJSONToCSV(
   } else if (options?.header) {
     header = Array.from(
       { length: maxColumns },
-      (_, i) => `${options?.header}${i + 1}`,
+      (_, i) => `${options?.header}${i + 1}`
     );
   }
 
@@ -115,7 +115,7 @@ export function convertJSONToCSV(
 
   if (header.length < maxColumns) {
     header = header.concat(
-      Array.from({ length: maxColumns - header.length }, (_, i) => `${i + 1}`),
+      Array.from({ length: maxColumns - header.length }, (_, i) => `${i + 1}`)
     );
   }
 

packages/docen/bundle.mjs

@@ -6,7 +6,7 @@ execSync("pnpm prepack", { stdio: "inherit" });
 execSync("pnpm ncc build dist/cli.cjs -o .cache/ncc -a", { stdio: "inherit" });
 
 const version = readFileSync("package.json", "utf-8").match(
-  /"version": "(.*?)"/,
+  /"version": "(.*?)"/
 )[1];
 
 execSync(`mv .cache/ncc/index.cjs .cache/ncc/docen-${version}.js`, {
@@ -17,5 +17,5 @@ execSync(
   `pnpm dlx pkg .cache/ncc/docen-${version}.js --out-path .cache/bundle -c pkg.config.json -C GZip`,
   {
     stdio: "inherit",
-  },
+  }
 );

packages/docen/package.json

@@ -4,9 +4,7 @@
   "description": "Programmatically and command-line implementation of document formatting, powered by Demo Macro.",
   "main": "dist/index.mjs",
   "types": "dist/index.d.ts",
-  "files": [
-    "dist"
-  ],
+  "files": ["dist"],
   "bin": {
     "docen": "dist/cli.mjs"
   },
@@ -24,10 +22,7 @@
     "type": "git",
     "url": "git+https://github.com/docenjs/docen.git"
   },
-  "keywords": [
-    "pandoc",
-    "mammoth"
-  ],
+  "keywords": ["pandoc", "mammoth"],
   "author": {
     "name": "Demo Macro",
     "email": "[email protected]",

packages/docen/src/text.ts

@@ -7,7 +7,7 @@ export async function extractText(
   source: DataType,
   options?: {
     sourceType?: string;
-  },
+  }
 ) {
   let text: string;
 

packages/docen/src/utils/file-type.ts

@@ -10,7 +10,7 @@ export async function detectFileType(source: Uint8Array) {
     const buffer = Buffer.from(source);
     const isTxt = Buffer.compare(
       buffer.subarray(0, 3),
-      Buffer.from([0xef, 0xbb, 0xbf]),
+      Buffer.from([0xef, 0xbb, 0xbf])
     );
     if (isTxt) {
       fileType = {

packages/docx/package.json

@@ -4,9 +4,7 @@
   "description": "Programmed implementation of docx format, powered by Demo Macro.",
   "main": "dist/index.mjs",
   "types": "dist/index.d.ts",
-  "files": [
-    "dist"
-  ],
+  "files": ["dist"],
   "exports": {
     ".": {
       "import": "./dist/index.mjs",
@@ -20,10 +18,7 @@
     "type": "git",
     "url": "git+https://github.com/docenjs/docen.git"
   },
-  "keywords": [
-    "pandoc",
-    "mammoth"
-  ],
+  "keywords": ["pandoc", "mammoth"],
   "author": {
     "name": "Demo Macro",
     "email": "[email protected]",

packages/docx/src/document.ts

@@ -3,13 +3,13 @@ import { type DataType, toUint8Array } from "undio";
 
 export async function getDocumentXML(source: DataType) {
   return JSZIP.loadAsync(toUint8Array(source)).then((zip) =>
-    zip?.file("word/document.xml")?.async("text"),
+    zip?.file("word/document.xml")?.async("text")
   );
 }
 
 export async function getStylesXML(source: DataType) {
   return JSZIP.loadAsync(toUint8Array(source)).then((zip) =>
-    zip?.file("word/styles.xml")?.async("text"),
+    zip?.file("word/styles.xml")?.async("text")
   );
 }
 

packages/image/README.md

@@ -0,0 +1,30 @@
+# @docen/image
+
+![npm version](https://img.shields.io/npm/v/@docen/image)
+![npm downloads](https://img.shields.io/npm/dw/@docen/image)
+![npm license](https://img.shields.io/npm/l/@docen/image)
+
+> Programmed implementation of image format, powered by Demo Macro.
+
+## Getting started
+
+```bash
+# npm
+$ npm install @docen/image
+
+# yarn
+$ yarn add @docen/image
+
+# pnpm
+$ pnpm add @docen/image
+```
+
+## Usage
+
+```ts
+
+```
+
+## License
+
+- [MIT](LICENSE) © [Demo Macro](https://imst.xyz/)

packages/image/build.config.ts

@@ -0,0 +1,12 @@
+import { defineBuildConfig } from "unbuild";
+
+export default defineBuildConfig({
+  declaration: true,
+  entries: ["src/index"],
+  rollup: {
+    emitCJS: true,
+    esbuild: {
+      minify: true,
+    },
+  },
+});

packages/image/package.json

@@ -0,0 +1,37 @@
+{
+  "name": "@docen/image",
+  "version": "0.0.7-edge.3",
+  "description": "Programmed implementation of image format, powered by Demo Macro.",
+  "main": "dist/index.mjs",
+  "types": "dist/index.d.ts",
+  "files": ["dist"],
+  "exports": {
+    ".": {
+      "import": "./dist/index.mjs",
+      "require": "./dist/index.cjs"
+    }
+  },
+  "scripts": {
+    "prepack": "unbuild"
+  },
+  "repository": {
+    "type": "git",
+    "url": "git+https://github.com/docenjs/docen.git"
+  },
+  "keywords": ["pandoc", "image"],
+  "author": {
+    "name": "Demo Macro",
+    "email": "[email protected]",
+    "url": "https://imst.xyz/"
+  },
+  "license": "MIT",
+  "bugs": {
+    "url": "https://github.com/docenjs/docen/issues"
+  },
+  "homepage": "https://github.com/docenjs/docen#readme",
+  "dependencies": {
+    "image-meta": "^0.2.0",
+    "tesseract.js": "5.1.0",
+    "undio": "0.2.0"
+  }
+}

packages/image/src/index.ts

@@ -0,0 +1 @@
+export * from "./ocr";

packages/image/src/ocr.ts

@@ -0,0 +1,31 @@
+import { imageMeta } from "image-meta";
+import { createWorker } from "tesseract.js";
+import { type DataType, toArrayBuffer, toUint8Array } from "undio";
+
+export async function convertImageToText(
+  source: DataType,
+  options?: {
+    // https://tesseract-ocr.github.io/tessdoc/Data-Files
+    lang?: string | string[];
+  }
+) {
+  const { type } = imageMeta(toUint8Array(source));
+
+  // bmp、jpg、png、pbm、webp
+  const isImageType =
+    type && ["bmp", "jpg", "png", "pbm", "webp"].includes(type);
+
+  if (isImageType) {
+    const worker = await createWorker(options?.lang);
+
+    const {
+      data: { text },
+    } = await worker.recognize(toArrayBuffer(source));
+
+    await worker.terminate();
+
+    return text;
+  }
+
+  throw new Error("Unsupported image type");
+}

packages/jsonp/package.json

@@ -4,9 +4,7 @@
   "description": "Programmed implementation of jsonp format, powered by Demo Macro.",
   "main": "dist/index.mjs",
   "types": "dist/index.d.ts",
-  "files": [
-    "dist"
-  ],
+  "files": ["dist"],
   "exports": {
     ".": {
       "import": "./dist/index.mjs",
@@ -20,12 +18,7 @@
     "type": "git",
     "url": "git+https://github.com/docenjs/docen.git"
   },
-  "keywords": [
-    "pandoc",
-    "jsonp",
-    "csv",
-    "json"
-  ],
+  "keywords": ["pandoc", "jsonp", "csv", "json"],
   "author": {
     "name": "Demo Macro",
     "email": "[email protected]",

packages/jsonp/src/index.ts

@@ -1,11 +1,11 @@
 export function convertJSONPToJSON(
   source: string,
-  options: { callbackName?: string; multiple?: boolean } = {},
+  options: { callbackName?: string; multiple?: boolean } = {}
 ): unknown {
   const multiple: boolean = options.multiple ?? false;
 
   const regGroup: RegExpMatchArray | null = source.match(
-    /(?<functionName>.+)\(.*\)/,
+    /(?<functionName>.+)\(.*\)/
   );
 
   const functionName: string | undefined = regGroup?.groups?.functionName;

packages/pdf/package.json

@@ -4,9 +4,7 @@
   "description": "Programmed implementation of pdf format, powered by Demo Macro.",
   "main": "dist/index.mjs",
   "types": "dist/index.d.ts",
-  "files": [
-    "dist"
-  ],
+  "files": ["dist"],
   "exports": {
     ".": {
       "import": "./dist/index.mjs",
@@ -20,11 +18,7 @@
     "type": "git",
     "url": "git+https://github.com/docenjs/docen.git"
   },
-  "keywords": [
-    "pandoc",
-    "unpdf",
-    "pdf"
-  ],
+  "keywords": ["pandoc", "unpdf", "pdf"],
   "author": {
     "name": "Demo Macro",
     "email": "[email protected]",