edgeparse 0.1.0

package/dist/chunk-2PGOFCEV.mjs

@@ -0,0 +1,47 @@
+var __require = /* @__PURE__ */ ((x) => typeof require !== "undefined" ? require : typeof Proxy !== "undefined" ? new Proxy(x, {
+  get: (a, b) => (typeof require !== "undefined" ? require : a)[b]
+}) : x)(function(x) {
+  if (typeof require !== "undefined") return require.apply(this, arguments);
+  throw Error('Dynamic require of "' + x + '" is not supported');
+});
+
+// src/index.ts
+function loadNative() {
+  const platforms = {
+    "linux-x64": "@edgeparse/pdf-linux-x64-gnu",
+    "linux-arm64": "@edgeparse/pdf-linux-arm64-gnu",
+    "darwin-x64": "@edgeparse/pdf-darwin-x64",
+    "darwin-arm64": "@edgeparse/pdf-darwin-arm64",
+    "win32-x64": "@edgeparse/pdf-win32-x64-msvc"
+  };
+  const key = `${process.platform}-${process.arch}`;
+  const pkg = platforms[key];
+  if (!pkg) throw new Error(`edgeparse: unsupported platform: ${key}`);
+  return __require(pkg);
+}
+var native;
+function getNative() {
+  if (!native) {
+    native = loadNative();
+  }
+  return native;
+}
+function convert(inputPath, options) {
+  const n = getNative();
+  return n.convert(inputPath, options ? {
+    format: options.format,
+    pages: options.pages,
+    password: options.password,
+    reading_order: options.readingOrder,
+    table_method: options.tableMethod,
+    image_output: options.imageOutput
+  } : void 0);
+}
+function version() {
+  return getNative().version();
+}
+
+export {
+  convert,
+  version
+};

package/dist/cli.d.mts

@@ -0,0 +1 @@
+#!/usr/bin/env node

package/dist/cli.d.ts

@@ -0,0 +1 @@
+#!/usr/bin/env node

package/dist/cli.js

@@ -0,0 +1,98 @@
+#!/usr/bin/env node
+"use strict";
+
+// src/cli.ts
+var import_node_util = require("util");
+var import_node_fs = require("fs");
+
+// src/index.ts
+function loadNative() {
+  const platforms = {
+    "linux-x64": "@edgeparse/pdf-linux-x64-gnu",
+    "linux-arm64": "@edgeparse/pdf-linux-arm64-gnu",
+    "darwin-x64": "@edgeparse/pdf-darwin-x64",
+    "darwin-arm64": "@edgeparse/pdf-darwin-arm64",
+    "win32-x64": "@edgeparse/pdf-win32-x64-msvc"
+  };
+  const key = `${process.platform}-${process.arch}`;
+  const pkg = platforms[key];
+  if (!pkg) throw new Error(`edgeparse: unsupported platform: ${key}`);
+  return require(pkg);
+}
+var native;
+function getNative() {
+  if (!native) {
+    native = loadNative();
+  }
+  return native;
+}
+function convert(inputPath2, options) {
+  const n = getNative();
+  return n.convert(inputPath2, options ? {
+    format: options.format,
+    pages: options.pages,
+    password: options.password,
+    reading_order: options.readingOrder,
+    table_method: options.tableMethod,
+    image_output: options.imageOutput
+  } : void 0);
+}
+function version() {
+  return getNative().version();
+}
+
+// src/cli.ts
+var { values, positionals } = (0, import_node_util.parseArgs)({
+  allowPositionals: true,
+  options: {
+    format: { type: "string", short: "f", default: "markdown" },
+    pages: { type: "string", short: "p" },
+    password: { type: "string" },
+    "reading-order": { type: "string", default: "xycut" },
+    "table-method": { type: "string", default: "default" },
+    "image-output": { type: "string", default: "off" },
+    output: { type: "string", short: "o" },
+    version: { type: "boolean", short: "v" },
+    help: { type: "boolean", short: "h" }
+  }
+});
+if (values.version) {
+  console.log(`edgeparse ${version()}`);
+  process.exit(0);
+}
+if (values.help || positionals.length === 0) {
+  console.log(`Usage: edgeparse [options] <input.pdf>
+
+Options:
+  -f, --format <fmt>         Output format: markdown, json, html, text (default: markdown)
+  -p, --pages <range>        Page range, e.g. "1,3,5-7"
+      --password <pw>        Password for encrypted PDFs
+      --reading-order <algo> Reading order: xycut (default) or off
+      --table-method <m>     Table method: default or cluster
+      --image-output <mode>  Image output: off (default), embedded, external
+  -o, --output <path>        Output file path (default: stdout)
+  -v, --version              Show version
+  -h, --help                 Show this help
+`);
+  process.exit(values.help ? 0 : 1);
+}
+var inputPath = positionals[0];
+try {
+  const result = convert(inputPath, {
+    format: values.format,
+    pages: values.pages,
+    password: values.password,
+    readingOrder: values["reading-order"],
+    tableMethod: values["table-method"],
+    imageOutput: values["image-output"]
+  });
+  if (values.output) {
+    (0, import_node_fs.writeFileSync)(values.output, result, "utf-8");
+  } else {
+    process.stdout.write(result);
+  }
+} catch (err) {
+  const msg = err instanceof Error ? err.message : String(err);
+  console.error(`edgeparse: ${msg}`);
+  process.exit(1);
+}

package/dist/cli.mjs

@@ -0,0 +1,63 @@
+#!/usr/bin/env node
+import {
+  convert,
+  version
+} from "./chunk-2PGOFCEV.mjs";
+
+// src/cli.ts
+import { parseArgs } from "util";
+import { writeFileSync } from "fs";
+var { values, positionals } = parseArgs({
+  allowPositionals: true,
+  options: {
+    format: { type: "string", short: "f", default: "markdown" },
+    pages: { type: "string", short: "p" },
+    password: { type: "string" },
+    "reading-order": { type: "string", default: "xycut" },
+    "table-method": { type: "string", default: "default" },
+    "image-output": { type: "string", default: "off" },
+    output: { type: "string", short: "o" },
+    version: { type: "boolean", short: "v" },
+    help: { type: "boolean", short: "h" }
+  }
+});
+if (values.version) {
+  console.log(`edgeparse ${version()}`);
+  process.exit(0);
+}
+if (values.help || positionals.length === 0) {
+  console.log(`Usage: edgeparse [options] <input.pdf>
+
+Options:
+  -f, --format <fmt>         Output format: markdown, json, html, text (default: markdown)
+  -p, --pages <range>        Page range, e.g. "1,3,5-7"
+      --password <pw>        Password for encrypted PDFs
+      --reading-order <algo> Reading order: xycut (default) or off
+      --table-method <m>     Table method: default or cluster
+      --image-output <mode>  Image output: off (default), embedded, external
+  -o, --output <path>        Output file path (default: stdout)
+  -v, --version              Show version
+  -h, --help                 Show this help
+`);
+  process.exit(values.help ? 0 : 1);
+}
+var inputPath = positionals[0];
+try {
+  const result = convert(inputPath, {
+    format: values.format,
+    pages: values.pages,
+    password: values.password,
+    readingOrder: values["reading-order"],
+    tableMethod: values["table-method"],
+    imageOutput: values["image-output"]
+  });
+  if (values.output) {
+    writeFileSync(values.output, result, "utf-8");
+  } else {
+    process.stdout.write(result);
+  }
+} catch (err) {
+  const msg = err instanceof Error ? err.message : String(err);
+  console.error(`edgeparse: ${msg}`);
+  process.exit(1);
+}

package/dist/index.d.mts

@@ -0,0 +1,29 @@
+interface ConvertOptions {
+    /** Output format. Valid values: "markdown", "json", "html", "text". Default: "markdown". */
+    format?: string;
+    /** Page range string, e.g. "1,3,5-7". */
+    pages?: string;
+    /** Password for encrypted PDFs. */
+    password?: string;
+    /** Reading order algorithm: "xycut" (default) or "off". */
+    readingOrder?: string;
+    /** Table detection method: "default" or "cluster". */
+    tableMethod?: string;
+    /** Image output mode: "off" (default), "embedded", or "external". */
+    imageOutput?: string;
+}
+
+/**
+ * Convert a PDF file and return the extracted content as a string.
+ *
+ * @param inputPath - Path to the PDF file.
+ * @param options   - Conversion options.
+ * @returns The extracted content as a string.
+ */
+declare function convert(inputPath: string, options?: ConvertOptions): string;
+/**
+ * Return the edgeparse version string.
+ */
+declare function version(): string;
+
+export { type ConvertOptions, convert, version };

package/dist/index.d.ts

@@ -0,0 +1,29 @@
+interface ConvertOptions {
+    /** Output format. Valid values: "markdown", "json", "html", "text". Default: "markdown". */
+    format?: string;
+    /** Page range string, e.g. "1,3,5-7". */
+    pages?: string;
+    /** Password for encrypted PDFs. */
+    password?: string;
+    /** Reading order algorithm: "xycut" (default) or "off". */
+    readingOrder?: string;
+    /** Table detection method: "default" or "cluster". */
+    tableMethod?: string;
+    /** Image output mode: "off" (default), "embedded", or "external". */
+    imageOutput?: string;
+}
+
+/**
+ * Convert a PDF file and return the extracted content as a string.
+ *
+ * @param inputPath - Path to the PDF file.
+ * @param options   - Conversion options.
+ * @returns The extracted content as a string.
+ */
+declare function convert(inputPath: string, options?: ConvertOptions): string;
+/**
+ * Return the edgeparse version string.
+ */
+declare function version(): string;
+
+export { type ConvertOptions, convert, version };

package/dist/index.js

@@ -0,0 +1,65 @@
+"use strict";
+var __defProp = Object.defineProperty;
+var __getOwnPropDesc = Object.getOwnPropertyDescriptor;
+var __getOwnPropNames = Object.getOwnPropertyNames;
+var __hasOwnProp = Object.prototype.hasOwnProperty;
+var __export = (target, all) => {
+  for (var name in all)
+    __defProp(target, name, { get: all[name], enumerable: true });
+};
+var __copyProps = (to, from, except, desc) => {
+  if (from && typeof from === "object" || typeof from === "function") {
+    for (let key of __getOwnPropNames(from))
+      if (!__hasOwnProp.call(to, key) && key !== except)
+        __defProp(to, key, { get: () => from[key], enumerable: !(desc = __getOwnPropDesc(from, key)) || desc.enumerable });
+  }
+  return to;
+};
+var __toCommonJS = (mod) => __copyProps(__defProp({}, "__esModule", { value: true }), mod);
+
+// src/index.ts
+var index_exports = {};
+__export(index_exports, {
+  convert: () => convert,
+  version: () => version
+});
+module.exports = __toCommonJS(index_exports);
+function loadNative() {
+  const platforms = {
+    "linux-x64": "@edgeparse/pdf-linux-x64-gnu",
+    "linux-arm64": "@edgeparse/pdf-linux-arm64-gnu",
+    "darwin-x64": "@edgeparse/pdf-darwin-x64",
+    "darwin-arm64": "@edgeparse/pdf-darwin-arm64",
+    "win32-x64": "@edgeparse/pdf-win32-x64-msvc"
+  };
+  const key = `${process.platform}-${process.arch}`;
+  const pkg = platforms[key];
+  if (!pkg) throw new Error(`edgeparse: unsupported platform: ${key}`);
+  return require(pkg);
+}
+var native;
+function getNative() {
+  if (!native) {
+    native = loadNative();
+  }
+  return native;
+}
+function convert(inputPath, options) {
+  const n = getNative();
+  return n.convert(inputPath, options ? {
+    format: options.format,
+    pages: options.pages,
+    password: options.password,
+    reading_order: options.readingOrder,
+    table_method: options.tableMethod,
+    image_output: options.imageOutput
+  } : void 0);
+}
+function version() {
+  return getNative().version();
+}
+// Annotate the CommonJS export names for ESM import in node:
+0 && (module.exports = {
+  convert,
+  version
+});

package/dist/index.mjs

@@ -0,0 +1,8 @@
+import {
+  convert,
+  version
+} from "./chunk-2PGOFCEV.mjs";
+export {
+  convert,
+  version
+};

package/npm/darwin-arm64/package.json

@@ -0,0 +1,15 @@
+{
+  "name": "edgeparse-darwin-arm64",
+  "version": "0.1.0",
+  "os": ["darwin"],
+  "cpu": ["arm64"],
+  "main": "edgeparse-node.darwin-arm64.node",
+  "files": ["edgeparse-node.darwin-arm64.node"],
+  "description": "edgeparse native addon for darwin-arm64",
+  "license": "Apache-2.0",
+  "repository": {
+    "type": "git",
+    "url": "https://github.com/opendataloader-project/edgeparse.git",
+    "directory": "sdks/node/npm/darwin-arm64"
+  }
+}

package/npm/darwin-x64/package.json

@@ -0,0 +1,15 @@
+{
+  "name": "edgeparse-darwin-x64",
+  "version": "0.1.0",
+  "os": ["darwin"],
+  "cpu": ["x64"],
+  "main": "edgeparse-node.darwin-x64.node",
+  "files": ["edgeparse-node.darwin-x64.node"],
+  "description": "edgeparse native addon for darwin-x64",
+  "license": "Apache-2.0",
+  "repository": {
+    "type": "git",
+    "url": "https://github.com/opendataloader-project/edgeparse.git",
+    "directory": "sdks/node/npm/darwin-x64"
+  }
+}

package/npm/linux-arm64-gnu/package.json

@@ -0,0 +1,15 @@
+{
+  "name": "edgeparse-linux-arm64-gnu",
+  "version": "0.1.0",
+  "os": ["linux"],
+  "cpu": ["arm64"],
+  "main": "edgeparse-node.linux-arm64-gnu.node",
+  "files": ["edgeparse-node.linux-arm64-gnu.node"],
+  "description": "edgeparse native addon for linux-arm64-gnu",
+  "license": "Apache-2.0",
+  "repository": {
+    "type": "git",
+    "url": "https://github.com/opendataloader-project/edgeparse.git",
+    "directory": "sdks/node/npm/linux-arm64-gnu"
+  }
+}

package/npm/linux-x64-gnu/package.json

@@ -0,0 +1,15 @@
+{
+  "name": "edgeparse-linux-x64-gnu",
+  "version": "0.1.0",
+  "os": ["linux"],
+  "cpu": ["x64"],
+  "main": "edgeparse-node.linux-x64-gnu.node",
+  "files": ["edgeparse-node.linux-x64-gnu.node"],
+  "description": "edgeparse native addon for linux-x64-gnu",
+  "license": "Apache-2.0",
+  "repository": {
+    "type": "git",
+    "url": "https://github.com/opendataloader-project/edgeparse.git",
+    "directory": "sdks/node/npm/linux-x64-gnu"
+  }
+}

package/npm/win32-x64-msvc/package.json

@@ -0,0 +1,15 @@
+{
+  "name": "edgeparse-win32-x64-msvc",
+  "version": "0.1.0",
+  "os": ["win32"],
+  "cpu": ["x64"],
+  "main": "edgeparse-node.win32-x64-msvc.node",
+  "files": ["edgeparse-node.win32-x64-msvc.node"],
+  "description": "edgeparse native addon for win32-x64-msvc",
+  "license": "Apache-2.0",
+  "repository": {
+    "type": "git",
+    "url": "https://github.com/opendataloader-project/edgeparse.git",
+    "directory": "sdks/node/npm/win32-x64-msvc"
+  }
+}

package/package.json

@@ -0,0 +1,60 @@
+{
+  "name": "edgeparse",
+  "version": "0.1.0",
+  "description": "High-performance PDF extraction — Rust engine, Node.js interface",
+  "main": "./dist/index.cjs",
+  "module": "./dist/index.js",
+  "types": "./dist/index.d.ts",
+  "exports": {
+    ".": {
+      "import": "./dist/index.js",
+      "require": "./dist/index.cjs",
+      "types": "./dist/index.d.ts"
+    }
+  },
+  "bin": {
+    "edgeparse": "./dist/cli.cjs"
+  },
+  "files": [
+    "dist/",
+    "npm/",
+    "README.md"
+  ],
+  "optionalDependencies": {
+    "edgeparse-darwin-arm64": "0.1.0",
+    "edgeparse-darwin-x64": "0.1.0",
+    "edgeparse-linux-arm64-gnu": "0.1.0",
+    "edgeparse-linux-x64-gnu": "0.1.0",
+    "edgeparse-win32-x64-msvc": "0.1.0"
+  },
+  "engines": {
+    "node": ">=18"
+  },
+  "license": "Apache-2.0",
+  "repository": {
+    "type": "git",
+    "url": "https://github.com/opendataloader-project/edgeparse.git",
+    "directory": "sdks/node"
+  },
+  "keywords": [
+    "pdf",
+    "extraction",
+    "markdown",
+    "json",
+    "rag",
+    "llm",
+    "rust"
+  ],
+  "scripts": {
+    "build:native": "cargo build --manifest-path ../../crates/edgeparse-node/Cargo.toml --release",
+    "build:ts": "npx tsup src/index.ts src/cli.ts --format cjs,esm --dts",
+    "build": "npm run build:native && npm run build:ts",
+    "test": "npx vitest run"
+  },
+  "devDependencies": {
+    "@types/node": "^25.5.0",
+    "tsup": "^8.0.0",
+    "typescript": "^5.0.0",
+    "vitest": "^3.0.0"
+  }
+}