@opendataloader/pdf 1.0.5 → 1.1.0

package/README.md

@@ -75,40 +75,42 @@ pip install -U opendataloader-pdf
 
 ### Usage
 
-- input_path can be either the path to a single document or the path to a folder.
-- If you don’t specify an output_folder, the output data will be saved in the same directory as the input document.
+input_path can be either the path to a single document or the path to a folder.
 
 ```python
 import opendataloader_pdf
 
-opendataloader_pdf.run(
-    input_path="path/to/document.pdf",
-    output_folder="path/to/output",
-    generate_markdown=True,
-    generate_html=True,
-    generate_annotated_pdf=True,
+opendataloader_pdf.convert(
+    input_path=["path/to/document.pdf", "path/to/folder"],
+    output_dir="path/to/output",
+    format=["json", "html", "pdf", "markdown"]
 )
 ```
 
-### Function: run()
+If you want to run it via CLI, you can use the following command on the terminal:
+
+```bash
+opendataloader-pdf path/to/document.pdf path/to/folder -o path/to/output -f json html pdf markdown
+```
+
+### Function: convert()
 
 The main function to process PDFs.
 
-| Parameter                | Type   | Required | Default      | Description                                                                                                                                 |
-|--------------------------| ------ | -------- |--------------|---------------------------------------------------------------------------------------------------------------------------------------------|
-| `input_path`             | `str`  | ✅ Yes    | —            | Path to the input PDF file or folder.                                                                                                       |
-| `output_folder`          | `str`  | No       | input folder | Path to the output folder.                                                                                                                  |
-| `password`               | `str`  | No       | `None`       | Password for the PDF file.                                                                                                                  |
-| `replace_invalid_chars`  | `str`  | No       | `" "`       | Character to replace invalid or unrecognized characters (e.g., �, \u0000)                                                                   |
-| `content_safety_off`     | `str`  | No       | `None`       | Disables one or more content safety filters. Accepts a comma-separated list of filter names. Arguments: all, hidden-text, off-page, tiny, hidden-ocg. |
-| `generate_markdown`      | `bool` | No       | `False`      | If `True`, generates a Markdown output file.                                                                                                |
-| `generate_html`          | `bool` | No       | `False`      | If `True`, generates an HTML output file.                                                                                                   |
-| `generate_annotated_pdf` | `bool` | No       | `False`      | If `True`, generates an annotated PDF output file.                                                                                          |
-| `keep_line_breaks`       | `bool` | No       | `False`      | If `True`, keeps line breaks in the output.                                                                                                 |
-| `html_in_markdown`       | `bool` | No       | `False`      | If `True`, uses HTML in the Markdown output.                                                                                                |
-| `add_image_to_markdown`  | `bool` | No       | `False`      | If `True`, adds images to the Markdown output.                                                                                              |
-| `no_json`                | `bool` | No       | `False`      | If `True`, disables the JSON output.                                                                                                        |
-| `debug`                  | `bool` | No       | `False`      | If `True`, prints CLI messages to the console during execution.                                                                             |
+| Parameter                | Type           | Required | Default      | Description                                                                                                                                 |
+|--------------------------|----------------| -------- |--------------|---------------------------------------------------------------------------------------------------------------------------------------------|
+| `input_path`             | `List[str]`     | ✅ Yes    | —            | One or more PDF file paths or directories to process.                                                                                       |
+| `output_dir`             | `Optional[str]` | No       | input folder | Directory where outputs are written.                                                                                                       |
+| `password`               | `Optional[str]` | No       | `None`       | Password used for encrypted PDFs.                                                                                                           |
+| `format`                 | `Optional[List[str]]` | No | `None`       | Output formats to generate (e.g. `"json"`, `"html"`, `"pdf"`, `"text"`, `"markdown"`, `"markdown-with-html"`, `"markdown-with-images"`).                                                             |
+| `quiet`                  | `bool`          | No       | `False`      | Suppresses CLI logging output when `True`.                                                                                                  |
+| `content_safety_off`     | `Optional[List[str]]` | No | `None`       | List of content safety filters to disable (e.g. `"all"`, `"hidden-text"`, `"off-page"`, `"tiny"`, `"hidden-ocg"`).                      |
+| `keep_line_breaks`       | `bool`          | No       | `False`      | Preserves line breaks in text output when `True`.                                                                                           |
+| `replace_invalid_chars`  | `Optional[str]` | No       | `None`       | Replacement character for invalid or unrecognized characters (e.g., �, `\u0000`).                                                           |
+
+### Function: run()
+
+Deprecated.
 
 <br/>
 
@@ -152,6 +154,24 @@ async function main() {
 main();
 ```
 
+If you want to run it via CLI, you can use the following command:
+
+```bash
+npx @opendataloader/pdf path/to/document.pdf -o path/to/output --markdown --html --pdf
+```
+
+or you can install it globally:
+
+```bash
+npm install -g @opendataloader/pdf
+```
+
+then run:
+
+```bash
+opendataloader-pdf path/to/document.pdf -o path/to/output --markdown --html --pdf
+```
+
 ### Function: run()
 
 `run(inputPath: string, options?: RunOptions): Promise<string>`
@@ -321,16 +341,23 @@ The images are extracted from PDF as individual files and stored in a subfolder
 ```
 Options:
 -o,--output-dir <arg>           Specifies the output directory for generated files
+-p,--password <arg>             Specifies the password for an encrypted PDF
+-f,--format <arg>               List of output formats to generate (json, text, html, pdf, markdown, markdown-with-html, markdown-with-images). Default: json
+-q,--quiet                      Suppresses console logging output
+--content-safety-off <arg>      Disables one or more content safety filters. Accepts a list of filter names. Arguments: all, hidden-text, off-page, tiny, hidden-ocg
 --keep-line-breaks              Preserves original line breaks in the extracted text
---content-safety-off <arg>      Disables one or more content safety filters. Accepts a comma-separated list of filter names. Arguments: all, hidden-text, off-page, tiny, hidden-ocg
---markdown-with-html            Sets the data extraction output format to Markdown with rendering complex elements like tables as HTML for better structure
---markdown-with-images          Sets the data extraction output format to Markdown with extracting images from the PDF and includes them as links
---markdown                      Sets the data extraction output format to Markdown
---html                          Sets the data extraction output format to HTML
+--replace-invalid-chars <arg>   Replaces invalid or unrecognized characters (e.g., �, \u0000) with the specified character
+```
+
+The legacy options (for backward compatibility):
+
+```
 --no-json                       Disables the JSON output format
--p,--password <arg>             Specifies the password for an encrypted PDF
+--html                          Sets the data extraction output format to HTML
 --pdf                           Generates a new PDF file where the extracted layout data is visualized as annotations
---replace-invalid-chars <arg>   Replaces invalid or unrecognized characters (e.g., �, \u0000) with the specified character
+--markdown                      Sets the data extraction output format to Markdown
+--markdown-with-html            Sets the data extraction output format to Markdown with rendering complex elements like tables as HTML for better structure
+--markdown-with-images          Sets the data extraction output format to Markdown with extracting images from the PDF and includes them as links
 ```
 
 ### Schema of the JSON output

package/dist/cli.cjs

@@ -0,0 +1,276 @@
+#!/usr/bin/env node
+"use strict";
+var __create = Object.create;
+var __defProp = Object.defineProperty;
+var __getOwnPropDesc = Object.getOwnPropertyDescriptor;
+var __getOwnPropNames = Object.getOwnPropertyNames;
+var __getProtoOf = Object.getPrototypeOf;
+var __hasOwnProp = Object.prototype.hasOwnProperty;
+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 __toESM = (mod, isNodeMode, target) => (target = mod != null ? __create(__getProtoOf(mod)) : {}, __copyProps(
+  // If the importer is in node compatibility mode or this is not an ESM
+  // file that has been converted to a CommonJS file using a Babel-
+  // compatible transform (i.e. "__esModule" has not been set), then set
+  // "default" to the CommonJS "module.exports" for node compatibility.
+  isNodeMode || !mod || !mod.__esModule ? __defProp(target, "default", { value: mod, enumerable: true }) : target,
+  mod
+));
+
+// src/index.ts
+var import_child_process = require("child_process");
+var path = __toESM(require("path"), 1);
+var fs = __toESM(require("fs"), 1);
+var import_url = require("url");
+var import_meta = {};
+var __filename = (0, import_url.fileURLToPath)(import_meta.url);
+var __dirname = path.dirname(__filename);
+var JAR_NAME = "opendataloader-pdf-cli.jar";
+function getRedactedCommandString(command, commandArgs) {
+  const commandArgsForLogging = [...commandArgs];
+  const passwordIndex = commandArgsForLogging.indexOf("--password");
+  if (passwordIndex > -1 && passwordIndex + 1 < commandArgsForLogging.length) {
+    commandArgsForLogging[passwordIndex + 1] = "[REDACTED]";
+  }
+  return `${command} ${commandArgsForLogging.join(" ")}`;
+}
+function run(inputPath, options = {}) {
+  return new Promise((resolve, reject) => {
+    if (!fs.existsSync(inputPath)) {
+      return reject(new Error(`Input file or folder not found: ${inputPath}`));
+    }
+    const args = [];
+    if (options.outputFolder) {
+      args.push("--output-dir", options.outputFolder);
+    }
+    if (options.password) {
+      args.push("--password", options.password);
+    }
+    if (options.replaceInvalidChars) {
+      args.push("--replace-invalid-chars", options.replaceInvalidChars);
+    }
+    if (options.generateMarkdown) {
+      args.push("--markdown");
+    }
+    if (options.generateHtml) {
+      args.push("--html");
+    }
+    if (options.generateAnnotatedPdf) {
+      args.push("--pdf");
+    }
+    if (options.keepLineBreaks) {
+      args.push("--keep-line-breaks");
+    }
+    if (options.contentSafetyOff) {
+      args.push("--content-safety-off", options.contentSafetyOff);
+    }
+    if (options.htmlInMarkdown) {
+      args.push("--markdown-with-html");
+    }
+    if (options.addImageToMarkdown) {
+      args.push("--markdown-with-images");
+    }
+    if (options.noJson) {
+      args.push("--no-json");
+    }
+    args.push(inputPath);
+    const jarPath = path.join(__dirname, "..", "lib", JAR_NAME);
+    if (!fs.existsSync(jarPath)) {
+      return reject(
+        new Error(`JAR file not found at ${jarPath}. Please run the build script first.`)
+      );
+    }
+    const command = "java";
+    const commandArgs = ["-jar", jarPath, ...args];
+    if (options.debug) {
+      console.error(`Running command: ${getRedactedCommandString(command, commandArgs)}`);
+    }
+    const javaProcess = (0, import_child_process.spawn)(command, commandArgs);
+    let stdout = "";
+    let stderr = "";
+    javaProcess.stdout.on("data", (data) => {
+      const chunk = data.toString();
+      if (options.debug) {
+        process.stdout.write(chunk);
+      }
+      stdout += chunk;
+    });
+    javaProcess.stderr.on("data", (data) => {
+      const chunk = data.toString();
+      if (options.debug) {
+        process.stderr.write(chunk);
+      }
+      stderr += chunk;
+    });
+    javaProcess.on("close", (code) => {
+      if (code === 0) {
+        resolve(stdout);
+      } else {
+        const error = new Error(
+          `The opendataloader-pdf CLI exited with code ${code}.
+
+${stderr}`
+        );
+        reject(error);
+      }
+    });
+    javaProcess.on("error", (err) => {
+      if (err.message.includes("ENOENT")) {
+        reject(
+          new Error(
+            "'java' command not found. Please ensure Java is installed and in your system's PATH."
+          )
+        );
+      } else {
+        reject(err);
+      }
+    });
+  });
+}
+
+// src/cli.ts
+function printHelp() {
+  console.log(`Usage: opendataloader-pdf [options] <input>`);
+  console.log("");
+  console.log("Options:");
+  console.log("  -o, --output-dir <path>      Directory where outputs are written");
+  console.log("  -p, --password <password>    Password for encrypted PDFs");
+  console.log("  --replace-invalid-chars <c>  Replacement character for invalid characters");
+  console.log("  --content-safety-off <mode>  Disable content safety filtering (provide mode)");
+  console.log("  --markdown                   Generate Markdown output");
+  console.log("  --html                       Generate HTML output");
+  console.log("  --pdf                        Generate annotated PDF output");
+  console.log("  --keep-line-breaks           Preserve line breaks in text output");
+  console.log("  --markdown-with-html         Allow raw HTML within Markdown output");
+  console.log("  --markdown-with-images       Embed images in Markdown output");
+  console.log("  --no-json                    Disable JSON output generation");
+  console.log("  --debug                      Stream CLI logs directly to stdout/stderr");
+  console.log("  -h, --help                   Show this message and exit");
+}
+function parseArgs(argv) {
+  const options = {};
+  let inputPath;
+  let showHelp = false;
+  const readValue = (currentIndex, option) => {
+    const nextValue = argv[currentIndex + 1];
+    if (!nextValue || nextValue.startsWith("-")) {
+      throw new Error(`Option ${option} requires a value.`);
+    }
+    return { value: nextValue, nextIndex: currentIndex + 1 };
+  };
+  for (let i = 0; i < argv.length; i += 1) {
+    const arg = argv[i];
+    switch (arg) {
+      case "--help":
+      case "-h":
+        showHelp = true;
+        i = argv.length;
+        break;
+      case "--output-dir":
+      case "-o": {
+        const { value, nextIndex } = readValue(i, arg);
+        options.outputFolder = value;
+        i = nextIndex;
+        break;
+      }
+      case "--password":
+      case "-p": {
+        const { value, nextIndex } = readValue(i, arg);
+        options.password = value;
+        i = nextIndex;
+        break;
+      }
+      case "--replace-invalid-chars": {
+        const { value, nextIndex } = readValue(i, arg);
+        options.replaceInvalidChars = value;
+        i = nextIndex;
+        break;
+      }
+      case "--content-safety-off": {
+        const { value, nextIndex } = readValue(i, arg);
+        options.contentSafetyOff = value;
+        i = nextIndex;
+        break;
+      }
+      case "--markdown":
+        options.generateMarkdown = true;
+        break;
+      case "--html":
+        options.generateHtml = true;
+        break;
+      case "--pdf":
+        options.generateAnnotatedPdf = true;
+        break;
+      case "--keep-line-breaks":
+        options.keepLineBreaks = true;
+        break;
+      case "--markdown-with-html":
+        options.htmlInMarkdown = true;
+        break;
+      case "--markdown-with-images":
+        options.addImageToMarkdown = true;
+        break;
+      case "--no-json":
+        options.noJson = true;
+        break;
+      case "--debug":
+        options.debug = true;
+        break;
+      default:
+        if (arg.startsWith("-")) {
+          throw new Error(`Unknown option: ${arg}`);
+        }
+        if (inputPath) {
+          throw new Error("Multiple input paths provided. Only one input path is allowed.");
+        }
+        inputPath = arg;
+    }
+  }
+  return { inputPath, options, showHelp };
+}
+async function main() {
+  let parsed;
+  try {
+    parsed = parseArgs(process.argv.slice(2));
+  } catch (err) {
+    const message = err instanceof Error ? err.message : String(err);
+    console.error(message);
+    console.error("Use '--help' to see available options.");
+    return 1;
+  }
+  if (parsed.showHelp) {
+    printHelp();
+    return 0;
+  }
+  if (!parsed.inputPath) {
+    console.error("Missing required input path.");
+    console.error("Use '--help' to see usage information.");
+    return 1;
+  }
+  try {
+    const output = await run(parsed.inputPath, parsed.options);
+    if (output && !parsed.options.debug) {
+      process.stdout.write(output);
+    }
+    if (output && !output.endsWith("\n") && !parsed.options.debug) {
+      process.stdout.write("\n");
+    }
+    return 0;
+  } catch (err) {
+    const message = err instanceof Error ? err.message : String(err);
+    console.error(message);
+    return 1;
+  }
+}
+main().then((code) => {
+  if (code !== 0) {
+    process.exit(code);
+  }
+});
+//# sourceMappingURL=cli.cjs.map

package/dist/cli.cjs.map

@@ -0,0 +1 @@
+{"version":3,"sources":["../src/index.ts","../src/cli.ts"],"sourcesContent":["import { spawn } from 'child_process';\nimport * as path from 'path';\nimport * as fs from 'fs';\nimport { fileURLToPath } from 'url';\n\nconst __filename = fileURLToPath(import.meta.url);\nconst __dirname = path.dirname(__filename);\n\nconst JAR_NAME = 'opendataloader-pdf-cli.jar';\n\nfunction getRedactedCommandString(command: string, commandArgs: string[]): string {\n  const commandArgsForLogging = [...commandArgs];\n  const passwordIndex = commandArgsForLogging.indexOf('--password');\n  if (passwordIndex > -1 && passwordIndex + 1 < commandArgsForLogging.length) {\n    commandArgsForLogging[passwordIndex + 1] = '[REDACTED]';\n  }\n  return `${command} ${commandArgsForLogging.join(' ')}`;\n}\n\nexport interface RunOptions {\n  outputFolder?: string;\n  password?: string;\n  replaceInvalidChars?: string;\n  generateMarkdown?: boolean;\n  generateHtml?: boolean;\n  generateAnnotatedPdf?: boolean;\n  keepLineBreaks?: boolean;\n  contentSafetyOff?: string;\n  htmlInMarkdown?: boolean;\n  addImageToMarkdown?: boolean;\n  noJson?: boolean;\n  debug?: boolean;\n}\n\nexport function run(inputPath: string, options: RunOptions = {}): Promise<string> {\n  return new Promise((resolve, reject) => {\n    if (!fs.existsSync(inputPath)) {\n      return reject(new Error(`Input file or folder not found: ${inputPath}`));\n    }\n\n    const args: string[] = [];\n    if (options.outputFolder) {\n      args.push('--output-dir', options.outputFolder);\n    }\n    if (options.password) {\n      args.push('--password', options.password);\n    }\n    if (options.replaceInvalidChars) {\n      args.push('--replace-invalid-chars', options.replaceInvalidChars);\n    }\n    if (options.generateMarkdown) {\n      args.push('--markdown');\n    }\n    if (options.generateHtml) {\n      args.push('--html');\n    }\n    if (options.generateAnnotatedPdf) {\n      args.push('--pdf');\n    }\n    if (options.keepLineBreaks) {\n      args.push('--keep-line-breaks');\n    }\n    if (options.contentSafetyOff) {\n      args.push('--content-safety-off', options.contentSafetyOff);\n    }\n    if (options.htmlInMarkdown) {\n      args.push('--markdown-with-html');\n    }\n    if (options.addImageToMarkdown) {\n      args.push('--markdown-with-images');\n    }\n    if (options.noJson) {\n      args.push('--no-json');\n    }\n\n    args.push(inputPath);\n\n    const jarPath = path.join(__dirname, '..', 'lib', JAR_NAME);\n\n    if (!fs.existsSync(jarPath)) {\n      return reject(\n        new Error(`JAR file not found at ${jarPath}. Please run the build script first.`),\n      );\n    }\n\n    const command = 'java';\n    const commandArgs = ['-jar', jarPath, ...args];\n\n    if (options.debug) {\n      console.error(`Running command: ${getRedactedCommandString(command, commandArgs)}`);\n    }\n\n    const javaProcess = spawn(command, commandArgs);\n\n    let stdout = '';\n    let stderr = '';\n\n    javaProcess.stdout.on('data', (data) => {\n      const chunk = data.toString();\n      if (options.debug) {\n        process.stdout.write(chunk);\n      }\n      stdout += chunk;\n    });\n\n    javaProcess.stderr.on('data', (data) => {\n      const chunk = data.toString();\n      if (options.debug) {\n        process.stderr.write(chunk);\n      }\n      stderr += chunk;\n    });\n\n    javaProcess.on('close', (code) => {\n      if (code === 0) {\n        resolve(stdout);\n      } else {\n        const error = new Error(\n          `The opendataloader-pdf CLI exited with code ${code}.\\n\\n${stderr}`,\n        );\n        reject(error);\n      }\n    });\n\n    javaProcess.on('error', (err) => {\n      if (err.message.includes('ENOENT')) {\n        reject(\n          new Error(\n            \"'java' command not found. Please ensure Java is installed and in your system's PATH.\",\n          ),\n        );\n      } else {\n        reject(err);\n      }\n    });\n  });\n}\n","#!/usr/bin/env node\nimport { run, RunOptions } from './index.js';\n\ninterface ParsedArgs {\n  inputPath?: string;\n  options: RunOptions;\n  showHelp: boolean;\n}\n\nfunction printHelp(): void {\n  console.log(`Usage: opendataloader-pdf [options] <input>`);\n  console.log('');\n  console.log('Options:');\n  console.log('  -o, --output-dir <path>      Directory where outputs are written');\n  console.log('  -p, --password <password>    Password for encrypted PDFs');\n  console.log('  --replace-invalid-chars <c>  Replacement character for invalid characters');\n  console.log('  --content-safety-off <mode>  Disable content safety filtering (provide mode)');\n  console.log('  --markdown                   Generate Markdown output');\n  console.log('  --html                       Generate HTML output');\n  console.log('  --pdf                        Generate annotated PDF output');\n  console.log('  --keep-line-breaks           Preserve line breaks in text output');\n  console.log('  --markdown-with-html         Allow raw HTML within Markdown output');\n  console.log('  --markdown-with-images       Embed images in Markdown output');\n  console.log('  --no-json                    Disable JSON output generation');\n  console.log('  --debug                      Stream CLI logs directly to stdout/stderr');\n  console.log('  -h, --help                   Show this message and exit');\n}\n\nfunction parseArgs(argv: string[]): ParsedArgs {\n  const options: RunOptions = {};\n  let inputPath: string | undefined;\n  let showHelp = false;\n\n  const readValue = (currentIndex: number, option: string): { value: string; nextIndex: number } => {\n    const nextValue = argv[currentIndex + 1];\n    if (!nextValue || nextValue.startsWith('-')) {\n      throw new Error(`Option ${option} requires a value.`);\n    }\n    return { value: nextValue, nextIndex: currentIndex + 1 };\n  };\n\n  for (let i = 0; i < argv.length; i += 1) {\n    const arg = argv[i];\n\n    switch (arg) {\n      case '--help':\n      case '-h':\n        showHelp = true;\n        i = argv.length; // exit loop\n        break;\n      case '--output-dir':\n      case '-o': {\n        const { value, nextIndex } = readValue(i, arg);\n        options.outputFolder = value;\n        i = nextIndex;\n        break;\n      }\n      case '--password':\n      case '-p': {\n        const { value, nextIndex } = readValue(i, arg);\n        options.password = value;\n        i = nextIndex;\n        break;\n      }\n      case '--replace-invalid-chars': {\n        const { value, nextIndex } = readValue(i, arg);\n        options.replaceInvalidChars = value;\n        i = nextIndex;\n        break;\n      }\n      case '--content-safety-off': {\n        const { value, nextIndex } = readValue(i, arg);\n        options.contentSafetyOff = value;\n        i = nextIndex;\n        break;\n      }\n      case '--markdown':\n        options.generateMarkdown = true;\n        break;\n      case '--html':\n        options.generateHtml = true;\n        break;\n      case '--pdf':\n        options.generateAnnotatedPdf = true;\n        break;\n      case '--keep-line-breaks':\n        options.keepLineBreaks = true;\n        break;\n      case '--markdown-with-html':\n        options.htmlInMarkdown = true;\n        break;\n      case '--markdown-with-images':\n        options.addImageToMarkdown = true;\n        break;\n      case '--no-json':\n        options.noJson = true;\n        break;\n      case '--debug':\n        options.debug = true;\n        break;\n      default:\n        if (arg.startsWith('-')) {\n          throw new Error(`Unknown option: ${arg}`);\n        }\n\n        if (inputPath) {\n          throw new Error('Multiple input paths provided. Only one input path is allowed.');\n        }\n        inputPath = arg;\n    }\n  }\n\n  return { inputPath, options, showHelp };\n}\n\nasync function main(): Promise<number> {\n  let parsed: ParsedArgs;\n\n  try {\n    parsed = parseArgs(process.argv.slice(2));\n  } catch (err) {\n    const message = err instanceof Error ? err.message : String(err);\n    console.error(message);\n    console.error(\"Use '--help' to see available options.\");\n    return 1;\n  }\n\n  if (parsed.showHelp) {\n    printHelp();\n    return 0;\n  }\n\n  if (!parsed.inputPath) {\n    console.error('Missing required input path.');\n    console.error(\"Use '--help' to see usage information.\");\n    return 1;\n  }\n\n  try {\n    const output = await run(parsed.inputPath, parsed.options);\n    if (output && !parsed.options.debug) {\n      process.stdout.write(output);\n    }\n    if (output && !output.endsWith('\\n') && !parsed.options.debug) {\n      process.stdout.write('\\n');\n    }\n    return 0;\n  } catch (err) {\n    const message = err instanceof Error ? err.message : String(err);\n    console.error(message);\n    return 1;\n  }\n}\n\nmain().then((code) => {\n  if (code !== 0) {\n    process.exit(code);\n  }\n});\n"],"mappings":";;;;;;;;;;;;;;;;;;;;;;;;;;AAAA,2BAAsB;AACtB,WAAsB;AACtB,SAAoB;AACpB,iBAA8B;AAH9B;AAKA,IAAM,iBAAa,0BAAc,YAAY,GAAG;AAChD,IAAM,YAAiB,aAAQ,UAAU;AAEzC,IAAM,WAAW;AAEjB,SAAS,yBAAyB,SAAiB,aAA+B;AAChF,QAAM,wBAAwB,CAAC,GAAG,WAAW;AAC7C,QAAM,gBAAgB,sBAAsB,QAAQ,YAAY;AAChE,MAAI,gBAAgB,MAAM,gBAAgB,IAAI,sBAAsB,QAAQ;AAC1E,0BAAsB,gBAAgB,CAAC,IAAI;AAAA,EAC7C;AACA,SAAO,GAAG,OAAO,IAAI,sBAAsB,KAAK,GAAG,CAAC;AACtD;AAiBO,SAAS,IAAI,WAAmB,UAAsB,CAAC,GAAoB;AAChF,SAAO,IAAI,QAAQ,CAAC,SAAS,WAAW;AACtC,QAAI,CAAI,cAAW,SAAS,GAAG;AAC7B,aAAO,OAAO,IAAI,MAAM,mCAAmC,SAAS,EAAE,CAAC;AAAA,IACzE;AAEA,UAAM,OAAiB,CAAC;AACxB,QAAI,QAAQ,cAAc;AACxB,WAAK,KAAK,gBAAgB,QAAQ,YAAY;AAAA,IAChD;AACA,QAAI,QAAQ,UAAU;AACpB,WAAK,KAAK,cAAc,QAAQ,QAAQ;AAAA,IAC1C;AACA,QAAI,QAAQ,qBAAqB;AAC/B,WAAK,KAAK,2BAA2B,QAAQ,mBAAmB;AAAA,IAClE;AACA,QAAI,QAAQ,kBAAkB;AAC5B,WAAK,KAAK,YAAY;AAAA,IACxB;AACA,QAAI,QAAQ,cAAc;AACxB,WAAK,KAAK,QAAQ;AAAA,IACpB;AACA,QAAI,QAAQ,sBAAsB;AAChC,WAAK,KAAK,OAAO;AAAA,IACnB;AACA,QAAI,QAAQ,gBAAgB;AAC1B,WAAK,KAAK,oBAAoB;AAAA,IAChC;AACA,QAAI,QAAQ,kBAAkB;AAC5B,WAAK,KAAK,wBAAwB,QAAQ,gBAAgB;AAAA,IAC5D;AACA,QAAI,QAAQ,gBAAgB;AAC1B,WAAK,KAAK,sBAAsB;AAAA,IAClC;AACA,QAAI,QAAQ,oBAAoB;AAC9B,WAAK,KAAK,wBAAwB;AAAA,IACpC;AACA,QAAI,QAAQ,QAAQ;AAClB,WAAK,KAAK,WAAW;AAAA,IACvB;AAEA,SAAK,KAAK,SAAS;AAEnB,UAAM,UAAe,UAAK,WAAW,MAAM,OAAO,QAAQ;AAE1D,QAAI,CAAI,cAAW,OAAO,GAAG;AAC3B,aAAO;AAAA,QACL,IAAI,MAAM,yBAAyB,OAAO,sCAAsC;AAAA,MAClF;AAAA,IACF;AAEA,UAAM,UAAU;AAChB,UAAM,cAAc,CAAC,QAAQ,SAAS,GAAG,IAAI;AAE7C,QAAI,QAAQ,OAAO;AACjB,cAAQ,MAAM,oBAAoB,yBAAyB,SAAS,WAAW,CAAC,EAAE;AAAA,IACpF;AAEA,UAAM,kBAAc,4BAAM,SAAS,WAAW;AAE9C,QAAI,SAAS;AACb,QAAI,SAAS;AAEb,gBAAY,OAAO,GAAG,QAAQ,CAAC,SAAS;AACtC,YAAM,QAAQ,KAAK,SAAS;AAC5B,UAAI,QAAQ,OAAO;AACjB,gBAAQ,OAAO,MAAM,KAAK;AAAA,MAC5B;AACA,gBAAU;AAAA,IACZ,CAAC;AAED,gBAAY,OAAO,GAAG,QAAQ,CAAC,SAAS;AACtC,YAAM,QAAQ,KAAK,SAAS;AAC5B,UAAI,QAAQ,OAAO;AACjB,gBAAQ,OAAO,MAAM,KAAK;AAAA,MAC5B;AACA,gBAAU;AAAA,IACZ,CAAC;AAED,gBAAY,GAAG,SAAS,CAAC,SAAS;AAChC,UAAI,SAAS,GAAG;AACd,gBAAQ,MAAM;AAAA,MAChB,OAAO;AACL,cAAM,QAAQ,IAAI;AAAA,UAChB,+CAA+C,IAAI;AAAA;AAAA,EAAQ,MAAM;AAAA,QACnE;AACA,eAAO,KAAK;AAAA,MACd;AAAA,IACF,CAAC;AAED,gBAAY,GAAG,SAAS,CAAC,QAAQ;AAC/B,UAAI,IAAI,QAAQ,SAAS,QAAQ,GAAG;AAClC;AAAA,UACE,IAAI;AAAA,YACF;AAAA,UACF;AAAA,QACF;AAAA,MACF,OAAO;AACL,eAAO,GAAG;AAAA,MACZ;AAAA,IACF,CAAC;AAAA,EACH,CAAC;AACH;;;AC/HA,SAAS,YAAkB;AACzB,UAAQ,IAAI,6CAA6C;AACzD,UAAQ,IAAI,EAAE;AACd,UAAQ,IAAI,UAAU;AACtB,UAAQ,IAAI,oEAAoE;AAChF,UAAQ,IAAI,4DAA4D;AACxE,UAAQ,IAAI,6EAA6E;AACzF,UAAQ,IAAI,gFAAgF;AAC5F,UAAQ,IAAI,yDAAyD;AACrE,UAAQ,IAAI,qDAAqD;AACjE,UAAQ,IAAI,8DAA8D;AAC1E,UAAQ,IAAI,oEAAoE;AAChF,UAAQ,IAAI,sEAAsE;AAClF,UAAQ,IAAI,gEAAgE;AAC5E,UAAQ,IAAI,+DAA+D;AAC3E,UAAQ,IAAI,0EAA0E;AACtF,UAAQ,IAAI,2DAA2D;AACzE;AAEA,SAAS,UAAU,MAA4B;AAC7C,QAAM,UAAsB,CAAC;AAC7B,MAAI;AACJ,MAAI,WAAW;AAEf,QAAM,YAAY,CAAC,cAAsB,WAAyD;AAChG,UAAM,YAAY,KAAK,eAAe,CAAC;AACvC,QAAI,CAAC,aAAa,UAAU,WAAW,GAAG,GAAG;AAC3C,YAAM,IAAI,MAAM,UAAU,MAAM,oBAAoB;AAAA,IACtD;AACA,WAAO,EAAE,OAAO,WAAW,WAAW,eAAe,EAAE;AAAA,EACzD;AAEA,WAAS,IAAI,GAAG,IAAI,KAAK,QAAQ,KAAK,GAAG;AACvC,UAAM,MAAM,KAAK,CAAC;AAElB,YAAQ,KAAK;AAAA,MACX,KAAK;AAAA,MACL,KAAK;AACH,mBAAW;AACX,YAAI,KAAK;AACT;AAAA,MACF,KAAK;AAAA,MACL,KAAK,MAAM;AACT,cAAM,EAAE,OAAO,UAAU,IAAI,UAAU,GAAG,GAAG;AAC7C,gBAAQ,eAAe;AACvB,YAAI;AACJ;AAAA,MACF;AAAA,MACA,KAAK;AAAA,MACL,KAAK,MAAM;AACT,cAAM,EAAE,OAAO,UAAU,IAAI,UAAU,GAAG,GAAG;AAC7C,gBAAQ,WAAW;AACnB,YAAI;AACJ;AAAA,MACF;AAAA,MACA,KAAK,2BAA2B;AAC9B,cAAM,EAAE,OAAO,UAAU,IAAI,UAAU,GAAG,GAAG;AAC7C,gBAAQ,sBAAsB;AAC9B,YAAI;AACJ;AAAA,MACF;AAAA,MACA,KAAK,wBAAwB;AAC3B,cAAM,EAAE,OAAO,UAAU,IAAI,UAAU,GAAG,GAAG;AAC7C,gBAAQ,mBAAmB;AAC3B,YAAI;AACJ;AAAA,MACF;AAAA,MACA,KAAK;AACH,gBAAQ,mBAAmB;AAC3B;AAAA,MACF,KAAK;AACH,gBAAQ,eAAe;AACvB;AAAA,MACF,KAAK;AACH,gBAAQ,uBAAuB;AAC/B;AAAA,MACF,KAAK;AACH,gBAAQ,iBAAiB;AACzB;AAAA,MACF,KAAK;AACH,gBAAQ,iBAAiB;AACzB;AAAA,MACF,KAAK;AACH,gBAAQ,qBAAqB;AAC7B;AAAA,MACF,KAAK;AACH,gBAAQ,SAAS;AACjB;AAAA,MACF,KAAK;AACH,gBAAQ,QAAQ;AAChB;AAAA,MACF;AACE,YAAI,IAAI,WAAW,GAAG,GAAG;AACvB,gBAAM,IAAI,MAAM,mBAAmB,GAAG,EAAE;AAAA,QAC1C;AAEA,YAAI,WAAW;AACb,gBAAM,IAAI,MAAM,gEAAgE;AAAA,QAClF;AACA,oBAAY;AAAA,IAChB;AAAA,EACF;AAEA,SAAO,EAAE,WAAW,SAAS,SAAS;AACxC;AAEA,eAAe,OAAwB;AACrC,MAAI;AAEJ,MAAI;AACF,aAAS,UAAU,QAAQ,KAAK,MAAM,CAAC,CAAC;AAAA,EAC1C,SAAS,KAAK;AACZ,UAAM,UAAU,eAAe,QAAQ,IAAI,UAAU,OAAO,GAAG;AAC/D,YAAQ,MAAM,OAAO;AACrB,YAAQ,MAAM,wCAAwC;AACtD,WAAO;AAAA,EACT;AAEA,MAAI,OAAO,UAAU;AACnB,cAAU;AACV,WAAO;AAAA,EACT;AAEA,MAAI,CAAC,OAAO,WAAW;AACrB,YAAQ,MAAM,8BAA8B;AAC5C,YAAQ,MAAM,wCAAwC;AACtD,WAAO;AAAA,EACT;AAEA,MAAI;AACF,UAAM,SAAS,MAAM,IAAI,OAAO,WAAW,OAAO,OAAO;AACzD,QAAI,UAAU,CAAC,OAAO,QAAQ,OAAO;AACnC,cAAQ,OAAO,MAAM,MAAM;AAAA,IAC7B;AACA,QAAI,UAAU,CAAC,OAAO,SAAS,IAAI,KAAK,CAAC,OAAO,QAAQ,OAAO;AAC7D,cAAQ,OAAO,MAAM,IAAI;AAAA,IAC3B;AACA,WAAO;AAAA,EACT,SAAS,KAAK;AACZ,UAAM,UAAU,eAAe,QAAQ,IAAI,UAAU,OAAO,GAAG;AAC/D,YAAQ,MAAM,OAAO;AACrB,WAAO;AAAA,EACT;AACF;AAEA,KAAK,EAAE,KAAK,CAAC,SAAS;AACpB,MAAI,SAAS,GAAG;AACd,YAAQ,KAAK,IAAI;AAAA,EACnB;AACF,CAAC;","names":[]}

package/dist/cli.d.cts

@@ -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 +1,252 @@
+#!/usr/bin/env node
+
+// src/index.ts
+import { spawn } from "child_process";
+import * as path from "path";
+import * as fs from "fs";
+import { fileURLToPath } from "url";
+var __filename = fileURLToPath(import.meta.url);
+var __dirname = path.dirname(__filename);
+var JAR_NAME = "opendataloader-pdf-cli.jar";
+function getRedactedCommandString(command, commandArgs) {
+  const commandArgsForLogging = [...commandArgs];
+  const passwordIndex = commandArgsForLogging.indexOf("--password");
+  if (passwordIndex > -1 && passwordIndex + 1 < commandArgsForLogging.length) {
+    commandArgsForLogging[passwordIndex + 1] = "[REDACTED]";
+  }
+  return `${command} ${commandArgsForLogging.join(" ")}`;
+}
+function run(inputPath, options = {}) {
+  return new Promise((resolve, reject) => {
+    if (!fs.existsSync(inputPath)) {
+      return reject(new Error(`Input file or folder not found: ${inputPath}`));
+    }
+    const args = [];
+    if (options.outputFolder) {
+      args.push("--output-dir", options.outputFolder);
+    }
+    if (options.password) {
+      args.push("--password", options.password);
+    }
+    if (options.replaceInvalidChars) {
+      args.push("--replace-invalid-chars", options.replaceInvalidChars);
+    }
+    if (options.generateMarkdown) {
+      args.push("--markdown");
+    }
+    if (options.generateHtml) {
+      args.push("--html");
+    }
+    if (options.generateAnnotatedPdf) {
+      args.push("--pdf");
+    }
+    if (options.keepLineBreaks) {
+      args.push("--keep-line-breaks");
+    }
+    if (options.contentSafetyOff) {
+      args.push("--content-safety-off", options.contentSafetyOff);
+    }
+    if (options.htmlInMarkdown) {
+      args.push("--markdown-with-html");
+    }
+    if (options.addImageToMarkdown) {
+      args.push("--markdown-with-images");
+    }
+    if (options.noJson) {
+      args.push("--no-json");
+    }
+    args.push(inputPath);
+    const jarPath = path.join(__dirname, "..", "lib", JAR_NAME);
+    if (!fs.existsSync(jarPath)) {
+      return reject(
+        new Error(`JAR file not found at ${jarPath}. Please run the build script first.`)
+      );
+    }
+    const command = "java";
+    const commandArgs = ["-jar", jarPath, ...args];
+    if (options.debug) {
+      console.error(`Running command: ${getRedactedCommandString(command, commandArgs)}`);
+    }
+    const javaProcess = spawn(command, commandArgs);
+    let stdout = "";
+    let stderr = "";
+    javaProcess.stdout.on("data", (data) => {
+      const chunk = data.toString();
+      if (options.debug) {
+        process.stdout.write(chunk);
+      }
+      stdout += chunk;
+    });
+    javaProcess.stderr.on("data", (data) => {
+      const chunk = data.toString();
+      if (options.debug) {
+        process.stderr.write(chunk);
+      }
+      stderr += chunk;
+    });
+    javaProcess.on("close", (code) => {
+      if (code === 0) {
+        resolve(stdout);
+      } else {
+        const error = new Error(
+          `The opendataloader-pdf CLI exited with code ${code}.
+
+${stderr}`
+        );
+        reject(error);
+      }
+    });
+    javaProcess.on("error", (err) => {
+      if (err.message.includes("ENOENT")) {
+        reject(
+          new Error(
+            "'java' command not found. Please ensure Java is installed and in your system's PATH."
+          )
+        );
+      } else {
+        reject(err);
+      }
+    });
+  });
+}
+
+// src/cli.ts
+function printHelp() {
+  console.log(`Usage: opendataloader-pdf [options] <input>`);
+  console.log("");
+  console.log("Options:");
+  console.log("  -o, --output-dir <path>      Directory where outputs are written");
+  console.log("  -p, --password <password>    Password for encrypted PDFs");
+  console.log("  --replace-invalid-chars <c>  Replacement character for invalid characters");
+  console.log("  --content-safety-off <mode>  Disable content safety filtering (provide mode)");
+  console.log("  --markdown                   Generate Markdown output");
+  console.log("  --html                       Generate HTML output");
+  console.log("  --pdf                        Generate annotated PDF output");
+  console.log("  --keep-line-breaks           Preserve line breaks in text output");
+  console.log("  --markdown-with-html         Allow raw HTML within Markdown output");
+  console.log("  --markdown-with-images       Embed images in Markdown output");
+  console.log("  --no-json                    Disable JSON output generation");
+  console.log("  --debug                      Stream CLI logs directly to stdout/stderr");
+  console.log("  -h, --help                   Show this message and exit");
+}
+function parseArgs(argv) {
+  const options = {};
+  let inputPath;
+  let showHelp = false;
+  const readValue = (currentIndex, option) => {
+    const nextValue = argv[currentIndex + 1];
+    if (!nextValue || nextValue.startsWith("-")) {
+      throw new Error(`Option ${option} requires a value.`);
+    }
+    return { value: nextValue, nextIndex: currentIndex + 1 };
+  };
+  for (let i = 0; i < argv.length; i += 1) {
+    const arg = argv[i];
+    switch (arg) {
+      case "--help":
+      case "-h":
+        showHelp = true;
+        i = argv.length;
+        break;
+      case "--output-dir":
+      case "-o": {
+        const { value, nextIndex } = readValue(i, arg);
+        options.outputFolder = value;
+        i = nextIndex;
+        break;
+      }
+      case "--password":
+      case "-p": {
+        const { value, nextIndex } = readValue(i, arg);
+        options.password = value;
+        i = nextIndex;
+        break;
+      }
+      case "--replace-invalid-chars": {
+        const { value, nextIndex } = readValue(i, arg);
+        options.replaceInvalidChars = value;
+        i = nextIndex;
+        break;
+      }
+      case "--content-safety-off": {
+        const { value, nextIndex } = readValue(i, arg);
+        options.contentSafetyOff = value;
+        i = nextIndex;
+        break;
+      }
+      case "--markdown":
+        options.generateMarkdown = true;
+        break;
+      case "--html":
+        options.generateHtml = true;
+        break;
+      case "--pdf":
+        options.generateAnnotatedPdf = true;
+        break;
+      case "--keep-line-breaks":
+        options.keepLineBreaks = true;
+        break;
+      case "--markdown-with-html":
+        options.htmlInMarkdown = true;
+        break;
+      case "--markdown-with-images":
+        options.addImageToMarkdown = true;
+        break;
+      case "--no-json":
+        options.noJson = true;
+        break;
+      case "--debug":
+        options.debug = true;
+        break;
+      default:
+        if (arg.startsWith("-")) {
+          throw new Error(`Unknown option: ${arg}`);
+        }
+        if (inputPath) {
+          throw new Error("Multiple input paths provided. Only one input path is allowed.");
+        }
+        inputPath = arg;
+    }
+  }
+  return { inputPath, options, showHelp };
+}
+async function main() {
+  let parsed;
+  try {
+    parsed = parseArgs(process.argv.slice(2));
+  } catch (err) {
+    const message = err instanceof Error ? err.message : String(err);
+    console.error(message);
+    console.error("Use '--help' to see available options.");
+    return 1;
+  }
+  if (parsed.showHelp) {
+    printHelp();
+    return 0;
+  }
+  if (!parsed.inputPath) {
+    console.error("Missing required input path.");
+    console.error("Use '--help' to see usage information.");
+    return 1;
+  }
+  try {
+    const output = await run(parsed.inputPath, parsed.options);
+    if (output && !parsed.options.debug) {
+      process.stdout.write(output);
+    }
+    if (output && !output.endsWith("\n") && !parsed.options.debug) {
+      process.stdout.write("\n");
+    }
+    return 0;
+  } catch (err) {
+    const message = err instanceof Error ? err.message : String(err);
+    console.error(message);
+    return 1;
+  }
+}
+main().then((code) => {
+  if (code !== 0) {
+    process.exit(code);
+  }
+});
+//# sourceMappingURL=cli.js.map

package/dist/cli.js.map

@@ -0,0 +1 @@
+{"version":3,"sources":["../src/index.ts","../src/cli.ts"],"sourcesContent":["import { spawn } from 'child_process';\nimport * as path from 'path';\nimport * as fs from 'fs';\nimport { fileURLToPath } from 'url';\n\nconst __filename = fileURLToPath(import.meta.url);\nconst __dirname = path.dirname(__filename);\n\nconst JAR_NAME = 'opendataloader-pdf-cli.jar';\n\nfunction getRedactedCommandString(command: string, commandArgs: string[]): string {\n  const commandArgsForLogging = [...commandArgs];\n  const passwordIndex = commandArgsForLogging.indexOf('--password');\n  if (passwordIndex > -1 && passwordIndex + 1 < commandArgsForLogging.length) {\n    commandArgsForLogging[passwordIndex + 1] = '[REDACTED]';\n  }\n  return `${command} ${commandArgsForLogging.join(' ')}`;\n}\n\nexport interface RunOptions {\n  outputFolder?: string;\n  password?: string;\n  replaceInvalidChars?: string;\n  generateMarkdown?: boolean;\n  generateHtml?: boolean;\n  generateAnnotatedPdf?: boolean;\n  keepLineBreaks?: boolean;\n  contentSafetyOff?: string;\n  htmlInMarkdown?: boolean;\n  addImageToMarkdown?: boolean;\n  noJson?: boolean;\n  debug?: boolean;\n}\n\nexport function run(inputPath: string, options: RunOptions = {}): Promise<string> {\n  return new Promise((resolve, reject) => {\n    if (!fs.existsSync(inputPath)) {\n      return reject(new Error(`Input file or folder not found: ${inputPath}`));\n    }\n\n    const args: string[] = [];\n    if (options.outputFolder) {\n      args.push('--output-dir', options.outputFolder);\n    }\n    if (options.password) {\n      args.push('--password', options.password);\n    }\n    if (options.replaceInvalidChars) {\n      args.push('--replace-invalid-chars', options.replaceInvalidChars);\n    }\n    if (options.generateMarkdown) {\n      args.push('--markdown');\n    }\n    if (options.generateHtml) {\n      args.push('--html');\n    }\n    if (options.generateAnnotatedPdf) {\n      args.push('--pdf');\n    }\n    if (options.keepLineBreaks) {\n      args.push('--keep-line-breaks');\n    }\n    if (options.contentSafetyOff) {\n      args.push('--content-safety-off', options.contentSafetyOff);\n    }\n    if (options.htmlInMarkdown) {\n      args.push('--markdown-with-html');\n    }\n    if (options.addImageToMarkdown) {\n      args.push('--markdown-with-images');\n    }\n    if (options.noJson) {\n      args.push('--no-json');\n    }\n\n    args.push(inputPath);\n\n    const jarPath = path.join(__dirname, '..', 'lib', JAR_NAME);\n\n    if (!fs.existsSync(jarPath)) {\n      return reject(\n        new Error(`JAR file not found at ${jarPath}. Please run the build script first.`),\n      );\n    }\n\n    const command = 'java';\n    const commandArgs = ['-jar', jarPath, ...args];\n\n    if (options.debug) {\n      console.error(`Running command: ${getRedactedCommandString(command, commandArgs)}`);\n    }\n\n    const javaProcess = spawn(command, commandArgs);\n\n    let stdout = '';\n    let stderr = '';\n\n    javaProcess.stdout.on('data', (data) => {\n      const chunk = data.toString();\n      if (options.debug) {\n        process.stdout.write(chunk);\n      }\n      stdout += chunk;\n    });\n\n    javaProcess.stderr.on('data', (data) => {\n      const chunk = data.toString();\n      if (options.debug) {\n        process.stderr.write(chunk);\n      }\n      stderr += chunk;\n    });\n\n    javaProcess.on('close', (code) => {\n      if (code === 0) {\n        resolve(stdout);\n      } else {\n        const error = new Error(\n          `The opendataloader-pdf CLI exited with code ${code}.\\n\\n${stderr}`,\n        );\n        reject(error);\n      }\n    });\n\n    javaProcess.on('error', (err) => {\n      if (err.message.includes('ENOENT')) {\n        reject(\n          new Error(\n            \"'java' command not found. Please ensure Java is installed and in your system's PATH.\",\n          ),\n        );\n      } else {\n        reject(err);\n      }\n    });\n  });\n}\n","#!/usr/bin/env node\nimport { run, RunOptions } from './index.js';\n\ninterface ParsedArgs {\n  inputPath?: string;\n  options: RunOptions;\n  showHelp: boolean;\n}\n\nfunction printHelp(): void {\n  console.log(`Usage: opendataloader-pdf [options] <input>`);\n  console.log('');\n  console.log('Options:');\n  console.log('  -o, --output-dir <path>      Directory where outputs are written');\n  console.log('  -p, --password <password>    Password for encrypted PDFs');\n  console.log('  --replace-invalid-chars <c>  Replacement character for invalid characters');\n  console.log('  --content-safety-off <mode>  Disable content safety filtering (provide mode)');\n  console.log('  --markdown                   Generate Markdown output');\n  console.log('  --html                       Generate HTML output');\n  console.log('  --pdf                        Generate annotated PDF output');\n  console.log('  --keep-line-breaks           Preserve line breaks in text output');\n  console.log('  --markdown-with-html         Allow raw HTML within Markdown output');\n  console.log('  --markdown-with-images       Embed images in Markdown output');\n  console.log('  --no-json                    Disable JSON output generation');\n  console.log('  --debug                      Stream CLI logs directly to stdout/stderr');\n  console.log('  -h, --help                   Show this message and exit');\n}\n\nfunction parseArgs(argv: string[]): ParsedArgs {\n  const options: RunOptions = {};\n  let inputPath: string | undefined;\n  let showHelp = false;\n\n  const readValue = (currentIndex: number, option: string): { value: string; nextIndex: number } => {\n    const nextValue = argv[currentIndex + 1];\n    if (!nextValue || nextValue.startsWith('-')) {\n      throw new Error(`Option ${option} requires a value.`);\n    }\n    return { value: nextValue, nextIndex: currentIndex + 1 };\n  };\n\n  for (let i = 0; i < argv.length; i += 1) {\n    const arg = argv[i];\n\n    switch (arg) {\n      case '--help':\n      case '-h':\n        showHelp = true;\n        i = argv.length; // exit loop\n        break;\n      case '--output-dir':\n      case '-o': {\n        const { value, nextIndex } = readValue(i, arg);\n        options.outputFolder = value;\n        i = nextIndex;\n        break;\n      }\n      case '--password':\n      case '-p': {\n        const { value, nextIndex } = readValue(i, arg);\n        options.password = value;\n        i = nextIndex;\n        break;\n      }\n      case '--replace-invalid-chars': {\n        const { value, nextIndex } = readValue(i, arg);\n        options.replaceInvalidChars = value;\n        i = nextIndex;\n        break;\n      }\n      case '--content-safety-off': {\n        const { value, nextIndex } = readValue(i, arg);\n        options.contentSafetyOff = value;\n        i = nextIndex;\n        break;\n      }\n      case '--markdown':\n        options.generateMarkdown = true;\n        break;\n      case '--html':\n        options.generateHtml = true;\n        break;\n      case '--pdf':\n        options.generateAnnotatedPdf = true;\n        break;\n      case '--keep-line-breaks':\n        options.keepLineBreaks = true;\n        break;\n      case '--markdown-with-html':\n        options.htmlInMarkdown = true;\n        break;\n      case '--markdown-with-images':\n        options.addImageToMarkdown = true;\n        break;\n      case '--no-json':\n        options.noJson = true;\n        break;\n      case '--debug':\n        options.debug = true;\n        break;\n      default:\n        if (arg.startsWith('-')) {\n          throw new Error(`Unknown option: ${arg}`);\n        }\n\n        if (inputPath) {\n          throw new Error('Multiple input paths provided. Only one input path is allowed.');\n        }\n        inputPath = arg;\n    }\n  }\n\n  return { inputPath, options, showHelp };\n}\n\nasync function main(): Promise<number> {\n  let parsed: ParsedArgs;\n\n  try {\n    parsed = parseArgs(process.argv.slice(2));\n  } catch (err) {\n    const message = err instanceof Error ? err.message : String(err);\n    console.error(message);\n    console.error(\"Use '--help' to see available options.\");\n    return 1;\n  }\n\n  if (parsed.showHelp) {\n    printHelp();\n    return 0;\n  }\n\n  if (!parsed.inputPath) {\n    console.error('Missing required input path.');\n    console.error(\"Use '--help' to see usage information.\");\n    return 1;\n  }\n\n  try {\n    const output = await run(parsed.inputPath, parsed.options);\n    if (output && !parsed.options.debug) {\n      process.stdout.write(output);\n    }\n    if (output && !output.endsWith('\\n') && !parsed.options.debug) {\n      process.stdout.write('\\n');\n    }\n    return 0;\n  } catch (err) {\n    const message = err instanceof Error ? err.message : String(err);\n    console.error(message);\n    return 1;\n  }\n}\n\nmain().then((code) => {\n  if (code !== 0) {\n    process.exit(code);\n  }\n});\n"],"mappings":";;;AAAA,SAAS,aAAa;AACtB,YAAY,UAAU;AACtB,YAAY,QAAQ;AACpB,SAAS,qBAAqB;AAE9B,IAAM,aAAa,cAAc,YAAY,GAAG;AAChD,IAAM,YAAiB,aAAQ,UAAU;AAEzC,IAAM,WAAW;AAEjB,SAAS,yBAAyB,SAAiB,aAA+B;AAChF,QAAM,wBAAwB,CAAC,GAAG,WAAW;AAC7C,QAAM,gBAAgB,sBAAsB,QAAQ,YAAY;AAChE,MAAI,gBAAgB,MAAM,gBAAgB,IAAI,sBAAsB,QAAQ;AAC1E,0BAAsB,gBAAgB,CAAC,IAAI;AAAA,EAC7C;AACA,SAAO,GAAG,OAAO,IAAI,sBAAsB,KAAK,GAAG,CAAC;AACtD;AAiBO,SAAS,IAAI,WAAmB,UAAsB,CAAC,GAAoB;AAChF,SAAO,IAAI,QAAQ,CAAC,SAAS,WAAW;AACtC,QAAI,CAAI,cAAW,SAAS,GAAG;AAC7B,aAAO,OAAO,IAAI,MAAM,mCAAmC,SAAS,EAAE,CAAC;AAAA,IACzE;AAEA,UAAM,OAAiB,CAAC;AACxB,QAAI,QAAQ,cAAc;AACxB,WAAK,KAAK,gBAAgB,QAAQ,YAAY;AAAA,IAChD;AACA,QAAI,QAAQ,UAAU;AACpB,WAAK,KAAK,cAAc,QAAQ,QAAQ;AAAA,IAC1C;AACA,QAAI,QAAQ,qBAAqB;AAC/B,WAAK,KAAK,2BAA2B,QAAQ,mBAAmB;AAAA,IAClE;AACA,QAAI,QAAQ,kBAAkB;AAC5B,WAAK,KAAK,YAAY;AAAA,IACxB;AACA,QAAI,QAAQ,cAAc;AACxB,WAAK,KAAK,QAAQ;AAAA,IACpB;AACA,QAAI,QAAQ,sBAAsB;AAChC,WAAK,KAAK,OAAO;AAAA,IACnB;AACA,QAAI,QAAQ,gBAAgB;AAC1B,WAAK,KAAK,oBAAoB;AAAA,IAChC;AACA,QAAI,QAAQ,kBAAkB;AAC5B,WAAK,KAAK,wBAAwB,QAAQ,gBAAgB;AAAA,IAC5D;AACA,QAAI,QAAQ,gBAAgB;AAC1B,WAAK,KAAK,sBAAsB;AAAA,IAClC;AACA,QAAI,QAAQ,oBAAoB;AAC9B,WAAK,KAAK,wBAAwB;AAAA,IACpC;AACA,QAAI,QAAQ,QAAQ;AAClB,WAAK,KAAK,WAAW;AAAA,IACvB;AAEA,SAAK,KAAK,SAAS;AAEnB,UAAM,UAAe,UAAK,WAAW,MAAM,OAAO,QAAQ;AAE1D,QAAI,CAAI,cAAW,OAAO,GAAG;AAC3B,aAAO;AAAA,QACL,IAAI,MAAM,yBAAyB,OAAO,sCAAsC;AAAA,MAClF;AAAA,IACF;AAEA,UAAM,UAAU;AAChB,UAAM,cAAc,CAAC,QAAQ,SAAS,GAAG,IAAI;AAE7C,QAAI,QAAQ,OAAO;AACjB,cAAQ,MAAM,oBAAoB,yBAAyB,SAAS,WAAW,CAAC,EAAE;AAAA,IACpF;AAEA,UAAM,cAAc,MAAM,SAAS,WAAW;AAE9C,QAAI,SAAS;AACb,QAAI,SAAS;AAEb,gBAAY,OAAO,GAAG,QAAQ,CAAC,SAAS;AACtC,YAAM,QAAQ,KAAK,SAAS;AAC5B,UAAI,QAAQ,OAAO;AACjB,gBAAQ,OAAO,MAAM,KAAK;AAAA,MAC5B;AACA,gBAAU;AAAA,IACZ,CAAC;AAED,gBAAY,OAAO,GAAG,QAAQ,CAAC,SAAS;AACtC,YAAM,QAAQ,KAAK,SAAS;AAC5B,UAAI,QAAQ,OAAO;AACjB,gBAAQ,OAAO,MAAM,KAAK;AAAA,MAC5B;AACA,gBAAU;AAAA,IACZ,CAAC;AAED,gBAAY,GAAG,SAAS,CAAC,SAAS;AAChC,UAAI,SAAS,GAAG;AACd,gBAAQ,MAAM;AAAA,MAChB,OAAO;AACL,cAAM,QAAQ,IAAI;AAAA,UAChB,+CAA+C,IAAI;AAAA;AAAA,EAAQ,MAAM;AAAA,QACnE;AACA,eAAO,KAAK;AAAA,MACd;AAAA,IACF,CAAC;AAED,gBAAY,GAAG,SAAS,CAAC,QAAQ;AAC/B,UAAI,IAAI,QAAQ,SAAS,QAAQ,GAAG;AAClC;AAAA,UACE,IAAI;AAAA,YACF;AAAA,UACF;AAAA,QACF;AAAA,MACF,OAAO;AACL,eAAO,GAAG;AAAA,MACZ;AAAA,IACF,CAAC;AAAA,EACH,CAAC;AACH;;;AC/HA,SAAS,YAAkB;AACzB,UAAQ,IAAI,6CAA6C;AACzD,UAAQ,IAAI,EAAE;AACd,UAAQ,IAAI,UAAU;AACtB,UAAQ,IAAI,oEAAoE;AAChF,UAAQ,IAAI,4DAA4D;AACxE,UAAQ,IAAI,6EAA6E;AACzF,UAAQ,IAAI,gFAAgF;AAC5F,UAAQ,IAAI,yDAAyD;AACrE,UAAQ,IAAI,qDAAqD;AACjE,UAAQ,IAAI,8DAA8D;AAC1E,UAAQ,IAAI,oEAAoE;AAChF,UAAQ,IAAI,sEAAsE;AAClF,UAAQ,IAAI,gEAAgE;AAC5E,UAAQ,IAAI,+DAA+D;AAC3E,UAAQ,IAAI,0EAA0E;AACtF,UAAQ,IAAI,2DAA2D;AACzE;AAEA,SAAS,UAAU,MAA4B;AAC7C,QAAM,UAAsB,CAAC;AAC7B,MAAI;AACJ,MAAI,WAAW;AAEf,QAAM,YAAY,CAAC,cAAsB,WAAyD;AAChG,UAAM,YAAY,KAAK,eAAe,CAAC;AACvC,QAAI,CAAC,aAAa,UAAU,WAAW,GAAG,GAAG;AAC3C,YAAM,IAAI,MAAM,UAAU,MAAM,oBAAoB;AAAA,IACtD;AACA,WAAO,EAAE,OAAO,WAAW,WAAW,eAAe,EAAE;AAAA,EACzD;AAEA,WAAS,IAAI,GAAG,IAAI,KAAK,QAAQ,KAAK,GAAG;AACvC,UAAM,MAAM,KAAK,CAAC;AAElB,YAAQ,KAAK;AAAA,MACX,KAAK;AAAA,MACL,KAAK;AACH,mBAAW;AACX,YAAI,KAAK;AACT;AAAA,MACF,KAAK;AAAA,MACL,KAAK,MAAM;AACT,cAAM,EAAE,OAAO,UAAU,IAAI,UAAU,GAAG,GAAG;AAC7C,gBAAQ,eAAe;AACvB,YAAI;AACJ;AAAA,MACF;AAAA,MACA,KAAK;AAAA,MACL,KAAK,MAAM;AACT,cAAM,EAAE,OAAO,UAAU,IAAI,UAAU,GAAG,GAAG;AAC7C,gBAAQ,WAAW;AACnB,YAAI;AACJ;AAAA,MACF;AAAA,MACA,KAAK,2BAA2B;AAC9B,cAAM,EAAE,OAAO,UAAU,IAAI,UAAU,GAAG,GAAG;AAC7C,gBAAQ,sBAAsB;AAC9B,YAAI;AACJ;AAAA,MACF;AAAA,MACA,KAAK,wBAAwB;AAC3B,cAAM,EAAE,OAAO,UAAU,IAAI,UAAU,GAAG,GAAG;AAC7C,gBAAQ,mBAAmB;AAC3B,YAAI;AACJ;AAAA,MACF;AAAA,MACA,KAAK;AACH,gBAAQ,mBAAmB;AAC3B;AAAA,MACF,KAAK;AACH,gBAAQ,eAAe;AACvB;AAAA,MACF,KAAK;AACH,gBAAQ,uBAAuB;AAC/B;AAAA,MACF,KAAK;AACH,gBAAQ,iBAAiB;AACzB;AAAA,MACF,KAAK;AACH,gBAAQ,iBAAiB;AACzB;AAAA,MACF,KAAK;AACH,gBAAQ,qBAAqB;AAC7B;AAAA,MACF,KAAK;AACH,gBAAQ,SAAS;AACjB;AAAA,MACF,KAAK;AACH,gBAAQ,QAAQ;AAChB;AAAA,MACF;AACE,YAAI,IAAI,WAAW,GAAG,GAAG;AACvB,gBAAM,IAAI,MAAM,mBAAmB,GAAG,EAAE;AAAA,QAC1C;AAEA,YAAI,WAAW;AACb,gBAAM,IAAI,MAAM,gEAAgE;AAAA,QAClF;AACA,oBAAY;AAAA,IAChB;AAAA,EACF;AAEA,SAAO,EAAE,WAAW,SAAS,SAAS;AACxC;AAEA,eAAe,OAAwB;AACrC,MAAI;AAEJ,MAAI;AACF,aAAS,UAAU,QAAQ,KAAK,MAAM,CAAC,CAAC;AAAA,EAC1C,SAAS,KAAK;AACZ,UAAM,UAAU,eAAe,QAAQ,IAAI,UAAU,OAAO,GAAG;AAC/D,YAAQ,MAAM,OAAO;AACrB,YAAQ,MAAM,wCAAwC;AACtD,WAAO;AAAA,EACT;AAEA,MAAI,OAAO,UAAU;AACnB,cAAU;AACV,WAAO;AAAA,EACT;AAEA,MAAI,CAAC,OAAO,WAAW;AACrB,YAAQ,MAAM,8BAA8B;AAC5C,YAAQ,MAAM,wCAAwC;AACtD,WAAO;AAAA,EACT;AAEA,MAAI;AACF,UAAM,SAAS,MAAM,IAAI,OAAO,WAAW,OAAO,OAAO;AACzD,QAAI,UAAU,CAAC,OAAO,QAAQ,OAAO;AACnC,cAAQ,OAAO,MAAM,MAAM;AAAA,IAC7B;AACA,QAAI,UAAU,CAAC,OAAO,SAAS,IAAI,KAAK,CAAC,OAAO,QAAQ,OAAO;AAC7D,cAAQ,OAAO,MAAM,IAAI;AAAA,IAC3B;AACA,WAAO;AAAA,EACT,SAAS,KAAK;AACZ,UAAM,UAAU,eAAe,QAAQ,IAAI,UAAU,OAAO,GAAG;AAC/D,YAAQ,MAAM,OAAO;AACrB,WAAO;AAAA,EACT;AACF;AAEA,KAAK,EAAE,KAAK,CAAC,SAAS;AACpB,MAAI,SAAS,GAAG;AACd,YAAQ,KAAK,IAAI;AAAA,EACnB;AACF,CAAC;","names":[]}

package/package.json

@@ -1,11 +1,14 @@
 {
   "name": "@opendataloader/pdf",
-  "version": "1.0.5",
+  "version": "1.1.0",
   "description": "A Node.js wrapper for the opendataloader-pdf Java CLI.",
   "main": "./dist/index.cjs",
   "module": "./dist/index.js",
   "types": "./dist/index.d.ts",
   "type": "module",
+  "bin": {
+    "opendataloader-pdf": "./dist/cli.js"
+  },
   "exports": {
     ".": {
       "import": "./dist/index.js",