@ragaeeb/ocr-js NPM

GitHub npm npm GitHub issues GitHub Release

The ocr-js project simplifies the process of performing OCR on documents sent to 3rd party services like Google Document AI.

ocr-js Usage Guide

This guide explains how to use the exported functions from ocr-js.

Installation

npm i @ragaeeb/ocr-js

Prerequisites

Node.js (v20 or higher) Google Cloud account with Document AI and Cloud Storage enabled Service account credentials with necessary permissions.

Configuration

Before using the functions, you need to initialize the configuration:

import { init } from './index';

init({
    bucketUri: 'gs://your-bucket-name',
    processorId: 'your-processor-id',
    processorVersion: 'your-processor-version',
    projectId: 'your-project-id',
    projectLocation: 'your-project-location',
});

bucketUri: The URI of your Google Cloud Storage bucket.
processorId: The ID of your Document AI processor.
processorVersion: The version of your processor.
projectId: Your Google Cloud project ID.
projectLocation: The location of your processor (e.g., us-central1).

Usage

Initialization

Initialize the configuration as shown above before calling any other functions.

Requesting OCR

To request OCR processing for a PDF file:

import { requestOCR } from './index';

const pdfFile = 'path/to/your/file.pdf'; const options = { language: 'en', // Specify language hints };

requestOCR(pdfFile, options) .then((result) => { console.log('OCR request successful:', result); }) .catch((error) => { console.error('Error requesting OCR:', error); });

Retrieving OCR Results

After the OCR processing is complete, retrieve the results:

import { getOCRResult } from './index';

const requestId = 'your-request-id'; // Use the requestId from the requestOCR result

getOCRResult(requestId)
    .then((pages) => {
        console.log('OCR results:', pages);
    })
    .catch((error) => {
        console.error('Error retrieving OCR results:', error);
    });