CV PDF to JSON

Extract and process CV data from PDF files with Claude AI's native PDF support. This library provides a robust pipeline for converting PDF resumes into structured JSON data.

Features

• Direct PDF processing using Claude AI's native PDF support
• Structured JSON output of CV data
• Support for processing single PDF file or directory of PDF files
• Support for saving structured JSON outputs
• Debug mode for detailed processing insights

Installation

npm install @racsodev/cv-pdf-to-json

Basic Usage

import { createPdfExtractor } from '@racsodev/cv-pdf-to-json'
import path from 'path'

// Initialize the PDF extractor
const extractor = createPdfExtractor({
  anthropicApiKey: process.env.ANTHROPIC_API_KEY || '',
  outputJsonPath: './outputs/json',
})

// Process a single file
const result = await extractor.process('path/to/cv.pdf')
console.log('Processing Result:', result)

// Process a directory
const results = await extractor.processDirectory('path/to/cvs')
console.log('Processing Results:', results)

Advanced Usage

For more control over the extraction process, you can use individual components:

import {
  DocumentProcessor,
  ClaudeProcessor,
  type CvData,
  type Experience,
  type Education,
  type Language,
  ContractType,
  LanguageLevel,
} from '@racsodev/cv-pdf-to-json'

// Initialize Claude AI processor with native PDF support
const processor = new ClaudeProcessor({
  apiKey: process.env.ANTHROPIC_API_KEY || '',
})

// Create document processor
const documentProcessor = new DocumentProcessor({
  processor,
  outputJsonPath: './outputs/json',
  debug: true,
})

// Process CV
async function processCV(pdfPath: string) {
  const result = await documentProcessor.process(pdfPath)

  if (result.success && result.data) {
    const cvData: CvData = result.data
    console.log('Extracted CV Data:', cvData)
  }

  return result
}

// Use the processor
const result = await processCV('path/to/cv.pdf')

Output Format

The processor returns data in the following format:

interface CvData {
  lastName: string
  firstName: string
  address: string
  email: string
  phone: string
  linkedin: string
  github: string
  personalWebsite: string
  professionalSummary: string
  school: string
  schoolLowerCase: string
  promotionYear: number
  professionalExperiences: Experience[]
  otherExperiences: Experience[]
  educations: Education[]
  hardSkills: string[]
  softSkills: string[]
  languages: Language[]
  publications: string[]
  distinctions: string[]
  hobbies: string[]
  references: string[]
}

interface Experience {
  companyName?: string
  title?: string
  location: string
  type: ContractType
  startDate: number
  endDate: number
  duration: number // in months
  ongoing: boolean
  description: string
  associatedSkills: string[]
}

interface Education {
  degree: string
  institution: string
  location: string
  startDate: number
  endDate: number
  duration: number // in months
  ongoing: boolean
  description: string
  associatedSkills: string[]
}

interface Language {
  language: string
  level: LanguageLevel
}

enum LanguageLevel {
  BASIC_KNOWLEDGE = 'BASIC_KNOWLEDGE',
  LIMITED_PROFESSIONAL = 'LIMITED_PROFESSIONAL',
  PROFESSIONAL = 'PROFESSIONAL',
  FULL_PROFESSIONAL = 'FULL_PROFESSIONAL',
  NATIVE_BILINGUAL = 'NATIVE_BILINGUAL',
}

enum ContractType {
  PERMANENT_CONTRACT = 'PERMANENT_CONTRACT',
  SELF_EMPLOYED = 'SELF_EMPLOYED',
  FREELANCE = 'FREELANCE',
  FIXED_TERM_CONTRACT = 'FIXED_TERM_CONTRACT',
  INTERNSHIP = 'INTERNSHIP',
  APPRENTICESHIP = 'APPRENTICESHIP',
  PERFORMING_ARTS_INTERMITTENT = 'PERFORMING_ARTS_INTERMITTENT',
  PART_TIME_PERMANENT = 'PART_TIME_PERMANENT',
  CIVIC_SERVICE = 'CIVIC_SERVICE',
  PART_TIME_FIXED_TERM = 'PART_TIME_FIXED_TERM',
  SUPPORTED_EMPLOYMENT = 'SUPPORTED_EMPLOYMENT',
  CIVIL_SERVANT = 'CIVIL_SERVANT',
  TEMPORARY_WORKER = 'TEMPORARY_WORKER',
  ASSOCIATIVE = 'ASSOCIATIVE',
}

Development Setup

1. Install dependencies:

npm install

2. Set up environment variables:

# Copy the example env file
cp .env.example .env

# Edit .env and add your Anthropic API key
ANTHROPIC_API_KEY=your_api_key_here

3. Process documents:

npm run process <file-path>

This will process the specified PDF file or directory and generate JSON outputs in the configured directory.

Project Structure

• src/ - Source code directory
  • processors/ - Document processing pipeline
    • Processor.ts - Base processor class
    • DocumentProcessor.ts - Main document processing logic
    • ClaudeProcessor.ts - Claude AI integration with native PDF support
  • types/ - TypeScript type definitions
  • utils/ - Utility functions for data processing and file handling

Requirements

• Node.js >= 18.4.2
• Anthropic API key for Claude AI integration

License

Apache-2.0