import Tesseract from 'tesseract.js'
import path from 'path'

export default defineEventHandler(async (event) => {
  const body = await readBody(event)
  const imageData = body?.image

  if (!imageData) {
    throw createError({
      statusCode: 400,
      statusMessage: 'Image data is required'
    })
  }

  // Validate base64 image format
  if (!imageData.startsWith('data:image/')) {
    throw createError({
      statusCode: 400,
      statusMessage: 'Invalid image format. Expected base64 data URL.'
    })
  }

  try {
    // Configure tesseract to use node_modules directly (no copying needed)
    const workerPath = path.resolve(process.cwd(), 'node_modules/tesseract.js/dist/worker.min.js')
    const corePath = path.resolve(process.cwd(), 'node_modules/tesseract.js-core')

    // Process OCR on server with better performance
    const result = await Tesseract.recognize(imageData, 'deu+eng', {
      workerPath,
      corePath,
      // Server-side processing - no logger needed
    })

    const text = result.data.text || ''

    // Split into lines and filter empty/short ones
    const lines = text.split('\n')
      .map(line => line.trim())
      .filter(line => line.length > 2)

    if (lines.length === 0) {
      return {
        success: false,
        message: 'No text detected in image',
        lines: []
      }
    }

    return {
      success: true,
      lines: lines,
      fullText: text,
      confidence: result.data.confidence || 0
    }
  } catch (error: any) {
    console.error('OCR processing error:', error)
    throw createError({
      statusCode: 500,
      statusMessage: 'OCR processing failed: ' + (error?.message || 'Unknown error')
    })
  }
})