import Foundation
import Vision
import AppKit

struct OcrLine: Encodable {
  let text: String
  let x: Double
  let y: Double
  let width: Double
  let height: Double
}

let args = CommandLine.arguments
guard args.count >= 2 else {
  fputs("usage: swift ocr_image_text.swift <image>\\n", stderr)
  exit(2)
}

let imageURL = URL(fileURLWithPath: args[1])
guard let image = NSImage(contentsOf: imageURL),
      let cgImage = image.cgImage(forProposedRect: nil, context: nil, hints: nil) else {
  fputs("failed to load image: \\(args[1])\\n", stderr)
  exit(1)
}

let request = VNRecognizeTextRequest()
request.recognitionLevel = .accurate
request.usesLanguageCorrection = true
request.recognitionLanguages = ["zh-Hans", "en-US"]
request.minimumTextHeight = 0.006

let handler = VNImageRequestHandler(cgImage: cgImage, options: [:])
try handler.perform([request])

let observations = request.results ?? []
let lines = observations.compactMap { observation -> OcrLine? in
  guard let candidate = observation.topCandidates(1).first else { return nil }
  let box = observation.boundingBox
  return OcrLine(
    text: candidate.string,
    x: Double(box.minX),
    y: Double(box.minY),
    width: Double(box.width),
    height: Double(box.height)
  )
}.sorted {
  let y0 = Int($0.y * 1000)
  let y1 = Int($1.y * 1000)
  if y0 != y1 { return y0 > y1 }
  return $0.x < $1.x
}

let data = try JSONEncoder().encode(lines)
FileHandle.standardOutput.write(data)