OCRController after file check and spell correction

OCRController.kt

import com.google.cloud.vision.v1.AnnotateImageRequest
import com.google.cloud.vision.v1.Feature
import com.google.cloud.vision.v1.Image
import com.google.cloud.vision.v1.ImageAnnotatorClient
import com.google.cloud.vision.v1.TextAnnotation
import com.google.protobuf.ByteString
import com.google.cloud.language.v1.LanguageServiceClient
import com.google.cloud.language.v1.Document
import com.google.cloud.language.v1.AnalyzeSyntaxRequest
import org.springframework.beans.factory.annotation.Value
import org.springframework.http.HttpStatus
import org.springframework.http.MediaType
import org.springframework.http.ResponseEntity
import org.springframework.validation.annotation.Validated
import org.springframework.web.bind.annotation.PostMapping
import org.springframework.web.bind.annotation.RequestMapping
import org.springframework.web.bind.annotation.RequestPart
import org.springframework.web.bind.annotation.RestController
import org.springframework.web.multipart.MultipartFile
import java.io.FileInputStream
import javax.validation.constraints.NotNull

@RestController
@RequestMapping("/api/ocr")
@Validated
class OCRController {
    @Value("\${google.cloud.credentials.path}")
    private lateinit var credentialsPath: String

    @Value("\${google.cloud.project-id}")
    private lateinit var projectId: String

    @PostMapping(consumes = [MediaType.MULTIPART_FORM_DATA_VALUE])
    fun ocr(@NotNull @RequestPart("file") file: MultipartFile): ResponseEntity<String> {
        if (!isSupportedImageType(file)) {
            return ResponseEntity.status(HttpStatus.BAD_REQUEST).body("Unsupported image type. Only PNG and JPEG images are allowed.")
        }

        val imageBytes = ByteString.copyFrom(file.bytes)
        val image = Image.newBuilder().setContent(imageBytes).build()

        val settings = ImageAnnotatorSettings.newBuilder()
            .setCredentialsProvider { FixedCredentialsProvider.create(ServiceAccountCredentials.fromStream(FileInputStream(credentialsPath))) }
            .build()

        ImageAnnotatorClient.create(settings).use { imageAnnotatorClient ->
            val imageRequest = AnnotateImageRequest.newBuilder()
                .addFeatures(Feature.newBuilder().setType(Feature.Type.TEXT_DETECTION))
                .setImage(image)
                .build()

            val response = imageAnnotatorClient.batchAnnotateImages(listOf(imageRequest))

            if (response.responsesCount > 0) {
                val annotation = response.responsesList[0].textAnnotationsList[0]
                val correctedText = correctSpelling(annotation.description)
                return ResponseEntity.ok(correctedText)
            }
        }

        return ResponseEntity.status(HttpStatus.BAD_REQUEST).body("OCR failed: No response received")
    }

    private fun isSupportedImageType(file: MultipartFile): Boolean {
        val contentType = file.contentType
        return contentType == MediaType.IMAGE_PNG_VALUE || contentType == MediaType.IMAGE_JPEG_VALUE
    }

    private fun correctSpelling(text: String): String {
        LanguageServiceClient.create().use { languageServiceClient ->
            val document = Document.newBuilder()
                .setContent(text)
                .setType(Document.Type.PLAIN_TEXT)
                .build()

            val syntaxRequest = AnalyzeSyntaxRequest.newBuilder()
                .setDocument(document)
                .setEncodingType(AnalyzeSyntaxRequest.EncodingType.UTF16)
                .build()

            val response = languageServiceClient.analyzeSyntax(syntaxRequest)

            val correctedText = StringBuilder(text)

            response.tokensList.forEach { token ->
                if (token.partOfSpeech.tag == "VERB" && token.lemma.isNotBlank()) {
                    val startIndex = token.text.beginOffset
                    val endIndex = startIndex + token.text.content.length
                    correctedText.replace(startIndex, endIndex, token.lemma)
                }
            }

            return correctedText.toString()
        }
    }
}