Coverage Summary for Class: GhostServerExtensionsKt (com.ghost.serialization.ktor)

 package com.ghost.serialization.ktor
 
 import com.ghost.protobuf.GhostProtobuf
 import com.ghost.serialization.Ghost
 import io.ktor.http.ContentType
 import io.ktor.http.HttpStatusCode
 import io.ktor.http.content.ByteArrayContent
 import io.ktor.server.application.ApplicationCall
 import io.ktor.server.response.respond
 import kotlin.reflect.KClass
 
 @PublishedApi
 internal const val ERROR_PREFIX = "Ghost serializer not found for class "
 @PublishedApi
 internal const val ERROR_SUFFIX = ". Make sure it is annotated with @GhostSerialization."
 
 /**
  * Serializes the [value] directly using Ghost Serializer and responds,
  * bypassing the ContentNegotiation pipeline entirely to maximize throughput.
  */
 suspend inline fun <reified T : Any> ApplicationCall.respondGhost(
     value: T,
     status: HttpStatusCode = HttpStatusCode.OK
 ) {
     val serializer = Ghost.getSerializer(T::class)
         ?: throw IllegalArgumentException("$ERROR_PREFIX${T::class.simpleName}$ERROR_SUFFIX")
 
     val bytes = Ghost.encodeToBytes(serializer, value)
     respond(ByteArrayContent(bytes, ContentType.Application.Json, status))
 }
 
 /**
  * Proto3-JSON variant of [respondGhost] — named for discoverability alongside [respondGhost]
  * and `bodyGhostProto`. Encoding is identical either way: proto3 wire correctness (int64
  * quoting, Base64 bytes, default-value omission) is generated into the
  * `@GhostProtoSerialization` serializer's own `serialize()` method, not a separate writer.
  */
 suspend inline fun <reified T : Any> ApplicationCall.respondGhostProto(
     value: T,
     status: HttpStatusCode = HttpStatusCode.OK
 ) {
     val bytes = GhostProtobuf.encodeToBytes(value)
     respond(ByteArrayContent(bytes, ContentType.Application.Json, status))
 }