import {z} from 'zod'

export type DeployPretrainedModelInputTimeSeries = {inputType: 'time-series', frequencyHz: number, windowLengthMs: number};

export type DeployPretrainedModelInputAudio = {inputType: 'audio', frequencyHz: number};

export type ImageInputScaling = '0..1' | '-1..1' | '-128..127' | '0..255' | 'torch' | 'bgr-subtract-imagenet-mean';

export type ImageInputResizeMode = 'squash' | 'fit-short' | 'fit-long' | 'crop';

export type DeployPretrainedModelInputImage = {inputType: 'image', inputScaling?: ImageInputScaling | undefined, resizeMode?: ImageInputResizeMode | undefined};

export type DeployPretrainedModelInputOther = {inputType: 'other'};

export type DeployPretrainedModelModelClassification = {modelType: 'classification', labels: Array<string>, addSoftmax?: boolean | undefined};

export type DeployPretrainedModelModelRegression = {modelType: 'regression'};

export type ObjectDetectionLastLayer = 'mobilenet-ssd' | 'fomo' | 'yolov2-akida' | 'yolov5' | 'yolov5v5-drpai' | 'yolox' | 'yolov7' | 'yolo-pro' | 'tao-retinanet' | 'tao-ssd' | 'tao-yolov3' | 'tao-yolov4' | 'yolov11' | 'yolov11-abs' | 'paddleocr-detector' | 'qc-face-det-lite' | 'qc-yolox';

export type DeployPretrainedModelModelObjectDetection = {modelType: 'object-detection', labels: Array<string>, lastLayer: ObjectDetectionLastLayer, minimumConfidence: number, thresholdValues?: {[key: string]: number} | undefined};

export type DeployPretrainedModelModelFreeform = {modelType: 'freeform'};

export type DeployPretrainedModelModelAnomaly = {modelType: 'anomaly'};

export type ThresholdValue = number | string;

export type DeployPretrainedModelModelVisualAnomaly = {modelType: 'visual-anomaly', thresholdValues?: {[key: string]: ThresholdValue} | undefined};

export type TestPretrainedModelRequest = {features: Array<number>, modelInfo: {input: DeployPretrainedModelInputTimeSeries | DeployPretrainedModelInputAudio | DeployPretrainedModelInputImage | DeployPretrainedModelInputOther, model: DeployPretrainedModelModelClassification | DeployPretrainedModelModelRegression | DeployPretrainedModelModelObjectDetection | DeployPretrainedModelModelFreeform | DeployPretrainedModelModelAnomaly | DeployPretrainedModelModelVisualAnomaly}};

export type CreateApiApiProjectIdPretrainedModelTestArgs = {projectId: number, impulseId?: number | undefined, body: TestPretrainedModelRequest};

export type CreateApiApiProjectIdPretrainedModelTestOptions = {baseUrl?: string | undefined, ApiKeyAuthentication?: string | undefined, fetch?: typeof fetch | undefined};

export const boundingBoxWithScore = z.object({label: z.string(), x: z.number(), y: z.number(), width: z.number(), height: z.number(), score: z.number()});

export const anomalyResult = z.object({boxes: z.array(boundingBoxWithScore).optional(), scores: z.array(z.array(z.number())).optional(), meanScore: z.number().optional(), maxScore: z.number().optional()});

export const testPretrainedModelResponse = z.object({success: z.boolean(), error: z.string().optional(), result: z.record(z.string(), z.number()).optional(), boundingBoxes: z.array(boundingBoxWithScore).optional(), freeformResult: z.object({outputTensors: z.array(z.object({shape: z.array(z.number().int()), dataType: z.enum(["int8", "uint8", "float32"]), data: z.array(z.number())}))}).optional(), anomalyResult: z.array(anomalyResult).optional()});

export const validateResponse = async <Output>(
  schema: {
    readonly '~standard': {
      readonly validate: (value: unknown) => ({ readonly value: Output; readonly issues?: undefined } | { readonly issues: ReadonlyArray<{ readonly message: string; readonly path?: ReadonlyArray<PropertyKey | { readonly key: PropertyKey }> | undefined }> }) | Promise<{ readonly value: Output; readonly issues?: undefined } | { readonly issues: ReadonlyArray<{ readonly message: string; readonly path?: ReadonlyArray<PropertyKey | { readonly key: PropertyKey }> | undefined }> }>
    }
  },
  value: unknown
): Promise<Output> => {
  const result = await schema['~standard'].validate(value)

  if (result.issues !== undefined) {
    const detail = result.issues
      .map(({ message, path }) =>
        path?.length
          ? `${path.map((segment) => String(typeof segment === 'object' ? segment.key : segment)).join('.')}: ${message}`
          : message
      )
      .join('; ')

    throw new Error(`Response does not match the schema: ${detail}`, { cause: { issues: result.issues } })
  }

  return result.value
};

/**
 * Test pretrained model
 * 
 * POST /api/{projectId}/pretrained-model/test
 * 
 * Also accepts auth: JWTAuthentication, JWTHttpHeaderAuthentication, OAuth2
 */
export const createApiApiProjectIdPretrainedModelTest = async (args: CreateApiApiProjectIdPretrainedModelTestArgs, options?: CreateApiApiProjectIdPretrainedModelTestOptions) => {
const {baseUrl = "https://studio.edgeimpulse.com/v1", fetch: fetchFn = fetch, ApiKeyAuthentication} = options ?? {}

const query = new URLSearchParams()
const headers = new Headers()
headers.set('content-type', "application/json")
headers.set('accept', "application/json")

if (args.impulseId != null) query.append("impulseId", String(args.impulseId))
if (ApiKeyAuthentication !== undefined) headers.set("x-api-key", ApiKeyAuthentication)

const url = `${baseUrl.replace(/\/+$/, '')}/api/${encodeURIComponent(String(args.projectId))}/pretrained-model/test${query.size > 0 ? `?${query}` : ''}`

const res = await fetchFn(url, {method: "POST", headers, body: JSON.stringify(args.body)})

if (!res.ok) throw new Error(`Request failed with status ${res.status}`, {cause: {status: res.status, body: await res.text()}})

return validateResponse(testPretrainedModelResponse, await res.json())
};