import {z} from 'zod'

export type TableParsingStrategy = 'tsr' | 'vlm' | 'unknown';

export type TableOutputMode = 'markdown' | 'html';

export type FormDetectionMode = 'vlm' | 'tsr' | 'unknown';

export type ChunkStrategy = 'page' | 'section' | 'fragment';

export type ModelProvider = 'tensorlake' | 'claude-3-5-sonnet-latest' | 'gpt-4o-mini';

export type Settings = {tableParsingMode?: TableParsingStrategy | undefined, tableOutputMode?: TableOutputMode | undefined, tableSummarization?: boolean | undefined, tableSummarizationPrompt?: (string | null) | undefined, figureSummarization?: boolean | undefined, figureSummarizationPrompt?: (string | null) | undefined, formDetectionMode?: FormDetectionMode | undefined, chunkStrategy?: (ChunkStrategy | null) | undefined, jsonSchema: unknown, structuredExtractionPrompt?: (string | null) | undefined, modelProvider?: (ModelProvider | null) | undefined, deliverWebhook?: boolean | undefined, detectSignature?: boolean | undefined, skewCorrection?: boolean | undefined, disableLayoutDetection?: boolean | undefined, structuredExtractionSkipOcr?: boolean | undefined, detectStrikethrough?: boolean | undefined};

export type UpdateDatasetRequest = {name?: (string | null) | undefined, description?: (string | null) | undefined, settings?: (Settings | null) | undefined};

export type UpdateApiDocumentsV1DatasetsDatasetNameArgs = {dataset_name: string, body: UpdateDatasetRequest};

export type UpdateApiDocumentsV1DatasetsDatasetNameOptions = {baseUrl?: string | undefined, bearerAuth?: string | undefined, fetch?: typeof fetch | undefined};

export const tableParsingStrategy = z.enum(["tsr", "vlm", "unknown"]);

export const tableOutputMode = z.enum(["markdown", "html"]);

export const formDetectionMode = z.enum(["vlm", "tsr", "unknown"]);

export const chunkStrategy = z.enum(["page", "section", "fragment"]);

export const modelProvider = z.enum(["tensorlake", "claude-3-5-sonnet-latest", "gpt-4o-mini"]);

export const settings = z.object({tableParsingMode: tableParsingStrategy.optional(), tableOutputMode: tableOutputMode.optional(), tableSummarization: z.boolean().optional(), tableSummarizationPrompt: z.string().nullable().optional(), figureSummarization: z.boolean().optional(), figureSummarizationPrompt: z.string().nullable().optional(), formDetectionMode: formDetectionMode.optional(), chunkStrategy: chunkStrategy.nullable().optional(), jsonSchema: z.unknown(), structuredExtractionPrompt: z.string().nullable().optional(), modelProvider: modelProvider.nullable().optional(), deliverWebhook: z.boolean().optional(), detectSignature: z.boolean().optional(), skewCorrection: z.boolean().optional(), disableLayoutDetection: z.boolean().optional(), structuredExtractionSkipOcr: z.boolean().optional(), detectStrikethrough: z.boolean().optional()});

export const datasetStatus = z.enum(["idle", "processing"]);

export const jobStatus = z.enum(["failure", "pending", "processing", "successful"]);

export const pageDatasetJobItem = z.object({items: z.array(z.object({id: z.string(), datasetId: z.string(), fileId: z.string().nullable().optional(), fileName: z.string().nullable().optional(), outputsUrl: z.string().nullable().optional(), status: jobStatus, errorMessage: z.string().nullable().optional(), createdAt: z.string(), updatedAt: z.string()})), hasMore: z.boolean(), prevCursor: z.string().nullable().optional(), nextCursor: z.string().nullable().optional()});

export const datasetAnalytics = z.object({totalJobs: z.number().int().gte(0), totalProcessingJobs: z.number().int().gte(0), totalErrorJobs: z.number().int().gte(0), totalSuccessfulJobs: z.number().int().gte(0), totalPendingJobs: z.number().int().gte(0)});

export const datasetResponse = z.object({id: z.string(), name: z.string(), description: z.string().nullable().optional(), settings: settings, status: datasetStatus.optional(), createdAt: z.string(), jobs: pageDatasetJobItem.optional(), analytics: datasetAnalytics});

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
};

/**
 * PUT /documents/v1/datasets/{dataset_name}
 */
export const updateApiDocumentsV1DatasetsDatasetName = async (args: UpdateApiDocumentsV1DatasetsDatasetNameArgs, options?: UpdateApiDocumentsV1DatasetsDatasetNameOptions) => {
const {baseUrl = "https://api.tensorlake.ai", fetch: fetchFn = fetch, bearerAuth} = options ?? {}

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

if (bearerAuth !== undefined) headers.set('authorization', `Bearer ${bearerAuth}`)

const url = `${baseUrl.replace(/\/+$/, '')}/documents/v1/datasets/${encodeURIComponent(String(args.dataset_name))}${query.size > 0 ? `?${query}` : ''}`

const res = await fetchFn(url, {method: "PUT", 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(datasetResponse, await res.json())
};