// File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.

package openai

import (
	"context"
	"net/http"
	"slices"

	"github.com/openai/openai-go/v3/internal/apijson"
	"github.com/openai/openai-go/v3/internal/requestconfig"
	"github.com/openai/openai-go/v3/option"
	"github.com/openai/openai-go/v3/packages/param"
	"github.com/openai/openai-go/v3/packages/respjson"
	"github.com/openai/openai-go/v3/shared/constant"
)

// Get a vector representation of a given input that can be easily consumed by
// machine learning models and algorithms.
//
// EmbeddingService contains methods and other services that help with interacting
// with the openai API.
//
// Note, unlike clients, this service does not read variables from the environment
// automatically. You should not instantiate this service directly, and instead use
// the [NewEmbeddingService] method instead.
type EmbeddingService struct {
	Options []option.RequestOption
}

// NewEmbeddingService generates a new service that applies the given options to
// each request. These options are applied after the parent client's options (if
// there is one), and before any request-specific options.
func NewEmbeddingService(opts ...option.RequestOption) (r EmbeddingService) {
	r = EmbeddingService{}
	r.Options = opts
	return
}

// Creates an embedding vector representing the input text.
func (r *EmbeddingService) New(ctx context.Context, body EmbeddingNewParams, opts ...option.RequestOption) (res *CreateEmbeddingResponse, err error) {
	opts = slices.Concat(r.Options, opts)
	path := "embeddings"
	err = requestconfig.ExecuteNewRequest(ctx, http.MethodPost, path, body, &res, opts...)
	return res, err
}

type CreateEmbeddingResponse struct {
	// The list of embeddings generated by the model.
	Data []Embedding `json:"data" api:"required"`
	// The name of the model used to generate the embedding.
	Model string `json:"model" api:"required"`
	// The object type, which is always "list".
	Object constant.List `json:"object" default:"list"`
	// The usage information for the request.
	Usage CreateEmbeddingResponseUsage `json:"usage" api:"required"`
	// JSON contains metadata for fields, check presence with [respjson.Field.Valid].
	JSON struct {
		Data        respjson.Field
		Model       respjson.Field
		Object      respjson.Field
		Usage       respjson.Field
		ExtraFields map[string]respjson.Field
		raw         string
	} `json:"-"`
}

// Returns the unmodified JSON received from the API
func (r CreateEmbeddingResponse) RawJSON() string { return r.JSON.raw }
func (r *CreateEmbeddingResponse) UnmarshalJSON(data []byte) error {
	return apijson.UnmarshalRoot(data, r)
}

// The usage information for the request.
type CreateEmbeddingResponseUsage struct {
	// The number of tokens used by the prompt.
	PromptTokens int64 `json:"prompt_tokens" api:"required"`
	// The total number of tokens used by the request.
	TotalTokens int64 `json:"total_tokens" api:"required"`
	// JSON contains metadata for fields, check presence with [respjson.Field.Valid].
	JSON struct {
		PromptTokens respjson.Field
		TotalTokens  respjson.Field
		ExtraFields  map[string]respjson.Field
		raw          string
	} `json:"-"`
}

// Returns the unmodified JSON received from the API
func (r CreateEmbeddingResponseUsage) RawJSON() string { return r.JSON.raw }
func (r *CreateEmbeddingResponseUsage) UnmarshalJSON(data []byte) error {
	return apijson.UnmarshalRoot(data, r)
}

// Represents an embedding vector returned by embedding endpoint.
type Embedding struct {
	// The embedding vector, which is a list of floats. The length of vector depends on
	// the model as listed in the
	// [embedding guide](https://platform.openai.com/docs/guides/embeddings).
	Embedding []float64 `json:"embedding" api:"required"`
	// The index of the embedding in the list of embeddings.
	Index int64 `json:"index" api:"required"`
	// The object type, which is always "embedding".
	Object constant.Embedding `json:"object" default:"embedding"`
	// JSON contains metadata for fields, check presence with [respjson.Field.Valid].
	JSON struct {
		Embedding   respjson.Field
		Index       respjson.Field
		Object      respjson.Field
		ExtraFields map[string]respjson.Field
		raw         string
	} `json:"-"`
}

// Returns the unmodified JSON received from the API
func (r Embedding) RawJSON() string { return r.JSON.raw }
func (r *Embedding) UnmarshalJSON(data []byte) error {
	return apijson.UnmarshalRoot(data, r)
}

type EmbeddingModel = string

const (
	EmbeddingModelTextEmbeddingAda002 EmbeddingModel = "text-embedding-ada-002"
	EmbeddingModelTextEmbedding3Small EmbeddingModel = "text-embedding-3-small"
	EmbeddingModelTextEmbedding3Large EmbeddingModel = "text-embedding-3-large"
)

type EmbeddingNewParams struct {
	// Input text to embed, encoded as a string or array of tokens. To embed multiple
	// inputs in a single request, pass an array of strings or array of token arrays.
	// The input must not exceed the max input tokens for the model (8192 tokens for
	// all embedding models), cannot be an empty string, and any array must be 2048
	// dimensions or less.
	// [Example Python code](https://cookbook.openai.com/examples/how_to_count_tokens_with_tiktoken)
	// for counting tokens. In addition to the per-input token limit, all embedding
	// models enforce a maximum of 300,000 tokens summed across all inputs in a single
	// request.
	Input EmbeddingNewParamsInputUnion `json:"input,omitzero" api:"required"`
	// ID of the model to use. You can use the
	// [List models](https://platform.openai.com/docs/api-reference/models/list) API to
	// see all of your available models, or see our
	// [Model overview](https://platform.openai.com/docs/models) for descriptions of
	// them.
	Model EmbeddingModel `json:"model,omitzero" api:"required"`
	// The number of dimensions the resulting output embeddings should have. Only
	// supported in `text-embedding-3` and later models.
	Dimensions param.Opt[int64] `json:"dimensions,omitzero"`
	// A unique identifier representing your end-user, which can help OpenAI to monitor
	// and detect abuse.
	// [Learn more](https://platform.openai.com/docs/guides/safety-best-practices#end-user-ids).
	User param.Opt[string] `json:"user,omitzero"`
	// The format to return the embeddings in. Can be either `float` or
	// [`base64`](https://pypi.org/project/pybase64/).
	//
	// Any of "float", "base64".
	EncodingFormat EmbeddingNewParamsEncodingFormat `json:"encoding_format,omitzero"`
	paramObj
}

func (r EmbeddingNewParams) MarshalJSON() (data []byte, err error) {
	type shadow EmbeddingNewParams
	return param.MarshalObject(r, (*shadow)(&r))
}
func (r *EmbeddingNewParams) UnmarshalJSON(data []byte) error {
	return apijson.UnmarshalRoot(data, r)
}

// Only one field can be non-zero.
//
// Use [param.IsOmitted] to confirm if a field is set.
type EmbeddingNewParamsInputUnion struct {
	OfString             param.Opt[string] `json:",omitzero,inline"`
	OfArrayOfStrings     []string          `json:",omitzero,inline"`
	OfArrayOfTokens      []int64           `json:",omitzero,inline"`
	OfArrayOfTokenArrays [][]int64         `json:",omitzero,inline"`
	paramUnion
}

func (u EmbeddingNewParamsInputUnion) MarshalJSON() ([]byte, error) {
	return param.MarshalUnion(u, u.OfString, u.OfArrayOfStrings, u.OfArrayOfTokens, u.OfArrayOfTokenArrays)
}
func (u *EmbeddingNewParamsInputUnion) UnmarshalJSON(data []byte) error {
	return apijson.UnmarshalRoot(data, u)
}

func (u *EmbeddingNewParamsInputUnion) asAny() any {
	if !param.IsOmitted(u.OfString) {
		return &u.OfString.Value
	} else if !param.IsOmitted(u.OfArrayOfStrings) {
		return &u.OfArrayOfStrings
	} else if !param.IsOmitted(u.OfArrayOfTokens) {
		return &u.OfArrayOfTokens
	} else if !param.IsOmitted(u.OfArrayOfTokenArrays) {
		return &u.OfArrayOfTokenArrays
	}
	return nil
}

// The format to return the embeddings in. Can be either `float` or
// [`base64`](https://pypi.org/project/pybase64/).
type EmbeddingNewParamsEncodingFormat string

const (
	EmbeddingNewParamsEncodingFormatFloat  EmbeddingNewParamsEncodingFormat = "float"
	EmbeddingNewParamsEncodingFormatBase64 EmbeddingNewParamsEncodingFormat = "base64"
)