first commit: one-api base code + SAAS plan document

2025-12-29 22:52:27 +08:00
commit cb7c48bfa7
564 changed files with 61468 additions and 0 deletions
--- a/relay/adaptor/tencent/adaptor.go
+++ b/relay/adaptor/tencent/adaptor.go
@@ -0,0 +1,105 @@
+package tencent
+
+import (
+	"errors"
+	"io"
+	"net/http"
+	"strconv"
+	"strings"
+
+	"github.com/gin-gonic/gin"
+
+	"github.com/songquanpeng/one-api/common/helper"
+	"github.com/songquanpeng/one-api/relay/adaptor"
+	"github.com/songquanpeng/one-api/relay/adaptor/openai"
+	"github.com/songquanpeng/one-api/relay/meta"
+	"github.com/songquanpeng/one-api/relay/model"
+	"github.com/songquanpeng/one-api/relay/relaymode"
+)
+
+// https://cloud.tencent.com/document/api/1729/101837
+
+type Adaptor struct {
+	Sign      string
+	Action    string
+	Version   string
+	Timestamp int64
+}
+
+func (a *Adaptor) Init(meta *meta.Meta) {
+	a.Action = "ChatCompletions"
+	a.Version = "2023-09-01"
+	a.Timestamp = helper.GetTimestamp()
+}
+
+func (a *Adaptor) GetRequestURL(meta *meta.Meta) (string, error) {
+	return meta.BaseURL + "/", nil
+}
+
+func (a *Adaptor) SetupRequestHeader(c *gin.Context, req *http.Request, meta *meta.Meta) error {
+	adaptor.SetupCommonRequestHeader(c, req, meta)
+	req.Header.Set("Authorization", a.Sign)
+	req.Header.Set("X-TC-Action", a.Action)
+	req.Header.Set("X-TC-Version", a.Version)
+	req.Header.Set("X-TC-Timestamp", strconv.FormatInt(a.Timestamp, 10))
+	return nil
+}
+
+func (a *Adaptor) ConvertRequest(c *gin.Context, relayMode int, request *model.GeneralOpenAIRequest) (any, error) {
+	if request == nil {
+		return nil, errors.New("request is nil")
+	}
+	apiKey := c.Request.Header.Get("Authorization")
+	apiKey = strings.TrimPrefix(apiKey, "Bearer ")
+	_, secretId, secretKey, err := ParseConfig(apiKey)
+	if err != nil {
+		return nil, err
+	}
+	var convertedRequest any
+	switch relayMode {
+	case relaymode.Embeddings:
+		a.Action = "GetEmbedding"
+		convertedRequest = ConvertEmbeddingRequest(*request)
+	default:
+		a.Action = "ChatCompletions"
+		convertedRequest = ConvertRequest(*request)
+	}
+	// we have to calculate the sign here
+	a.Sign = GetSign(convertedRequest, a, secretId, secretKey)
+	return convertedRequest, nil
+}
+
+func (a *Adaptor) ConvertImageRequest(request *model.ImageRequest) (any, error) {
+	if request == nil {
+		return nil, errors.New("request is nil")
+	}
+	return request, nil
+}
+
+func (a *Adaptor) DoRequest(c *gin.Context, meta *meta.Meta, requestBody io.Reader) (*http.Response, error) {
+	return adaptor.DoRequestHelper(a, c, meta, requestBody)
+}
+
+func (a *Adaptor) DoResponse(c *gin.Context, resp *http.Response, meta *meta.Meta) (usage *model.Usage, err *model.ErrorWithStatusCode) {
+	if meta.IsStream {
+		var responseText string
+		err, responseText = StreamHandler(c, resp)
+		usage = openai.ResponseText2Usage(responseText, meta.ActualModelName, meta.PromptTokens)
+	} else {
+		switch meta.Mode {
+		case relaymode.Embeddings:
+			err, usage = EmbeddingHandler(c, resp)
+		default:
+			err, usage = Handler(c, resp)
+		}
+	}
+	return
+}
+
+func (a *Adaptor) GetModelList() []string {
+	return ModelList
+}
+
+func (a *Adaptor) GetChannelName() string {
+	return "tencent"
+}
--- a/relay/adaptor/tencent/constants.go
+++ b/relay/adaptor/tencent/constants.go
@@ -0,0 +1,10 @@
+package tencent
+
+var ModelList = []string{
+	"hunyuan-lite",
+	"hunyuan-standard",
+	"hunyuan-standard-256K",
+	"hunyuan-pro",
+	"hunyuan-vision",
+	"hunyuan-embedding",
+}
--- a/relay/adaptor/tencent/main.go
+++ b/relay/adaptor/tencent/main.go
@@ -0,0 +1,307 @@
+package tencent
+
+import (
+	"bufio"
+	"crypto/hmac"
+	"crypto/sha256"
+	"encoding/hex"
+	"encoding/json"
+	"errors"
+	"fmt"
+	"io"
+	"net/http"
+	"strconv"
+	"strings"
+	"time"
+
+	"github.com/gin-gonic/gin"
+
+	"github.com/songquanpeng/one-api/common"
+	"github.com/songquanpeng/one-api/common/conv"
+	"github.com/songquanpeng/one-api/common/ctxkey"
+	"github.com/songquanpeng/one-api/common/helper"
+	"github.com/songquanpeng/one-api/common/logger"
+	"github.com/songquanpeng/one-api/common/random"
+	"github.com/songquanpeng/one-api/common/render"
+	"github.com/songquanpeng/one-api/relay/adaptor/openai"
+	"github.com/songquanpeng/one-api/relay/constant"
+	"github.com/songquanpeng/one-api/relay/model"
+)
+
+func ConvertRequest(request model.GeneralOpenAIRequest) *ChatRequest {
+	messages := make([]*Message, 0, len(request.Messages))
+	for i := 0; i < len(request.Messages); i++ {
+		message := request.Messages[i]
+		messages = append(messages, &Message{
+			Content: message.StringContent(),
+			Role:    message.Role,
+		})
+	}
+	return &ChatRequest{
+		Model:       &request.Model,
+		Stream:      &request.Stream,
+		Messages:    messages,
+		TopP:        request.TopP,
+		Temperature: request.Temperature,
+	}
+}
+
+func ConvertEmbeddingRequest(request model.GeneralOpenAIRequest) *EmbeddingRequest {
+	return &EmbeddingRequest{
+		InputList: request.ParseInput(),
+	}
+}
+
+func EmbeddingHandler(c *gin.Context, resp *http.Response) (*model.ErrorWithStatusCode, *model.Usage) {
+	var tencentResponseP EmbeddingResponseP
+	err := json.NewDecoder(resp.Body).Decode(&tencentResponseP)
+	if err != nil {
+		return openai.ErrorWrapper(err, "unmarshal_response_body_failed", http.StatusInternalServerError), nil
+	}
+
+	err = resp.Body.Close()
+	if err != nil {
+		return openai.ErrorWrapper(err, "close_response_body_failed", http.StatusInternalServerError), nil
+	}
+
+	tencentResponse := tencentResponseP.Response
+	if tencentResponse.Error.Code != "" {
+		return &model.ErrorWithStatusCode{
+			Error: model.Error{
+				Message: tencentResponse.Error.Message,
+				Code:    tencentResponse.Error.Code,
+			},
+			StatusCode: resp.StatusCode,
+		}, nil
+	}
+	requestModel := c.GetString(ctxkey.RequestModel)
+	fullTextResponse := embeddingResponseTencent2OpenAI(&tencentResponse)
+	fullTextResponse.Model = requestModel
+	jsonResponse, err := json.Marshal(fullTextResponse)
+	if err != nil {
+		return openai.ErrorWrapper(err, "marshal_response_body_failed", http.StatusInternalServerError), nil
+	}
+	c.Writer.Header().Set("Content-Type", "application/json")
+	c.Writer.WriteHeader(resp.StatusCode)
+	_, err = c.Writer.Write(jsonResponse)
+	return nil, &fullTextResponse.Usage
+}
+
+func embeddingResponseTencent2OpenAI(response *EmbeddingResponse) *openai.EmbeddingResponse {
+	openAIEmbeddingResponse := openai.EmbeddingResponse{
+		Object: "list",
+		Data:   make([]openai.EmbeddingResponseItem, 0, len(response.Data)),
+		Model:  "hunyuan-embedding",
+		Usage:  model.Usage{TotalTokens: response.EmbeddingUsage.TotalTokens},
+	}
+
+	for _, item := range response.Data {
+		openAIEmbeddingResponse.Data = append(openAIEmbeddingResponse.Data, openai.EmbeddingResponseItem{
+			Object:    item.Object,
+			Index:     item.Index,
+			Embedding: item.Embedding,
+		})
+	}
+	return &openAIEmbeddingResponse
+}
+
+func responseTencent2OpenAI(response *ChatResponse) *openai.TextResponse {
+	fullTextResponse := openai.TextResponse{
+		Id:      response.ReqID,
+		Object:  "chat.completion",
+		Created: helper.GetTimestamp(),
+		Usage: model.Usage{
+			PromptTokens:     response.Usage.PromptTokens,
+			CompletionTokens: response.Usage.CompletionTokens,
+			TotalTokens:      response.Usage.TotalTokens,
+		},
+	}
+	if len(response.Choices) > 0 {
+		choice := openai.TextResponseChoice{
+			Index: 0,
+			Message: model.Message{
+				Role:    "assistant",
+				Content: response.Choices[0].Messages.Content,
+			},
+			FinishReason: response.Choices[0].FinishReason,
+		}
+		fullTextResponse.Choices = append(fullTextResponse.Choices, choice)
+	}
+	return &fullTextResponse
+}
+
+func streamResponseTencent2OpenAI(TencentResponse *ChatResponse) *openai.ChatCompletionsStreamResponse {
+	response := openai.ChatCompletionsStreamResponse{
+		Id:      fmt.Sprintf("chatcmpl-%s", random.GetUUID()),
+		Object:  "chat.completion.chunk",
+		Created: helper.GetTimestamp(),
+		Model:   "tencent-hunyuan",
+	}
+	if len(TencentResponse.Choices) > 0 {
+		var choice openai.ChatCompletionsStreamResponseChoice
+		choice.Delta.Content = TencentResponse.Choices[0].Delta.Content
+		if TencentResponse.Choices[0].FinishReason == "stop" {
+			choice.FinishReason = &constant.StopFinishReason
+		}
+		response.Choices = append(response.Choices, choice)
+	}
+	return &response
+}
+
+func StreamHandler(c *gin.Context, resp *http.Response) (*model.ErrorWithStatusCode, string) {
+	var responseText string
+	scanner := bufio.NewScanner(resp.Body)
+	scanner.Split(bufio.ScanLines)
+
+	common.SetEventStreamHeaders(c)
+
+	for scanner.Scan() {
+		data := scanner.Text()
+		if len(data) < 5 || !strings.HasPrefix(data, "data:") {
+			continue
+		}
+		data = strings.TrimPrefix(data, "data:")
+
+		var tencentResponse ChatResponse
+		err := json.Unmarshal([]byte(data), &tencentResponse)
+		if err != nil {
+			logger.SysError("error unmarshalling stream response: " + err.Error())
+			continue
+		}
+
+		response := streamResponseTencent2OpenAI(&tencentResponse)
+		if len(response.Choices) != 0 {
+			responseText += conv.AsString(response.Choices[0].Delta.Content)
+		}
+
+		err = render.ObjectData(c, response)
+		if err != nil {
+			logger.SysError(err.Error())
+		}
+	}
+
+	if err := scanner.Err(); err != nil {
+		logger.SysError("error reading stream: " + err.Error())
+	}
+
+	render.Done(c)
+
+	err := resp.Body.Close()
+	if err != nil {
+		return openai.ErrorWrapper(err, "close_response_body_failed", http.StatusInternalServerError), ""
+	}
+
+	return nil, responseText
+}
+
+func Handler(c *gin.Context, resp *http.Response) (*model.ErrorWithStatusCode, *model.Usage) {
+	var TencentResponse ChatResponse
+	var responseP ChatResponseP
+	responseBody, err := io.ReadAll(resp.Body)
+	if err != nil {
+		return openai.ErrorWrapper(err, "read_response_body_failed", http.StatusInternalServerError), nil
+	}
+	err = resp.Body.Close()
+	if err != nil {
+		return openai.ErrorWrapper(err, "close_response_body_failed", http.StatusInternalServerError), nil
+	}
+	err = json.Unmarshal(responseBody, &responseP)
+	if err != nil {
+		return openai.ErrorWrapper(err, "unmarshal_response_body_failed", http.StatusInternalServerError), nil
+	}
+	TencentResponse = responseP.Response
+	if TencentResponse.Error.Code != "" {
+		return &model.ErrorWithStatusCode{
+			Error: model.Error{
+				Message: TencentResponse.Error.Message,
+				Code:    TencentResponse.Error.Code,
+			},
+			StatusCode: resp.StatusCode,
+		}, nil
+	}
+	fullTextResponse := responseTencent2OpenAI(&TencentResponse)
+	fullTextResponse.Model = "hunyuan"
+	jsonResponse, err := json.Marshal(fullTextResponse)
+	if err != nil {
+		return openai.ErrorWrapper(err, "marshal_response_body_failed", http.StatusInternalServerError), nil
+	}
+	c.Writer.Header().Set("Content-Type", "application/json")
+	c.Writer.WriteHeader(resp.StatusCode)
+	_, err = c.Writer.Write(jsonResponse)
+	if err != nil {
+		return openai.ErrorWrapper(err, "write_response_body_failed", http.StatusInternalServerError), nil
+	}
+	return nil, &fullTextResponse.Usage
+}
+
+func ParseConfig(config string) (appId int64, secretId string, secretKey string, err error) {
+	parts := strings.Split(config, "|")
+	if len(parts) != 3 {
+		err = errors.New("invalid tencent config")
+		return
+	}
+	appId, err = strconv.ParseInt(parts[0], 10, 64)
+	secretId = parts[1]
+	secretKey = parts[2]
+	return
+}
+
+func sha256hex(s string) string {
+	b := sha256.Sum256([]byte(s))
+	return hex.EncodeToString(b[:])
+}
+
+func hmacSha256(s, key string) string {
+	hashed := hmac.New(sha256.New, []byte(key))
+	hashed.Write([]byte(s))
+	return string(hashed.Sum(nil))
+}
+
+func GetSign(req any, adaptor *Adaptor, secId, secKey string) string {
+	// build canonical request string
+	host := "hunyuan.tencentcloudapi.com"
+	httpRequestMethod := "POST"
+	canonicalURI := "/"
+	canonicalQueryString := ""
+	canonicalHeaders := fmt.Sprintf("content-type:%s\nhost:%s\nx-tc-action:%s\n",
+		"application/json", host, strings.ToLower(adaptor.Action))
+	signedHeaders := "content-type;host;x-tc-action"
+	payload, _ := json.Marshal(req)
+	hashedRequestPayload := sha256hex(string(payload))
+	canonicalRequest := fmt.Sprintf("%s\n%s\n%s\n%s\n%s\n%s",
+		httpRequestMethod,
+		canonicalURI,
+		canonicalQueryString,
+		canonicalHeaders,
+		signedHeaders,
+		hashedRequestPayload)
+	// build string to sign
+	algorithm := "TC3-HMAC-SHA256"
+	requestTimestamp := strconv.FormatInt(adaptor.Timestamp, 10)
+	timestamp, _ := strconv.ParseInt(requestTimestamp, 10, 64)
+	t := time.Unix(timestamp, 0).UTC()
+	// must be the format 2006-01-02, ref to package time for more info
+	date := t.Format("2006-01-02")
+	credentialScope := fmt.Sprintf("%s/%s/tc3_request", date, "hunyuan")
+	hashedCanonicalRequest := sha256hex(canonicalRequest)
+	string2sign := fmt.Sprintf("%s\n%s\n%s\n%s",
+		algorithm,
+		requestTimestamp,
+		credentialScope,
+		hashedCanonicalRequest)
+
+	// sign string
+	secretDate := hmacSha256(date, "TC3"+secKey)
+	secretService := hmacSha256("hunyuan", secretDate)
+	secretKey := hmacSha256("tc3_request", secretService)
+	signature := hex.EncodeToString([]byte(hmacSha256(string2sign, secretKey)))
+
+	// build authorization
+	authorization := fmt.Sprintf("%s Credential=%s/%s, SignedHeaders=%s, Signature=%s",
+		algorithm,
+		secId,
+		credentialScope,
+		signedHeaders,
+		signature)
+	return authorization
+}
--- a/relay/adaptor/tencent/model.go
+++ b/relay/adaptor/tencent/model.go
@@ -0,0 +1,101 @@
+package tencent
+
+type Message struct {
+	Role    string `json:"Role"`
+	Content string `json:"Content"`
+}
+
+type ChatRequest struct {
+	// 模型名称，可选值包括 hunyuan-lite、hunyuan-standard、hunyuan-standard-256K、hunyuan-pro。
+	// 各模型介绍请阅读 [产品概述](https://cloud.tencent.com/document/product/1729/104753) 中的说明。
+	//
+	// 注意：
+	// 不同的模型计费不同，请根据 [购买指南](https://cloud.tencent.com/document/product/1729/97731) 按需调用。
+	Model *string `json:"Model"`
+	// 聊天上下文信息。
+	// 说明：
+	// 1. 长度最多为 40，按对话时间从旧到新在数组中排列。
+	// 2. Message.Role 可选值：system、user、assistant。
+	// 其中，system 角色可选，如存在则必须位于列表的最开始。user 和 assistant 需交替出现（一问一答），以 user 提问开始和结束，且 Content 不能为空。Role 的顺序示例：[system（可选） user assistant user assistant user ...]。
+	// 3. Messages 中 Content 总长度不能超过模型输入长度上限（可参考 [产品概述](https://cloud.tencent.com/document/product/1729/104753) 文档），超过则会截断最前面的内容，只保留尾部内容。
+	Messages []*Message `json:"Messages"`
+	// 流式调用开关。
+	// 说明：
+	// 1. 未传值时默认为非流式调用（false）。
+	// 2. 流式调用时以 SSE 协议增量返回结果（返回值取 Choices[n].Delta 中的值，需要拼接增量数据才能获得完整结果）。
+	// 3. 非流式调用时：
+	// 调用方式与普通 HTTP 请求无异。
+	// 接口响应耗时较长，**如需更低时延建议设置为 true**。
+	// 只返回一次最终结果（返回值取 Choices[n].Message 中的值）。
+	//
+	// 注意：
+	// 通过 SDK 调用时，流式和非流式调用需用**不同的方式**获取返回值，具体参考 SDK 中的注释或示例（在各语言 SDK 代码仓库的 examples/hunyuan/v20230901/ 目录中）。
+	Stream *bool `json:"Stream"`
+	// 说明：
+	// 1. 影响输出文本的多样性，取值越大，生成文本的多样性越强。
+	// 2. 取值区间为 [0.0, 1.0]，未传值时使用各模型推荐值。
+	// 3. 非必要不建议使用，不合理的取值会影响效果。
+	TopP *float64 `json:"TopP,omitempty"`
+	// 说明：
+	// 1. 较高的数值会使输出更加随机，而较低的数值会使其更加集中和确定。
+	// 2. 取值区间为 [0.0, 2.0]，未传值时使用各模型推荐值。
+	// 3. 非必要不建议使用，不合理的取值会影响效果。
+	Temperature *float64 `json:"Temperature,omitempty"`
+}
+
+type Error struct {
+	Code    string `json:"Code"`
+	Message string `json:"Message"`
+}
+
+type Usage struct {
+	PromptTokens     int `json:"PromptTokens"`
+	CompletionTokens int `json:"CompletionTokens"`
+	TotalTokens      int `json:"TotalTokens"`
+}
+
+type ResponseChoices struct {
+	FinishReason string  `json:"FinishReason,omitempty"` // 流式结束标志位，为 stop 则表示尾包
+	Messages     Message `json:"Message,omitempty"`      // 内容，同步模式返回内容，流模式为 null 输出 content 内容总数最多支持 1024token。
+	Delta        Message `json:"Delta,omitempty"`        // 内容，流模式返回内容，同步模式为 null 输出 content 内容总数最多支持 1024token。
+}
+
+type ChatResponse struct {
+	Choices []ResponseChoices `json:"Choices,omitempty"`   // 结果
+	Created int64             `json:"Created,omitempty"`   // unix 时间戳的字符串
+	Id      string            `json:"Id,omitempty"`        // 会话 id
+	Usage   Usage             `json:"Usage,omitempty"`     // token 数量
+	Error   Error             `json:"Error,omitempty"`     // 错误信息 注意：此字段可能返回 null，表示取不到有效值
+	Note    string            `json:"Note,omitempty"`      // 注释
+	ReqID   string            `json:"RequestId,omitempty"` // 唯一请求 Id，每次请求都会返回。用于反馈接口入参
+}
+
+type ChatResponseP struct {
+	Response ChatResponse `json:"Response,omitempty"`
+}
+
+type EmbeddingRequest struct {
+	InputList []string `json:"InputList"`
+}
+
+type EmbeddingData struct {
+	Embedding []float64 `json:"Embedding"`
+	Index     int       `json:"Index"`
+	Object    string    `json:"Object"`
+}
+
+type EmbeddingUsage struct {
+	PromptTokens int `json:"PromptTokens"`
+	TotalTokens  int `json:"TotalTokens"`
+}
+
+type EmbeddingResponse struct {
+	Data           []EmbeddingData `json:"Data"`
+	EmbeddingUsage EmbeddingUsage  `json:"Usage,omitempty"`
+	RequestId      string          `json:"RequestId,omitempty"`
+	Error          Error           `json:"Error,omitempty"`
+}
+
+type EmbeddingResponseP struct {
+	Response EmbeddingResponse `json:"Response,omitempty"`
+}