From f83894c83f480fbb06150dddf6dca11e1022ae7c Mon Sep 17 00:00:00 2001 From: wagxuebing <65780532+lynnssb@users.noreply.github.com> Date: Wed, 12 Jun 2024 23:12:58 +0800 Subject: [PATCH 01/65] fix: xunfei interface call 4001 error (#1499) Co-authored-by: lynnssb --- relay/adaptor/xunfei/model.go | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/relay/adaptor/xunfei/model.go b/relay/adaptor/xunfei/model.go index 97a43154..5ee17cb0 100644 --- a/relay/adaptor/xunfei/model.go +++ b/relay/adaptor/xunfei/model.go @@ -26,7 +26,7 @@ type ChatRequest struct { Message struct { Text []Message `json:"text"` } `json:"message"` - Functions struct { + Functions *struct { Text []model.Function `json:"text,omitempty"` } `json:"functions,omitempty"` } `json:"payload"` From c1971870fa4d2f6c7c406fa644bc5b55737461c4 Mon Sep 17 00:00:00 2001 From: Zhong Liu Date: Thu, 13 Jun 2024 00:07:26 +0800 Subject: [PATCH 02/65] fix: support for Spark Lite model (#1526) * fix: Support for Spark Lite model * fix: fix panic * fix: fix xunfei version config --------- Co-authored-by: JustSong <39998050+songquanpeng@users.noreply.github.com> Co-authored-by: JustSong --- middleware/distributor.go | 42 ++++++++++---------- model/channel.go | 2 +- relay/adaptor/xunfei/adaptor.go | 16 ++++---- relay/adaptor/xunfei/main.go | 25 ++++++------ relay/adaptor/xunfei/model.go | 8 ++-- web/default/src/pages/Channel/EditChannel.js | 3 -- 6 files changed, 48 insertions(+), 48 deletions(-) diff --git a/middleware/distributor.go b/middleware/distributor.go index d0fd7ba5..0c4b04c3 100644 --- a/middleware/distributor.go +++ b/middleware/distributor.go @@ -67,26 +67,28 @@ func SetupContextForSelectedChannel(c *gin.Context, channel *model.Channel, mode c.Set(ctxkey.BaseURL, channel.GetBaseURL()) cfg, _ := channel.LoadConfig() // this is for backward compatibility - switch channel.Type { - case channeltype.Azure: - if cfg.APIVersion == "" { - cfg.APIVersion = channel.Other - } - case channeltype.Xunfei: - if cfg.APIVersion == "" { - cfg.APIVersion = channel.Other - } - case channeltype.Gemini: - if cfg.APIVersion == "" { - cfg.APIVersion = channel.Other - } - case channeltype.AIProxyLibrary: - if cfg.LibraryID == "" { - cfg.LibraryID = channel.Other - } - case channeltype.Ali: - if cfg.Plugin == "" { - cfg.Plugin = channel.Other + if channel.Other != nil { + switch channel.Type { + case channeltype.Azure: + if cfg.APIVersion == "" { + cfg.APIVersion = *channel.Other + } + case channeltype.Xunfei: + if cfg.APIVersion == "" { + cfg.APIVersion = *channel.Other + } + case channeltype.Gemini: + if cfg.APIVersion == "" { + cfg.APIVersion = *channel.Other + } + case channeltype.AIProxyLibrary: + if cfg.LibraryID == "" { + cfg.LibraryID = *channel.Other + } + case channeltype.Ali: + if cfg.Plugin == "" { + cfg.Plugin = *channel.Other + } } } c.Set(ctxkey.Config, cfg) diff --git a/model/channel.go b/model/channel.go index ec52683e..a92f402c 100644 --- a/model/channel.go +++ b/model/channel.go @@ -27,7 +27,7 @@ type Channel struct { TestTime int64 `json:"test_time" gorm:"bigint"` ResponseTime int `json:"response_time"` // in milliseconds BaseURL *string `json:"base_url" gorm:"column:base_url;default:''"` - Other string `json:"other"` // DEPRECATED: please save config to field Config + Other *string `json:"other"` // DEPRECATED: please save config to field Config Balance float64 `json:"balance"` // in USD BalanceUpdatedTime int64 `json:"balance_updated_time" gorm:"bigint"` Models string `json:"models"` diff --git a/relay/adaptor/xunfei/adaptor.go b/relay/adaptor/xunfei/adaptor.go index 3af97831..b5967f26 100644 --- a/relay/adaptor/xunfei/adaptor.go +++ b/relay/adaptor/xunfei/adaptor.go @@ -27,14 +27,6 @@ func (a *Adaptor) GetRequestURL(meta *meta.Meta) (string, error) { func (a *Adaptor) SetupRequestHeader(c *gin.Context, req *http.Request, meta *meta.Meta) error { adaptor.SetupCommonRequestHeader(c, req, meta) - version := parseAPIVersionByModelName(meta.ActualModelName) - if version == "" { - version = a.meta.Config.APIVersion - } - if version == "" { - version = "v1.1" - } - a.meta.Config.APIVersion = version // check DoResponse for auth part return nil } @@ -69,6 +61,14 @@ func (a *Adaptor) DoResponse(c *gin.Context, resp *http.Response, meta *meta.Met if a.request == nil { return nil, openai.ErrorWrapper(errors.New("request is nil"), "request_is_nil", http.StatusBadRequest) } + version := parseAPIVersionByModelName(meta.ActualModelName) + if version == "" { + version = a.meta.Config.APIVersion + } + if version == "" { + version = "v1.1" + } + a.meta.Config.APIVersion = version if meta.IsStream { err, usage = StreamHandler(c, meta, *a.request, splits[0], splits[1], splits[2]) } else { diff --git a/relay/adaptor/xunfei/main.go b/relay/adaptor/xunfei/main.go index c3e768b7..00122d5e 100644 --- a/relay/adaptor/xunfei/main.go +++ b/relay/adaptor/xunfei/main.go @@ -5,7 +5,14 @@ import ( "crypto/sha256" "encoding/base64" "encoding/json" + "errors" "fmt" + "io" + "net/http" + "net/url" + "strings" + "time" + "github.com/gin-gonic/gin" "github.com/gorilla/websocket" "github.com/songquanpeng/one-api/common" @@ -16,11 +23,6 @@ import ( "github.com/songquanpeng/one-api/relay/constant" "github.com/songquanpeng/one-api/relay/meta" "github.com/songquanpeng/one-api/relay/model" - "io" - "net/http" - "net/url" - "strings" - "time" ) // https://console.xfyun.cn/services/cbm @@ -28,11 +30,7 @@ import ( func requestOpenAI2Xunfei(request model.GeneralOpenAIRequest, xunfeiAppId string, domain string) *ChatRequest { messages := make([]Message, 0, len(request.Messages)) - var lastToolCalls []model.Tool for _, message := range request.Messages { - if message.ToolCalls != nil { - lastToolCalls = message.ToolCalls - } messages = append(messages, Message{ Role: message.Role, Content: message.StringContent(), @@ -45,9 +43,10 @@ func requestOpenAI2Xunfei(request model.GeneralOpenAIRequest, xunfeiAppId string xunfeiRequest.Parameter.Chat.TopK = request.N xunfeiRequest.Parameter.Chat.MaxTokens = request.MaxTokens xunfeiRequest.Payload.Message.Text = messages - if len(lastToolCalls) != 0 { - for _, toolCall := range lastToolCalls { - xunfeiRequest.Payload.Functions.Text = append(xunfeiRequest.Payload.Functions.Text, toolCall.Function) + + if strings.HasPrefix(domain, "generalv3") { + xunfeiRequest.Payload.Functions = &Functions{ + Text: request.Tools, } } @@ -203,7 +202,7 @@ func Handler(c *gin.Context, meta *meta.Meta, textRequest model.GeneralOpenAIReq } } if len(xunfeiResponse.Payload.Choices.Text) == 0 { - return openai.ErrorWrapper(err, "xunfei_empty_response_detected", http.StatusInternalServerError), nil + return openai.ErrorWrapper(errors.New("xunfei empty response detected"), "xunfei_empty_response_detected", http.StatusInternalServerError), nil } xunfeiResponse.Payload.Choices.Text[0].Content = content diff --git a/relay/adaptor/xunfei/model.go b/relay/adaptor/xunfei/model.go index 5ee17cb0..681c0c05 100644 --- a/relay/adaptor/xunfei/model.go +++ b/relay/adaptor/xunfei/model.go @@ -9,6 +9,10 @@ type Message struct { Content string `json:"content"` } +type Functions struct { + Text []model.Tool `json:"text,omitempty"` +} + type ChatRequest struct { Header struct { AppId string `json:"app_id"` @@ -26,9 +30,7 @@ type ChatRequest struct { Message struct { Text []Message `json:"text"` } `json:"message"` - Functions *struct { - Text []model.Function `json:"text,omitempty"` - } `json:"functions,omitempty"` + Functions *Functions `json:"functions,omitempty"` } `json:"payload"` } diff --git a/web/default/src/pages/Channel/EditChannel.js b/web/default/src/pages/Channel/EditChannel.js index ac0fed60..fb80a726 100644 --- a/web/default/src/pages/Channel/EditChannel.js +++ b/web/default/src/pages/Channel/EditChannel.js @@ -181,9 +181,6 @@ const EditChannel = () => { if (localInputs.type === 3 && localInputs.other === '') { localInputs.other = '2024-03-01-preview'; } - if (localInputs.type === 18 && localInputs.other === '') { - localInputs.other = 'v2.1'; - } let res; localInputs.models = localInputs.models.join(','); localInputs.group = localInputs.groups.join(','); From 6aafb7a99ed9ccd8905ade480f133e0c9bb138d6 Mon Sep 17 00:00:00 2001 From: Buer <42402987+MartialBE@users.noreply.github.com> Date: Thu, 13 Jun 2024 00:08:49 +0800 Subject: [PATCH 03/65] fix: channel edit settings key error (#1496) --- web/berry/src/views/Channel/component/EditModal.js | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/web/berry/src/views/Channel/component/EditModal.js b/web/berry/src/views/Channel/component/EditModal.js index 4f7f216d..29bd3de2 100644 --- a/web/berry/src/views/Channel/component/EditModal.js +++ b/web/berry/src/views/Channel/component/EditModal.js @@ -163,7 +163,7 @@ const EditModal = ({ open, channelId, onCancel, onOk }) => { values.other = 'v2.1'; } if (values.key === '') { - if (values.config.ak !== '' && values.config.sk !== '' && values.config.region !== '') { + if (values.config.ak && values.config.sk && values.config.region) { values.key = `${values.config.ak}|${values.config.sk}|${values.config.region}`; } } From f74577141c881256c8385963ea28166c42b3bfc0 Mon Sep 17 00:00:00 2001 From: jinjianming <57084209+jinjianming@users.noreply.github.com> Date: Thu, 13 Jun 2024 00:20:48 +0800 Subject: [PATCH 04/65] fix: fix default token not created in some cases (#1510) MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit * 修复git、微信等用户注册不会创建默认令牌问题 修复git、微信等用户注册不会创建默认令牌问题 * 修复git、微信等用户注册不会创建默认令牌问题 删除普通用户注册代码 * fix: do not block if error happened --------- Co-authored-by: JustSong --- controller/user.go | 26 +------------------------- model/user.go | 17 +++++++++++++++++ 2 files changed, 18 insertions(+), 25 deletions(-) diff --git a/controller/user.go b/controller/user.go index 9ab37b5a..e79881c2 100644 --- a/controller/user.go +++ b/controller/user.go @@ -6,8 +6,6 @@ import ( "github.com/songquanpeng/one-api/common" "github.com/songquanpeng/one-api/common/config" "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/model" "net/http" @@ -111,7 +109,6 @@ func Logout(c *gin.Context) { } func Register(c *gin.Context) { - ctx := c.Request.Context() if !config.RegisterEnabled { c.JSON(http.StatusOK, gin.H{ "message": "管理员关闭了新用户注册", @@ -176,28 +173,7 @@ func Register(c *gin.Context) { }) return } - go func() { - err := user.ValidateAndFill() - if err != nil { - logger.Errorf(ctx, "user.ValidateAndFill failed: %w", err) - return - } - cleanToken := model.Token{ - UserId: user.Id, - Name: "default", - Key: random.GenerateKey(), - CreatedTime: helper.GetTimestamp(), - AccessedTime: helper.GetTimestamp(), - ExpiredTime: -1, - RemainQuota: -1, - UnlimitedQuota: true, - } - err = cleanToken.Insert() - if err != nil { - logger.Errorf(ctx, "cleanToken.Insert failed: %w", err) - return - } - }() + c.JSON(http.StatusOK, gin.H{ "success": true, "message": "", diff --git a/model/user.go b/model/user.go index 1dc633b1..924d72f9 100644 --- a/model/user.go +++ b/model/user.go @@ -6,6 +6,7 @@ import ( "github.com/songquanpeng/one-api/common" "github.com/songquanpeng/one-api/common/blacklist" "github.com/songquanpeng/one-api/common/config" + "github.com/songquanpeng/one-api/common/helper" "github.com/songquanpeng/one-api/common/logger" "github.com/songquanpeng/one-api/common/random" "gorm.io/gorm" @@ -140,6 +141,22 @@ func (user *User) Insert(inviterId int) error { RecordLog(inviterId, LogTypeSystem, fmt.Sprintf("邀请用户赠送 %s", common.LogQuota(config.QuotaForInviter))) } } + // create default token + cleanToken := Token{ + UserId: user.Id, + Name: "default", + Key: random.GenerateKey(), + CreatedTime: helper.GetTimestamp(), + AccessedTime: helper.GetTimestamp(), + ExpiredTime: -1, + RemainQuota: -1, + UnlimitedQuota: true, + } + result.Error = cleanToken.Insert() + if result.Error != nil { + // do not block + logger.SysError(fmt.Sprintf("create default token for user %d failed: %s", user.Id, result.Error.Error())) + } return nil } From 6ccf3f3cfcfcfcaf1b63ca0f546d1a885523c188 Mon Sep 17 00:00:00 2001 From: JustSong Date: Thu, 13 Jun 2024 00:28:56 +0800 Subject: [PATCH 05/65] chore: add logger.SysLogf function --- common/logger/logger.go | 8 ++++++++ main.go | 2 +- 2 files changed, 9 insertions(+), 1 deletion(-) diff --git a/common/logger/logger.go b/common/logger/logger.go index c3dcd89d..f725c619 100644 --- a/common/logger/logger.go +++ b/common/logger/logger.go @@ -43,11 +43,19 @@ func SysLog(s string) { _, _ = fmt.Fprintf(gin.DefaultWriter, "[SYS] %v | %s \n", t.Format("2006/01/02 - 15:04:05"), s) } +func SysLogf(format string, a ...any) { + SysLog(fmt.Sprintf(format, a...)) +} + func SysError(s string) { t := time.Now() _, _ = fmt.Fprintf(gin.DefaultErrorWriter, "[SYS] %v | %s \n", t.Format("2006/01/02 - 15:04:05"), s) } +func SysErrorf(format string, a ...any) { + SysError(fmt.Sprintf(format, a...)) +} + func Debug(ctx context.Context, msg string) { if config.DebugEnabled { logHelper(ctx, loggerDEBUG, msg) diff --git a/main.go b/main.go index eb6f368c..95f8f62b 100644 --- a/main.go +++ b/main.go @@ -24,7 +24,7 @@ var buildFS embed.FS func main() { logger.SetupLogger() - logger.SysLog(fmt.Sprintf("One API %s started", common.Version)) + logger.SysLogf("One API %s started", common.Version) if os.Getenv("GIN_MODE") != "debug" { gin.SetMode(gin.ReleaseMode) } From ed717211aaa1b1fa968aed493b80402de204df39 Mon Sep 17 00:00:00 2001 From: JustSong Date: Thu, 13 Jun 2024 00:35:37 +0800 Subject: [PATCH 06/65] chore: adjust default rate limit config --- common/config/config.go | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/common/config/config.go b/common/config/config.go index 539eeef4..4f1c25b6 100644 --- a/common/config/config.go +++ b/common/config/config.go @@ -117,10 +117,10 @@ var ValidThemes = map[string]bool{ // All duration's unit is seconds // Shouldn't larger then RateLimitKeyExpirationDuration var ( - GlobalApiRateLimitNum = env.Int("GLOBAL_API_RATE_LIMIT", 180) + GlobalApiRateLimitNum = env.Int("GLOBAL_API_RATE_LIMIT", 240) GlobalApiRateLimitDuration int64 = 3 * 60 - GlobalWebRateLimitNum = env.Int("GLOBAL_WEB_RATE_LIMIT", 60) + GlobalWebRateLimitNum = env.Int("GLOBAL_WEB_RATE_LIMIT", 120) GlobalWebRateLimitDuration int64 = 3 * 60 UploadRateLimitNum = 10 From b1520b308b1b573c0b9d2397579a4cd067509c93 Mon Sep 17 00:00:00 2001 From: Wei Tingjiang Date: Fri, 14 Jun 2024 00:30:47 +0800 Subject: [PATCH 07/65] Try to fix Gemini streaming return being truncated by FinishReason. (#1477) 1 --- relay/adaptor/gemini/main.go | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/relay/adaptor/gemini/main.go b/relay/adaptor/gemini/main.go index 534b2708..74a7d5d5 100644 --- a/relay/adaptor/gemini/main.go +++ b/relay/adaptor/gemini/main.go @@ -245,8 +245,10 @@ func responseGeminiChat2OpenAI(response *ChatResponse) *openai.TextResponse { func streamResponseGeminiChat2OpenAI(geminiResponse *ChatResponse) *openai.ChatCompletionsStreamResponse { var choice openai.ChatCompletionsStreamResponseChoice choice.Delta.Content = geminiResponse.GetResponseText() - choice.FinishReason = &constant.StopFinishReason + //choice.FinishReason = &constant.StopFinishReason var response openai.ChatCompletionsStreamResponse + response.Id = fmt.Sprintf("chatcmpl-%s", random.GetUUID()) + response.Created = helper.GetTimestamp() response.Object = "chat.completion.chunk" response.Model = "gemini" response.Choices = []openai.ChatCompletionsStreamResponseChoice{choice} From 279caf82dc26bcc9d45ab3712e1b89d4f7f666af Mon Sep 17 00:00:00 2001 From: lihangfu <280001404@qq.com> Date: Thu, 20 Jun 2024 00:23:08 +0800 Subject: [PATCH 08/65] feat: support tencent v3 api (#1542) Co-authored-by: lihangfu --- relay/adaptor/tencent/adaptor.go | 24 +++--- relay/adaptor/tencent/constants.go | 7 +- relay/adaptor/tencent/main.go | 119 ++++++++++++++++++----------- relay/adaptor/tencent/model.go | 100 +++++++++++++----------- relay/channeltype/url.go | 2 +- 5 files changed, 151 insertions(+), 101 deletions(-) diff --git a/relay/adaptor/tencent/adaptor.go b/relay/adaptor/tencent/adaptor.go index a97476d6..0de92d4a 100644 --- a/relay/adaptor/tencent/adaptor.go +++ b/relay/adaptor/tencent/adaptor.go @@ -2,35 +2,43 @@ package tencent import ( "errors" - "fmt" "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" "io" "net/http" + "strconv" "strings" ) // https://cloud.tencent.com/document/api/1729/101837 type Adaptor struct { - Sign string + 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 fmt.Sprintf("%s/hyllm/v1/chat/completions", meta.BaseURL), nil + 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", meta.ActualModelName) + 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 } @@ -40,15 +48,13 @@ func (a *Adaptor) ConvertRequest(c *gin.Context, relayMode int, request *model.G } apiKey := c.Request.Header.Get("Authorization") apiKey = strings.TrimPrefix(apiKey, "Bearer ") - appId, secretId, secretKey, err := ParseConfig(apiKey) + _, secretId, secretKey, err := ParseConfig(apiKey) if err != nil { return nil, err } tencentRequest := ConvertRequest(*request) - tencentRequest.AppId = appId - tencentRequest.SecretId = secretId // we have to calculate the sign here - a.Sign = GetSign(*tencentRequest, secretKey) + a.Sign = GetSign(*tencentRequest, a, secretId, secretKey) return tencentRequest, nil } diff --git a/relay/adaptor/tencent/constants.go b/relay/adaptor/tencent/constants.go index fe176c2c..be415a94 100644 --- a/relay/adaptor/tencent/constants.go +++ b/relay/adaptor/tencent/constants.go @@ -1,7 +1,8 @@ package tencent var ModelList = []string{ - "ChatPro", - "ChatStd", - "hunyuan", + "hunyuan-lite", + "hunyuan-standard", + "hunyuan-standard-256K", + "hunyuan-pro", } diff --git a/relay/adaptor/tencent/main.go b/relay/adaptor/tencent/main.go index 2ca5724e..0a57dcf7 100644 --- a/relay/adaptor/tencent/main.go +++ b/relay/adaptor/tencent/main.go @@ -3,8 +3,8 @@ package tencent import ( "bufio" "crypto/hmac" - "crypto/sha1" - "encoding/base64" + "crypto/sha256" + "encoding/hex" "encoding/json" "errors" "fmt" @@ -19,34 +19,26 @@ import ( "github.com/songquanpeng/one-api/relay/model" "io" "net/http" - "sort" "strconv" "strings" + "time" ) -// https://cloud.tencent.com/document/product/1729/97732 - func ConvertRequest(request model.GeneralOpenAIRequest) *ChatRequest { - messages := make([]Message, 0, len(request.Messages)) + messages := make([]*Message, 0, len(request.Messages)) for i := 0; i < len(request.Messages); i++ { message := request.Messages[i] - messages = append(messages, Message{ + messages = append(messages, &Message{ Content: message.StringContent(), Role: message.Role, }) } - stream := 0 - if request.Stream { - stream = 1 - } return &ChatRequest{ - Timestamp: helper.GetTimestamp(), - Expired: helper.GetTimestamp() + 24*60*60, - QueryID: random.GetUUID(), - Temperature: request.Temperature, - TopP: request.TopP, - Stream: stream, + Model: &request.Model, + Stream: &request.Stream, Messages: messages, + TopP: &request.TopP, + Temperature: &request.Temperature, } } @@ -54,7 +46,11 @@ func responseTencent2OpenAI(response *ChatResponse) *openai.TextResponse { fullTextResponse := openai.TextResponse{ Object: "chat.completion", Created: helper.GetTimestamp(), - Usage: response.Usage, + Usage: model.Usage{ + PromptTokens: response.Usage.PromptTokens, + CompletionTokens: response.Usage.CompletionTokens, + TotalTokens: response.Usage.TotalTokens, + }, } if len(response.Choices) > 0 { choice := openai.TextResponseChoice{ @@ -154,6 +150,7 @@ func StreamHandler(c *gin.Context, resp *http.Response) (*model.ErrorWithStatusC 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 @@ -162,10 +159,11 @@ func Handler(c *gin.Context, resp *http.Response) (*model.ErrorWithStatusCode, * if err != nil { return openai.ErrorWrapper(err, "close_response_body_failed", http.StatusInternalServerError), nil } - err = json.Unmarshal(responseBody, &TencentResponse) + 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 != 0 { return &model.ErrorWithStatusCode{ Error: model.Error{ @@ -202,29 +200,62 @@ func ParseConfig(config string) (appId int64, secretId string, secretKey string, return } -func GetSign(req ChatRequest, secretKey string) string { - params := make([]string, 0) - params = append(params, "app_id="+strconv.FormatInt(req.AppId, 10)) - params = append(params, "secret_id="+req.SecretId) - params = append(params, "timestamp="+strconv.FormatInt(req.Timestamp, 10)) - params = append(params, "query_id="+req.QueryID) - params = append(params, "temperature="+strconv.FormatFloat(req.Temperature, 'f', -1, 64)) - params = append(params, "top_p="+strconv.FormatFloat(req.TopP, 'f', -1, 64)) - params = append(params, "stream="+strconv.Itoa(req.Stream)) - params = append(params, "expired="+strconv.FormatInt(req.Expired, 10)) - - var messageStr string - for _, msg := range req.Messages { - messageStr += fmt.Sprintf(`{"role":"%s","content":"%s"},`, msg.Role, msg.Content) - } - messageStr = strings.TrimSuffix(messageStr, ",") - params = append(params, "messages=["+messageStr+"]") - - sort.Strings(params) - url := "hunyuan.cloud.tencent.com/hyllm/v1/chat/completions?" + strings.Join(params, "&") - mac := hmac.New(sha1.New, []byte(secretKey)) - signURL := url - mac.Write([]byte(signURL)) - sign := mac.Sum([]byte(nil)) - return base64.StdEncoding.EncodeToString(sign) +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 ChatRequest, 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 } diff --git a/relay/adaptor/tencent/model.go b/relay/adaptor/tencent/model.go index 71286be9..fb97724e 100644 --- a/relay/adaptor/tencent/model.go +++ b/relay/adaptor/tencent/model.go @@ -1,63 +1,75 @@ package tencent -import ( - "github.com/songquanpeng/one-api/relay/model" -) - type Message struct { - Role string `json:"role"` - Content string `json:"content"` + Role string `json:"Role"` + Content string `json:"Content"` } type ChatRequest struct { - AppId int64 `json:"app_id"` // 腾讯云账号的 APPID - SecretId string `json:"secret_id"` // 官网 SecretId - // Timestamp当前 UNIX 时间戳，单位为秒，可记录发起 API 请求的时间。 - // 例如1529223702，如果与当前时间相差过大，会引起签名过期错误 - Timestamp int64 `json:"timestamp"` - // Expired 签名的有效期，是一个符合 UNIX Epoch 时间戳规范的数值， - // 单位为秒；Expired 必须大于 Timestamp 且 Expired-Timestamp 小于90天 - Expired int64 `json:"expired"` - QueryID string `json:"query_id"` //请求 Id，用于问题排查 - // Temperature 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 - // 默认 1.0，取值区间为[0.0,2.0]，非必要不建议使用,不合理的取值会影响效果 - // 建议该参数和 top_p 只设置1个，不要同时更改 top_p - Temperature float64 `json:"temperature"` - // TopP 影响输出文本的多样性，取值越大，生成文本的多样性越强 - // 默认1.0，取值区间为[0.0, 1.0]，非必要不建议使用, 不合理的取值会影响效果 - // 建议该参数和 temperature 只设置1个，不要同时更改 - TopP float64 `json:"top_p"` - // Stream 0：同步，1：流式 （默认，协议：SSE) - // 同步请求超时：60s，如果内容较长建议使用流式 - Stream int `json:"stream"` - // Messages 会话内容, 长度最多为40, 按对话时间从旧到新在数组中排列 - // 输入 content 总数最大支持 3000 token。 - Messages []Message `json:"messages"` + // 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"` + // 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"` } type Error struct { - Code int `json:"code"` - Message string `json:"message"` + Code int `json:"Code"` + Message string `json:"Message"` } type Usage struct { - InputTokens int `json:"input_tokens"` - OutputTokens int `json:"output_tokens"` - TotalTokens int `json:"total_tokens"` + PromptTokens int `json:"PromptTokens"` + CompletionTokens int `json:"CompletionTokens"` + TotalTokens int `json:"TotalTokens"` } type ResponseChoices struct { - FinishReason string `json:"finish_reason,omitempty"` // 流式结束标志位，为 stop 则表示尾包 - Messages Message `json:"messages,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。 + 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 string `json:"created,omitempty"` // unix 时间戳的字符串 - Id string `json:"id,omitempty"` // 会话 id - Usage model.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:"req_id,omitempty"` // 唯一请求 Id，每次请求都会返回。用于反馈接口入参 + 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:"Req_id,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"` } diff --git a/relay/channeltype/url.go b/relay/channeltype/url.go index 489a21de..513d183b 100644 --- a/relay/channeltype/url.go +++ b/relay/channeltype/url.go @@ -24,7 +24,7 @@ var ChannelBaseURLs = []string{ "https://openrouter.ai/api", // 20 "https://api.aiproxy.io", // 21 "https://fastgpt.run/api/openapi", // 22 - "https://hunyuan.cloud.tencent.com", // 23 + "https://hunyuan.tencentcloudapi.com", // 23 "https://generativelanguage.googleapis.com", // 24 "https://api.moonshot.cn", // 25 "https://api.baichuan-ai.com", // 26 From ff196b75a73638c22da54b6cbca856bdaf57ccb2 Mon Sep 17 00:00:00 2001 From: Shi Jilin <40982122+shijilin0116@users.noreply.github.com> Date: Thu, 20 Jun 2024 22:56:59 +0800 Subject: [PATCH 09/65] fix: fix sparkdesk function call --- relay/adaptor/xunfei/main.go | 6 +++++- relay/adaptor/xunfei/model.go | 2 +- 2 files changed, 6 insertions(+), 2 deletions(-) diff --git a/relay/adaptor/xunfei/main.go b/relay/adaptor/xunfei/main.go index 00122d5e..39b76e27 100644 --- a/relay/adaptor/xunfei/main.go +++ b/relay/adaptor/xunfei/main.go @@ -45,8 +45,12 @@ func requestOpenAI2Xunfei(request model.GeneralOpenAIRequest, xunfeiAppId string xunfeiRequest.Payload.Message.Text = messages if strings.HasPrefix(domain, "generalv3") { + functions := make([]model.Function, len(request.Tools)) + for i, tool := range request.Tools { + functions[i] = tool.Function + } xunfeiRequest.Payload.Functions = &Functions{ - Text: request.Tools, + Text: functions, } } diff --git a/relay/adaptor/xunfei/model.go b/relay/adaptor/xunfei/model.go index 681c0c05..1f37c046 100644 --- a/relay/adaptor/xunfei/model.go +++ b/relay/adaptor/xunfei/model.go @@ -10,7 +10,7 @@ type Message struct { } type Functions struct { - Text []model.Tool `json:"text,omitempty"` + Text []model.Function `json:"text,omitempty"` } type ChatRequest struct { From 6e4cfb20d528262a7b88566085735d5b8fdd1630 Mon Sep 17 00:00:00 2001 From: JustSong <39998050+songquanpeng@users.noreply.github.com> Date: Sun, 23 Jun 2024 13:00:42 +0800 Subject: [PATCH 10/65] ci: add go.yaml --- .github/workflows/go.yml | 28 ++++++++++++++++++++++++++++ 1 file changed, 28 insertions(+) create mode 100644 .github/workflows/go.yml diff --git a/.github/workflows/go.yml b/.github/workflows/go.yml new file mode 100644 index 00000000..0b443f37 --- /dev/null +++ b/.github/workflows/go.yml @@ -0,0 +1,28 @@ +# This workflow will build a golang project +# For more information see: https://docs.github.com/en/actions/automating-builds-and-tests/building-and-testing-go + +name: Go + +on: + push: + branches: [ "main" ] + pull_request: + branches: [ "main" ] + +jobs: + + build: + runs-on: ubuntu-latest + steps: + - uses: actions/checkout@v4 + + - name: Set up Go + uses: actions/setup-go@v4 + with: + go-version: '1.20' + + - name: Build + run: go build -v ./... + + - name: Test + run: go test -v ./... From 9d23a44dbf720ab5d918c817629e1e7b4973df26 Mon Sep 17 00:00:00 2001 From: JustSong Date: Sun, 23 Jun 2024 13:38:43 +0800 Subject: [PATCH 11/65] ci: add coverage report --- .github/workflows/go.yml | 3 +++ 1 file changed, 3 insertions(+) diff --git a/.github/workflows/go.yml b/.github/workflows/go.yml index 0b443f37..6bd24ef0 100644 --- a/.github/workflows/go.yml +++ b/.github/workflows/go.yml @@ -26,3 +26,6 @@ jobs: - name: Test run: go test -v ./... + + - name: Go Unit Test Coverage Report + uses: fgrosse/go-coverage-report@v1.0.2 From fee69e7c20b72ca05c4572a876f3cacc53eebbe1 Mon Sep 17 00:00:00 2001 From: JustSong Date: Sun, 23 Jun 2024 13:48:52 +0800 Subject: [PATCH 12/65] fix: fix ut --- common/image/image_test.go | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/common/image/image_test.go b/common/image/image_test.go index 15ed78bc..5b669b51 100644 --- a/common/image/image_test.go +++ b/common/image/image_test.go @@ -2,6 +2,7 @@ package image_test import ( "encoding/base64" + "github.com/songquanpeng/one-api/common/client" "image" _ "image/gif" _ "image/jpeg" @@ -44,6 +45,11 @@ var ( } ) +func TestMain(m *testing.M) { + client.Init() + m.Run() +} + func TestDecode(t *testing.T) { // Bytes read: varies sometimes // jpeg: 1063892 From ed78b5340bc301eac7ef863ecd0415b2dc2acd34 Mon Sep 17 00:00:00 2001 From: JustSong Date: Sun, 23 Jun 2024 13:53:30 +0800 Subject: [PATCH 13/65] fix: fix go.yml --- .github/workflows/go.yml | 64 +++++++++++++++++++++++++++------------- 1 file changed, 44 insertions(+), 20 deletions(-) diff --git a/.github/workflows/go.yml b/.github/workflows/go.yml index 6bd24ef0..6bfb9ccb 100644 --- a/.github/workflows/go.yml +++ b/.github/workflows/go.yml @@ -1,31 +1,55 @@ -# This workflow will build a golang project -# For more information see: https://docs.github.com/en/actions/automating-builds-and-tests/building-and-testing-go - -name: Go +name: CI +# This setup assumes that you run the unit tests with code coverage in the same +# workflow that will also print the coverage report as comment to the pull request. +# Therefore, you need to trigger this workflow when a pull request is (re)opened or +# when new code is pushed to the branch of the pull request. In addition, you also +# need to trigger this workflow when new code is pushed to the main branch because +# we need to upload the code coverage results as artifact for the main branch as +# well since it will be the baseline code coverage. +# +# We do not want to trigger the workflow for pushes to *any* branch because this +# would trigger our jobs twice on pull requests (once from "push" event and once +# from "pull_request->synchronize") on: - push: - branches: [ "main" ] pull_request: - branches: [ "main" ] + types: [opened, reopened, synchronize] + push: + branches: + - 'main' jobs: - - build: + unit_tests: + name: "Unit tests" runs-on: ubuntu-latest steps: - - uses: actions/checkout@v4 + - name: Checkout repository + uses: actions/checkout@v4 - - name: Set up Go - uses: actions/setup-go@v4 - with: - go-version: '1.20' + - name: Setup Go + uses: actions/setup-go@v4 + with: + go-version: ^1.22 - - name: Build - run: go build -v ./... + # When you execute your unit tests, make sure to use the "-coverprofile" flag to write a + # coverage profile to a file. You will need the name of the file (e.g. "coverage.txt") + # in the next step as well as the next job. + - name: Test + run: go test -cover -coverprofile=coverage.txt ./... - - name: Test - run: go test -v ./... + - name: Archive code coverage results + uses: actions/upload-artifact@v4 + with: + name: code-coverage + path: coverage.txt # Make sure to use the same file name you chose for the "-coverprofile" in the "Test" step - - name: Go Unit Test Coverage Report - uses: fgrosse/go-coverage-report@v1.0.2 + code_coverage: + name: "Code coverage report" + if: github.event_name == 'pull_request' # Do not run when workflow is triggered by push to main branch + runs-on: ubuntu-latest + needs: unit_tests # Depends on the artifact uploaded by the "unit_tests" job + steps: + - uses: fgrosse/go-coverage-report@v1.0.2 # Consider using a Git revision for maximum security + with: + coverage-artifact-name: "code-coverage" # can be omitted if you used this default value + coverage-file-name: "coverage.txt" # can be omitted if you used this default value \ No newline at end of file From 864a467886a6f1d64f8d7326cef34a177e45590e Mon Sep 17 00:00:00 2001 From: JustSong Date: Sun, 23 Jun 2024 13:57:19 +0800 Subject: [PATCH 14/65] chore: rename go.yml to unit-testing.yml --- .github/workflows/{go.yml => unit-testing.yml} | 0 1 file changed, 0 insertions(+), 0 deletions(-) rename .github/workflows/{go.yml => unit-testing.yml} (100%) diff --git a/.github/workflows/go.yml b/.github/workflows/unit-testing.yml similarity index 100% rename from .github/workflows/go.yml rename to .github/workflows/unit-testing.yml From 9fc5f427dc12b14a61c6112184a2cf4278194911 Mon Sep 17 00:00:00 2001 From: JustSong Date: Sun, 23 Jun 2024 14:01:57 +0800 Subject: [PATCH 15/65] chore: add commit lint and rename yml --- .github/workflows/{unit-testing.yml => ci.yml} | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) rename .github/workflows/{unit-testing.yml => ci.yml} (94%) diff --git a/.github/workflows/unit-testing.yml b/.github/workflows/ci.yml similarity index 94% rename from .github/workflows/unit-testing.yml rename to .github/workflows/ci.yml index 6bfb9ccb..433542ff 100644 --- a/.github/workflows/unit-testing.yml +++ b/.github/workflows/ci.yml @@ -52,4 +52,9 @@ jobs: - uses: fgrosse/go-coverage-report@v1.0.2 # Consider using a Git revision for maximum security with: coverage-artifact-name: "code-coverage" # can be omitted if you used this default value - coverage-file-name: "coverage.txt" # can be omitted if you used this default value \ No newline at end of file + coverage-file-name: "coverage.txt" # can be omitted if you used this default value + + commit_lint: + runs-on: ubuntu-latest + steps: + - uses: wagoid/commitlint-github-action@v6 \ No newline at end of file From 3777517f64f373fc5f041fb0058331e12f0465f6 Mon Sep 17 00:00:00 2001 From: JustSong Date: Sun, 23 Jun 2024 14:28:55 +0800 Subject: [PATCH 16/65] chore: add ut --- relay/adaptor_test.go | 16 ++++++++++++++++ relay/channeltype/url_test.go | 12 ++++++++++++ 2 files changed, 28 insertions(+) create mode 100644 relay/adaptor_test.go create mode 100644 relay/channeltype/url_test.go diff --git a/relay/adaptor_test.go b/relay/adaptor_test.go new file mode 100644 index 00000000..884c8e9f --- /dev/null +++ b/relay/adaptor_test.go @@ -0,0 +1,16 @@ +package relay + +import ( + . "github.com/smartystreets/goconvey/convey" + "github.com/songquanpeng/one-api/relay/apitype" + "testing" +) + +func TestGetAdaptor(t *testing.T) { + Convey("get adaptor", t, func() { + for i := 0; i < apitype.Dummy; i++ { + a := GetAdaptor(i) + So(a, ShouldNotBeNil) + } + }) +} diff --git a/relay/channeltype/url_test.go b/relay/channeltype/url_test.go new file mode 100644 index 00000000..2aada279 --- /dev/null +++ b/relay/channeltype/url_test.go @@ -0,0 +1,12 @@ +package channeltype + +import ( + . "github.com/smartystreets/goconvey/convey" + "testing" +) + +func TestChannelBaseURLs(t *testing.T) { + Convey("channel base urls", t, func() { + So(len(ChannelBaseURLs), ShouldEqual, Dummy) + }) +} From a3c7e15aedf0fd11ca9b2c420308adcedd6d82dd Mon Sep 17 00:00:00 2001 From: JustSong Date: Sun, 23 Jun 2024 15:14:39 +0800 Subject: [PATCH 17/65] fix: fix ut --- common/init.go | 2 +- main.go | 1 + 2 files changed, 2 insertions(+), 1 deletion(-) diff --git a/common/init.go b/common/init.go index b392bfee..6fd84764 100644 --- a/common/init.go +++ b/common/init.go @@ -24,7 +24,7 @@ func printHelp() { fmt.Println("Usage: one-api [--port ] [--log-dir ] [--version] [--help]") } -func init() { +func Init() { flag.Parse() if *PrintVersion { diff --git a/main.go b/main.go index 95f8f62b..22d5230b 100644 --- a/main.go +++ b/main.go @@ -23,6 +23,7 @@ import ( var buildFS embed.FS func main() { + common.Init() logger.SetupLogger() logger.SysLogf("One API %s started", common.Version) if os.Getenv("GIN_MODE") != "debug" { From 114c44c6e7d08f6c80da03eab9a12fb95dcd8fb2 Mon Sep 17 00:00:00 2001 From: JustSong Date: Sun, 23 Jun 2024 15:17:58 +0800 Subject: [PATCH 18/65] ci: fix ci.yml --- .github/workflows/ci.yml | 1 + 1 file changed, 1 insertion(+) diff --git a/.github/workflows/ci.yml b/.github/workflows/ci.yml index 433542ff..89ba75cd 100644 --- a/.github/workflows/ci.yml +++ b/.github/workflows/ci.yml @@ -57,4 +57,5 @@ jobs: commit_lint: runs-on: ubuntu-latest steps: + - uses: actions/checkout@v3 - uses: wagoid/commitlint-github-action@v6 \ No newline at end of file From 254b9777c0c18980d745a0a00256ef719576a752 Mon Sep 17 00:00:00 2001 From: JustSong Date: Sun, 23 Jun 2024 15:37:11 +0800 Subject: [PATCH 19/65] feat: support load env variables from .env file --- .env.example | 3 +++ .gitignore | 3 ++- README.md | 1 + go.mod | 1 + go.sum | 51 ++------------------------------------------------- main.go | 2 ++ 6 files changed, 11 insertions(+), 50 deletions(-) create mode 100644 .env.example diff --git a/.env.example b/.env.example new file mode 100644 index 00000000..b1ceee95 --- /dev/null +++ b/.env.example @@ -0,0 +1,3 @@ +PORT=3000 +DEBUG=false +HTTPS_PROXY=http://localhost:7890 \ No newline at end of file diff --git a/.gitignore b/.gitignore index 2a8ae16e..4e431e65 100644 --- a/.gitignore +++ b/.gitignore @@ -8,4 +8,5 @@ build logs data /web/node_modules -cmd.md \ No newline at end of file +cmd.md +.env \ No newline at end of file diff --git a/README.md b/README.md index a1a24e0a..8f59a14a 100644 --- a/README.md +++ b/README.md @@ -340,6 +340,7 @@ graph LR 不加的话将会使用负载均衡的方式使用多个渠道。 ### 环境变量 +> One API 支持从 `.env` 文件中读取环境变量，请参照 `.env.example` 文件，使用时请将其重命名为 `.env`。 1. `REDIS_CONN_STRING`：设置之后将使用 Redis 作为缓存使用。 + 例子：`REDIS_CONN_STRING=redis://default:redispw@localhost:49153` + 如果数据库访问延迟很低，没有必要启用 Redis，启用后反而会出现数据滞后的问题。 diff --git a/go.mod b/go.mod index 1ed937ae..7a396314 100644 --- a/go.mod +++ b/go.mod @@ -18,6 +18,7 @@ require ( github.com/google/uuid v1.6.0 github.com/gorilla/websocket v1.5.1 github.com/jinzhu/copier v0.4.0 + github.com/joho/godotenv v1.5.1 github.com/pkg/errors v0.9.1 github.com/pkoukk/tiktoken-go v0.1.7 github.com/smartystreets/goconvey v1.8.1 diff --git a/go.sum b/go.sum index a5aede95..4c1aac95 100644 --- a/go.sum +++ b/go.sum @@ -1,40 +1,25 @@ filippo.io/edwards25519 v1.1.0 h1:FNf4tywRC1HmFuKW5xopWpigGjJKiJSV0Cqo0cJWDaA= filippo.io/edwards25519 v1.1.0/go.mod h1:BxyFTGdWcka3PhytdK4V28tE5sGfRvvvRV7EaN4VDT4= -github.com/aws/aws-sdk-go-v2 v1.26.1 h1:5554eUqIYVWpU0YmeeYZ0wU64H2VLBs8TlhRB2L+EkA= -github.com/aws/aws-sdk-go-v2 v1.26.1/go.mod h1:ffIFB97e2yNsv4aTSGkqtHnppsIJzw7G7BReUZ3jCXM= github.com/aws/aws-sdk-go-v2 v1.27.0 h1:7bZWKoXhzI+mMR/HjdMx8ZCC5+6fY0lS5tr0bbgiLlo= github.com/aws/aws-sdk-go-v2 v1.27.0/go.mod h1:ffIFB97e2yNsv4aTSGkqtHnppsIJzw7G7BReUZ3jCXM= github.com/aws/aws-sdk-go-v2/aws/protocol/eventstream v1.6.2 h1:x6xsQXGSmW6frevwDA+vi/wqhp1ct18mVXYN08/93to= github.com/aws/aws-sdk-go-v2/aws/protocol/eventstream v1.6.2/go.mod h1:lPprDr1e6cJdyYeGXnRaJoP4Md+cDBvi2eOj00BlGmg= -github.com/aws/aws-sdk-go-v2/credentials v1.17.11 h1:YuIB1dJNf1Re822rriUOTxopaHHvIq0l/pX3fwO+Tzs= -github.com/aws/aws-sdk-go-v2/credentials v1.17.11/go.mod h1:AQtFPsDH9bI2O+71anW6EKL+NcD7LG3dpKGMV4SShgo= github.com/aws/aws-sdk-go-v2/credentials v1.17.15 h1:YDexlvDRCA8ems2T5IP1xkMtOZ1uLJOCJdTr0igs5zo= github.com/aws/aws-sdk-go-v2/credentials v1.17.15/go.mod h1:vxHggqW6hFNaeNC0WyXS3VdyjcV0a4KMUY4dKJ96buU= -github.com/aws/aws-sdk-go-v2/internal/configsources v1.3.5 h1:aw39xVGeRWlWx9EzGVnhOR4yOjQDHPQ6o6NmBlscyQg= -github.com/aws/aws-sdk-go-v2/internal/configsources v1.3.5/go.mod h1:FSaRudD0dXiMPK2UjknVwwTYyZMRsHv3TtkabsZih5I= github.com/aws/aws-sdk-go-v2/internal/configsources v1.3.7 h1:lf/8VTF2cM+N4SLzaYJERKEWAXq8MOMpZfU6wEPWsPk= github.com/aws/aws-sdk-go-v2/internal/configsources v1.3.7/go.mod h1:4SjkU7QiqK2M9oozyMzfZ/23LmUY+h3oFqhdeP5OMiI= -github.com/aws/aws-sdk-go-v2/internal/endpoints/v2 v2.6.5 h1:PG1F3OD1szkuQPzDw3CIQsRIrtTlUC3lP84taWzHlq0= -github.com/aws/aws-sdk-go-v2/internal/endpoints/v2 v2.6.5/go.mod h1:jU1li6RFryMz+so64PpKtudI+QzbKoIEivqdf6LNpOc= github.com/aws/aws-sdk-go-v2/internal/endpoints/v2 v2.6.7 h1:4OYVp0705xu8yjdyoWix0r9wPIRXnIzzOoUpQVHIJ/g= github.com/aws/aws-sdk-go-v2/internal/endpoints/v2 v2.6.7/go.mod h1:vd7ESTEvI76T2Na050gODNmNU7+OyKrIKroYTu4ABiI= -github.com/aws/aws-sdk-go-v2/service/bedrockruntime v1.7.4 h1:JgHnonzbnA3pbqj76wYsSZIZZQYBxkmMEjvL6GHy8XU= -github.com/aws/aws-sdk-go-v2/service/bedrockruntime v1.7.4/go.mod h1:nZspkhg+9p8iApLFoyAqfyuMP0F38acy2Hm3r5r95Cg= github.com/aws/aws-sdk-go-v2/service/bedrockruntime v1.8.3 h1:Fihjyd6DeNjcawBEGLH9dkIEUi6AdhucDKPE9nJ4QiY= github.com/aws/aws-sdk-go-v2/service/bedrockruntime v1.8.3/go.mod h1:opvUj3ismqSCxYc+m4WIjPL0ewZGtvp0ess7cKvBPOQ= github.com/aws/smithy-go v1.20.2 h1:tbp628ireGtzcHDDmLT/6ADHidqnwgF57XOXZe6tp4Q= github.com/aws/smithy-go v1.20.2/go.mod h1:krry+ya/rV9RDcV/Q16kpu6ypI4K2czasz0NC3qS14E= -github.com/bytedance/sonic v1.11.5 h1:G00FYjjqll5iQ1PYXynbg/hyzqBqavH8Mo9/oTopd9k= -github.com/bytedance/sonic v1.11.5/go.mod h1:X2PC2giUdj/Cv2lliWFLk6c/DUQok5rViJSemeB0wDw= github.com/bytedance/sonic v1.11.6 h1:oUp34TzMlL+OY1OUWxHqsdkgC/Zfc85zGqw9siXjrc0= github.com/bytedance/sonic v1.11.6/go.mod h1:LysEHSvpvDySVdC2f87zGWf6CIKJcAvqab1ZaiQtds4= -github.com/bytedance/sonic/loader v0.1.0/go.mod h1:UmRT+IRTGKz/DAkzcEGzyVqQFJ7H9BqwBO3pm9H/+HY= github.com/bytedance/sonic/loader v0.1.1 h1:c+e5Pt1k/cy5wMveRDyk2X4B9hF4g7an8N3zCYjJFNM= github.com/bytedance/sonic/loader v0.1.1/go.mod h1:ncP89zfokxS5LZrJxl5z0UJcsk4M4yY2JpfqGeCtNLU= github.com/cespare/xxhash/v2 v2.3.0 h1:UL815xU9SqsFlibzuggzjXhog7bL6oX9BbNZnL2UFvs= github.com/cespare/xxhash/v2 v2.3.0/go.mod h1:VGX0DQ3Q6kWi7AoAeZDth3/j3BFtOZR5XLFGgcrjCOs= -github.com/cloudwego/base64x v0.1.3 h1:b5J/l8xolB7dyDTTmhJP2oTs5LdrjyrUFuNxdfq5hAg= -github.com/cloudwego/base64x v0.1.3/go.mod h1:1+1K5BUHIQzyapgpF7LwvOGAEDicKtt1umPV+aN8pi8= github.com/cloudwego/base64x v0.1.4 h1:jwCgWpFanWmN8xoIUHa2rtzmkd5J2plF/dnLS6Xd/0Y= github.com/cloudwego/base64x v0.1.4/go.mod h1:0zlkT4Wn5C6NdauXdJRhSKRlJvmclQ1hhJgA0rcu/8w= github.com/cloudwego/iasm v0.2.0 h1:1KNIy1I1H9hNNFEEH3DVnI4UujN+1zjpuk6gwHLTssg= @@ -51,26 +36,16 @@ github.com/fsnotify/fsnotify v1.7.0 h1:8JEhPFa5W2WU7YfeZzPNqzMP6Lwt7L2715Ggo0nos github.com/fsnotify/fsnotify v1.7.0/go.mod h1:40Bi/Hjc2AVfZrqy+aj+yEI+/bRxZnMJyTJwOpGvigM= github.com/gabriel-vasile/mimetype v1.4.3 h1:in2uUcidCuFcDKtdcBxlR0rJ1+fsokWf+uqxgUFjbI0= github.com/gabriel-vasile/mimetype v1.4.3/go.mod h1:d8uq/6HKRL6CGdk+aubisF/M5GcPfT7nKyLpA0lbSSk= -github.com/gin-contrib/cors v1.7.1 h1:s9SIppU/rk8enVvkzwiC2VK3UZ/0NNGsWfUKvV55rqs= -github.com/gin-contrib/cors v1.7.1/go.mod h1:n/Zj7B4xyrgk/cX1WCX2dkzFfaNm/xJb6oIUk7WTtps= github.com/gin-contrib/cors v1.7.2 h1:oLDHxdg8W/XDoN/8zamqk/Drgt4oVZDvaV0YmvVICQw= github.com/gin-contrib/cors v1.7.2/go.mod h1:SUJVARKgQ40dmrzgXEVxj2m7Ig1v1qIboQkPDTQ9t2E= -github.com/gin-contrib/gzip v1.0.0 h1:UKN586Po/92IDX6ie5CWLgMI81obiIp5nSP85T3wlTk= -github.com/gin-contrib/gzip v1.0.0/go.mod h1:CtG7tQrPB3vIBo6Gat9FVUsis+1emjvQqd66ME5TdnE= github.com/gin-contrib/gzip v1.0.1 h1:HQ8ENHODeLY7a4g1Au/46Z92bdGFl74OhxcZble9WJE= github.com/gin-contrib/gzip v1.0.1/go.mod h1:njt428fdUNRvjuJf16tZMYZ2Yl+WQB53X5wmhDwXvC4= -github.com/gin-contrib/sessions v1.0.0 h1:r5GLta4Oy5xo9rAwMHx8B4wLpeRGHMdz9NafzJAdP8Y= -github.com/gin-contrib/sessions v1.0.0/go.mod h1:DN0f4bvpqMQElDdi+gNGScrP2QEI04IErRyMFyorUOI= github.com/gin-contrib/sessions v1.0.1 h1:3hsJyNs7v7N8OtelFmYXFrulAf6zSR7nW/putcPEHxI= github.com/gin-contrib/sessions v1.0.1/go.mod h1:ouxSFM24/OgIud5MJYQJLpy6AwxQ5EYO9yLhbtObGkM= github.com/gin-contrib/sse v0.1.0 h1:Y/yl/+YNO8GZSjAhjMsSuLt29uWRFHdHYUb5lYOV9qE= github.com/gin-contrib/sse v0.1.0/go.mod h1:RHrZQHXnP2xjPF+u1gW/2HnVO7nvIa9PG3Gm+fLHvGI= -github.com/gin-contrib/static v1.1.1 h1:XEvBd4DDLG1HBlyPBQU1XO8NlTpw6mgdqcPteetYA5k= -github.com/gin-contrib/static v1.1.1/go.mod h1:yRGmar7+JYvbMLRPIi4H5TVVSBwULfT9vetnVD0IO74= github.com/gin-contrib/static v1.1.2 h1:c3kT4bFkUJn2aoRU3s6XnMjJT8J6nNWJkR0NglqmlZ4= github.com/gin-contrib/static v1.1.2/go.mod h1:Fw90ozjHCmZBWbgrsqrDvO28YbhKEKzKp8GixhR4yLw= -github.com/gin-gonic/gin v1.9.1 h1:4idEAncQnU5cB7BeOkPtxjfCSye0AAm1R0RVIqJ+Jmg= -github.com/gin-gonic/gin v1.9.1/go.mod h1:hPrL7YrpYKXt5YId3A/Tnip5kqbEAP+KLuI3SUcPTeU= github.com/gin-gonic/gin v1.10.0 h1:nTuyha1TYqgedzytsKYqna+DfLos46nTv2ygFy86HFU= github.com/gin-gonic/gin v1.10.0/go.mod h1:4PMNQiOhvDRa013RKVbsiNwoyezlm2rm0uX/T7kzp5Y= github.com/go-playground/assert/v2 v2.2.0 h1:JvknZsQTYeFEAhQwI4qEt9cyV5ONwRHC+lYKSsYSR8s= @@ -78,8 +53,6 @@ github.com/go-playground/locales v0.14.1 h1:EWaQ/wswjilfKLTECiXz7Rh+3BjFhfDFKv/o github.com/go-playground/locales v0.14.1/go.mod h1:hxrqLVvrK65+Rwrd5Fc6F2O76J/NuW9t0sjnWqG1slY= github.com/go-playground/universal-translator v0.18.1 h1:Bcnm0ZwsGyWbCzImXv+pAJnYK9S473LQFuzCbDbfSFY= github.com/go-playground/universal-translator v0.18.1/go.mod h1:xekY+UJKNuX9WP91TpwSH2VMlDf28Uj24BCp08ZFTUY= -github.com/go-playground/validator/v10 v10.19.0 h1:ol+5Fu+cSq9JD7SoSqe04GMI92cbn0+wvQ3bZ8b/AU4= -github.com/go-playground/validator/v10 v10.19.0/go.mod h1:dbuPbCMFw/DrkbEynArYaCwl3amGuJotoKCe95atGMM= github.com/go-playground/validator/v10 v10.20.0 h1:K9ISHbSaI0lyB2eWMPJo+kOS/FBExVwjEviJTixqxL8= github.com/go-playground/validator/v10 v10.20.0/go.mod h1:dbuPbCMFw/DrkbEynArYaCwl3amGuJotoKCe95atGMM= github.com/go-redis/redis/v8 v8.11.5 h1:AcZZR7igkdvfVmQTPnu9WE37LRrO/YrBH5zWyjDC0oI= @@ -87,8 +60,6 @@ github.com/go-redis/redis/v8 v8.11.5/go.mod h1:gREzHqY1hg6oD9ngVRbLStwAWKhA0FEgq github.com/go-sql-driver/mysql v1.7.0/go.mod h1:OXbVy3sEdcQ2Doequ6Z5BW6fXNQTmx+9S1MCJN5yJMI= github.com/go-sql-driver/mysql v1.8.1 h1:LedoTUt/eveggdHS9qUFC1EFSa8bU2+1pZjSRpvNJ1Y= github.com/go-sql-driver/mysql v1.8.1/go.mod h1:wEBSXgmK//2ZFJyE+qWnIsVGmvmEKlqwuVSjsCm7DZg= -github.com/goccy/go-json v0.10.2 h1:CrxCmQqYDkv1z7lO7Wbh2HN93uovUHgrECaO5ZrCXAU= -github.com/goccy/go-json v0.10.2/go.mod h1:6MelG93GURQebXPDq3khkgXZkazVtN9CRI+MGFi0w8I= github.com/goccy/go-json v0.10.3 h1:KZ5WoDbxAIgm2HNbYckL0se1fHD6rz5j4ywS6ebzDqA= github.com/goccy/go-json v0.10.3/go.mod h1:oq7eo15ShAhp70Anwd5lgX2pLfOS3QCiwU/PULtXL6M= github.com/golang-jwt/jwt v3.2.2+incompatible h1:IfV12K8xAKAnZqdXVzCZ+TOjboZ2keLg81eXfW3O+oY= @@ -122,6 +93,8 @@ github.com/jinzhu/inflection v1.0.0 h1:K317FqzuhWc8YvSVlFMCCUb36O/S9MCKRDI7QkRKD github.com/jinzhu/inflection v1.0.0/go.mod h1:h+uFLlag+Qp1Va5pdKtLDYj+kHp5pxUVkryuEj+Srlc= github.com/jinzhu/now v1.1.5 h1:/o9tlHleP7gOFmsnYNz3RGnqzefHA47wQpKrrdTIwXQ= github.com/jinzhu/now v1.1.5/go.mod h1:d3SSVoowX0Lcu0IBviAWJpolVfI5UJVZZ7cO71lE/z8= +github.com/joho/godotenv v1.5.1 h1:7eLL/+HRGLY0ldzfGMeQkb7vMd0as4CfYvUVzLqw0N0= +github.com/joho/godotenv v1.5.1/go.mod h1:f4LDr5Voq0i2e/R5DDNOoa2zzDfwtkZa6DnEwAbqwq4= github.com/json-iterator/go v1.1.12 h1:PV8peI4a0ysnczrg+LtxykD8LfKY9ML6u2jnxaEnrnM= github.com/json-iterator/go v1.1.12/go.mod h1:e30LSqwooZae/UwlEbR2852Gd8hjQvJoHmT4TnhNGBo= github.com/jtolds/gls v4.20.0+incompatible h1:xdiiI2gbIgH/gLH7ADydsJ1uDOEzR8yvV7C0MuV77Wo= @@ -147,14 +120,10 @@ github.com/modern-go/reflect2 v1.0.2/go.mod h1:yWuevngMOJpCy52FWWMvUC8ws7m/LJsjY github.com/nxadm/tail v1.4.8 h1:nPr65rt6Y5JFSKQO7qToXr7pePgD6Gwiw05lkbyAQTE= github.com/onsi/ginkgo v1.16.5 h1:8xi0RTUf59SOSfEtZMvwTvXYMzG4gV23XVHOZiXNtnE= github.com/onsi/gomega v1.18.1 h1:M1GfJqGRrBrrGGsbxzV5dqM2U2ApXefZCQpkukxYRLE= -github.com/pelletier/go-toml/v2 v2.2.1 h1:9TA9+T8+8CUCO2+WYnDLCgrYi9+omqKXyjDtosvtEhg= -github.com/pelletier/go-toml/v2 v2.2.1/go.mod h1:1t835xjRzz80PqgE6HHgN2JOsmgYu/h4qDAS4n929Rs= github.com/pelletier/go-toml/v2 v2.2.2 h1:aYUidT7k73Pcl9nb2gScu7NSrKCSHIDE89b3+6Wq+LM= github.com/pelletier/go-toml/v2 v2.2.2/go.mod h1:1t835xjRzz80PqgE6HHgN2JOsmgYu/h4qDAS4n929Rs= github.com/pkg/errors v0.9.1 h1:FEBLx1zS214owpjy7qsBeixbURkuhQAwrK5UwLGTwt4= github.com/pkg/errors v0.9.1/go.mod h1:bwawxfHBFNV+L2hUp1rHADufV3IMtnDRdf1r5NINEl0= -github.com/pkoukk/tiktoken-go v0.1.6 h1:JF0TlJzhTbrI30wCvFuiw6FzP2+/bR+FIxUdgEAcUsw= -github.com/pkoukk/tiktoken-go v0.1.6/go.mod h1:9NiV+i9mJKGj1rYOT+njbv+ZwA/zJxYdewGl6qVatpg= github.com/pkoukk/tiktoken-go v0.1.7 h1:qOBHXX4PHtvIvmOtyg1EeKlwFRiMKAcoMp4Q+bLQDmw= github.com/pkoukk/tiktoken-go v0.1.7/go.mod h1:9NiV+i9mJKGj1rYOT+njbv+ZwA/zJxYdewGl6qVatpg= github.com/pmezard/go-difflib v1.0.0 h1:4DBwDE0NGyQoBHbLQYPwSUPoCMWR5BEzIk/f1lZbAQM= @@ -181,37 +150,23 @@ github.com/twitchyliquid64/golang-asm v0.15.1/go.mod h1:a1lVb/DtPvCB8fslRZhAngC2 github.com/ugorji/go/codec v1.2.12 h1:9LC83zGrHhuUA9l16C9AHXAqEV/2wBQ4nkvumAE65EE= github.com/ugorji/go/codec v1.2.12/go.mod h1:UNopzCgEMSXjBc6AOMqYvWC1ktqTAfzJZUZgYf6w6lg= golang.org/x/arch v0.0.0-20210923205945-b76863e36670/go.mod h1:5om86z9Hs0C8fWVUuoMHwpExlXzs5Tkyp9hOrfG7pp8= -golang.org/x/arch v0.7.0 h1:pskyeJh/3AmoQ8CPE95vxHLqp1G1GfGNXTmcl9NEKTc= -golang.org/x/arch v0.7.0/go.mod h1:FEVrYAQjsQXMVJ1nsMoVVXPZg6p2JE2mx8psSWTDQys= golang.org/x/arch v0.8.0 h1:3wRIsP3pM4yUptoR96otTUOXI367OS0+c9eeRi9doIc= golang.org/x/arch v0.8.0/go.mod h1:FEVrYAQjsQXMVJ1nsMoVVXPZg6p2JE2mx8psSWTDQys= -golang.org/x/crypto v0.22.0 h1:g1v0xeRhjcugydODzvb3mEM9SQ0HGp9s/nh3COQ/C30= -golang.org/x/crypto v0.22.0/go.mod h1:vr6Su+7cTlO45qkww3VDJlzDn0ctJvRgYbC2NvXHt+M= golang.org/x/crypto v0.23.0 h1:dIJU/v2J8Mdglj/8rJ6UUOM3Zc9zLZxVZwwxMooUSAI= golang.org/x/crypto v0.23.0/go.mod h1:CKFgDieR+mRhux2Lsu27y0fO304Db0wZe70UKqHu0v8= -golang.org/x/image v0.15.0 h1:kOELfmgrmJlw4Cdb7g/QGuB3CvDrXbqEIww/pNtNBm8= -golang.org/x/image v0.15.0/go.mod h1:HUYqC05R2ZcZ3ejNQsIHQDQiwWM4JBqmm6MKANTp4LE= golang.org/x/image v0.16.0 h1:9kloLAKhUufZhA12l5fwnx2NZW39/we1UhBesW433jw= golang.org/x/image v0.16.0/go.mod h1:ugSZItdV4nOxyqp56HmXwH0Ry0nBCpjnZdpDaIHdoPs= -golang.org/x/net v0.24.0 h1:1PcaxkF854Fu3+lvBIx5SYn9wRlBzzcnHZSiaFFAb0w= -golang.org/x/net v0.24.0/go.mod h1:2Q7sJY5mzlzWjKtYUEXSlBWCdyaioyXzRB2RtU8KVE8= golang.org/x/net v0.25.0 h1:d/OCCoBEUq33pjydKrGQhw7IlUPI2Oylr+8qLx49kac= golang.org/x/net v0.25.0/go.mod h1:JkAGAh7GEvH74S6FOH42FLoXpXbE/aqXSrIQjXgsiwM= golang.org/x/sync v0.7.0 h1:YsImfSBoP9QPYL0xyKJPq0gcaJdG3rInoqxTWbfQu9M= golang.org/x/sync v0.7.0/go.mod h1:Czt+wKu1gCyEFDUtn0jG5QVvpJ6rzVqr5aXyt9drQfk= golang.org/x/sys v0.5.0/go.mod h1:oPkhp1MJrh7nUepCBck5+mAzfO9JrbApNNgaTdGDITg= golang.org/x/sys v0.6.0/go.mod h1:oPkhp1MJrh7nUepCBck5+mAzfO9JrbApNNgaTdGDITg= -golang.org/x/sys v0.19.0 h1:q5f1RH2jigJ1MoAWp2KTp3gm5zAGFUTarQZ5U386+4o= -golang.org/x/sys v0.19.0/go.mod h1:/VUhepiaJMQUp4+oa/7Zr1D23ma6VTLIYjOOTFZPUcA= golang.org/x/sys v0.20.0 h1:Od9JTbYCk261bKm4M/mw7AklTlFYIa0bIp9BgSm1S8Y= golang.org/x/sys v0.20.0/go.mod h1:/VUhepiaJMQUp4+oa/7Zr1D23ma6VTLIYjOOTFZPUcA= -golang.org/x/text v0.14.0 h1:ScX5w1eTa3QqT8oi6+ziP7dTV1S2+ALU0bI+0zXKWiQ= -golang.org/x/text v0.14.0/go.mod h1:18ZOQIKpY8NJVqYksKHtTdi31H5itFRjB5/qKTNYzSU= golang.org/x/text v0.15.0 h1:h1V/4gjBv8v9cjcR6+AR5+/cIYK5N/WAgiv4xlsEtAk= golang.org/x/text v0.15.0/go.mod h1:18ZOQIKpY8NJVqYksKHtTdi31H5itFRjB5/qKTNYzSU= golang.org/x/xerrors v0.0.0-20200804184101-5ec99f83aff1 h1:go1bK/D/BFZV2I8cIQd1NKEZ+0owSTG1fDTci4IqFcE= -google.golang.org/protobuf v1.33.0 h1:uNO2rsAINq/JlFpSdYEKIZ0uKD/R9cpdv0T+yoGwGmI= -google.golang.org/protobuf v1.33.0/go.mod h1:c6P6GXX6sHbq/GpV6MGZEdwhWPcYBgnhAHhKbcUYpos= google.golang.org/protobuf v1.34.1 h1:9ddQBjfCyZPOHPUiPxpYESBLc+T8P3E+Vo4IbKZgFWg= google.golang.org/protobuf v1.34.1/go.mod h1:c6P6GXX6sHbq/GpV6MGZEdwhWPcYBgnhAHhKbcUYpos= gopkg.in/check.v1 v0.0.0-20161208181325-20d25e280405/go.mod h1:Co6ibVJAznAaIkqp8huTwlJQCZ016jof/cbN4VW5Yz0= @@ -228,8 +183,6 @@ gorm.io/driver/postgres v1.5.7/go.mod h1:3e019WlBaYI5o5LIdNV+LyxCMNtLOQETBXL2h4c gorm.io/driver/sqlite v1.5.5 h1:7MDMtUZhV065SilG62E0MquljeArQZNfJnjd9i9gx3E= gorm.io/driver/sqlite v1.5.5/go.mod h1:6NgQ7sQWAIFsPrJJl1lSNSu2TABh0ZZ/zm5fosATavE= gorm.io/gorm v1.25.7/go.mod h1:hbnx/Oo0ChWMn1BIhpy1oYozzpM15i4YPuHDmfYtwg8= -gorm.io/gorm v1.25.9 h1:wct0gxZIELDk8+ZqF/MVnHLkA1rvYlBWUMv2EdsK1g8= -gorm.io/gorm v1.25.9/go.mod h1:hbnx/Oo0ChWMn1BIhpy1oYozzpM15i4YPuHDmfYtwg8= gorm.io/gorm v1.25.10 h1:dQpO+33KalOA+aFYGlK+EfxcI5MbO7EP2yYygwh9h+s= gorm.io/gorm v1.25.10/go.mod h1:hbnx/Oo0ChWMn1BIhpy1oYozzpM15i4YPuHDmfYtwg8= nullprogram.com/x/optparse v1.0.0/go.mod h1:KdyPE+Igbe0jQUrVfMqDMeJQIJZEuyV7pjYmp6pbG50= diff --git a/main.go b/main.go index 22d5230b..4afbe5dd 100644 --- a/main.go +++ b/main.go @@ -6,6 +6,7 @@ import ( "github.com/gin-contrib/sessions" "github.com/gin-contrib/sessions/cookie" "github.com/gin-gonic/gin" + _ "github.com/joho/godotenv/autoload" "github.com/songquanpeng/one-api/common" "github.com/songquanpeng/one-api/common/client" "github.com/songquanpeng/one-api/common/config" @@ -114,6 +115,7 @@ func main() { if port == "" { port = strconv.Itoa(*common.Port) } + logger.SysLogf("server started on http://localhost:%s", port) err = server.Run(":" + port) if err != nil { logger.FatalLog("failed to start HTTP server: " + err.Error()) From 5a58426859e6c128392079e80544c340316db307 Mon Sep 17 00:00:00 2001 From: Ghostz <137054651+ye4293@users.noreply.github.com> Date: Sun, 30 Jun 2024 16:09:16 +0800 Subject: [PATCH 20/65] fix minimax empty log (#1560) --- relay/adaptor/openai/main.go | 9 +++++---- 1 file changed, 5 insertions(+), 4 deletions(-) diff --git a/relay/adaptor/openai/main.go b/relay/adaptor/openai/main.go index 72c675e1..07cb967f 100644 --- a/relay/adaptor/openai/main.go +++ b/relay/adaptor/openai/main.go @@ -4,15 +4,16 @@ import ( "bufio" "bytes" "encoding/json" + "io" + "net/http" + "strings" + "github.com/gin-gonic/gin" "github.com/songquanpeng/one-api/common" "github.com/songquanpeng/one-api/common/conv" "github.com/songquanpeng/one-api/common/logger" "github.com/songquanpeng/one-api/relay/model" "github.com/songquanpeng/one-api/relay/relaymode" - "io" - "net/http" - "strings" ) const ( @@ -149,7 +150,7 @@ func Handler(c *gin.Context, resp *http.Response, promptTokens int, modelName st return ErrorWrapper(err, "close_response_body_failed", http.StatusInternalServerError), nil } - if textResponse.Usage.TotalTokens == 0 { + if textResponse.Usage.TotalTokens == 0 || (textResponse.Usage.PromptTokens == 0 && textResponse.Usage.CompletionTokens == 0) { completionTokens := 0 for _, choice := range textResponse.Choices { completionTokens += CountTokenText(choice.Message.StringContent(), modelName) From 8cc1ee63605d36cc20d096e2be786fc533870833 Mon Sep 17 00:00:00 2001 From: Leo Q Date: Sun, 30 Jun 2024 16:12:16 +0800 Subject: [PATCH 21/65] ci: use codecov to upload coverage report (#1583) --- .github/workflows/ci.yml | 10 ++++------ 1 file changed, 4 insertions(+), 6 deletions(-) diff --git a/.github/workflows/ci.yml b/.github/workflows/ci.yml index 89ba75cd..698acdf1 100644 --- a/.github/workflows/ci.yml +++ b/.github/workflows/ci.yml @@ -45,17 +45,15 @@ jobs: code_coverage: name: "Code coverage report" - if: github.event_name == 'pull_request' # Do not run when workflow is triggered by push to main branch runs-on: ubuntu-latest needs: unit_tests # Depends on the artifact uploaded by the "unit_tests" job steps: - - uses: fgrosse/go-coverage-report@v1.0.2 # Consider using a Git revision for maximum security - with: - coverage-artifact-name: "code-coverage" # can be omitted if you used this default value - coverage-file-name: "coverage.txt" # can be omitted if you used this default value + - uses: codecov/codecov-action@v4 + with: + use_oidc: true commit_lint: runs-on: ubuntu-latest steps: - uses: actions/checkout@v3 - - uses: wagoid/commitlint-github-action@v6 \ No newline at end of file + - uses: wagoid/commitlint-github-action@v6 From 34cb147a744e717404ebccd566cdf1b753ef78a1 Mon Sep 17 00:00:00 2001 From: igophper <34326532+igophper@users.noreply.github.com> Date: Sun, 30 Jun 2024 16:13:43 +0800 Subject: [PATCH 22/65] refactor: replace hardcoded string with ctxkey constant (#1579) MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Co-authored-by: 江杭辉 --- common/ctxkey/key.go | 1 + common/gin.go | 7 +++---- controller/relay.go | 2 +- 3 files changed, 5 insertions(+), 5 deletions(-) diff --git a/common/ctxkey/key.go b/common/ctxkey/key.go index 6c640870..90556b3a 100644 --- a/common/ctxkey/key.go +++ b/common/ctxkey/key.go @@ -19,4 +19,5 @@ const ( TokenName = "token_name" BaseURL = "base_url" AvailableModels = "available_models" + KeyRequestBody = "key_request_body" ) diff --git a/common/gin.go b/common/gin.go index b6ef96a6..549d3279 100644 --- a/common/gin.go +++ b/common/gin.go @@ -4,14 +4,13 @@ import ( "bytes" "encoding/json" "github.com/gin-gonic/gin" + "github.com/songquanpeng/one-api/common/ctxkey" "io" "strings" ) -const KeyRequestBody = "key_request_body" - func GetRequestBody(c *gin.Context) ([]byte, error) { - requestBody, _ := c.Get(KeyRequestBody) + requestBody, _ := c.Get(ctxkey.KeyRequestBody) if requestBody != nil { return requestBody.([]byte), nil } @@ -20,7 +19,7 @@ func GetRequestBody(c *gin.Context) ([]byte, error) { return nil, err } _ = c.Request.Body.Close() - c.Set(KeyRequestBody, requestBody) + c.Set(ctxkey.KeyRequestBody, requestBody) return requestBody.([]byte), nil } diff --git a/controller/relay.go b/controller/relay.go index 5d8ac690..932e023b 100644 --- a/controller/relay.go +++ b/controller/relay.go @@ -48,7 +48,7 @@ func Relay(c *gin.Context) { logger.Debugf(ctx, "request body: %s", string(requestBody)) } channelId := c.GetInt(ctxkey.ChannelId) - userId := c.GetInt("id") + userId := c.GetInt(ctxkey.Id) bizErr := relayHelper(c, relayMode) if bizErr == nil { monitor.Emit(channelId, true) From b70a07e814c5907e044f45dac32cb02ab1e51efc Mon Sep 17 00:00:00 2001 From: JustSong Date: Sun, 30 Jun 2024 16:19:49 +0800 Subject: [PATCH 23/65] fix: fix ci --- .github/workflows/ci.yml | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/.github/workflows/ci.yml b/.github/workflows/ci.yml index 698acdf1..30ac5f82 100644 --- a/.github/workflows/ci.yml +++ b/.github/workflows/ci.yml @@ -50,7 +50,7 @@ jobs: steps: - uses: codecov/codecov-action@v4 with: - use_oidc: true + token: ${{ secrets.CODECOV_TOKEN }} commit_lint: runs-on: ubuntu-latest From f25aaf7752a6f1719f445bb3d2d62863774e626b Mon Sep 17 00:00:00 2001 From: "dependabot[bot]" <49699333+dependabot[bot]@users.noreply.github.com> Date: Sun, 30 Jun 2024 16:21:48 +0800 Subject: [PATCH 24/65] chore(deps): bump golang.org/x/image from 0.16.0 to 0.18.0 (#1568) Bumps [golang.org/x/image](https://github.com/golang/image) from 0.16.0 to 0.18.0. - [Commits](https://github.com/golang/image/compare/v0.16.0...v0.18.0) --- updated-dependencies: - dependency-name: golang.org/x/image dependency-type: direct:production ... Signed-off-by: dependabot[bot] Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com> --- go.mod | 4 ++-- go.sum | 8 ++++---- 2 files changed, 6 insertions(+), 6 deletions(-) diff --git a/go.mod b/go.mod index 7a396314..2d0df03f 100644 --- a/go.mod +++ b/go.mod @@ -24,7 +24,7 @@ require ( github.com/smartystreets/goconvey v1.8.1 github.com/stretchr/testify v1.9.0 golang.org/x/crypto v0.23.0 - golang.org/x/image v0.16.0 + golang.org/x/image v0.18.0 gorm.io/driver/mysql v1.5.6 gorm.io/driver/postgres v1.5.7 gorm.io/driver/sqlite v1.5.5 @@ -80,7 +80,7 @@ require ( golang.org/x/net v0.25.0 // indirect golang.org/x/sync v0.7.0 // indirect golang.org/x/sys v0.20.0 // indirect - golang.org/x/text v0.15.0 // indirect + golang.org/x/text v0.16.0 // indirect google.golang.org/protobuf v1.34.1 // indirect gopkg.in/yaml.v3 v3.0.1 // indirect ) diff --git a/go.sum b/go.sum index 4c1aac95..ab04845c 100644 --- a/go.sum +++ b/go.sum @@ -154,8 +154,8 @@ golang.org/x/arch v0.8.0 h1:3wRIsP3pM4yUptoR96otTUOXI367OS0+c9eeRi9doIc= golang.org/x/arch v0.8.0/go.mod h1:FEVrYAQjsQXMVJ1nsMoVVXPZg6p2JE2mx8psSWTDQys= golang.org/x/crypto v0.23.0 h1:dIJU/v2J8Mdglj/8rJ6UUOM3Zc9zLZxVZwwxMooUSAI= golang.org/x/crypto v0.23.0/go.mod h1:CKFgDieR+mRhux2Lsu27y0fO304Db0wZe70UKqHu0v8= -golang.org/x/image v0.16.0 h1:9kloLAKhUufZhA12l5fwnx2NZW39/we1UhBesW433jw= -golang.org/x/image v0.16.0/go.mod h1:ugSZItdV4nOxyqp56HmXwH0Ry0nBCpjnZdpDaIHdoPs= +golang.org/x/image v0.18.0 h1:jGzIakQa/ZXI1I0Fxvaa9W7yP25TqT6cHIHn+6CqvSQ= +golang.org/x/image v0.18.0/go.mod h1:4yyo5vMFQjVjUcVk4jEQcU9MGy/rulF5WvUILseCM2E= golang.org/x/net v0.25.0 h1:d/OCCoBEUq33pjydKrGQhw7IlUPI2Oylr+8qLx49kac= golang.org/x/net v0.25.0/go.mod h1:JkAGAh7GEvH74S6FOH42FLoXpXbE/aqXSrIQjXgsiwM= golang.org/x/sync v0.7.0 h1:YsImfSBoP9QPYL0xyKJPq0gcaJdG3rInoqxTWbfQu9M= @@ -164,8 +164,8 @@ golang.org/x/sys v0.5.0/go.mod h1:oPkhp1MJrh7nUepCBck5+mAzfO9JrbApNNgaTdGDITg= golang.org/x/sys v0.6.0/go.mod h1:oPkhp1MJrh7nUepCBck5+mAzfO9JrbApNNgaTdGDITg= golang.org/x/sys v0.20.0 h1:Od9JTbYCk261bKm4M/mw7AklTlFYIa0bIp9BgSm1S8Y= golang.org/x/sys v0.20.0/go.mod h1:/VUhepiaJMQUp4+oa/7Zr1D23ma6VTLIYjOOTFZPUcA= -golang.org/x/text v0.15.0 h1:h1V/4gjBv8v9cjcR6+AR5+/cIYK5N/WAgiv4xlsEtAk= -golang.org/x/text v0.15.0/go.mod h1:18ZOQIKpY8NJVqYksKHtTdi31H5itFRjB5/qKTNYzSU= +golang.org/x/text v0.16.0 h1:a94ExnEXNtEwYLGJSIUxnWoxoRz/ZcCsV63ROupILh4= +golang.org/x/text v0.16.0/go.mod h1:GhwF1Be+LQoKShO3cGOHzqOgRrGaYc9AvblQOmPVHnI= golang.org/x/xerrors v0.0.0-20200804184101-5ec99f83aff1 h1:go1bK/D/BFZV2I8cIQd1NKEZ+0owSTG1fDTci4IqFcE= google.golang.org/protobuf v1.34.1 h1:9ddQBjfCyZPOHPUiPxpYESBLc+T8P3E+Vo4IbKZgFWg= google.golang.org/protobuf v1.34.1/go.mod h1:c6P6GXX6sHbq/GpV6MGZEdwhWPcYBgnhAHhKbcUYpos= From ae1cd29f943b31d4c12dffecd166b621b1ac2400 Mon Sep 17 00:00:00 2001 From: shaoyun Date: Sun, 30 Jun 2024 16:25:25 +0800 Subject: [PATCH 25/65] feat: added support for Claude Sonnet 3.5 (#1567) --- relay/adaptor/anthropic/constants.go | 1 + relay/adaptor/aws/main.go | 13 +++++++------ relay/billing/ratio/model.go | 13 +++++++------ web/air/src/pages/Channel/EditChannel.js | 2 +- 4 files changed, 16 insertions(+), 13 deletions(-) diff --git a/relay/adaptor/anthropic/constants.go b/relay/adaptor/anthropic/constants.go index cadcedc8..143d1efc 100644 --- a/relay/adaptor/anthropic/constants.go +++ b/relay/adaptor/anthropic/constants.go @@ -5,4 +5,5 @@ var ModelList = []string{ "claude-3-haiku-20240307", "claude-3-sonnet-20240229", "claude-3-opus-20240229", + "claude-3-5-sonnet-20240620", } diff --git a/relay/adaptor/aws/main.go b/relay/adaptor/aws/main.go index 0776f985..5d29597c 100644 --- a/relay/adaptor/aws/main.go +++ b/relay/adaptor/aws/main.go @@ -33,12 +33,13 @@ func wrapErr(err error) *relaymodel.ErrorWithStatusCode { // https://docs.aws.amazon.com/bedrock/latest/userguide/model-ids.html var awsModelIDMap = map[string]string{ - "claude-instant-1.2": "anthropic.claude-instant-v1", - "claude-2.0": "anthropic.claude-v2", - "claude-2.1": "anthropic.claude-v2:1", - "claude-3-sonnet-20240229": "anthropic.claude-3-sonnet-20240229-v1:0", - "claude-3-opus-20240229": "anthropic.claude-3-opus-20240229-v1:0", - "claude-3-haiku-20240307": "anthropic.claude-3-haiku-20240307-v1:0", + "claude-instant-1.2": "anthropic.claude-instant-v1", + "claude-2.0": "anthropic.claude-v2", + "claude-2.1": "anthropic.claude-v2:1", + "claude-3-sonnet-20240229": "anthropic.claude-3-sonnet-20240229-v1:0", + "claude-3-5-sonnet-20240620": "anthropic.claude-3-5-sonnet-20240620-v1:0", + "claude-3-opus-20240229": "anthropic.claude-3-opus-20240229-v1:0", + "claude-3-haiku-20240307": "anthropic.claude-3-haiku-20240307-v1:0", } func awsModelID(requestModel string) (string, error) { diff --git a/relay/billing/ratio/model.go b/relay/billing/ratio/model.go index 3b289499..b1a8a5b4 100644 --- a/relay/billing/ratio/model.go +++ b/relay/billing/ratio/model.go @@ -70,12 +70,13 @@ var ModelRatio = map[string]float64{ "dall-e-2": 0.02 * USD, // $0.016 - $0.020 / image "dall-e-3": 0.04 * USD, // $0.040 - $0.120 / image // https://www.anthropic.com/api#pricing - "claude-instant-1.2": 0.8 / 1000 * USD, - "claude-2.0": 8.0 / 1000 * USD, - "claude-2.1": 8.0 / 1000 * USD, - "claude-3-haiku-20240307": 0.25 / 1000 * USD, - "claude-3-sonnet-20240229": 3.0 / 1000 * USD, - "claude-3-opus-20240229": 15.0 / 1000 * USD, + "claude-instant-1.2": 0.8 / 1000 * USD, + "claude-2.0": 8.0 / 1000 * USD, + "claude-2.1": 8.0 / 1000 * USD, + "claude-3-haiku-20240307": 0.25 / 1000 * USD, + "claude-3-sonnet-20240229": 3.0 / 1000 * USD, + "claude-3-5-sonnet-20240620": 3.0 / 1000 * USD, + "claude-3-opus-20240229": 15.0 / 1000 * USD, // https://cloud.baidu.com/doc/WENXINWORKSHOP/s/hlrk4akp7 "ERNIE-4.0-8K": 0.120 * RMB, "ERNIE-3.5-8K": 0.012 * RMB, diff --git a/web/air/src/pages/Channel/EditChannel.js b/web/air/src/pages/Channel/EditChannel.js index efb2cee8..d63fa8fa 100644 --- a/web/air/src/pages/Channel/EditChannel.js +++ b/web/air/src/pages/Channel/EditChannel.js @@ -63,7 +63,7 @@ const EditChannel = (props) => { let localModels = []; switch (value) { case 14: - localModels = ["claude-instant-1.2", "claude-2", "claude-2.0", "claude-2.1", "claude-3-opus-20240229", "claude-3-sonnet-20240229", "claude-3-haiku-20240307"]; + localModels = ["claude-instant-1.2", "claude-2", "claude-2.0", "claude-2.1", "claude-3-opus-20240229", "claude-3-sonnet-20240229", "claude-3-haiku-20240307", "claude-3-5-sonnet-20240620"]; break; case 11: localModels = ['PaLM-2']; From b21b3b5b460502a40217b3973cf6ee5f44c916f9 Mon Sep 17 00:00:00 2001 From: zijiren <84728412+zijiren233@users.noreply.github.com> Date: Sun, 30 Jun 2024 18:36:33 +0800 Subject: [PATCH 26/65] refactor: abusing goroutines and channel (#1561) * refactor: abusing goroutines * fix: trim data prefix * refactor: move functions to render package * refactor: add back trim & flush --------- Co-authored-by: JustSong --- common/render/render.go | 29 +++++++ relay/adaptor/aiproxy/main.go | 97 +++++++++++------------ relay/adaptor/ali/main.go | 91 ++++++++++----------- relay/adaptor/anthropic/main.go | 102 ++++++++++++------------ relay/adaptor/baidu/main.go | 96 ++++++++++------------ relay/adaptor/cloudflare/main.go | 122 +++++++++++++--------------- relay/adaptor/cohere/main.go | 101 +++++++++++------------- relay/adaptor/coze/main.go | 113 ++++++++++++-------------- relay/adaptor/gemini/main.go | 91 +++++++++------------ relay/adaptor/ollama/main.go | 75 +++++++++--------- relay/adaptor/openai/main.go | 131 +++++++++++++------------------ relay/adaptor/palm/palm.go | 93 +++++++++++----------- relay/adaptor/tencent/main.go | 98 ++++++++++------------- relay/adaptor/zhipu/main.go | 103 +++++++++++------------- 14 files changed, 614 insertions(+), 728 deletions(-) create mode 100644 common/render/render.go diff --git a/common/render/render.go b/common/render/render.go new file mode 100644 index 00000000..646b3777 --- /dev/null +++ b/common/render/render.go @@ -0,0 +1,29 @@ +package render + +import ( + "encoding/json" + "fmt" + "github.com/gin-gonic/gin" + "github.com/songquanpeng/one-api/common" + "strings" +) + +func StringData(c *gin.Context, str string) { + str = strings.TrimPrefix(str, "data: ") + str = strings.TrimSuffix(str, "\r") + c.Render(-1, common.CustomEvent{Data: "data: " + str}) + c.Writer.Flush() +} + +func ObjectData(c *gin.Context, object interface{}) error { + jsonData, err := json.Marshal(object) + if err != nil { + return fmt.Errorf("error marshalling object: %w", err) + } + StringData(c, string(jsonData)) + return nil +} + +func Done(c *gin.Context) { + StringData(c, "[DONE]") +} diff --git a/relay/adaptor/aiproxy/main.go b/relay/adaptor/aiproxy/main.go index 01a568f6..d64b6809 100644 --- a/relay/adaptor/aiproxy/main.go +++ b/relay/adaptor/aiproxy/main.go @@ -4,6 +4,12 @@ import ( "bufio" "encoding/json" "fmt" + "github.com/songquanpeng/one-api/common/render" + "io" + "net/http" + "strconv" + "strings" + "github.com/gin-gonic/gin" "github.com/songquanpeng/one-api/common" "github.com/songquanpeng/one-api/common/helper" @@ -12,10 +18,6 @@ import ( "github.com/songquanpeng/one-api/relay/adaptor/openai" "github.com/songquanpeng/one-api/relay/constant" "github.com/songquanpeng/one-api/relay/model" - "io" - "net/http" - "strconv" - "strings" ) // https://docs.aiproxy.io/dev/library#使用已经定制好的知识库进行对话问答 @@ -89,6 +91,7 @@ func streamResponseAIProxyLibrary2OpenAI(response *LibraryStreamResponse) *opena func StreamHandler(c *gin.Context, resp *http.Response) (*model.ErrorWithStatusCode, *model.Usage) { var usage model.Usage + var documents []LibraryDocument scanner := bufio.NewScanner(resp.Body) scanner.Split(func(data []byte, atEOF bool) (advance int, token []byte, err error) { if atEOF && len(data) == 0 { @@ -102,60 +105,48 @@ func StreamHandler(c *gin.Context, resp *http.Response) (*model.ErrorWithStatusC } return 0, nil, nil }) - dataChan := make(chan string) - stopChan := make(chan bool) - go func() { - for scanner.Scan() { - data := scanner.Text() - if len(data) < 5 { // ignore blank line or wrong format - continue - } - if data[:5] != "data:" { - continue - } - data = data[5:] - dataChan <- data - } - stopChan <- true - }() + common.SetEventStreamHeaders(c) - var documents []LibraryDocument - c.Stream(func(w io.Writer) bool { - select { - case data := <-dataChan: - var AIProxyLibraryResponse LibraryStreamResponse - err := json.Unmarshal([]byte(data), &AIProxyLibraryResponse) - if err != nil { - logger.SysError("error unmarshalling stream response: " + err.Error()) - return true - } - if len(AIProxyLibraryResponse.Documents) != 0 { - documents = AIProxyLibraryResponse.Documents - } - response := streamResponseAIProxyLibrary2OpenAI(&AIProxyLibraryResponse) - jsonResponse, err := json.Marshal(response) - if err != nil { - logger.SysError("error marshalling stream response: " + err.Error()) - return true - } - c.Render(-1, common.CustomEvent{Data: "data: " + string(jsonResponse)}) - return true - case <-stopChan: - response := documentsAIProxyLibrary(documents) - jsonResponse, err := json.Marshal(response) - if err != nil { - logger.SysError("error marshalling stream response: " + err.Error()) - return true - } - c.Render(-1, common.CustomEvent{Data: "data: " + string(jsonResponse)}) - c.Render(-1, common.CustomEvent{Data: "data: [DONE]"}) - return false + + for scanner.Scan() { + data := scanner.Text() + if len(data) < 5 || data[:5] != "data:" { + continue } - }) - err := resp.Body.Close() + data = data[5:] + + var AIProxyLibraryResponse LibraryStreamResponse + err := json.Unmarshal([]byte(data), &AIProxyLibraryResponse) + if err != nil { + logger.SysError("error unmarshalling stream response: " + err.Error()) + continue + } + if len(AIProxyLibraryResponse.Documents) != 0 { + documents = AIProxyLibraryResponse.Documents + } + response := streamResponseAIProxyLibrary2OpenAI(&AIProxyLibraryResponse) + 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()) + } + + response := documentsAIProxyLibrary(documents) + err := render.ObjectData(c, response) + if err != nil { + logger.SysError(err.Error()) + } + render.Done(c) + + err = resp.Body.Close() if err != nil { return openai.ErrorWrapper(err, "close_response_body_failed", http.StatusInternalServerError), nil } + return nil, &usage } diff --git a/relay/adaptor/ali/main.go b/relay/adaptor/ali/main.go index 0462c26b..f9039dbe 100644 --- a/relay/adaptor/ali/main.go +++ b/relay/adaptor/ali/main.go @@ -3,15 +3,17 @@ package ali import ( "bufio" "encoding/json" + "github.com/songquanpeng/one-api/common/render" + "io" + "net/http" + "strings" + "github.com/gin-gonic/gin" "github.com/songquanpeng/one-api/common" "github.com/songquanpeng/one-api/common/helper" "github.com/songquanpeng/one-api/common/logger" "github.com/songquanpeng/one-api/relay/adaptor/openai" "github.com/songquanpeng/one-api/relay/model" - "io" - "net/http" - "strings" ) // https://help.aliyun.com/document_detail/613695.html?spm=a2c4g.2399480.0.0.1adb778fAdzP9w#341800c0f8w0r @@ -181,56 +183,43 @@ func StreamHandler(c *gin.Context, resp *http.Response) (*model.ErrorWithStatusC } return 0, nil, nil }) - dataChan := make(chan string) - stopChan := make(chan bool) - go func() { - for scanner.Scan() { - data := scanner.Text() - if len(data) < 5 { // ignore blank line or wrong format - continue - } - if data[:5] != "data:" { - continue - } - data = data[5:] - dataChan <- data - } - stopChan <- true - }() + common.SetEventStreamHeaders(c) - //lastResponseText := "" - c.Stream(func(w io.Writer) bool { - select { - case data := <-dataChan: - var aliResponse ChatResponse - err := json.Unmarshal([]byte(data), &aliResponse) - if err != nil { - logger.SysError("error unmarshalling stream response: " + err.Error()) - return true - } - if aliResponse.Usage.OutputTokens != 0 { - usage.PromptTokens = aliResponse.Usage.InputTokens - usage.CompletionTokens = aliResponse.Usage.OutputTokens - usage.TotalTokens = aliResponse.Usage.InputTokens + aliResponse.Usage.OutputTokens - } - response := streamResponseAli2OpenAI(&aliResponse) - if response == nil { - return true - } - //response.Choices[0].Delta.Content = strings.TrimPrefix(response.Choices[0].Delta.Content, lastResponseText) - //lastResponseText = aliResponse.Output.Text - jsonResponse, err := json.Marshal(response) - if err != nil { - logger.SysError("error marshalling stream response: " + err.Error()) - return true - } - c.Render(-1, common.CustomEvent{Data: "data: " + string(jsonResponse)}) - return true - case <-stopChan: - c.Render(-1, common.CustomEvent{Data: "data: [DONE]"}) - return false + + for scanner.Scan() { + data := scanner.Text() + if len(data) < 5 || data[:5] != "data:" { + continue } - }) + data = data[5:] + + var aliResponse ChatResponse + err := json.Unmarshal([]byte(data), &aliResponse) + if err != nil { + logger.SysError("error unmarshalling stream response: " + err.Error()) + continue + } + if aliResponse.Usage.OutputTokens != 0 { + usage.PromptTokens = aliResponse.Usage.InputTokens + usage.CompletionTokens = aliResponse.Usage.OutputTokens + usage.TotalTokens = aliResponse.Usage.InputTokens + aliResponse.Usage.OutputTokens + } + response := streamResponseAli2OpenAI(&aliResponse) + if response == nil { + continue + } + 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), nil diff --git a/relay/adaptor/anthropic/main.go b/relay/adaptor/anthropic/main.go index a8de185c..c817a9d1 100644 --- a/relay/adaptor/anthropic/main.go +++ b/relay/adaptor/anthropic/main.go @@ -4,6 +4,7 @@ import ( "bufio" "encoding/json" "fmt" + "github.com/songquanpeng/one-api/common/render" "io" "net/http" "strings" @@ -169,64 +170,59 @@ func StreamHandler(c *gin.Context, resp *http.Response) (*model.ErrorWithStatusC } return 0, nil, nil }) - dataChan := make(chan string) - stopChan := make(chan bool) - go func() { - for scanner.Scan() { - data := scanner.Text() - if len(data) < 6 { - continue - } - if !strings.HasPrefix(data, "data:") { - continue - } - data = strings.TrimPrefix(data, "data:") - dataChan <- data - } - stopChan <- true - }() + common.SetEventStreamHeaders(c) + var usage model.Usage var modelName string var id string - c.Stream(func(w io.Writer) bool { - select { - case data := <-dataChan: - // some implementations may add \r at the end of data - data = strings.TrimSpace(data) - var claudeResponse StreamResponse - err := json.Unmarshal([]byte(data), &claudeResponse) - if err != nil { - logger.SysError("error unmarshalling stream response: " + err.Error()) - return true - } - response, meta := StreamResponseClaude2OpenAI(&claudeResponse) - if meta != nil { - usage.PromptTokens += meta.Usage.InputTokens - usage.CompletionTokens += meta.Usage.OutputTokens - modelName = meta.Model - id = fmt.Sprintf("chatcmpl-%s", meta.Id) - return true - } - if response == nil { - return true - } - response.Id = id - response.Model = modelName - response.Created = createdTime - jsonStr, err := json.Marshal(response) - if err != nil { - logger.SysError("error marshalling stream response: " + err.Error()) - return true - } - c.Render(-1, common.CustomEvent{Data: "data: " + string(jsonStr)}) - return true - case <-stopChan: - c.Render(-1, common.CustomEvent{Data: "data: [DONE]"}) - return false + + for scanner.Scan() { + data := scanner.Text() + if len(data) < 6 || !strings.HasPrefix(data, "data:") { + continue } - }) - _ = resp.Body.Close() + data = strings.TrimPrefix(data, "data:") + data = strings.TrimSpace(data) + + var claudeResponse StreamResponse + err := json.Unmarshal([]byte(data), &claudeResponse) + if err != nil { + logger.SysError("error unmarshalling stream response: " + err.Error()) + continue + } + + response, meta := StreamResponseClaude2OpenAI(&claudeResponse) + if meta != nil { + usage.PromptTokens += meta.Usage.InputTokens + usage.CompletionTokens += meta.Usage.OutputTokens + modelName = meta.Model + id = fmt.Sprintf("chatcmpl-%s", meta.Id) + continue + } + if response == nil { + continue + } + + response.Id = id + response.Model = modelName + response.Created = createdTime + 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), nil + } return nil, &usage } diff --git a/relay/adaptor/baidu/main.go b/relay/adaptor/baidu/main.go index b816e0f4..ebe70c32 100644 --- a/relay/adaptor/baidu/main.go +++ b/relay/adaptor/baidu/main.go @@ -5,6 +5,13 @@ import ( "encoding/json" "errors" "fmt" + "github.com/songquanpeng/one-api/common/render" + "io" + "net/http" + "strings" + "sync" + "time" + "github.com/gin-gonic/gin" "github.com/songquanpeng/one-api/common" "github.com/songquanpeng/one-api/common/client" @@ -12,11 +19,6 @@ import ( "github.com/songquanpeng/one-api/relay/adaptor/openai" "github.com/songquanpeng/one-api/relay/constant" "github.com/songquanpeng/one-api/relay/model" - "io" - "net/http" - "strings" - "sync" - "time" ) // https://cloud.baidu.com/doc/WENXINWORKSHOP/s/flfmc9do2 @@ -137,59 +139,41 @@ func embeddingResponseBaidu2OpenAI(response *EmbeddingResponse) *openai.Embeddin func StreamHandler(c *gin.Context, resp *http.Response) (*model.ErrorWithStatusCode, *model.Usage) { var usage model.Usage scanner := bufio.NewScanner(resp.Body) - scanner.Split(func(data []byte, atEOF bool) (advance int, token []byte, err error) { - if atEOF && len(data) == 0 { - return 0, nil, nil - } - if i := strings.Index(string(data), "\n"); i >= 0 { - return i + 1, data[0:i], nil - } - if atEOF { - return len(data), data, nil - } - return 0, nil, nil - }) - dataChan := make(chan string) - stopChan := make(chan bool) - go func() { - for scanner.Scan() { - data := scanner.Text() - if len(data) < 6 { // ignore blank line or wrong format - continue - } - data = data[6:] - dataChan <- data - } - stopChan <- true - }() + scanner.Split(bufio.ScanLines) + common.SetEventStreamHeaders(c) - c.Stream(func(w io.Writer) bool { - select { - case data := <-dataChan: - var baiduResponse ChatStreamResponse - err := json.Unmarshal([]byte(data), &baiduResponse) - if err != nil { - logger.SysError("error unmarshalling stream response: " + err.Error()) - return true - } - if baiduResponse.Usage.TotalTokens != 0 { - usage.TotalTokens = baiduResponse.Usage.TotalTokens - usage.PromptTokens = baiduResponse.Usage.PromptTokens - usage.CompletionTokens = baiduResponse.Usage.TotalTokens - baiduResponse.Usage.PromptTokens - } - response := streamResponseBaidu2OpenAI(&baiduResponse) - jsonResponse, err := json.Marshal(response) - if err != nil { - logger.SysError("error marshalling stream response: " + err.Error()) - return true - } - c.Render(-1, common.CustomEvent{Data: "data: " + string(jsonResponse)}) - return true - case <-stopChan: - c.Render(-1, common.CustomEvent{Data: "data: [DONE]"}) - return false + + for scanner.Scan() { + data := scanner.Text() + if len(data) < 6 { + continue } - }) + data = data[6:] + + var baiduResponse ChatStreamResponse + err := json.Unmarshal([]byte(data), &baiduResponse) + if err != nil { + logger.SysError("error unmarshalling stream response: " + err.Error()) + continue + } + if baiduResponse.Usage.TotalTokens != 0 { + usage.TotalTokens = baiduResponse.Usage.TotalTokens + usage.PromptTokens = baiduResponse.Usage.PromptTokens + usage.CompletionTokens = baiduResponse.Usage.TotalTokens - baiduResponse.Usage.PromptTokens + } + response := streamResponseBaidu2OpenAI(&baiduResponse) + 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), nil diff --git a/relay/adaptor/cloudflare/main.go b/relay/adaptor/cloudflare/main.go index f6d496f7..c76520a2 100644 --- a/relay/adaptor/cloudflare/main.go +++ b/relay/adaptor/cloudflare/main.go @@ -2,8 +2,8 @@ package cloudflare import ( "bufio" - "bytes" "encoding/json" + "github.com/songquanpeng/one-api/common/render" "io" "net/http" "strings" @@ -17,21 +17,20 @@ import ( ) func ConvertRequest(textRequest model.GeneralOpenAIRequest) *Request { - var promptBuilder strings.Builder - for _, message := range textRequest.Messages { - promptBuilder.WriteString(message.StringContent()) - promptBuilder.WriteString("\n") // 添加换行符来分隔每个消息 - } + var promptBuilder strings.Builder + for _, message := range textRequest.Messages { + promptBuilder.WriteString(message.StringContent()) + promptBuilder.WriteString("\n") // 添加换行符来分隔每个消息 + } - return &Request{ - MaxTokens: textRequest.MaxTokens, - Prompt: promptBuilder.String(), - Stream: textRequest.Stream, - Temperature: textRequest.Temperature, - } + return &Request{ + MaxTokens: textRequest.MaxTokens, + Prompt: promptBuilder.String(), + Stream: textRequest.Stream, + Temperature: textRequest.Temperature, + } } - func ResponseCloudflare2OpenAI(cloudflareResponse *Response) *openai.TextResponse { choice := openai.TextResponseChoice{ Index: 0, @@ -63,67 +62,54 @@ func StreamResponseCloudflare2OpenAI(cloudflareResponse *StreamResponse) *openai func StreamHandler(c *gin.Context, resp *http.Response, promptTokens int, modelName string) (*model.ErrorWithStatusCode, *model.Usage) { scanner := bufio.NewScanner(resp.Body) - scanner.Split(func(data []byte, atEOF bool) (advance int, token []byte, err error) { - if atEOF && len(data) == 0 { - return 0, nil, nil - } - if i := bytes.IndexByte(data, '\n'); i >= 0 { - return i + 1, data[0:i], nil - } - if atEOF { - return len(data), data, nil - } - return 0, nil, nil - }) + scanner.Split(bufio.ScanLines) - dataChan := make(chan string) - stopChan := make(chan bool) - go func() { - for scanner.Scan() { - data := scanner.Text() - if len(data) < len("data: ") { - continue - } - data = strings.TrimPrefix(data, "data: ") - dataChan <- data - } - stopChan <- true - }() common.SetEventStreamHeaders(c) id := helper.GetResponseID(c) responseModel := c.GetString("original_model") var responseText string - c.Stream(func(w io.Writer) bool { - select { - case data := <-dataChan: - // some implementations may add \r at the end of data - data = strings.TrimSuffix(data, "\r") - var cloudflareResponse StreamResponse - err := json.Unmarshal([]byte(data), &cloudflareResponse) - if err != nil { - logger.SysError("error unmarshalling stream response: " + err.Error()) - return true - } - response := StreamResponseCloudflare2OpenAI(&cloudflareResponse) - if response == nil { - return true - } - responseText += cloudflareResponse.Response - response.Id = id - response.Model = responseModel - jsonStr, err := json.Marshal(response) - if err != nil { - logger.SysError("error marshalling stream response: " + err.Error()) - return true - } - c.Render(-1, common.CustomEvent{Data: "data: " + string(jsonStr)}) - return true - case <-stopChan: - c.Render(-1, common.CustomEvent{Data: "data: [DONE]"}) - return false + + for scanner.Scan() { + data := scanner.Text() + if len(data) < len("data: ") { + continue } - }) - _ = resp.Body.Close() + data = strings.TrimPrefix(data, "data: ") + data = strings.TrimSuffix(data, "\r") + + var cloudflareResponse StreamResponse + err := json.Unmarshal([]byte(data), &cloudflareResponse) + if err != nil { + logger.SysError("error unmarshalling stream response: " + err.Error()) + continue + } + + response := StreamResponseCloudflare2OpenAI(&cloudflareResponse) + if response == nil { + continue + } + + responseText += cloudflareResponse.Response + response.Id = id + response.Model = responseModel + + 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), nil + } + usage := openai.ResponseText2Usage(responseText, responseModel, promptTokens) return nil, usage } diff --git a/relay/adaptor/cohere/main.go b/relay/adaptor/cohere/main.go index 4bc3fa8d..45db437b 100644 --- a/relay/adaptor/cohere/main.go +++ b/relay/adaptor/cohere/main.go @@ -2,9 +2,9 @@ package cohere import ( "bufio" - "bytes" "encoding/json" "fmt" + "github.com/songquanpeng/one-api/common/render" "io" "net/http" "strings" @@ -134,66 +134,53 @@ func ResponseCohere2OpenAI(cohereResponse *Response) *openai.TextResponse { func StreamHandler(c *gin.Context, resp *http.Response) (*model.ErrorWithStatusCode, *model.Usage) { createdTime := helper.GetTimestamp() scanner := bufio.NewScanner(resp.Body) - scanner.Split(func(data []byte, atEOF bool) (advance int, token []byte, err error) { - if atEOF && len(data) == 0 { - return 0, nil, nil - } - if i := bytes.IndexByte(data, '\n'); i >= 0 { - return i + 1, data[0:i], nil - } - if atEOF { - return len(data), data, nil - } - return 0, nil, nil - }) + scanner.Split(bufio.ScanLines) - dataChan := make(chan string) - stopChan := make(chan bool) - go func() { - for scanner.Scan() { - data := scanner.Text() - dataChan <- data - } - stopChan <- true - }() common.SetEventStreamHeaders(c) var usage model.Usage - c.Stream(func(w io.Writer) bool { - select { - case data := <-dataChan: - // some implementations may add \r at the end of data - data = strings.TrimSuffix(data, "\r") - var cohereResponse StreamResponse - err := json.Unmarshal([]byte(data), &cohereResponse) - if err != nil { - logger.SysError("error unmarshalling stream response: " + err.Error()) - return true - } - response, meta := StreamResponseCohere2OpenAI(&cohereResponse) - if meta != nil { - usage.PromptTokens += meta.Meta.Tokens.InputTokens - usage.CompletionTokens += meta.Meta.Tokens.OutputTokens - return true - } - if response == nil { - return true - } - response.Id = fmt.Sprintf("chatcmpl-%d", createdTime) - response.Model = c.GetString("original_model") - response.Created = createdTime - jsonStr, err := json.Marshal(response) - if err != nil { - logger.SysError("error marshalling stream response: " + err.Error()) - return true - } - c.Render(-1, common.CustomEvent{Data: "data: " + string(jsonStr)}) - return true - case <-stopChan: - c.Render(-1, common.CustomEvent{Data: "data: [DONE]"}) - return false + + for scanner.Scan() { + data := scanner.Text() + data = strings.TrimSuffix(data, "\r") + + var cohereResponse StreamResponse + err := json.Unmarshal([]byte(data), &cohereResponse) + if err != nil { + logger.SysError("error unmarshalling stream response: " + err.Error()) + continue } - }) - _ = resp.Body.Close() + + response, meta := StreamResponseCohere2OpenAI(&cohereResponse) + if meta != nil { + usage.PromptTokens += meta.Meta.Tokens.InputTokens + usage.CompletionTokens += meta.Meta.Tokens.OutputTokens + continue + } + if response == nil { + continue + } + + response.Id = fmt.Sprintf("chatcmpl-%d", createdTime) + response.Model = c.GetString("original_model") + response.Created = createdTime + + 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), nil + } + return nil, &usage } diff --git a/relay/adaptor/coze/main.go b/relay/adaptor/coze/main.go index 721c5d13..d0402a76 100644 --- a/relay/adaptor/coze/main.go +++ b/relay/adaptor/coze/main.go @@ -4,6 +4,11 @@ import ( "bufio" "encoding/json" "fmt" + "github.com/songquanpeng/one-api/common/render" + "io" + "net/http" + "strings" + "github.com/gin-gonic/gin" "github.com/songquanpeng/one-api/common" "github.com/songquanpeng/one-api/common/conv" @@ -12,9 +17,6 @@ import ( "github.com/songquanpeng/one-api/relay/adaptor/coze/constant/messagetype" "github.com/songquanpeng/one-api/relay/adaptor/openai" "github.com/songquanpeng/one-api/relay/model" - "io" - "net/http" - "strings" ) // https://www.coze.com/open @@ -109,69 +111,54 @@ func StreamHandler(c *gin.Context, resp *http.Response) (*model.ErrorWithStatusC var responseText string createdTime := helper.GetTimestamp() scanner := bufio.NewScanner(resp.Body) - scanner.Split(func(data []byte, atEOF bool) (advance int, token []byte, err error) { - if atEOF && len(data) == 0 { - return 0, nil, nil - } - if i := strings.Index(string(data), "\n"); i >= 0 { - return i + 1, data[0:i], nil - } - if atEOF { - return len(data), data, nil - } - return 0, nil, nil - }) - dataChan := make(chan string) - stopChan := make(chan bool) - go func() { - for scanner.Scan() { - data := scanner.Text() - if len(data) < 5 { - continue - } - if !strings.HasPrefix(data, "data:") { - continue - } - data = strings.TrimPrefix(data, "data:") - dataChan <- data - } - stopChan <- true - }() + scanner.Split(bufio.ScanLines) + common.SetEventStreamHeaders(c) var modelName string - c.Stream(func(w io.Writer) bool { - select { - case data := <-dataChan: - // some implementations may add \r at the end of data - data = strings.TrimSuffix(data, "\r") - var cozeResponse StreamResponse - err := json.Unmarshal([]byte(data), &cozeResponse) - if err != nil { - logger.SysError("error unmarshalling stream response: " + err.Error()) - return true - } - response, _ := StreamResponseCoze2OpenAI(&cozeResponse) - if response == nil { - return true - } - for _, choice := range response.Choices { - responseText += conv.AsString(choice.Delta.Content) - } - response.Model = modelName - response.Created = createdTime - jsonStr, err := json.Marshal(response) - if err != nil { - logger.SysError("error marshalling stream response: " + err.Error()) - return true - } - c.Render(-1, common.CustomEvent{Data: "data: " + string(jsonStr)}) - return true - case <-stopChan: - c.Render(-1, common.CustomEvent{Data: "data: [DONE]"}) - return false + + for scanner.Scan() { + data := scanner.Text() + if len(data) < 5 || !strings.HasPrefix(data, "data:") { + continue } - }) - _ = resp.Body.Close() + data = strings.TrimPrefix(data, "data:") + data = strings.TrimSuffix(data, "\r") + + var cozeResponse StreamResponse + err := json.Unmarshal([]byte(data), &cozeResponse) + if err != nil { + logger.SysError("error unmarshalling stream response: " + err.Error()) + continue + } + + response, _ := StreamResponseCoze2OpenAI(&cozeResponse) + if response == nil { + continue + } + + for _, choice := range response.Choices { + responseText += conv.AsString(choice.Delta.Content) + } + response.Model = modelName + response.Created = createdTime + + 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), nil + } + return nil, &responseText } diff --git a/relay/adaptor/gemini/main.go b/relay/adaptor/gemini/main.go index 74a7d5d5..51fd6aa8 100644 --- a/relay/adaptor/gemini/main.go +++ b/relay/adaptor/gemini/main.go @@ -4,6 +4,7 @@ import ( "bufio" "encoding/json" "fmt" + "github.com/songquanpeng/one-api/common/render" "io" "net/http" "strings" @@ -275,64 +276,50 @@ func embeddingResponseGemini2OpenAI(response *EmbeddingResponse) *openai.Embeddi func StreamHandler(c *gin.Context, resp *http.Response) (*model.ErrorWithStatusCode, string) { responseText := "" scanner := bufio.NewScanner(resp.Body) - scanner.Split(func(data []byte, atEOF bool) (advance int, token []byte, err error) { - if atEOF && len(data) == 0 { - return 0, nil, nil - } - if i := strings.Index(string(data), "\n"); i >= 0 { - return i + 1, data[0:i], nil - } - if atEOF { - return len(data), data, nil - } - return 0, nil, nil - }) - dataChan := make(chan string) - stopChan := make(chan bool) - go func() { - for scanner.Scan() { - data := scanner.Text() - data = strings.TrimSpace(data) - if !strings.HasPrefix(data, "data: ") { - continue - } - data = strings.TrimPrefix(data, "data: ") - data = strings.TrimSuffix(data, "\"") - dataChan <- data - } - stopChan <- true - }() + scanner.Split(bufio.ScanLines) + common.SetEventStreamHeaders(c) - c.Stream(func(w io.Writer) bool { - select { - case data := <-dataChan: - var geminiResponse ChatResponse - err := json.Unmarshal([]byte(data), &geminiResponse) - if err != nil { - logger.SysError("error unmarshalling stream response: " + err.Error()) - return true - } - response := streamResponseGeminiChat2OpenAI(&geminiResponse) - if response == nil { - return true - } - responseText += response.Choices[0].Delta.StringContent() - jsonResponse, err := json.Marshal(response) - if err != nil { - logger.SysError("error marshalling stream response: " + err.Error()) - return true - } - c.Render(-1, common.CustomEvent{Data: "data: " + string(jsonResponse)}) - return true - case <-stopChan: - c.Render(-1, common.CustomEvent{Data: "data: [DONE]"}) - return false + + for scanner.Scan() { + data := scanner.Text() + data = strings.TrimSpace(data) + if !strings.HasPrefix(data, "data: ") { + continue } - }) + data = strings.TrimPrefix(data, "data: ") + data = strings.TrimSuffix(data, "\"") + + var geminiResponse ChatResponse + err := json.Unmarshal([]byte(data), &geminiResponse) + if err != nil { + logger.SysError("error unmarshalling stream response: " + err.Error()) + continue + } + + response := streamResponseGeminiChat2OpenAI(&geminiResponse) + if response == nil { + continue + } + + responseText += response.Choices[0].Delta.StringContent() + + 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 } diff --git a/relay/adaptor/ollama/main.go b/relay/adaptor/ollama/main.go index c5fe08e6..936a7e14 100644 --- a/relay/adaptor/ollama/main.go +++ b/relay/adaptor/ollama/main.go @@ -5,12 +5,14 @@ import ( "context" "encoding/json" "fmt" - "github.com/songquanpeng/one-api/common/helper" - "github.com/songquanpeng/one-api/common/random" + "github.com/songquanpeng/one-api/common/render" "io" "net/http" "strings" + "github.com/songquanpeng/one-api/common/helper" + "github.com/songquanpeng/one-api/common/random" + "github.com/gin-gonic/gin" "github.com/songquanpeng/one-api/common" "github.com/songquanpeng/one-api/common/image" @@ -105,54 +107,51 @@ func StreamHandler(c *gin.Context, resp *http.Response) (*model.ErrorWithStatusC return 0, nil, nil } if i := strings.Index(string(data), "}\n"); i >= 0 { - return i + 2, data[0:i], nil + return i + 2, data[0 : i+1], nil } if atEOF { return len(data), data, nil } return 0, nil, nil }) - dataChan := make(chan string) - stopChan := make(chan bool) - go func() { - for scanner.Scan() { - data := strings.TrimPrefix(scanner.Text(), "}") - dataChan <- data + "}" - } - stopChan <- true - }() + common.SetEventStreamHeaders(c) - c.Stream(func(w io.Writer) bool { - select { - case data := <-dataChan: - var ollamaResponse ChatResponse - err := json.Unmarshal([]byte(data), &ollamaResponse) - if err != nil { - logger.SysError("error unmarshalling stream response: " + err.Error()) - return true - } - if ollamaResponse.EvalCount != 0 { - usage.PromptTokens = ollamaResponse.PromptEvalCount - usage.CompletionTokens = ollamaResponse.EvalCount - usage.TotalTokens = ollamaResponse.PromptEvalCount + ollamaResponse.EvalCount - } - response := streamResponseOllama2OpenAI(&ollamaResponse) - jsonResponse, err := json.Marshal(response) - if err != nil { - logger.SysError("error marshalling stream response: " + err.Error()) - return true - } - c.Render(-1, common.CustomEvent{Data: "data: " + string(jsonResponse)}) - return true - case <-stopChan: - c.Render(-1, common.CustomEvent{Data: "data: [DONE]"}) - return false + + for scanner.Scan() { + data := strings.TrimPrefix(scanner.Text(), "}") + data = data + "}" + + var ollamaResponse ChatResponse + err := json.Unmarshal([]byte(data), &ollamaResponse) + if err != nil { + logger.SysError("error unmarshalling stream response: " + err.Error()) + continue } - }) + + if ollamaResponse.EvalCount != 0 { + usage.PromptTokens = ollamaResponse.PromptEvalCount + usage.CompletionTokens = ollamaResponse.EvalCount + usage.TotalTokens = ollamaResponse.PromptEvalCount + ollamaResponse.EvalCount + } + + response := streamResponseOllama2OpenAI(&ollamaResponse) + 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), nil } + return nil, &usage } diff --git a/relay/adaptor/openai/main.go b/relay/adaptor/openai/main.go index 07cb967f..1d534644 100644 --- a/relay/adaptor/openai/main.go +++ b/relay/adaptor/openai/main.go @@ -4,6 +4,7 @@ import ( "bufio" "bytes" "encoding/json" + "github.com/songquanpeng/one-api/common/render" "io" "net/http" "strings" @@ -25,88 +26,68 @@ const ( func StreamHandler(c *gin.Context, resp *http.Response, relayMode int) (*model.ErrorWithStatusCode, string, *model.Usage) { responseText := "" scanner := bufio.NewScanner(resp.Body) - scanner.Split(func(data []byte, atEOF bool) (advance int, token []byte, err error) { - if atEOF && len(data) == 0 { - return 0, nil, nil - } - if i := strings.Index(string(data), "\n"); i >= 0 { - return i + 1, data[0:i], nil - } - if atEOF { - return len(data), data, nil - } - return 0, nil, nil - }) - dataChan := make(chan string) - stopChan := make(chan bool) + scanner.Split(bufio.ScanLines) var usage *model.Usage - go func() { - for scanner.Scan() { - data := scanner.Text() - if len(data) < dataPrefixLength { // ignore blank line or wrong format - continue - } - if data[:dataPrefixLength] != dataPrefix && data[:dataPrefixLength] != done { - continue - } - if strings.HasPrefix(data[dataPrefixLength:], done) { - dataChan <- data - continue - } - switch relayMode { - case relaymode.ChatCompletions: - var streamResponse ChatCompletionsStreamResponse - err := json.Unmarshal([]byte(data[dataPrefixLength:]), &streamResponse) - if err != nil { - logger.SysError("error unmarshalling stream response: " + err.Error()) - dataChan <- data // if error happened, pass the data to client - continue // just ignore the error - } - if len(streamResponse.Choices) == 0 { - // but for empty choice, we should not pass it to client, this is for azure - continue // just ignore empty choice - } - dataChan <- data - for _, choice := range streamResponse.Choices { - responseText += conv.AsString(choice.Delta.Content) - } - if streamResponse.Usage != nil { - usage = streamResponse.Usage - } - case relaymode.Completions: - dataChan <- data - var streamResponse CompletionsStreamResponse - err := json.Unmarshal([]byte(data[dataPrefixLength:]), &streamResponse) - if err != nil { - logger.SysError("error unmarshalling stream response: " + err.Error()) - continue - } - for _, choice := range streamResponse.Choices { - responseText += choice.Text - } - } - } - stopChan <- true - }() + common.SetEventStreamHeaders(c) - c.Stream(func(w io.Writer) bool { - select { - case data := <-dataChan: - if strings.HasPrefix(data, "data: [DONE]") { - data = data[:12] - } - // some implementations may add \r at the end of data - data = strings.TrimSuffix(data, "\r") - c.Render(-1, common.CustomEvent{Data: data}) - return true - case <-stopChan: - return false + + for scanner.Scan() { + data := scanner.Text() + if len(data) < dataPrefixLength { // ignore blank line or wrong format + continue } - }) + if data[:dataPrefixLength] != dataPrefix && data[:dataPrefixLength] != done { + continue + } + if strings.HasPrefix(data[dataPrefixLength:], done) { + render.StringData(c, data) + continue + } + switch relayMode { + case relaymode.ChatCompletions: + var streamResponse ChatCompletionsStreamResponse + err := json.Unmarshal([]byte(data[dataPrefixLength:]), &streamResponse) + if err != nil { + logger.SysError("error unmarshalling stream response: " + err.Error()) + render.StringData(c, data) // if error happened, pass the data to client + continue // just ignore the error + } + if len(streamResponse.Choices) == 0 { + // but for empty choice, we should not pass it to client, this is for azure + continue // just ignore empty choice + } + render.StringData(c, data) + for _, choice := range streamResponse.Choices { + responseText += conv.AsString(choice.Delta.Content) + } + if streamResponse.Usage != nil { + usage = streamResponse.Usage + } + case relaymode.Completions: + render.StringData(c, data) + var streamResponse CompletionsStreamResponse + err := json.Unmarshal([]byte(data[dataPrefixLength:]), &streamResponse) + if err != nil { + logger.SysError("error unmarshalling stream response: " + err.Error()) + continue + } + for _, choice := range streamResponse.Choices { + responseText += choice.Text + } + } + } + + if err := scanner.Err(); err != nil { + logger.SysError("error reading stream: " + err.Error()) + } + + render.Done(c) + err := resp.Body.Close() if err != nil { return ErrorWrapper(err, "close_response_body_failed", http.StatusInternalServerError), "", nil } + return nil, responseText, usage } diff --git a/relay/adaptor/palm/palm.go b/relay/adaptor/palm/palm.go index 1e60e7cd..d31784ec 100644 --- a/relay/adaptor/palm/palm.go +++ b/relay/adaptor/palm/palm.go @@ -3,6 +3,10 @@ package palm import ( "encoding/json" "fmt" + "github.com/songquanpeng/one-api/common/render" + "io" + "net/http" + "github.com/gin-gonic/gin" "github.com/songquanpeng/one-api/common" "github.com/songquanpeng/one-api/common/helper" @@ -11,8 +15,6 @@ import ( "github.com/songquanpeng/one-api/relay/adaptor/openai" "github.com/songquanpeng/one-api/relay/constant" "github.com/songquanpeng/one-api/relay/model" - "io" - "net/http" ) // https://developers.generativeai.google/api/rest/generativelanguage/models/generateMessage#request-body @@ -77,58 +79,51 @@ func StreamHandler(c *gin.Context, resp *http.Response) (*model.ErrorWithStatusC responseText := "" responseId := fmt.Sprintf("chatcmpl-%s", random.GetUUID()) createdTime := helper.GetTimestamp() - dataChan := make(chan string) - stopChan := make(chan bool) - go func() { - responseBody, err := io.ReadAll(resp.Body) - if err != nil { - logger.SysError("error reading stream response: " + err.Error()) - stopChan <- true - return - } - err = resp.Body.Close() - if err != nil { - logger.SysError("error closing stream response: " + err.Error()) - stopChan <- true - return - } - var palmResponse ChatResponse - err = json.Unmarshal(responseBody, &palmResponse) - if err != nil { - logger.SysError("error unmarshalling stream response: " + err.Error()) - stopChan <- true - return - } - fullTextResponse := streamResponsePaLM2OpenAI(&palmResponse) - fullTextResponse.Id = responseId - fullTextResponse.Created = createdTime - if len(palmResponse.Candidates) > 0 { - responseText = palmResponse.Candidates[0].Content - } - jsonResponse, err := json.Marshal(fullTextResponse) - if err != nil { - logger.SysError("error marshalling stream response: " + err.Error()) - stopChan <- true - return - } - dataChan <- string(jsonResponse) - stopChan <- true - }() + common.SetEventStreamHeaders(c) - c.Stream(func(w io.Writer) bool { - select { - case data := <-dataChan: - c.Render(-1, common.CustomEvent{Data: "data: " + data}) - return true - case <-stopChan: - c.Render(-1, common.CustomEvent{Data: "data: [DONE]"}) - return false + + responseBody, err := io.ReadAll(resp.Body) + if err != nil { + logger.SysError("error reading stream response: " + err.Error()) + err := resp.Body.Close() + if err != nil { + return openai.ErrorWrapper(err, "close_response_body_failed", http.StatusInternalServerError), "" } - }) - err := resp.Body.Close() + return openai.ErrorWrapper(err, "read_response_body_failed", http.StatusInternalServerError), "" + } + + err = resp.Body.Close() if err != nil { return openai.ErrorWrapper(err, "close_response_body_failed", http.StatusInternalServerError), "" } + + var palmResponse ChatResponse + err = json.Unmarshal(responseBody, &palmResponse) + if err != nil { + logger.SysError("error unmarshalling stream response: " + err.Error()) + return openai.ErrorWrapper(err, "unmarshal_response_body_failed", http.StatusInternalServerError), "" + } + + fullTextResponse := streamResponsePaLM2OpenAI(&palmResponse) + fullTextResponse.Id = responseId + fullTextResponse.Created = createdTime + if len(palmResponse.Candidates) > 0 { + responseText = palmResponse.Candidates[0].Content + } + + jsonResponse, err := json.Marshal(fullTextResponse) + if err != nil { + logger.SysError("error marshalling stream response: " + err.Error()) + return openai.ErrorWrapper(err, "marshal_response_body_failed", http.StatusInternalServerError), "" + } + + err = render.ObjectData(c, string(jsonResponse)) + if err != nil { + logger.SysError(err.Error()) + } + + render.Done(c) + return nil, responseText } diff --git a/relay/adaptor/tencent/main.go b/relay/adaptor/tencent/main.go index 0a57dcf7..365e33ae 100644 --- a/relay/adaptor/tencent/main.go +++ b/relay/adaptor/tencent/main.go @@ -8,6 +8,13 @@ import ( "encoding/json" "errors" "fmt" + "github.com/songquanpeng/one-api/common/render" + "io" + "net/http" + "strconv" + "strings" + "time" + "github.com/gin-gonic/gin" "github.com/songquanpeng/one-api/common" "github.com/songquanpeng/one-api/common/conv" @@ -17,11 +24,6 @@ import ( "github.com/songquanpeng/one-api/relay/adaptor/openai" "github.com/songquanpeng/one-api/relay/constant" "github.com/songquanpeng/one-api/relay/model" - "io" - "net/http" - "strconv" - "strings" - "time" ) func ConvertRequest(request model.GeneralOpenAIRequest) *ChatRequest { @@ -87,64 +89,46 @@ func streamResponseTencent2OpenAI(TencentResponse *ChatResponse) *openai.ChatCom func StreamHandler(c *gin.Context, resp *http.Response) (*model.ErrorWithStatusCode, string) { var responseText string scanner := bufio.NewScanner(resp.Body) - scanner.Split(func(data []byte, atEOF bool) (advance int, token []byte, err error) { - if atEOF && len(data) == 0 { - return 0, nil, nil - } - if i := strings.Index(string(data), "\n"); i >= 0 { - return i + 1, data[0:i], nil - } - if atEOF { - return len(data), data, nil - } - return 0, nil, nil - }) - dataChan := make(chan string) - stopChan := make(chan bool) - go func() { - for scanner.Scan() { - data := scanner.Text() - if len(data) < 5 { // ignore blank line or wrong format - continue - } - if data[:5] != "data:" { - continue - } - data = data[5:] - dataChan <- data - } - stopChan <- true - }() + scanner.Split(bufio.ScanLines) + common.SetEventStreamHeaders(c) - c.Stream(func(w io.Writer) bool { - select { - case data := <-dataChan: - var TencentResponse ChatResponse - err := json.Unmarshal([]byte(data), &TencentResponse) - if err != nil { - logger.SysError("error unmarshalling stream response: " + err.Error()) - return true - } - response := streamResponseTencent2OpenAI(&TencentResponse) - if len(response.Choices) != 0 { - responseText += conv.AsString(response.Choices[0].Delta.Content) - } - jsonResponse, err := json.Marshal(response) - if err != nil { - logger.SysError("error marshalling stream response: " + err.Error()) - return true - } - c.Render(-1, common.CustomEvent{Data: "data: " + string(jsonResponse)}) - return true - case <-stopChan: - c.Render(-1, common.CustomEvent{Data: "data: [DONE]"}) - return false + + 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 } diff --git a/relay/adaptor/zhipu/main.go b/relay/adaptor/zhipu/main.go index 74a1a05e..ab3a5678 100644 --- a/relay/adaptor/zhipu/main.go +++ b/relay/adaptor/zhipu/main.go @@ -3,6 +3,13 @@ package zhipu import ( "bufio" "encoding/json" + "github.com/songquanpeng/one-api/common/render" + "io" + "net/http" + "strings" + "sync" + "time" + "github.com/gin-gonic/gin" "github.com/golang-jwt/jwt" "github.com/songquanpeng/one-api/common" @@ -11,11 +18,6 @@ import ( "github.com/songquanpeng/one-api/relay/adaptor/openai" "github.com/songquanpeng/one-api/relay/constant" "github.com/songquanpeng/one-api/relay/model" - "io" - "net/http" - "strings" - "sync" - "time" ) // https://open.bigmodel.cn/doc/api#chatglm_std @@ -155,66 +157,55 @@ func StreamHandler(c *gin.Context, resp *http.Response) (*model.ErrorWithStatusC } return 0, nil, nil }) - dataChan := make(chan string) - metaChan := make(chan string) - stopChan := make(chan bool) - go func() { - for scanner.Scan() { - data := scanner.Text() - lines := strings.Split(data, "\n") - for i, line := range lines { - if len(line) < 5 { + + common.SetEventStreamHeaders(c) + + for scanner.Scan() { + data := scanner.Text() + lines := strings.Split(data, "\n") + for i, line := range lines { + if len(line) < 5 { + continue + } + if strings.HasPrefix(line, "data:") { + dataSegment := line[5:] + if i != len(lines)-1 { + dataSegment += "\n" + } + response := streamResponseZhipu2OpenAI(dataSegment) + err := render.ObjectData(c, response) + if err != nil { + logger.SysError("error marshalling stream response: " + err.Error()) + } + } else if strings.HasPrefix(line, "meta:") { + metaSegment := line[5:] + var zhipuResponse StreamMetaResponse + err := json.Unmarshal([]byte(metaSegment), &zhipuResponse) + if err != nil { + logger.SysError("error unmarshalling stream response: " + err.Error()) continue } - if line[:5] == "data:" { - dataChan <- line[5:] - if i != len(lines)-1 { - dataChan <- "\n" - } - } else if line[:5] == "meta:" { - metaChan <- line[5:] + response, zhipuUsage := streamMetaResponseZhipu2OpenAI(&zhipuResponse) + err = render.ObjectData(c, response) + if err != nil { + logger.SysError("error marshalling stream response: " + err.Error()) } + usage = zhipuUsage } } - stopChan <- true - }() - common.SetEventStreamHeaders(c) - c.Stream(func(w io.Writer) bool { - select { - case data := <-dataChan: - response := streamResponseZhipu2OpenAI(data) - jsonResponse, err := json.Marshal(response) - if err != nil { - logger.SysError("error marshalling stream response: " + err.Error()) - return true - } - c.Render(-1, common.CustomEvent{Data: "data: " + string(jsonResponse)}) - return true - case data := <-metaChan: - var zhipuResponse StreamMetaResponse - err := json.Unmarshal([]byte(data), &zhipuResponse) - if err != nil { - logger.SysError("error unmarshalling stream response: " + err.Error()) - return true - } - response, zhipuUsage := streamMetaResponseZhipu2OpenAI(&zhipuResponse) - jsonResponse, err := json.Marshal(response) - if err != nil { - logger.SysError("error marshalling stream response: " + err.Error()) - return true - } - usage = zhipuUsage - c.Render(-1, common.CustomEvent{Data: "data: " + string(jsonResponse)}) - return true - case <-stopChan: - c.Render(-1, common.CustomEvent{Data: "data: [DONE]"}) - return false - } - }) + } + + 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), nil } + return nil, usage } From d0369b114f6b9a34a926979b309ac1fd052db698 Mon Sep 17 00:00:00 2001 From: lihangfu <280001404@qq.com> Date: Sun, 30 Jun 2024 19:37:07 +0800 Subject: [PATCH 27/65] feat: support spark4.0 ultra (#1569) MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit * feat: 支持v3最新协议的腾讯混元（#1452） * feat: 支持Spark4.0 Ultra --------- Co-authored-by: lihangfu --- relay/adaptor/xunfei/constants.go | 1 + relay/adaptor/xunfei/main.go | 2 ++ relay/billing/ratio/model.go | 1 + web/air/src/pages/Channel/EditChannel.js | 2 +- web/berry/src/views/Channel/type/Config.js | 2 +- 5 files changed, 6 insertions(+), 2 deletions(-) diff --git a/relay/adaptor/xunfei/constants.go b/relay/adaptor/xunfei/constants.go index 31dcec71..12a56210 100644 --- a/relay/adaptor/xunfei/constants.go +++ b/relay/adaptor/xunfei/constants.go @@ -6,4 +6,5 @@ var ModelList = []string{ "SparkDesk-v2.1", "SparkDesk-v3.1", "SparkDesk-v3.5", + "SparkDesk-v4.0", } diff --git a/relay/adaptor/xunfei/main.go b/relay/adaptor/xunfei/main.go index 39b76e27..7cf413a4 100644 --- a/relay/adaptor/xunfei/main.go +++ b/relay/adaptor/xunfei/main.go @@ -290,6 +290,8 @@ func apiVersion2domain(apiVersion string) string { return "generalv3" case "v3.5": return "generalv3.5" + case "v4.0": + return "4.0Ultra" } return "general" + apiVersion } diff --git a/relay/billing/ratio/model.go b/relay/billing/ratio/model.go index b1a8a5b4..56d31e13 100644 --- a/relay/billing/ratio/model.go +++ b/relay/billing/ratio/model.go @@ -125,6 +125,7 @@ var ModelRatio = map[string]float64{ "SparkDesk-v2.1": 1.2858, // ￥0.018 / 1k tokens "SparkDesk-v3.1": 1.2858, // ￥0.018 / 1k tokens "SparkDesk-v3.5": 1.2858, // ￥0.018 / 1k tokens + "SparkDesk-v4.0": 1.2858, // ￥0.018 / 1k tokens "360GPT_S2_V9": 0.8572, // ¥0.012 / 1k tokens "embedding-bert-512-v1": 0.0715, // ¥0.001 / 1k tokens "embedding_s1_v1": 0.0715, // ¥0.001 / 1k tokens diff --git a/web/air/src/pages/Channel/EditChannel.js b/web/air/src/pages/Channel/EditChannel.js index d63fa8fa..73fd2da2 100644 --- a/web/air/src/pages/Channel/EditChannel.js +++ b/web/air/src/pages/Channel/EditChannel.js @@ -78,7 +78,7 @@ const EditChannel = (props) => { localModels = ['chatglm_pro', 'chatglm_std', 'chatglm_lite']; break; case 18: - localModels = ['SparkDesk', 'SparkDesk-v1.1', 'SparkDesk-v2.1', 'SparkDesk-v3.1', 'SparkDesk-v3.5']; + localModels = ['SparkDesk', 'SparkDesk-v1.1', 'SparkDesk-v2.1', 'SparkDesk-v3.1', 'SparkDesk-v3.5', 'SparkDesk-v4.0']; break; case 19: localModels = ['360GPT_S2_V9', 'embedding-bert-512-v1', 'embedding_s1_v1', 'semantic_similarity_s1_v1']; diff --git a/web/berry/src/views/Channel/type/Config.js b/web/berry/src/views/Channel/type/Config.js index 88e1ea92..51b7c6c4 100644 --- a/web/berry/src/views/Channel/type/Config.js +++ b/web/berry/src/views/Channel/type/Config.js @@ -91,7 +91,7 @@ const typeConfig = { other: '版本号' }, input: { - models: ['SparkDesk', 'SparkDesk-v1.1', 'SparkDesk-v2.1', 'SparkDesk-v3.1', 'SparkDesk-v3.5'] + models: ['SparkDesk', 'SparkDesk-v1.1', 'SparkDesk-v2.1', 'SparkDesk-v3.1', 'SparkDesk-v3.5', 'SparkDesk-v4.0'] }, prompt: { key: '按照如下格式输入：APPID|APISecret|APIKey', From c135d74f136813ff26731c2e78bcf2fc3dc3daed Mon Sep 17 00:00:00 2001 From: Shi Jilin <40982122+shijilin0116@users.noreply.github.com> Date: Sun, 30 Jun 2024 19:38:02 +0800 Subject: [PATCH 28/65] feat: support Spark4.0 Ultra (#1575) MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit * fix: fix SparkDesk Function Call (修复 Spark Pro/Max函数调用只会返回普通对话回答而不是Function Call回答的问题 * feat: support Spark4.0 Ultra --- relay/adaptor/xunfei/main.go | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/relay/adaptor/xunfei/main.go b/relay/adaptor/xunfei/main.go index 7cf413a4..ef6120e5 100644 --- a/relay/adaptor/xunfei/main.go +++ b/relay/adaptor/xunfei/main.go @@ -44,7 +44,7 @@ func requestOpenAI2Xunfei(request model.GeneralOpenAIRequest, xunfeiAppId string xunfeiRequest.Parameter.Chat.MaxTokens = request.MaxTokens xunfeiRequest.Payload.Message.Text = messages - if strings.HasPrefix(domain, "generalv3") { + if strings.HasPrefix(domain, "generalv3") || domain == "4.0Ultra" { functions := make([]model.Function, len(request.Tools)) for i, tool := range request.Tools { functions[i] = tool.Function From fecaece71b700b43ba11c161a3f8a971af204971 Mon Sep 17 00:00:00 2001 From: igophper <34326532+igophper@users.noreply.github.com> Date: Sun, 30 Jun 2024 19:52:33 +0800 Subject: [PATCH 29/65] fix: fix size not support during image generation (#1564) Fixes #1224, #1068 --- relay/controller/helper.go | 72 -------------------------------- relay/controller/image.go | 84 +++++++++++++++++++++++++++++++++++--- 2 files changed, 78 insertions(+), 78 deletions(-) diff --git a/relay/controller/helper.go b/relay/controller/helper.go index dccff486..c47cb558 100644 --- a/relay/controller/helper.go +++ b/relay/controller/helper.go @@ -40,78 +40,6 @@ func getAndValidateTextRequest(c *gin.Context, relayMode int) (*relaymodel.Gener return textRequest, nil } -func getImageRequest(c *gin.Context, relayMode int) (*relaymodel.ImageRequest, error) { - imageRequest := &relaymodel.ImageRequest{} - err := common.UnmarshalBodyReusable(c, imageRequest) - if err != nil { - return nil, err - } - if imageRequest.N == 0 { - imageRequest.N = 1 - } - if imageRequest.Size == "" { - imageRequest.Size = "1024x1024" - } - if imageRequest.Model == "" { - imageRequest.Model = "dall-e-2" - } - return imageRequest, nil -} - -func isValidImageSize(model string, size string) bool { - if model == "cogview-3" { - return true - } - _, ok := billingratio.ImageSizeRatios[model][size] - return ok -} - -func getImageSizeRatio(model string, size string) float64 { - ratio, ok := billingratio.ImageSizeRatios[model][size] - if !ok { - return 1 - } - return ratio -} - -func validateImageRequest(imageRequest *relaymodel.ImageRequest, meta *meta.Meta) *relaymodel.ErrorWithStatusCode { - // model validation - hasValidSize := isValidImageSize(imageRequest.Model, imageRequest.Size) - if !hasValidSize { - return openai.ErrorWrapper(errors.New("size not supported for this image model"), "size_not_supported", http.StatusBadRequest) - } - // check prompt length - if imageRequest.Prompt == "" { - return openai.ErrorWrapper(errors.New("prompt is required"), "prompt_missing", http.StatusBadRequest) - } - if len(imageRequest.Prompt) > billingratio.ImagePromptLengthLimitations[imageRequest.Model] { - return openai.ErrorWrapper(errors.New("prompt is too long"), "prompt_too_long", http.StatusBadRequest) - } - // Number of generated images validation - if !isWithinRange(imageRequest.Model, imageRequest.N) { - // channel not azure - if meta.ChannelType != channeltype.Azure { - return openai.ErrorWrapper(errors.New("invalid value of n"), "n_not_within_range", http.StatusBadRequest) - } - } - return nil -} - -func getImageCostRatio(imageRequest *relaymodel.ImageRequest) (float64, error) { - if imageRequest == nil { - return 0, errors.New("imageRequest is nil") - } - imageCostRatio := getImageSizeRatio(imageRequest.Model, imageRequest.Size) - if imageRequest.Quality == "hd" && imageRequest.Model == "dall-e-3" { - if imageRequest.Size == "1024x1024" { - imageCostRatio *= 2 - } else { - imageCostRatio *= 1.5 - } - } - return imageCostRatio, nil -} - func getPromptTokens(textRequest *relaymodel.GeneralOpenAIRequest, relayMode int) int { switch relayMode { case relaymode.ChatCompletions: diff --git a/relay/controller/image.go b/relay/controller/image.go index 691c7c0e..e6245226 100644 --- a/relay/controller/image.go +++ b/relay/controller/image.go @@ -7,6 +7,7 @@ import ( "errors" "fmt" "github.com/gin-gonic/gin" + "github.com/songquanpeng/one-api/common" "github.com/songquanpeng/one-api/common/ctxkey" "github.com/songquanpeng/one-api/common/logger" "github.com/songquanpeng/one-api/model" @@ -20,13 +21,84 @@ import ( "net/http" ) -func isWithinRange(element string, value int) bool { - if _, ok := billingratio.ImageGenerationAmounts[element]; !ok { - return false +func getImageRequest(c *gin.Context, relayMode int) (*relaymodel.ImageRequest, error) { + imageRequest := &relaymodel.ImageRequest{} + err := common.UnmarshalBodyReusable(c, imageRequest) + if err != nil { + return nil, err } - min := billingratio.ImageGenerationAmounts[element][0] - max := billingratio.ImageGenerationAmounts[element][1] - return value >= min && value <= max + if imageRequest.N == 0 { + imageRequest.N = 1 + } + if imageRequest.Size == "" { + imageRequest.Size = "1024x1024" + } + if imageRequest.Model == "" { + imageRequest.Model = "dall-e-2" + } + return imageRequest, nil +} + +func isValidImageSize(model string, size string) bool { + if model == "cogview-3" || billingratio.ImageSizeRatios[model] == nil { + return true + } + _, ok := billingratio.ImageSizeRatios[model][size] + return ok +} + +func isValidImagePromptLength(model string, promptLength int) bool { + maxPromptLength, ok := billingratio.ImagePromptLengthLimitations[model] + return !ok || promptLength <= maxPromptLength +} + +func isWithinRange(element string, value int) bool { + amounts, ok := billingratio.ImageGenerationAmounts[element] + return !ok || (value >= amounts[0] && value <= amounts[1]) +} + +func getImageSizeRatio(model string, size string) float64 { + if ratio, ok := billingratio.ImageSizeRatios[model][size]; ok { + return ratio + } + return 1 +} + +func validateImageRequest(imageRequest *relaymodel.ImageRequest, meta *meta.Meta) *relaymodel.ErrorWithStatusCode { + // check prompt length + if imageRequest.Prompt == "" { + return openai.ErrorWrapper(errors.New("prompt is required"), "prompt_missing", http.StatusBadRequest) + } + + // model validation + if !isValidImageSize(imageRequest.Model, imageRequest.Size) { + return openai.ErrorWrapper(errors.New("size not supported for this image model"), "size_not_supported", http.StatusBadRequest) + } + + if !isValidImagePromptLength(imageRequest.Model, len(imageRequest.Prompt)) { + return openai.ErrorWrapper(errors.New("prompt is too long"), "prompt_too_long", http.StatusBadRequest) + } + + // Number of generated images validation + if !isWithinRange(imageRequest.Model, imageRequest.N) { + return openai.ErrorWrapper(errors.New("invalid value of n"), "n_not_within_range", http.StatusBadRequest) + } + return nil +} + +func getImageCostRatio(imageRequest *relaymodel.ImageRequest) (float64, error) { + if imageRequest == nil { + return 0, errors.New("imageRequest is nil") + } + imageCostRatio := getImageSizeRatio(imageRequest.Model, imageRequest.Size) + if imageRequest.Quality == "hd" && imageRequest.Model == "dall-e-3" { + if imageRequest.Size == "1024x1024" { + imageCostRatio *= 2 + } else { + imageCostRatio *= 1.5 + } + } + return imageCostRatio, nil } func RelayImageHelper(c *gin.Context, relayMode int) *relaymodel.ErrorWithStatusCode { From d936817de9866b3e8b6bf1a0f741a2a4eb6c3bd4 Mon Sep 17 00:00:00 2001 From: Darkside Date: Sun, 30 Jun 2024 19:57:30 +0800 Subject: [PATCH 30/65] docs: add related projects (#1562) MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Co-authored-by: 成达 --- README.en.md | 10 ++++++---- README.md | 12 +++++++----- 2 files changed, 13 insertions(+), 9 deletions(-) diff --git a/README.en.md b/README.en.md index bce47353..db96a858 100644 --- a/README.en.md +++ b/README.en.md @@ -101,7 +101,7 @@ Nginx reference configuration: ``` server{ server_name openai.justsong.cn; # Modify your domain name accordingly - + location / { client_max_body_size 64m; proxy_http_version 1.1; @@ -132,12 +132,12 @@ The initial account username is `root` and password is `123456`. 1. Download the executable file from [GitHub Releases](https://github.com/songquanpeng/one-api/releases/latest) or compile from source: ```shell git clone https://github.com/songquanpeng/one-api.git - + # Build the frontend cd one-api/web/default npm install npm run build - + # Build the backend cd ../.. go mod download @@ -287,7 +287,9 @@ If the channel ID is not provided, load balancing will be used to distribute the + Double-check that your interface address and API Key are correct. ## Related Projects -[FastGPT](https://github.com/labring/FastGPT): Knowledge question answering system based on the LLM +* [FastGPT](https://github.com/labring/FastGPT): Knowledge question answering system based on the LLM +* [VChart](https://github.com/VisActor/VChart): More than just a cross-platform charting library, but also an expressive data storyteller. +* [VMind](https://github.com/VisActor/VMind): Not just automatic, but also fantastic. Open-source solution for intelligent visualization. ## Note This project is an open-source project. Please use it in compliance with OpenAI's [Terms of Use](https://openai.com/policies/terms-of-use) and **applicable laws and regulations**. It must not be used for illegal purposes. diff --git a/README.md b/README.md index 8f59a14a..b5168264 100644 --- a/README.md +++ b/README.md @@ -53,7 +53,7 @@ _✨ 通过标准的 OpenAI API 格式访问所有的大模型，开箱即用 > [!NOTE] > 本项目为开源项目，使用者必须在遵循 OpenAI 的[使用条款](https://openai.com/policies/terms-of-use)以及**法律法规**的情况下使用，不得用于非法用途。 -> +> > 根据[《生成式人工智能服务管理暂行办法》](http://www.cac.gov.cn/2023-07/13/c_1690898327029107.htm)的要求，请勿对中国地区公众提供一切未经备案的生成式人工智能服务。 > [!WARNING] @@ -144,7 +144,7 @@ Nginx 的参考配置： ``` server{ server_name openai.justsong.cn; # 请根据实际情况修改你的域名 - + location / { client_max_body_size 64m; proxy_http_version 1.1; @@ -189,12 +189,12 @@ docker-compose ps 1. 从 [GitHub Releases](https://github.com/songquanpeng/one-api/releases/latest) 下载可执行文件或者从源码编译： ```shell git clone https://github.com/songquanpeng/one-api.git - + # 构建前端 cd one-api/web/default npm install npm run build - + # 构建后端 cd ../.. go mod download @@ -321,7 +321,7 @@ Render 可以直接部署 docker 镜像，不需要 fork 仓库：https://dashbo 例如对于 OpenAI 的官方库： ```bash OPENAI_API_KEY="sk-xxxxxx" -OPENAI_API_BASE="https://:/v1" +OPENAI_API_BASE="https://:/v1" ``` ```mermaid @@ -448,6 +448,8 @@ https://openai.justsong.cn ## 相关项目 * [FastGPT](https://github.com/labring/FastGPT): 基于 LLM 大语言模型的知识库问答系统 * [ChatGPT Next Web](https://github.com/Yidadaa/ChatGPT-Next-Web): 一键拥有你自己的跨平台 ChatGPT 应用 +* [VChart](https://github.com/VisActor/VChart): 不只是开箱即用的多端图表库，更是生动灵活的数据故事讲述者。 +* [VMind](https://github.com/VisActor/VMind): 不仅自动，还很智能。开源智能可视化解决方案。 ## 注意 From 1ce1e529ee547989cdbcc1ab04163fbce608d887 Mon Sep 17 00:00:00 2001 From: Leo Q Date: Tue, 2 Jul 2024 00:05:47 +0800 Subject: [PATCH 31/65] ci: skip archive, upload directly (#1586) --- .github/workflows/ci.yml | 16 ++-------------- 1 file changed, 2 insertions(+), 14 deletions(-) diff --git a/.github/workflows/ci.yml b/.github/workflows/ci.yml index 30ac5f82..36798711 100644 --- a/.github/workflows/ci.yml +++ b/.github/workflows/ci.yml @@ -36,21 +36,9 @@ jobs: # in the next step as well as the next job. - name: Test run: go test -cover -coverprofile=coverage.txt ./... - - - name: Archive code coverage results - uses: actions/upload-artifact@v4 + - uses: codecov/codecov-action@v4 with: - name: code-coverage - path: coverage.txt # Make sure to use the same file name you chose for the "-coverprofile" in the "Test" step - - code_coverage: - name: "Code coverage report" - runs-on: ubuntu-latest - needs: unit_tests # Depends on the artifact uploaded by the "unit_tests" job - steps: - - uses: codecov/codecov-action@v4 - with: - token: ${{ secrets.CODECOV_TOKEN }} + token: ${{ secrets.CODECOV_TOKEN }} commit_lint: runs-on: ubuntu-latest From 0fc07ea55897a9d74380da2767b9bfa25e71cbd3 Mon Sep 17 00:00:00 2001 From: Mikey Date: Tue, 2 Jul 2024 00:12:01 +0800 Subject: [PATCH 32/65] feat: add support for Claude 3 tool use (function calling) (#1587) * feat: add tool support for AWS & Claude * fix: add {} for openai compatibility in streaming tool_use --- relay/adaptor/anthropic/main.go | 121 +++++++++++++++++++++++++++++-- relay/adaptor/anthropic/model.go | 21 ++++++ relay/adaptor/aws/main.go | 24 +++++- relay/adaptor/aws/model.go | 3 + relay/model/message.go | 9 ++- relay/model/tool.go | 4 +- 6 files changed, 168 insertions(+), 14 deletions(-) diff --git a/relay/adaptor/anthropic/main.go b/relay/adaptor/anthropic/main.go index c817a9d1..d3e306c8 100644 --- a/relay/adaptor/anthropic/main.go +++ b/relay/adaptor/anthropic/main.go @@ -29,12 +29,30 @@ func stopReasonClaude2OpenAI(reason *string) string { return "stop" case "max_tokens": return "length" + case "tool_use": + return "tool_calls" default: return *reason } } func ConvertRequest(textRequest model.GeneralOpenAIRequest) *Request { + claudeTools := make([]Tool, 0, len(textRequest.Tools)) + + for _, tool := range textRequest.Tools { + if params, ok := tool.Function.Parameters.(map[string]any); ok { + claudeTools = append(claudeTools, Tool{ + Name: tool.Function.Name, + Description: tool.Function.Description, + InputSchema: InputSchema{ + Type: params["type"].(string), + Properties: params["properties"], + Required: params["required"], + }, + }) + } + } + claudeRequest := Request{ Model: textRequest.Model, MaxTokens: textRequest.MaxTokens, @@ -42,6 +60,24 @@ func ConvertRequest(textRequest model.GeneralOpenAIRequest) *Request { TopP: textRequest.TopP, TopK: textRequest.TopK, Stream: textRequest.Stream, + Tools: claudeTools, + } + if len(claudeTools) > 0 { + claudeToolChoice := struct { + Type string `json:"type"` + Name string `json:"name,omitempty"` + }{Type: "auto"} // default value https://docs.anthropic.com/en/docs/build-with-claude/tool-use#controlling-claudes-output + if choice, ok := textRequest.ToolChoice.(map[string]any); ok { + if function, ok := choice["function"].(map[string]any); ok { + claudeToolChoice.Type = "tool" + claudeToolChoice.Name = function["name"].(string) + } + } else if toolChoiceType, ok := textRequest.ToolChoice.(string); ok { + if toolChoiceType == "any" { + claudeToolChoice.Type = toolChoiceType + } + } + claudeRequest.ToolChoice = claudeToolChoice } if claudeRequest.MaxTokens == 0 { claudeRequest.MaxTokens = 4096 @@ -64,7 +100,24 @@ func ConvertRequest(textRequest model.GeneralOpenAIRequest) *Request { if message.IsStringContent() { content.Type = "text" content.Text = message.StringContent() + if message.Role == "tool" { + claudeMessage.Role = "user" + content.Type = "tool_result" + content.Content = content.Text + content.Text = "" + content.ToolUseId = message.ToolCallId + } claudeMessage.Content = append(claudeMessage.Content, content) + for i := range message.ToolCalls { + inputParam := make(map[string]any) + _ = json.Unmarshal([]byte(message.ToolCalls[i].Function.Arguments.(string)), &inputParam) + claudeMessage.Content = append(claudeMessage.Content, Content{ + Type: "tool_use", + Id: message.ToolCalls[i].Id, + Name: message.ToolCalls[i].Function.Name, + Input: inputParam, + }) + } claudeRequest.Messages = append(claudeRequest.Messages, claudeMessage) continue } @@ -97,16 +150,35 @@ func StreamResponseClaude2OpenAI(claudeResponse *StreamResponse) (*openai.ChatCo var response *Response var responseText string var stopReason string + tools := make([]model.Tool, 0) + switch claudeResponse.Type { case "message_start": return nil, claudeResponse.Message case "content_block_start": if claudeResponse.ContentBlock != nil { responseText = claudeResponse.ContentBlock.Text + if claudeResponse.ContentBlock.Type == "tool_use" { + tools = append(tools, model.Tool{ + Id: claudeResponse.ContentBlock.Id, + Type: "function", + Function: model.Function{ + Name: claudeResponse.ContentBlock.Name, + Arguments: "", + }, + }) + } } case "content_block_delta": if claudeResponse.Delta != nil { responseText = claudeResponse.Delta.Text + if claudeResponse.Delta.Type == "input_json_delta" { + tools = append(tools, model.Tool{ + Function: model.Function{ + Arguments: claudeResponse.Delta.PartialJson, + }, + }) + } } case "message_delta": if claudeResponse.Usage != nil { @@ -120,6 +192,10 @@ func StreamResponseClaude2OpenAI(claudeResponse *StreamResponse) (*openai.ChatCo } var choice openai.ChatCompletionsStreamResponseChoice choice.Delta.Content = responseText + if len(tools) > 0 { + choice.Delta.Content = nil // compatible with other OpenAI derivative applications, like LobeOpenAICompatibleFactory ... + choice.Delta.ToolCalls = tools + } choice.Delta.Role = "assistant" finishReason := stopReasonClaude2OpenAI(&stopReason) if finishReason != "null" { @@ -136,12 +212,27 @@ func ResponseClaude2OpenAI(claudeResponse *Response) *openai.TextResponse { if len(claudeResponse.Content) > 0 { responseText = claudeResponse.Content[0].Text } + tools := make([]model.Tool, 0) + for _, v := range claudeResponse.Content { + if v.Type == "tool_use" { + args, _ := json.Marshal(v.Input) + tools = append(tools, model.Tool{ + Id: v.Id, + Type: "function", // compatible with other OpenAI derivative applications + Function: model.Function{ + Name: v.Name, + Arguments: string(args), + }, + }) + } + } choice := openai.TextResponseChoice{ Index: 0, Message: model.Message{ - Role: "assistant", - Content: responseText, - Name: nil, + Role: "assistant", + Content: responseText, + Name: nil, + ToolCalls: tools, }, FinishReason: stopReasonClaude2OpenAI(claudeResponse.StopReason), } @@ -176,6 +267,7 @@ func StreamHandler(c *gin.Context, resp *http.Response) (*model.ErrorWithStatusC var usage model.Usage var modelName string var id string + var lastToolCallChoice openai.ChatCompletionsStreamResponseChoice for scanner.Scan() { data := scanner.Text() @@ -196,9 +288,20 @@ func StreamHandler(c *gin.Context, resp *http.Response) (*model.ErrorWithStatusC if meta != nil { usage.PromptTokens += meta.Usage.InputTokens usage.CompletionTokens += meta.Usage.OutputTokens - modelName = meta.Model - id = fmt.Sprintf("chatcmpl-%s", meta.Id) - continue + if len(meta.Id) > 0 { // only message_start has an id, otherwise it's a finish_reason event. + modelName = meta.Model + id = fmt.Sprintf("chatcmpl-%s", meta.Id) + continue + } else { // finish_reason case + if len(lastToolCallChoice.Delta.ToolCalls) > 0 { + lastArgs := &lastToolCallChoice.Delta.ToolCalls[len(lastToolCallChoice.Delta.ToolCalls)-1].Function + if len(lastArgs.Arguments.(string)) == 0 { // compatible with OpenAI sending an empty object `{}` when no arguments. + lastArgs.Arguments = "{}" + response.Choices[len(response.Choices)-1].Delta.Content = nil + response.Choices[len(response.Choices)-1].Delta.ToolCalls = lastToolCallChoice.Delta.ToolCalls + } + } + } } if response == nil { continue @@ -207,6 +310,12 @@ func StreamHandler(c *gin.Context, resp *http.Response) (*model.ErrorWithStatusC response.Id = id response.Model = modelName response.Created = createdTime + + for _, choice := range response.Choices { + if len(choice.Delta.ToolCalls) > 0 { + lastToolCallChoice = choice + } + } err = render.ObjectData(c, response) if err != nil { logger.SysError(err.Error()) diff --git a/relay/adaptor/anthropic/model.go b/relay/adaptor/anthropic/model.go index 32b187cd..47f76629 100644 --- a/relay/adaptor/anthropic/model.go +++ b/relay/adaptor/anthropic/model.go @@ -16,6 +16,12 @@ type Content struct { Type string `json:"type"` Text string `json:"text,omitempty"` Source *ImageSource `json:"source,omitempty"` + // tool_calls + Id string `json:"id,omitempty"` + Name string `json:"name,omitempty"` + Input any `json:"input,omitempty"` + Content string `json:"content,omitempty"` + ToolUseId string `json:"tool_use_id,omitempty"` } type Message struct { @@ -23,6 +29,18 @@ type Message struct { Content []Content `json:"content"` } +type Tool struct { + Name string `json:"name"` + Description string `json:"description,omitempty"` + InputSchema InputSchema `json:"input_schema"` +} + +type InputSchema struct { + Type string `json:"type"` + Properties any `json:"properties,omitempty"` + Required any `json:"required,omitempty"` +} + type Request struct { Model string `json:"model"` Messages []Message `json:"messages"` @@ -33,6 +51,8 @@ type Request struct { Temperature float64 `json:"temperature,omitempty"` TopP float64 `json:"top_p,omitempty"` TopK int `json:"top_k,omitempty"` + Tools []Tool `json:"tools,omitempty"` + ToolChoice any `json:"tool_choice,omitempty"` //Metadata `json:"metadata,omitempty"` } @@ -61,6 +81,7 @@ type Response struct { type Delta struct { Type string `json:"type"` Text string `json:"text"` + PartialJson string `json:"partial_json,omitempty"` StopReason *string `json:"stop_reason"` StopSequence *string `json:"stop_sequence"` } diff --git a/relay/adaptor/aws/main.go b/relay/adaptor/aws/main.go index 5d29597c..72f40ddc 100644 --- a/relay/adaptor/aws/main.go +++ b/relay/adaptor/aws/main.go @@ -6,6 +6,7 @@ import ( "encoding/json" "fmt" "github.com/songquanpeng/one-api/common/ctxkey" + "github.com/songquanpeng/one-api/relay/adaptor/openai" "io" "net/http" @@ -143,6 +144,8 @@ func StreamHandler(c *gin.Context, awsCli *bedrockruntime.Client) (*relaymodel.E c.Writer.Header().Set("Content-Type", "text/event-stream") var usage relaymodel.Usage var id string + var lastToolCallChoice openai.ChatCompletionsStreamResponseChoice + c.Stream(func(w io.Writer) bool { event, ok := <-stream.Events() if !ok { @@ -163,8 +166,19 @@ func StreamHandler(c *gin.Context, awsCli *bedrockruntime.Client) (*relaymodel.E if meta != nil { usage.PromptTokens += meta.Usage.InputTokens usage.CompletionTokens += meta.Usage.OutputTokens - id = fmt.Sprintf("chatcmpl-%s", meta.Id) - return true + if len(meta.Id) > 0 { // only message_start has an id, otherwise it's a finish_reason event. + id = fmt.Sprintf("chatcmpl-%s", meta.Id) + return true + } else { // finish_reason case + if len(lastToolCallChoice.Delta.ToolCalls) > 0 { + lastArgs := &lastToolCallChoice.Delta.ToolCalls[len(lastToolCallChoice.Delta.ToolCalls)-1].Function + if len(lastArgs.Arguments.(string)) == 0 { // compatible with OpenAI sending an empty object `{}` when no arguments. + lastArgs.Arguments = "{}" + response.Choices[len(response.Choices)-1].Delta.Content = nil + response.Choices[len(response.Choices)-1].Delta.ToolCalls = lastToolCallChoice.Delta.ToolCalls + } + } + } } if response == nil { return true @@ -172,6 +186,12 @@ func StreamHandler(c *gin.Context, awsCli *bedrockruntime.Client) (*relaymodel.E response.Id = id response.Model = c.GetString(ctxkey.OriginalModel) response.Created = createdTime + + for _, choice := range response.Choices { + if len(choice.Delta.ToolCalls) > 0 { + lastToolCallChoice = choice + } + } jsonStr, err := json.Marshal(response) if err != nil { logger.SysError("error marshalling stream response: " + err.Error()) diff --git a/relay/adaptor/aws/model.go b/relay/adaptor/aws/model.go index bcbfb584..6d00b688 100644 --- a/relay/adaptor/aws/model.go +++ b/relay/adaptor/aws/model.go @@ -9,9 +9,12 @@ type Request struct { // AnthropicVersion should be "bedrock-2023-05-31" AnthropicVersion string `json:"anthropic_version"` Messages []anthropic.Message `json:"messages"` + System string `json:"system,omitempty"` MaxTokens int `json:"max_tokens,omitempty"` Temperature float64 `json:"temperature,omitempty"` TopP float64 `json:"top_p,omitempty"` TopK int `json:"top_k,omitempty"` StopSequences []string `json:"stop_sequences,omitempty"` + Tools []anthropic.Tool `json:"tools,omitempty"` + ToolChoice any `json:"tool_choice,omitempty"` } diff --git a/relay/model/message.go b/relay/model/message.go index 32a1055b..b908f989 100644 --- a/relay/model/message.go +++ b/relay/model/message.go @@ -1,10 +1,11 @@ package model type Message struct { - Role string `json:"role,omitempty"` - Content any `json:"content,omitempty"` - Name *string `json:"name,omitempty"` - ToolCalls []Tool `json:"tool_calls,omitempty"` + Role string `json:"role,omitempty"` + Content any `json:"content,omitempty"` + Name *string `json:"name,omitempty"` + ToolCalls []Tool `json:"tool_calls,omitempty"` + ToolCallId string `json:"tool_call_id,omitempty"` } func (m Message) IsStringContent() bool { diff --git a/relay/model/tool.go b/relay/model/tool.go index 253dca35..75dbb8f7 100644 --- a/relay/model/tool.go +++ b/relay/model/tool.go @@ -2,13 +2,13 @@ package model type Tool struct { Id string `json:"id,omitempty"` - Type string `json:"type"` + Type string `json:"type,omitempty"` // when splicing claude tools stream messages, it is empty Function Function `json:"function"` } type Function struct { Description string `json:"description,omitempty"` - Name string `json:"name"` + Name string `json:"name,omitempty"` // when splicing claude tools stream messages, it is empty Parameters any `json:"parameters,omitempty"` // request Arguments any `json:"arguments,omitempty"` // response } From 274fcf3d76299e1e56a670a4c203e130d3561a0e Mon Sep 17 00:00:00 2001 From: igophper <34326532+igophper@users.noreply.github.com> Date: Wed, 3 Jul 2024 20:50:40 +0800 Subject: [PATCH 33/65] refactor: init db (#1590) MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Co-authored-by: 江杭辉 --- main.go | 22 ++--- model/main.go | 219 ++++++++++++++++++++++++++++++++------------------ 2 files changed, 150 insertions(+), 91 deletions(-) diff --git a/main.go b/main.go index 4afbe5dd..67a3cd95 100644 --- a/main.go +++ b/main.go @@ -27,27 +27,19 @@ func main() { common.Init() logger.SetupLogger() logger.SysLogf("One API %s started", common.Version) - if os.Getenv("GIN_MODE") != "debug" { + + if os.Getenv("GIN_MODE") != gin.DebugMode { gin.SetMode(gin.ReleaseMode) } if config.DebugEnabled { logger.SysLog("running in debug mode") } - var err error + // Initialize SQL Database - model.DB, err = model.InitDB("SQL_DSN") - if err != nil { - logger.FatalLog("failed to initialize database: " + err.Error()) - } - if os.Getenv("LOG_SQL_DSN") != "" { - logger.SysLog("using secondary database for table logs") - model.LOG_DB, err = model.InitDB("LOG_SQL_DSN") - if err != nil { - logger.FatalLog("failed to initialize secondary database: " + err.Error()) - } - } else { - model.LOG_DB = model.DB - } + model.InitDB() + model.InitLogDB() + + var err error err = model.CreateRootAccountIfNeed() if err != nil { logger.FatalLog("database init error: " + err.Error()) diff --git a/model/main.go b/model/main.go index 4b5323c4..11752404 100644 --- a/model/main.go +++ b/model/main.go @@ -1,6 +1,7 @@ package model import ( + "database/sql" "fmt" "github.com/songquanpeng/one-api/common" "github.com/songquanpeng/one-api/common/config" @@ -60,90 +61,156 @@ func CreateRootAccountIfNeed() error { } func chooseDB(envName string) (*gorm.DB, error) { - if os.Getenv(envName) != "" { - dsn := os.Getenv(envName) - if strings.HasPrefix(dsn, "postgres://") { - // Use PostgreSQL - logger.SysLog("using PostgreSQL as database") - common.UsingPostgreSQL = true - return gorm.Open(postgres.New(postgres.Config{ - DSN: dsn, - PreferSimpleProtocol: true, // disables implicit prepared statement usage - }), &gorm.Config{ - PrepareStmt: true, // precompile SQL - }) - } + dsn := os.Getenv(envName) + + switch { + case strings.HasPrefix(dsn, "postgres://"): + // Use PostgreSQL + return openPostgreSQL(dsn) + case dsn != "": // Use MySQL - logger.SysLog("using MySQL as database") - common.UsingMySQL = true - return gorm.Open(mysql.Open(dsn), &gorm.Config{ - PrepareStmt: true, // precompile SQL - }) + return openMySQL(dsn) + default: + // Use SQLite + return openSQLite() } - // Use SQLite - logger.SysLog("SQL_DSN not set, using SQLite as database") - common.UsingSQLite = true - config := fmt.Sprintf("?_busy_timeout=%d", common.SQLiteBusyTimeout) - return gorm.Open(sqlite.Open(common.SQLitePath+config), &gorm.Config{ +} + +func openPostgreSQL(dsn string) (*gorm.DB, error) { + logger.SysLog("using PostgreSQL as database") + common.UsingPostgreSQL = true + return gorm.Open(postgres.New(postgres.Config{ + DSN: dsn, + PreferSimpleProtocol: true, // disables implicit prepared statement usage + }), &gorm.Config{ PrepareStmt: true, // precompile SQL }) } -func InitDB(envName string) (db *gorm.DB, err error) { - db, err = chooseDB(envName) - if err == nil { - if config.DebugSQLEnabled { - db = db.Debug() - } - sqlDB, err := db.DB() - if err != nil { - return nil, err - } - sqlDB.SetMaxIdleConns(env.Int("SQL_MAX_IDLE_CONNS", 100)) - sqlDB.SetMaxOpenConns(env.Int("SQL_MAX_OPEN_CONNS", 1000)) - sqlDB.SetConnMaxLifetime(time.Second * time.Duration(env.Int("SQL_MAX_LIFETIME", 60))) +func openMySQL(dsn string) (*gorm.DB, error) { + logger.SysLog("using MySQL as database") + common.UsingMySQL = true + return gorm.Open(mysql.Open(dsn), &gorm.Config{ + PrepareStmt: true, // precompile SQL + }) +} - if !config.IsMasterNode { - return db, err - } - if common.UsingMySQL { - _, _ = sqlDB.Exec("DROP INDEX idx_channels_key ON channels;") // TODO: delete this line when most users have upgraded - } - logger.SysLog("database migration started") - err = db.AutoMigrate(&Channel{}) - if err != nil { - return nil, err - } - err = db.AutoMigrate(&Token{}) - if err != nil { - return nil, err - } - err = db.AutoMigrate(&User{}) - if err != nil { - return nil, err - } - err = db.AutoMigrate(&Option{}) - if err != nil { - return nil, err - } - err = db.AutoMigrate(&Redemption{}) - if err != nil { - return nil, err - } - err = db.AutoMigrate(&Ability{}) - if err != nil { - return nil, err - } - err = db.AutoMigrate(&Log{}) - if err != nil { - return nil, err - } - logger.SysLog("database migrated") - return db, err - } else { - logger.FatalLog(err) +func openSQLite() (*gorm.DB, error) { + logger.SysLog("SQL_DSN not set, using SQLite as database") + common.UsingSQLite = true + dsn := fmt.Sprintf("%s?_busy_timeout=%d", common.SQLitePath, common.SQLiteBusyTimeout) + return gorm.Open(sqlite.Open(dsn), &gorm.Config{ + PrepareStmt: true, // precompile SQL + }) +} + +func InitDB() { + var err error + DB, err = chooseDB("SQL_DSN") + if err != nil { + logger.FatalLog("failed to initialize database: " + err.Error()) + return } - return db, err + + sqlDB := setDBConns(DB) + + if !config.IsMasterNode { + return + } + + if common.UsingMySQL { + _, _ = sqlDB.Exec("DROP INDEX idx_channels_key ON channels;") // TODO: delete this line when most users have upgraded + } + + logger.SysLog("database migration started") + if err = migrateDB(); err != nil { + logger.FatalLog("failed to migrate database: " + err.Error()) + return + } + logger.SysLog("database migrated") +} + +func migrateDB() error { + var err error + if err = DB.AutoMigrate(&Channel{}); err != nil { + return err + } + if err = DB.AutoMigrate(&Token{}); err != nil { + return err + } + if err = DB.AutoMigrate(&User{}); err != nil { + return err + } + if err = DB.AutoMigrate(&Option{}); err != nil { + return err + } + if err = DB.AutoMigrate(&Redemption{}); err != nil { + return err + } + if err = DB.AutoMigrate(&Ability{}); err != nil { + return err + } + if err = DB.AutoMigrate(&Log{}); err != nil { + return err + } + if err = DB.AutoMigrate(&Channel{}); err != nil { + return err + } + return nil +} + +func InitLogDB() { + if os.Getenv("LOG_SQL_DSN") == "" { + LOG_DB = DB + return + } + + logger.SysLog("using secondary database for table logs") + var err error + LOG_DB, err = chooseDB("LOG_SQL_DSN") + if err != nil { + logger.FatalLog("failed to initialize secondary database: " + err.Error()) + return + } + + setDBConns(LOG_DB) + + if !config.IsMasterNode { + return + } + + logger.SysLog("secondary database migration started") + err = migrateLOGDB() + if err != nil { + logger.FatalLog("failed to migrate secondary database: " + err.Error()) + return + } + logger.SysLog("secondary database migrated") +} + +func migrateLOGDB() error { + var err error + if err = LOG_DB.AutoMigrate(&Log{}); err != nil { + return err + } + return nil +} + +func setDBConns(db *gorm.DB) *sql.DB { + if config.DebugSQLEnabled { + db = db.Debug() + } + + sqlDB, err := db.DB() + if err != nil { + logger.FatalLog("failed to connect database: " + err.Error()) + return nil + } + + sqlDB.SetMaxIdleConns(env.Int("SQL_MAX_IDLE_CONNS", 100)) + sqlDB.SetMaxOpenConns(env.Int("SQL_MAX_OPEN_CONNS", 1000)) + sqlDB.SetConnMaxLifetime(time.Second * time.Duration(env.Int("SQL_MAX_LIFETIME", 60))) + return sqlDB } func closeDB(db *gorm.DB) error { From c4fe57c16512372b84f0765c78d3e0b2d1eef912 Mon Sep 17 00:00:00 2001 From: LinZeliang Date: Wed, 3 Jul 2024 20:53:29 +0800 Subject: [PATCH 34/65] feat: support one or more log file (#1400) Co-authored-by: Laisky.Cai --- common/config/config.go | 3 +++ common/logger/logger.go | 7 ++++++- 2 files changed, 9 insertions(+), 1 deletion(-) diff --git a/common/config/config.go b/common/config/config.go index 4f1c25b6..3f321c87 100644 --- a/common/config/config.go +++ b/common/config/config.go @@ -145,6 +145,9 @@ var InitialRootToken = os.Getenv("INITIAL_ROOT_TOKEN") var GeminiVersion = env.String("GEMINI_VERSION", "v1") + +var OnlyOneLogFile = env.Bool("ONLY_ONE_LOG_FILE", false) + var RelayProxy = env.String("RELAY_PROXY", "") var UserContentRequestProxy = env.String("USER_CONTENT_REQUEST_PROXY", "") var UserContentRequestTimeout = env.Int("USER_CONTENT_REQUEST_TIMEOUT", 30) diff --git a/common/logger/logger.go b/common/logger/logger.go index f725c619..d1022932 100644 --- a/common/logger/logger.go +++ b/common/logger/logger.go @@ -27,7 +27,12 @@ var setupLogOnce sync.Once func SetupLogger() { setupLogOnce.Do(func() { if LogDir != "" { - logPath := filepath.Join(LogDir, fmt.Sprintf("oneapi-%s.log", time.Now().Format("20060102"))) + var logPath string + if config.OnlyOneLogFile { + logPath = filepath.Join(LogDir, "oneapi.log") + } else { + logPath = filepath.Join(LogDir, fmt.Sprintf("oneapi-%s.log", time.Now().Format("20060102"))) + } fd, err := os.OpenFile(logPath, os.O_APPEND|os.O_CREATE|os.O_WRONLY, 0644) if err != nil { log.Fatal("failed to open log file") From ec6ad248104045d7b67effc72867d9f4a31e55fe Mon Sep 17 00:00:00 2001 From: Leo Q Date: Wed, 3 Jul 2024 22:23:49 +0800 Subject: [PATCH 35/65] feat: support smtp without auth (#1101) --- common/message/email.go | 29 ++++++++++++++++++++++------- 1 file changed, 22 insertions(+), 7 deletions(-) diff --git a/common/message/email.go b/common/message/email.go index b06782db..187ac8c3 100644 --- a/common/message/email.go +++ b/common/message/email.go @@ -6,11 +6,16 @@ import ( "encoding/base64" "fmt" "github.com/songquanpeng/one-api/common/config" + "net" "net/smtp" "strings" "time" ) +func shouldAuth() bool { + return config.SMTPAccount != "" || config.SMTPToken != "" +} + func SendEmail(subject string, receiver string, content string) error { if receiver == "" { return fmt.Errorf("receiver is empty") @@ -41,16 +46,24 @@ func SendEmail(subject string, receiver string, content string) error { "Date: %s\r\n"+ "Content-Type: text/html; charset=UTF-8\r\n\r\n%s\r\n", receiver, config.SystemName, config.SMTPFrom, encodedSubject, messageId, time.Now().Format(time.RFC1123Z), content)) + auth := smtp.PlainAuth("", config.SMTPAccount, config.SMTPToken, config.SMTPServer) addr := fmt.Sprintf("%s:%d", config.SMTPServer, config.SMTPPort) to := strings.Split(receiver, ";") - if config.SMTPPort == 465 { - tlsConfig := &tls.Config{ - InsecureSkipVerify: true, - ServerName: config.SMTPServer, + if config.SMTPPort == 465 || !shouldAuth() { + // need advanced client + var conn net.Conn + var err error + if config.SMTPPort == 465 { + tlsConfig := &tls.Config{ + InsecureSkipVerify: true, + ServerName: config.SMTPServer, + } + conn, err = tls.Dial("tcp", fmt.Sprintf("%s:%d", config.SMTPServer, config.SMTPPort), tlsConfig) + } else { + conn, err = net.Dial("tcp", fmt.Sprintf("%s:%d", config.SMTPServer, config.SMTPPort)) } - conn, err := tls.Dial("tcp", fmt.Sprintf("%s:%d", config.SMTPServer, config.SMTPPort), tlsConfig) if err != nil { return err } @@ -59,8 +72,10 @@ func SendEmail(subject string, receiver string, content string) error { return err } defer client.Close() - if err = client.Auth(auth); err != nil { - return err + if shouldAuth() { + if err = client.Auth(auth); err != nil { + return err + } } if err = client.Mail(config.SMTPFrom); err != nil { return err From 273be557975b758c4e6ee36165daeab772895b58 Mon Sep 17 00:00:00 2001 From: Leo Q Date: Thu, 4 Jul 2024 08:35:41 +0800 Subject: [PATCH 36/65] feat(ui): show available models for air theme (#1595) MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit * feat(ui): air 主题显示可用模型 * chore: 改为全角括号 --- web/air/src/components/PersonalSetting.js | 28 +++++++++++------------ 1 file changed, 14 insertions(+), 14 deletions(-) diff --git a/web/air/src/components/PersonalSetting.js b/web/air/src/components/PersonalSetting.js index 45a5b776..ef4acf14 100644 --- a/web/air/src/components/PersonalSetting.js +++ b/web/air/src/components/PersonalSetting.js @@ -47,7 +47,7 @@ const PersonalSetting = () => { const [countdown, setCountdown] = useState(30); const [affLink, setAffLink] = useState(''); const [systemToken, setSystemToken] = useState(''); - // const [models, setModels] = useState([]); + const [models, setModels] = useState([]); const [openTransfer, setOpenTransfer] = useState(false); const [transferAmount, setTransferAmount] = useState(0); @@ -72,7 +72,7 @@ const PersonalSetting = () => { console.log(userState); } ); - // loadModels().then(); + loadModels().then(); getAffLink().then(); setTransferAmount(getQuotaPerUnit()); }, []); @@ -127,16 +127,16 @@ const PersonalSetting = () => { } }; - // const loadModels = async () => { - // let res = await API.get(`/api/user/models`); - // const { success, message, data } = res.data; - // if (success) { - // setModels(data); - // console.log(data); - // } else { - // showError(message); - // } - // }; + const loadModels = async () => { + let res = await API.get(`/api/user/available_models`); + const { success, message, data } = res.data; + if (success) { + setModels(data); + console.log(data); + } else { + showError(message); + } + }; const handleAffLinkClick = async (e) => { e.target.select(); @@ -344,7 +344,7 @@ const PersonalSetting = () => { } > 调用信息 - {/* 可用模型 +

可用模型（可点击复制）

{models.map((model) => ( @@ -355,7 +355,7 @@ const PersonalSetting = () => { ))} -
*/} + {/* Date: Fri, 5 Jul 2024 18:05:16 +0800 Subject: [PATCH 37/65] feat: support test specific model (#1600) --- controller/channel-test.go | 36 ++++++----- web/default/src/components/ChannelsTable.js | 70 +++++++++++++++++---- 2 files changed, 77 insertions(+), 29 deletions(-) diff --git a/controller/channel-test.go b/controller/channel-test.go index b8c41819..f8327284 100644 --- a/controller/channel-test.go +++ b/controller/channel-test.go @@ -14,6 +14,7 @@ import ( "sync" "time" + "github.com/gin-gonic/gin" "github.com/songquanpeng/one-api/common/config" "github.com/songquanpeng/one-api/common/ctxkey" "github.com/songquanpeng/one-api/common/logger" @@ -27,15 +28,15 @@ import ( "github.com/songquanpeng/one-api/relay/meta" relaymodel "github.com/songquanpeng/one-api/relay/model" "github.com/songquanpeng/one-api/relay/relaymode" - - "github.com/gin-gonic/gin" ) -func buildTestRequest() *relaymodel.GeneralOpenAIRequest { +func buildTestRequest(model string) *relaymodel.GeneralOpenAIRequest { + if model == "" { + model = "gpt-3.5-turbo" + } testRequest := &relaymodel.GeneralOpenAIRequest{ MaxTokens: 2, - Stream: false, - Model: "gpt-3.5-turbo", + Model: model, } testMessage := relaymodel.Message{ Role: "user", @@ -45,7 +46,7 @@ func buildTestRequest() *relaymodel.GeneralOpenAIRequest { return testRequest } -func testChannel(channel *model.Channel) (err error, openaiErr *relaymodel.Error) { +func testChannel(channel *model.Channel, request *relaymodel.GeneralOpenAIRequest) (err error, openaiErr *relaymodel.Error) { w := httptest.NewRecorder() c, _ := gin.CreateTestContext(w) c.Request = &http.Request{ @@ -68,12 +69,8 @@ func testChannel(channel *model.Channel) (err error, openaiErr *relaymodel.Error return fmt.Errorf("invalid api type: %d, adaptor is nil", apiType), nil } adaptor.Init(meta) - var modelName string - modelList := adaptor.GetModelList() + modelName := request.Model modelMap := channel.GetModelMapping() - if len(modelList) != 0 { - modelName = modelList[0] - } if modelName == "" || !strings.Contains(channel.Models, modelName) { modelNames := strings.Split(channel.Models, ",") if len(modelNames) > 0 { @@ -83,9 +80,8 @@ func testChannel(channel *model.Channel) (err error, openaiErr *relaymodel.Error modelName = modelMap[modelName] } } - request := buildTestRequest() + meta.OriginModelName, meta.ActualModelName = request.Model, modelName request.Model = modelName - meta.OriginModelName, meta.ActualModelName = modelName, modelName convertedRequest, err := adaptor.ConvertRequest(c, relaymode.ChatCompletions, request) if err != nil { return err, nil @@ -139,10 +135,15 @@ func TestChannel(c *gin.Context) { }) return } + model := c.Query("model") + testRequest := buildTestRequest(model) tik := time.Now() - err, _ = testChannel(channel) + err, _ = testChannel(channel, testRequest) tok := time.Now() milliseconds := tok.Sub(tik).Milliseconds() + if err != nil { + milliseconds = 0 + } go channel.UpdateResponseTime(milliseconds) consumedTime := float64(milliseconds) / 1000.0 if err != nil { @@ -150,6 +151,7 @@ func TestChannel(c *gin.Context) { "success": false, "message": err.Error(), "time": consumedTime, + "model": model, }) return } @@ -157,6 +159,7 @@ func TestChannel(c *gin.Context) { "success": true, "message": "", "time": consumedTime, + "model": model, }) return } @@ -187,11 +190,12 @@ func testChannels(notify bool, scope string) error { for _, channel := range channels { isChannelEnabled := channel.Status == model.ChannelStatusEnabled tik := time.Now() - err, openaiErr := testChannel(channel) + testRequest := buildTestRequest("") + err, openaiErr := testChannel(channel, testRequest) tok := time.Now() milliseconds := tok.Sub(tik).Milliseconds() if isChannelEnabled && milliseconds > disableThreshold { - err = errors.New(fmt.Sprintf("响应时间 %.2fs 超过阈值 %.2fs", float64(milliseconds)/1000.0, float64(disableThreshold)/1000.0)) + err = fmt.Errorf("响应时间 %.2fs 超过阈值 %.2fs", float64(milliseconds)/1000.0, float64(disableThreshold)/1000.0) if config.AutomaticDisableChannelEnabled { monitor.DisableChannel(channel.Id, channel.Name, err.Error()) } else { diff --git a/web/default/src/components/ChannelsTable.js b/web/default/src/components/ChannelsTable.js index 1258ca5a..6025b7d9 100644 --- a/web/default/src/components/ChannelsTable.js +++ b/web/default/src/components/ChannelsTable.js @@ -1,5 +1,5 @@ import React, { useEffect, useState } from 'react'; -import { Button, Form, Input, Label, Message, Pagination, Popup, Table } from 'semantic-ui-react'; +import { Button, Dropdown, Form, Input, Label, Message, Pagination, Popup, Table } from 'semantic-ui-react'; import { Link } from 'react-router-dom'; import { API, @@ -70,13 +70,33 @@ const ChannelsTable = () => { const res = await API.get(`/api/channel/?p=${startIdx}`); const { success, message, data } = res.data; if (success) { - if (startIdx === 0) { - setChannels(data); - } else { - let newChannels = [...channels]; - newChannels.splice(startIdx * ITEMS_PER_PAGE, data.length, ...data); - setChannels(newChannels); - } + let localChannels = data.map((channel) => { + if (channel.models === '') { + channel.models = []; + channel.test_model = ""; + } else { + channel.models = channel.models.split(','); + if (channel.models.length > 0) { + channel.test_model = channel.models[0]; + } + channel.model_options = channel.models.map((model) => { + return { + key: model, + text: model, + value: model, + } + }) + console.log('channel', channel) + } + return channel; + }); + if (startIdx === 0) { + setChannels(localChannels); + } else { + let newChannels = [...channels]; + newChannels.splice(startIdx * ITEMS_PER_PAGE, data.length, ...localChannels); + setChannels(newChannels); + } } else { showError(message); } @@ -225,19 +245,31 @@ const ChannelsTable = () => { setSearching(false); }; - const testChannel = async (id, name, idx) => { - const res = await API.get(`/api/channel/test/${id}/`); - const { success, message, time } = res.data; + const switchTestModel = async (idx, model) => { + let newChannels = [...channels]; + let realIdx = (activePage - 1) * ITEMS_PER_PAGE + idx; + newChannels[realIdx].test_model = model; + setChannels(newChannels); + }; + + const testChannel = async (id, name, idx, m) => { + const res = await API.get(`/api/channel/test/${id}?model=${m}`); + const { success, message, time, model } = res.data; if (success) { let newChannels = [...channels]; let realIdx = (activePage - 1) * ITEMS_PER_PAGE + idx; newChannels[realIdx].response_time = time * 1000; newChannels[realIdx].test_time = Date.now() / 1000; setChannels(newChannels); - showInfo(`渠道 ${name} 测试成功，耗时 ${time.toFixed(2)} 秒。`); + showInfo(`渠道 ${name} 测试成功，模型 ${model}，耗时 ${time.toFixed(2)} 秒。`); } else { showError(message); } + let newChannels = [...channels]; + let realIdx = (activePage - 1) * ITEMS_PER_PAGE + idx; + newChannels[realIdx].response_time = time * 1000; + newChannels[realIdx].test_time = Date.now() / 1000; + setChannels(newChannels); }; const testChannels = async (scope) => { @@ -405,6 +437,7 @@ const ChannelsTable = () => { > 优先级 + 测试模型 操作 @@ -459,13 +492,24 @@ const ChannelsTable = () => { basic /> + + { + switchTestModel(idx, data.value); + }} + /> +
- - - 填入 - } - placeholder='输入自定义模型名称' - value={customModel} - onChange={(e, { value }) => { - setCustomModel(value); - }} - onKeyDown={(e) => { - if (e.key === 'Enter') { - addCustomModel(); - e.preventDefault(); - } - }} - /> -
- - - + { + inputs.type !== 43 && ( + + { + copy(value).then(); + }} + selection + onChange={handleInputChange} + value={inputs.models} + autoComplete='new-password' + options={modelOptions} + /> + + ) + } + { + inputs.type !== 43 && ( +
+ + + + 填入 + } + placeholder='输入自定义模型名称' + value={customModel} + onChange={(e, { value }) => { + setCustomModel(value); + }} + onKeyDown={(e) => { + if (e.key === 'Enter') { + addCustomModel(); + e.preventDefault(); + } + }} + /> +
+ ) + } + { + inputs.type !== 43 && ( + + + + ) + } { inputs.type === 33 && ( From 36039e329e0cef61056b34ab364ee9072ca08501 Mon Sep 17 00:00:00 2001 From: Junyan Qin <1010553892@qq.com> Date: Tue, 6 Aug 2024 23:33:43 +0800 Subject: [PATCH 59/65] docs: update introduction for QChatGPT (#1707) --- README.md | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/README.md b/README.md index 987fde7d..f8225121 100644 --- a/README.md +++ b/README.md @@ -251,9 +251,9 @@ docker run --name chatgpt-web -d -p 3002:3002 -e OPENAI_API_BASE_URL=https://ope #### QChatGPT - QQ机器人 项目主页：https://github.com/RockChinQ/QChatGPT -根据文档完成部署后，在`config.py`设置配置项`openai_config`的`reverse_proxy`为 One API 后端地址，设置`api_key`为 One API 生成的key，并在配置项`completion_api_params`的`model`参数设置为 One API 支持的模型名称。 +根据[文档](https://qchatgpt.rockchin.top)完成部署后，在 `data/provider.json`设置`requester.openai-chat-completions.base-url`为 One API 实例地址，并填写 API Key 到 `keys.openai` 组中，设置 `model` 为要使用的模型名称。 -可安装 [Switcher 插件](https://github.com/RockChinQ/Switcher)在运行时切换所使用的模型。 +运行期间可以通过`!model`命令查看、切换可用模型。 ### 部署到第三方平台
From 402fcf7f79e22697a00a10f4e9dc176799cb28ba Mon Sep 17 00:00:00 2001 From: Shenghang Tsai Date: Tue, 6 Aug 2024 23:42:25 +0800 Subject: [PATCH 60/65] feat: add SiliconFlow (#1717) * Add SiliconFlow * Update README.md * Update README.md * Update channel.constants.js * Update ChannelConstants.js * Update channel.constants.js * Update ChannelConstants.js * Update compatible.go * Update README.md --- README.md | 1 + relay/adaptor/openai/compatible.go | 4 +++ relay/adaptor/siliconflow/constants.go | 36 +++++++++++++++++++ relay/channeltype/define.go | 1 + relay/channeltype/url.go | 1 + web/air/src/constants/channel.constants.js | 1 + web/berry/src/constants/ChannelConstants.js | 6 ++++ .../src/constants/channel.constants.js | 1 + 8 files changed, 51 insertions(+) create mode 100644 relay/adaptor/siliconflow/constants.go diff --git a/README.md b/README.md index f8225121..5f9947b0 100644 --- a/README.md +++ b/README.md @@ -89,6 +89,7 @@ _✨ 通过标准的 OpenAI API 格式访问所有的大模型，开箱即用 + [x] [DeepL](https://www.deepl.com/) + [x] [together.ai](https://www.together.ai/) + [x] [novita.ai](https://www.novita.ai/) + + [x] [硅基流动 SiliconCloud](https://siliconflow.cn/siliconcloud) 2. 支持配置镜像以及众多[第三方代理服务](https://iamazing.cn/page/openai-api-third-party-services)。 3. 支持通过**负载均衡**的方式访问多个渠道。 4. 支持 **stream 模式**，可以通过流式传输实现打字机效果。 diff --git a/relay/adaptor/openai/compatible.go b/relay/adaptor/openai/compatible.go index 3445249c..0512f05c 100644 --- a/relay/adaptor/openai/compatible.go +++ b/relay/adaptor/openai/compatible.go @@ -13,6 +13,7 @@ import ( "github.com/songquanpeng/one-api/relay/adaptor/novita" "github.com/songquanpeng/one-api/relay/adaptor/stepfun" "github.com/songquanpeng/one-api/relay/adaptor/togetherai" + "github.com/songquanpeng/one-api/relay/adaptor/siliconflow" "github.com/songquanpeng/one-api/relay/channeltype" ) @@ -30,6 +31,7 @@ var CompatibleChannels = []int{ channeltype.DeepSeek, channeltype.TogetherAI, channeltype.Novita, + channeltype.SiliconFlow, } func GetCompatibleChannelMeta(channelType int) (string, []string) { @@ -60,6 +62,8 @@ func GetCompatibleChannelMeta(channelType int) (string, []string) { return "doubao", doubao.ModelList case channeltype.Novita: return "novita", novita.ModelList + case channeltype.SiliconFlow: + return "siliconflow", siliconflow.ModelList default: return "openai", ModelList } diff --git a/relay/adaptor/siliconflow/constants.go b/relay/adaptor/siliconflow/constants.go new file mode 100644 index 00000000..0bf54761 --- /dev/null +++ b/relay/adaptor/siliconflow/constants.go @@ -0,0 +1,36 @@ +package siliconflow + +// https://docs.siliconflow.cn/docs/getting-started + +var ModelList = []string{ + "deepseek-ai/deepseek-llm-67b-chat", + "Qwen/Qwen1.5-14B-Chat", + "Qwen/Qwen1.5-7B-Chat", + "Qwen/Qwen1.5-110B-Chat", + "Qwen/Qwen1.5-32B-Chat", + "01-ai/Yi-1.5-6B-Chat", + "01-ai/Yi-1.5-9B-Chat-16K", + "01-ai/Yi-1.5-34B-Chat-16K", + "THUDM/chatglm3-6b", + "deepseek-ai/DeepSeek-V2-Chat", + "THUDM/glm-4-9b-chat", + "Qwen/Qwen2-72B-Instruct", + "Qwen/Qwen2-7B-Instruct", + "Qwen/Qwen2-57B-A14B-Instruct", + "deepseek-ai/DeepSeek-Coder-V2-Instruct", + "Qwen/Qwen2-1.5B-Instruct", + "internlm/internlm2_5-7b-chat", + "BAAI/bge-large-en-v1.5", + "BAAI/bge-large-zh-v1.5", + "Pro/Qwen/Qwen2-7B-Instruct", + "Pro/Qwen/Qwen2-1.5B-Instruct", + "Pro/Qwen/Qwen1.5-7B-Chat", + "Pro/THUDM/glm-4-9b-chat", + "Pro/THUDM/chatglm3-6b", + "Pro/01-ai/Yi-1.5-9B-Chat-16K", + "Pro/01-ai/Yi-1.5-6B-Chat", + "Pro/google/gemma-2-9b-it", + "Pro/internlm/internlm2_5-7b-chat", + "Pro/meta-llama/Meta-Llama-3-8B-Instruct", + "Pro/mistralai/Mistral-7B-Instruct-v0.2", +} diff --git a/relay/channeltype/define.go b/relay/channeltype/define.go index e3b0c98e..a261cff8 100644 --- a/relay/channeltype/define.go +++ b/relay/channeltype/define.go @@ -45,5 +45,6 @@ const ( Novita VertextAI Proxy + SiliconFlow Dummy ) diff --git a/relay/channeltype/url.go b/relay/channeltype/url.go index b5026713..8727faea 100644 --- a/relay/channeltype/url.go +++ b/relay/channeltype/url.go @@ -45,6 +45,7 @@ var ChannelBaseURLs = []string{ "https://api.novita.ai/v3/openai", // 41 "", // 42 "", // 43 + "https://api.siliconflow.cn", // 44 } func init() { diff --git a/web/air/src/constants/channel.constants.js b/web/air/src/constants/channel.constants.js index 18293f5f..04fe94f1 100644 --- a/web/air/src/constants/channel.constants.js +++ b/web/air/src/constants/channel.constants.js @@ -29,6 +29,7 @@ export const CHANNEL_OPTIONS = [ { key: 39, text: 'together.ai', value: 39, color: 'blue' }, { key: 42, text: 'VertexAI', value: 42, color: 'blue' }, { key: 43, text: 'Proxy', value: 43, color: 'blue' }, + { key: 44, text: 'SiliconFlow', value: 44, color: 'blue' }, { key: 8, text: '自定义渠道', value: 8, color: 'pink' }, { key: 22, text: '知识库：FastGPT', value: 22, color: 'blue' }, { key: 21, text: '知识库：AI Proxy', value: 21, color: 'purple' }, diff --git a/web/berry/src/constants/ChannelConstants.js b/web/berry/src/constants/ChannelConstants.js index acfda37b..98ea7ca5 100644 --- a/web/berry/src/constants/ChannelConstants.js +++ b/web/berry/src/constants/ChannelConstants.js @@ -173,6 +173,12 @@ export const CHANNEL_OPTIONS = { value: 43, color: 'primary' }, + 44: { + key: 44, + text: 'SiliconFlow', + value: 44, + color: 'primary' + }, 41: { key: 41, text: 'Novita', diff --git a/web/default/src/constants/channel.constants.js b/web/default/src/constants/channel.constants.js index b2a71016..04e361a1 100644 --- a/web/default/src/constants/channel.constants.js +++ b/web/default/src/constants/channel.constants.js @@ -29,6 +29,7 @@ export const CHANNEL_OPTIONS = [ { key: 39, text: 'together.ai', value: 39, color: 'blue' }, { key: 42, text: 'VertexAI', value: 42, color: 'blue' }, { key: 43, text: 'Proxy', value: 43, color: 'blue' }, + { key: 44, text: 'SiliconFlow', value: 44, color: 'blue' }, { key: 8, text: '自定义渠道', value: 8, color: 'pink' }, { key: 22, text: '知识库：FastGPT', value: 22, color: 'blue' }, { key: 21, text: '知识库：AI Proxy', value: 21, color: 'purple' }, From e7e99e558afab7423f70f6ebd6a9949fcb616dbd Mon Sep 17 00:00:00 2001 From: SLKun Date: Tue, 6 Aug 2024 23:43:20 +0800 Subject: [PATCH 61/65] feat: update Ollama embedding API to latest version with multi-text embedding support (#1715) --- relay/adaptor/ollama/adaptor.go | 2 +- relay/adaptor/ollama/main.go | 25 +++++++++++++++++-------- relay/adaptor/ollama/model.go | 12 ++++++++---- 3 files changed, 26 insertions(+), 13 deletions(-) diff --git a/relay/adaptor/ollama/adaptor.go b/relay/adaptor/ollama/adaptor.go index 66702c5d..ad1f8983 100644 --- a/relay/adaptor/ollama/adaptor.go +++ b/relay/adaptor/ollama/adaptor.go @@ -24,7 +24,7 @@ func (a *Adaptor) GetRequestURL(meta *meta.Meta) (string, error) { // https://github.com/ollama/ollama/blob/main/docs/api.md fullRequestURL := fmt.Sprintf("%s/api/chat", meta.BaseURL) if meta.Mode == relaymode.Embeddings { - fullRequestURL = fmt.Sprintf("%s/api/embeddings", meta.BaseURL) + fullRequestURL = fmt.Sprintf("%s/api/embed", meta.BaseURL) } return fullRequestURL, nil } diff --git a/relay/adaptor/ollama/main.go b/relay/adaptor/ollama/main.go index 936a7e14..6a1d334d 100644 --- a/relay/adaptor/ollama/main.go +++ b/relay/adaptor/ollama/main.go @@ -157,8 +157,15 @@ func StreamHandler(c *gin.Context, resp *http.Response) (*model.ErrorWithStatusC func ConvertEmbeddingRequest(request model.GeneralOpenAIRequest) *EmbeddingRequest { return &EmbeddingRequest{ - Model: request.Model, - Prompt: strings.Join(request.ParseInput(), " "), + Model: request.Model, + Input: request.ParseInput(), + Options: &Options{ + Seed: int(request.Seed), + Temperature: request.Temperature, + TopP: request.TopP, + FrequencyPenalty: request.FrequencyPenalty, + PresencePenalty: request.PresencePenalty, + }, } } @@ -201,15 +208,17 @@ func embeddingResponseOllama2OpenAI(response *EmbeddingResponse) *openai.Embeddi openAIEmbeddingResponse := openai.EmbeddingResponse{ Object: "list", Data: make([]openai.EmbeddingResponseItem, 0, 1), - Model: "text-embedding-v1", + Model: response.Model, Usage: model.Usage{TotalTokens: 0}, } - openAIEmbeddingResponse.Data = append(openAIEmbeddingResponse.Data, openai.EmbeddingResponseItem{ - Object: `embedding`, - Index: 0, - Embedding: response.Embedding, - }) + for i, embedding := range response.Embeddings { + openAIEmbeddingResponse.Data = append(openAIEmbeddingResponse.Data, openai.EmbeddingResponseItem{ + Object: `embedding`, + Index: i, + Embedding: embedding, + }) + } return &openAIEmbeddingResponse } diff --git a/relay/adaptor/ollama/model.go b/relay/adaptor/ollama/model.go index 8baf56a0..29430e1c 100644 --- a/relay/adaptor/ollama/model.go +++ b/relay/adaptor/ollama/model.go @@ -37,11 +37,15 @@ type ChatResponse struct { } type EmbeddingRequest struct { - Model string `json:"model"` - Prompt string `json:"prompt"` + Model string `json:"model"` + Input []string `json:"input"` + // Truncate bool `json:"truncate,omitempty"` + Options *Options `json:"options,omitempty"` + // KeepAlive string `json:"keep_alive,omitempty"` } type EmbeddingResponse struct { - Error string `json:"error,omitempty"` - Embedding []float64 `json:"embedding,omitempty"` + Error string `json:"error,omitempty"` + Model string `json:"model"` + Embeddings [][]float64 `json:"embeddings"` } From b4bfa418a8cfde3fe0dec14e90e4d43feed3cfe1 Mon Sep 17 00:00:00 2001 From: longkeyy Date: Tue, 6 Aug 2024 23:43:33 +0800 Subject: [PATCH 62/65] feat: update gemini model and price (#1705) --- relay/adaptor/gemini/constants.go | 3 +-- relay/adaptor/groq/constants.go | 9 +++++++-- relay/billing/ratio/model.go | 27 +++++++++++++++------------ 3 files changed, 23 insertions(+), 16 deletions(-) diff --git a/relay/adaptor/gemini/constants.go b/relay/adaptor/gemini/constants.go index f65e6bfc..b0f84dfc 100644 --- a/relay/adaptor/gemini/constants.go +++ b/relay/adaptor/gemini/constants.go @@ -3,6 +3,5 @@ package gemini // https://ai.google.dev/models/gemini var ModelList = []string{ - "gemini-pro", "gemini-1.0-pro-001", "gemini-1.5-pro", - "gemini-pro-vision", "gemini-1.0-pro-vision-001", "embedding-001", "text-embedding-004", + "gemini-pro", "gemini-1.0-pro", "gemini-1.5-flash", "gemini-1.5-pro", "text-embedding-004", "aqa", } diff --git a/relay/adaptor/groq/constants.go b/relay/adaptor/groq/constants.go index 1aa2574b..559851ee 100644 --- a/relay/adaptor/groq/constants.go +++ b/relay/adaptor/groq/constants.go @@ -4,9 +4,14 @@ package groq var ModelList = []string{ "gemma-7b-it", - "llama2-7b-2048", - "llama2-70b-4096", "mixtral-8x7b-32768", "llama3-8b-8192", "llama3-70b-8192", + "gemma2-9b-it", + "llama-3.1-405b-reasoning", + "llama-3.1-70b-versatile", + "llama-3.1-8b-instant", + "llama3-groq-70b-8192-tool-use-preview", + "llama3-groq-8b-8192-tool-use-preview", + "whisper-large-v3", } diff --git a/relay/billing/ratio/model.go b/relay/billing/ratio/model.go index 799fa90f..7bc6cd54 100644 --- a/relay/billing/ratio/model.go +++ b/relay/billing/ratio/model.go @@ -98,12 +98,11 @@ var ModelRatio = map[string]float64{ "bge-large-en": 0.002 * RMB, "tao-8k": 0.002 * RMB, // https://ai.google.dev/pricing - "PaLM-2": 1, - "gemini-pro": 1, // $0.00025 / 1k characters -> $0.001 / 1k tokens - "gemini-pro-vision": 1, // $0.00025 / 1k characters -> $0.001 / 1k tokens - "gemini-1.0-pro-vision-001": 1, - "gemini-1.0-pro-001": 1, - "gemini-1.5-pro": 1, + "gemini-pro": 1, // $0.00025 / 1k characters -> $0.001 / 1k tokens + "gemini-1.0-pro": 1, + "gemini-1.5-flash": 1, + "gemini-1.5-pro": 1, + "aqa": 1, // https://open.bigmodel.cn/pricing "glm-4": 0.1 * RMB, "glm-4v": 0.1 * RMB, @@ -158,12 +157,16 @@ var ModelRatio = map[string]float64{ "mistral-large-latest": 8.0 / 1000 * USD, "mistral-embed": 0.1 / 1000 * USD, // https://wow.groq.com/#:~:text=inquiries%C2%A0here.-,Model,-Current%20Speed - "llama3-70b-8192": 0.59 / 1000 * USD, - "mixtral-8x7b-32768": 0.27 / 1000 * USD, - "llama3-8b-8192": 0.05 / 1000 * USD, - "gemma-7b-it": 0.1 / 1000 * USD, - "llama2-70b-4096": 0.64 / 1000 * USD, - "llama2-7b-2048": 0.1 / 1000 * USD, + "gemma-7b-it": 0.07 / 1000000 * USD, + "mixtral-8x7b-32768": 0.24 / 1000000 * USD, + "llama3-8b-8192": 0.05 / 1000000 * USD, + "llama3-70b-8192": 0.59 / 1000000 * USD, + "gemma2-9b-it": 0.20 / 1000000 * USD, + "llama-3.1-405b-reasoning": 0.89 / 1000000 * USD, + "llama-3.1-70b-versatile": 0.59 / 1000000 * USD, + "llama-3.1-8b-instant": 0.05 / 1000000 * USD, + "llama3-groq-70b-8192-tool-use-preview": 0.89 / 1000000 * USD, + "llama3-groq-8b-8192-tool-use-preview": 0.19 / 1000000 * USD, // https://platform.lingyiwanwu.com/docs#-计费单元 "yi-34b-chat-0205": 2.5 / 1000 * RMB, "yi-34b-chat-200k": 12.0 / 1000 * RMB, From 04bb3ef3923ba4b0931f0940e65f06b29cd53df8 Mon Sep 17 00:00:00 2001 From: MotorBottle <71703952+MotorBottle@users.noreply.github.com> Date: Tue, 6 Aug 2024 23:44:37 +0800 Subject: [PATCH 63/65] feat: add Max Tokens and Context Window Setting Options for Ollama Channel (#1694) * Update main.go with max_tokens param * Update model.go with max_tokens param * Update model.go * Update main.go * Update main.go * Adds num_ctx param for Ollama Channel * Added num_ctx param for ollama adapter * Added num_ctx param for ollama adapter * Improved data process logic --- relay/adaptor/ollama/main.go | 8 ++++++-- relay/adaptor/ollama/model.go | 2 ++ relay/model/general.go | 1 + 3 files changed, 9 insertions(+), 2 deletions(-) diff --git a/relay/adaptor/ollama/main.go b/relay/adaptor/ollama/main.go index 6a1d334d..43317ff6 100644 --- a/relay/adaptor/ollama/main.go +++ b/relay/adaptor/ollama/main.go @@ -31,6 +31,8 @@ func ConvertRequest(request model.GeneralOpenAIRequest) *ChatRequest { TopP: request.TopP, FrequencyPenalty: request.FrequencyPenalty, PresencePenalty: request.PresencePenalty, + NumPredict: request.MaxTokens, + NumCtx: request.NumCtx, }, Stream: request.Stream, } @@ -118,8 +120,10 @@ func StreamHandler(c *gin.Context, resp *http.Response) (*model.ErrorWithStatusC common.SetEventStreamHeaders(c) for scanner.Scan() { - data := strings.TrimPrefix(scanner.Text(), "}") - data = data + "}" + data := scanner.Text() + if strings.HasPrefix(data, "}") { + data = strings.TrimPrefix(data, "}") + "}" + } var ollamaResponse ChatResponse err := json.Unmarshal([]byte(data), &ollamaResponse) diff --git a/relay/adaptor/ollama/model.go b/relay/adaptor/ollama/model.go index 29430e1c..7039984f 100644 --- a/relay/adaptor/ollama/model.go +++ b/relay/adaptor/ollama/model.go @@ -7,6 +7,8 @@ type Options struct { TopP float64 `json:"top_p,omitempty"` FrequencyPenalty float64 `json:"frequency_penalty,omitempty"` PresencePenalty float64 `json:"presence_penalty,omitempty"` + NumPredict int `json:"num_predict,omitempty"` + NumCtx int `json:"num_ctx,omitempty"` } type Message struct { diff --git a/relay/model/general.go b/relay/model/general.go index 229a61c1..c34c1c2d 100644 --- a/relay/model/general.go +++ b/relay/model/general.go @@ -29,6 +29,7 @@ type GeneralOpenAIRequest struct { Dimensions int `json:"dimensions,omitempty"` Instruction string `json:"instruction,omitempty"` Size string `json:"size,omitempty"` + NumCtx int `json:"num_ctx,omitempty"` } func (r GeneralOpenAIRequest) ParseInput() []string { From 2af6f6a166604f346a1a326fb481cf1741eadd31 Mon Sep 17 00:00:00 2001 From: TAKO <20227709+HynoR@users.noreply.github.com> Date: Tue, 6 Aug 2024 23:45:15 +0800 Subject: [PATCH 64/65] feat: add Cloudflare New Free Model Llama 3.1 8b (#1703) --- relay/adaptor/cloudflare/constant.go | 1 + 1 file changed, 1 insertion(+) diff --git a/relay/adaptor/cloudflare/constant.go b/relay/adaptor/cloudflare/constant.go index dee79a76..54052aa6 100644 --- a/relay/adaptor/cloudflare/constant.go +++ b/relay/adaptor/cloudflare/constant.go @@ -1,6 +1,7 @@ package cloudflare var ModelList = []string{ + "@cf/meta/llama-3.1-8b-instruct", "@cf/meta/llama-2-7b-chat-fp16", "@cf/meta/llama-2-7b-chat-int8", "@cf/mistral/mistral-7b-instruct-v0.1", From f9774698e9a59823a65c0ede475e28c29638f396 Mon Sep 17 00:00:00 2001 From: longkeyy Date: Tue, 6 Aug 2024 23:51:08 +0800 Subject: [PATCH 65/65] feat: synchronize with the official release of the groq model (#1677) update groq add gemma2-9b-it llama3.1 family fixup price k/token -> m/token