restruct: restruct channel factories

adapter/adapter.go

@@ -4,11 +4,12 @@ import (
 	"chat/adapter/azure"
 	"chat/adapter/baichuan"
 	"chat/adapter/bing"
-	"chat/adapter/chatgpt"
 	"chat/adapter/claude"
+	"chat/adapter/common"
 	"chat/adapter/dashscope"
 	"chat/adapter/hunyuan"
 	"chat/adapter/midjourney"
+	"chat/adapter/openai"
 	"chat/adapter/palm2"
 	"chat/adapter/skylark"
 	"chat/adapter/slack"
@@ -16,171 +17,33 @@ import (
 	"chat/adapter/zhinao"
 	"chat/adapter/zhipuai"
 	"chat/globals"
-	"chat/utils"
 	"fmt"
 )
 
-type RequestProps struct {
+var channelFactories = map[string]adaptercommon.FactoryCreator{
-	MaxRetries *int
+	globals.OpenAIChannelType:      openai.NewChatInstanceFromConfig,
-	Current    int
+	globals.AzureOpenAIChannelType: azure.NewChatInstanceFromConfig,
-	Group      string
+	globals.ClaudeChannelType:      claude.NewChatInstanceFromConfig,
+	globals.SlackChannelType:       slack.NewChatInstanceFromConfig,
+	globals.BingChannelType:        bing.NewChatInstanceFromConfig,
+	globals.PalmChannelType:        palm2.NewChatInstanceFromConfig,
+	globals.SparkdeskChannelType:   sparkdesk.NewChatInstanceFromConfig,
+	globals.ChatGLMChannelType:     zhipuai.NewChatInstanceFromConfig,
+	globals.QwenChannelType:        dashscope.NewChatInstanceFromConfig,
+	globals.HunyuanChannelType:     hunyuan.NewChatInstanceFromConfig,
+	globals.BaichuanChannelType:    baichuan.NewChatInstanceFromConfig,
+	globals.SkylarkChannelType:     skylark.NewChatInstanceFromConfig,
+	globals.ZhinaoChannelType:      zhinao.NewChatInstanceFromConfig,
+	globals.MidjourneyChannelType:  midjourney.NewChatInstanceFromConfig,
 }
 
-type ChatProps struct {
+func createChatRequest(conf globals.ChannelConfig, props *adaptercommon.ChatProps, hook globals.Hook) error {
-	RequestProps
+	props.Model = conf.GetModelReflect(props.OriginalModel)
 
-	Model             string
+	factoryType := conf.GetType()
-	Message           []globals.Message
+	if factory, ok := channelFactories[factoryType]; ok {
-	MaxTokens         *int
+		return factory(conf).CreateStreamChatRequest(props, hook)
-	PresencePenalty   *float32
-	FrequencyPenalty  *float32
-	RepetitionPenalty *float32
-	Temperature       *float32
-	TopP              *float32
-	TopK              *int
-	Tools             *globals.FunctionTools
-	ToolChoice        *interface{}
-	Buffer            utils.Buffer
-}
-
-func createChatRequest(conf globals.ChannelConfig, props *ChatProps, hook globals.Hook) error {
-	model := conf.GetModelReflect(props.Model)
-
-	switch conf.GetType() {
-	case globals.OpenAIChannelType:
-		return chatgpt.NewChatInstanceFromConfig(conf).CreateStreamChatRequest(&chatgpt.ChatProps{
-			Model:            model,
-			Message:          props.Message,
-			Token:            props.MaxTokens,
-			PresencePenalty:  props.PresencePenalty,
-			FrequencyPenalty: props.FrequencyPenalty,
-			Temperature:      props.Temperature,
-			TopP:             props.TopP,
-			Tools:            props.Tools,
-			ToolChoice:       props.ToolChoice,
-			Buffer:           props.Buffer,
-		}, hook)
-
-	case globals.AzureOpenAIChannelType:
-		return azure.NewChatInstanceFromConfig(conf).CreateStreamChatRequest(&azure.ChatProps{
-			Model:            model,
-			Message:          props.Message,
-			Token:            props.MaxTokens,
-			PresencePenalty:  props.PresencePenalty,
-			FrequencyPenalty: props.FrequencyPenalty,
-			Temperature:      props.Temperature,
-			TopP:             props.TopP,
-			Tools:            props.Tools,
-			ToolChoice:       props.ToolChoice,
-			Buffer:           props.Buffer,
-		}, hook)
-
-	case globals.ClaudeChannelType:
-		return claude.NewChatInstanceFromConfig(conf).CreateStreamChatRequest(&claude.ChatProps{
-			Model:       model,
-			Message:     props.Message,
-			Token:       props.MaxTokens,
-			TopP:        props.TopP,
-			TopK:        props.TopK,
-			Temperature: props.Temperature,
-		}, hook)
-
-	case globals.SlackChannelType:
-		return slack.NewChatInstanceFromConfig(conf).CreateStreamChatRequest(&slack.ChatProps{
-			Message: props.Message,
-		}, hook)
-
-	case globals.BingChannelType:
-		return bing.NewChatInstanceFromConfig(conf).CreateStreamChatRequest(&bing.ChatProps{
-			Model:   model,
-			Message: props.Message,
-		}, hook)
-
-	case globals.PalmChannelType:
-		return palm2.NewChatInstanceFromConfig(conf).CreateStreamChatRequest(&palm2.ChatProps{
-			Model:   model,
-			Message: props.Message,
-		}, hook)
-
-	case globals.SparkdeskChannelType:
-		return sparkdesk.NewChatInstance(conf, model).CreateStreamChatRequest(&sparkdesk.ChatProps{
-			Model:       model,
-			Message:     props.Message,
-			Token:       props.MaxTokens,
-			Temperature: props.Temperature,
-			TopK:        props.TopK,
-			Tools:       props.Tools,
-			Buffer:      props.Buffer,
-		}, hook)
-
-	case globals.ChatGLMChannelType:
-		return zhipuai.NewChatInstanceFromConfig(conf).CreateStreamChatRequest(&zhipuai.ChatProps{
-			Model:       model,
-			Message:     props.Message,
-			Temperature: props.Temperature,
-			TopP:        props.TopP,
-		}, hook)
-
-	case globals.QwenChannelType:
-		return dashscope.NewChatInstanceFromConfig(conf).CreateStreamChatRequest(&dashscope.ChatProps{
-			Model:             model,
-			Message:           props.Message,
-			Token:             props.MaxTokens,
-			Temperature:       props.Temperature,
-			TopP:              props.TopP,
-			TopK:              props.TopK,
-			RepetitionPenalty: props.RepetitionPenalty,
-		}, hook)
-
-	case globals.HunyuanChannelType:
-		return hunyuan.NewChatInstanceFromConfig(conf).CreateStreamChatRequest(&hunyuan.ChatProps{
-			Model:       model,
-			Message:     props.Message,
-			Temperature: props.Temperature,
-			TopP:        props.TopP,
-		}, hook)
-
-	case globals.BaichuanChannelType:
-		return baichuan.NewChatInstanceFromConfig(conf).CreateStreamChatRequest(&baichuan.ChatProps{
-			Model:       model,
-			Message:     props.Message,
-			TopP:        props.TopP,
-			TopK:        props.TopK,
-			Temperature: props.Temperature,
-		}, hook)
-
-	case globals.SkylarkChannelType:
-		return skylark.NewChatInstanceFromConfig(conf).CreateStreamChatRequest(&skylark.ChatProps{
-			Model:            model,
-			Message:          props.Message,
-			Token:            props.MaxTokens,
-			TopP:             props.TopP,
-			TopK:             props.TopK,
-			Temperature:      props.Temperature,
-			FrequencyPenalty: props.FrequencyPenalty,
-			PresencePenalty:  props.PresencePenalty,
-			RepeatPenalty:    props.RepetitionPenalty,
-			Tools:            props.Tools,
-		}, hook)
-
-	case globals.ZhinaoChannelType:
-		return zhinao.NewChatInstanceFromConfig(conf).CreateStreamChatRequest(&zhinao.ChatProps{
-			Model:             model,
-			Message:           props.Message,
-			Token:             props.MaxTokens,
-			TopP:              props.TopP,
-			TopK:              props.TopK,
-			Temperature:       props.Temperature,
-			RepetitionPenalty: props.RepetitionPenalty,
-		}, hook)
-
-	case globals.MidjourneyChannelType:
-		return midjourney.NewChatInstanceFromConfig(conf).CreateStreamChatRequest(&midjourney.ChatProps{
-			Model:    model,
-			Messages: props.Message,
-		}, hook)
-
-	default:
-		return fmt.Errorf("unknown channel type %s (model: %s)", conf.GetType(), props.Model)
 	}
+
+	return fmt.Errorf("unknown channel type %s (channel #%d)", conf.GetType(), conf.GetId())
 }

adapter/azure/chat.go

@@ -1,6 +1,7 @@
 package azure
 
 import (
+	adaptercommon "chat/adapter/common"
 	"chat/globals"
 	"chat/utils"
 	"errors"
@@ -8,20 +9,7 @@ import (
 	"strings"
 )
 
-type ChatProps struct {
+func (c *ChatInstance) GetChatEndpoint(props *adaptercommon.ChatProps) string {
-	Model            string
-	Message          []globals.Message
-	Token            *int
-	PresencePenalty  *float32
-	FrequencyPenalty *float32
-	Temperature      *float32
-	TopP             *float32
-	Tools            *globals.FunctionTools
-	ToolChoice       *interface{}
-	Buffer           utils.Buffer
-}
-
-func (c *ChatInstance) GetChatEndpoint(props *ChatProps) string {
 	model := strings.ReplaceAll(props.Model, ".", "")
 	if props.Model == globals.GPT3TurboInstruct {
 		return fmt.Sprintf("%s/openai/deployments/%s/completions?api-version=%s", c.GetResource(), model, c.GetEndpoint())
@@ -37,7 +25,7 @@ func (c *ChatInstance) GetCompletionPrompt(messages []globals.Message) string {
 	return result
 }
 
-func (c *ChatInstance) GetLatestPrompt(props *ChatProps) string {
+func (c *ChatInstance) GetLatestPrompt(props *adaptercommon.ChatProps) string {
 	if len(props.Message) == 0 {
 		return ""
 	}
@@ -45,19 +33,19 @@ func (c *ChatInstance) GetLatestPrompt(props *ChatProps) string {
 	return props.Message[len(props.Message)-1].Content
 }
 
-func (c *ChatInstance) GetChatBody(props *ChatProps, stream bool) interface{} {
+func (c *ChatInstance) GetChatBody(props *adaptercommon.ChatProps, stream bool) interface{} {
 	if props.Model == globals.GPT3TurboInstruct {
 		// for completions
 		return CompletionRequest{
 			Prompt:   c.GetCompletionPrompt(props.Message),
-			MaxToken: props.Token,
+			MaxToken: props.MaxTokens,
 			Stream:   stream,
 		}
 	}
 
 	return ChatRequest{
 		Messages:         formatMessages(props),
-		MaxToken:         props.Token,
+		MaxToken:         props.MaxTokens,
 		Stream:           stream,
 		PresencePenalty:  props.PresencePenalty,
 		FrequencyPenalty: props.FrequencyPenalty,
@@ -68,8 +56,8 @@ func (c *ChatInstance) GetChatBody(props *ChatProps, stream bool) interface{} {
 	}
 }
 
-// CreateChatRequest is the native http request body for chatgpt
+// CreateChatRequest is the native http request body for openai
-func (c *ChatInstance) CreateChatRequest(props *ChatProps) (string, error) {
+func (c *ChatInstance) CreateChatRequest(props *adaptercommon.ChatProps) (string, error) {
 	if globals.IsOpenAIDalleModel(props.Model) {
 		return c.CreateImage(props)
 	}
@@ -81,20 +69,20 @@ func (c *ChatInstance) CreateChatRequest(props *ChatProps) (string, error) {
 	)
 
 	if err != nil || res == nil {
-		return "", fmt.Errorf("chatgpt error: %s", err.Error())
+		return "", fmt.Errorf("openai error: %s", err.Error())
 	}
 
 	data := utils.MapToStruct[ChatResponse](res)
 	if data == nil {
-		return "", fmt.Errorf("chatgpt error: cannot parse response")
+		return "", fmt.Errorf("openai error: cannot parse response")
 	} else if data.Error.Message != "" {
-		return "", fmt.Errorf("chatgpt error: %s", data.Error.Message)
+		return "", fmt.Errorf("openai error: %s", data.Error.Message)
 	}
 	return data.Choices[0].Message.Content, nil
 }
 
-// CreateStreamChatRequest is the stream response body for chatgpt
+// CreateStreamChatRequest is the stream response body for openai
-func (c *ChatInstance) CreateStreamChatRequest(props *ChatProps, callback globals.Hook) error {
+func (c *ChatInstance) CreateStreamChatRequest(props *adaptercommon.ChatProps, callback globals.Hook) error {
 	if globals.IsOpenAIDalleModel(props.Model) {
 		if url, err := c.CreateImage(props); err != nil {
 			return err

adapter/azure/image.go

@@ -1,6 +1,7 @@
 package azure
 
 import (
+	adaptercommon "chat/adapter/common"
 	"chat/globals"
 	"chat/utils"
 	"fmt"
@@ -32,21 +33,21 @@ func (c *ChatInstance) CreateImageRequest(props ImageProps) (string, error) {
 			N: 1,
 		})
 	if err != nil || res == nil {
-		return "", fmt.Errorf("chatgpt error: %s", err.Error())
+		return "", fmt.Errorf("openai error: %s", err.Error())
 	}
 
 	data := utils.MapToStruct[ImageResponse](res)
 	if data == nil {
-		return "", fmt.Errorf("chatgpt error: cannot parse response")
+		return "", fmt.Errorf("openai error: cannot parse response")
 	} else if data.Error.Message != "" {
-		return "", fmt.Errorf("chatgpt error: %s", data.Error.Message)
+		return "", fmt.Errorf("openai error: %s", data.Error.Message)
 	}
 
 	return data.Data[0].Url, nil
 }
 
 // CreateImage will create a dalle image from prompt, return markdown of image
-func (c *ChatInstance) CreateImage(props *ChatProps) (string, error) {
+func (c *ChatInstance) CreateImage(props *adaptercommon.ChatProps) (string, error) {
 	url, err := c.CreateImageRequest(ImageProps{
 		Model:  props.Model,
 		Prompt: c.GetLatestPrompt(props),

adapter/azure/processor.go

@@ -1,6 +1,7 @@
 package azure
 
 import (
+	adaptercommon "chat/adapter/common"
 	"chat/globals"
 	"chat/utils"
 	"errors"
@@ -8,7 +9,7 @@ import (
 	"regexp"
 )
 
-func formatMessages(props *ChatProps) interface{} {
+func formatMessages(props *adaptercommon.ChatProps) interface{} {
 	if globals.IsVisionModel(props.Model) {
 		return utils.Each[globals.Message, Message](props.Message, func(message globals.Message) Message {
 			if message.Role == globals.User {
@@ -120,7 +121,7 @@ func (c *ChatInstance) ProcessLine(data string, isCompletionType bool) (*globals
 			}, nil
 		}
 
-		globals.Warn(fmt.Sprintf("chatgpt error: cannot parse completion response: %s", data))
+		globals.Warn(fmt.Sprintf("openai error: cannot parse completion response: %s", data))
 		return &globals.Chunk{Content: ""}, errors.New("parser error: cannot parse completion response")
 	}
 
@@ -129,9 +130,9 @@ func (c *ChatInstance) ProcessLine(data string, isCompletionType bool) (*globals
 	}
 
 	if form := processChatErrorResponse(data); form != nil {
-		return &globals.Chunk{Content: ""}, errors.New(fmt.Sprintf("chatgpt error: %s (type: %s)", form.Error.Message, form.Error.Type))
+		return &globals.Chunk{Content: ""}, errors.New(fmt.Sprintf("openai error: %s (type: %s)", form.Error.Message, form.Error.Type))
 	}
 
-	globals.Warn(fmt.Sprintf("chatgpt error: cannot parse chat completion response: %s", data))
+	globals.Warn(fmt.Sprintf("openai error: cannot parse chat completion response: %s", data))
 	return &globals.Chunk{Content: ""}, errors.New("parser error: cannot parse chat completion response")
 }

adapter/azure/struct.go

@@ -1,6 +1,7 @@
 package azure
 
 import (
+	factory "chat/adapter/common"
 	"chat/globals"
 )
 
@@ -42,7 +43,7 @@ func NewChatInstance(endpoint, apiKey string, resource string) *ChatInstance {
 	}
 }
 
-func NewChatInstanceFromConfig(conf globals.ChannelConfig) *ChatInstance {
+func NewChatInstanceFromConfig(conf globals.ChannelConfig) factory.Factory {
 	param := conf.SplitRandomSecret(2)
 	return NewChatInstance(
 		conf.GetEndpoint(),

adapter/azure/types.go

@@ -24,7 +24,7 @@ type Message struct {
 	ToolCalls    *globals.ToolCalls    `json:"tool_calls,omitempty"`    // only `assistant` role
 }
 
-// ChatRequest is the request body for chatgpt
+// ChatRequest is the request body for openai
 type ChatRequest struct {
 	Model            string                 `json:"model"`
 	Messages         interface{}            `json:"messages"`
@@ -38,7 +38,7 @@ type ChatRequest struct {
 	ToolChoice       *interface{}           `json:"tool_choice,omitempty"` // string or object
 }
 
-// CompletionRequest is the request body for chatgpt completion
+// CompletionRequest is the request body for openai completion
 type CompletionRequest struct {
 	Model    string `json:"model"`
 	Prompt   string `json:"prompt"`
@@ -46,7 +46,7 @@ type CompletionRequest struct {
 	Stream   bool   `json:"stream"`
 }
 
-// ChatResponse is the native http request body for chatgpt
+// ChatResponse is the native http request body for openai
 type ChatResponse struct {
 	ID      string `json:"id"`
 	Object  string `json:"object"`
@@ -62,7 +62,7 @@ type ChatResponse struct {
 	} `json:"error"`
 }
 
-// ChatStreamResponse is the stream response body for chatgpt
+// ChatStreamResponse is the stream response body for openai
 type ChatStreamResponse struct {
 	ID      string `json:"id"`
 	Object  string `json:"object"`
@@ -75,7 +75,7 @@ type ChatStreamResponse struct {
 	} `json:"choices"`
 }
 
-// CompletionResponse is the native http request body / stream response body for chatgpt completion
+// CompletionResponse is the native http request body / stream response body for openai completion
 type CompletionResponse struct {
 	ID      string `json:"id"`
 	Object  string `json:"object"`
@@ -96,7 +96,7 @@ type ChatStreamErrorResponse struct {
 
 type ImageSize string
 
-// ImageRequest is the request body for chatgpt dalle image generation
+// ImageRequest is the request body for openai dalle image generation
 type ImageRequest struct {
 	Model  string    `json:"model"`
 	Prompt string    `json:"prompt"`

adapter/baichuan/chat.go

@@ -1,20 +1,13 @@
 package baichuan
 
 import (
+	adaptercommon "chat/adapter/common"
 	"chat/globals"
 	"chat/utils"
 	"errors"
 	"fmt"
 )
 
-type ChatProps struct {
-	Model       string
-	Message     []globals.Message
-	TopP        *float32
-	TopK        *int
-	Temperature *float32
-}
-
 func (c *ChatInstance) GetChatEndpoint() string {
 	return fmt.Sprintf("%s/v1/chat/completions", c.GetEndpoint())
 }
@@ -38,7 +31,7 @@ func (c *ChatInstance) GetMessages(messages []globals.Message) []globals.Message
 	return messages
 }
 
-func (c *ChatInstance) GetChatBody(props *ChatProps, stream bool) ChatRequest {
+func (c *ChatInstance) GetChatBody(props *adaptercommon.ChatProps, stream bool) ChatRequest {
 	return ChatRequest{
 		Model:       c.GetModel(props.Model),
 		Messages:    c.GetMessages(props.Message),
@@ -50,7 +43,7 @@ func (c *ChatInstance) GetChatBody(props *ChatProps, stream bool) ChatRequest {
 }
 
 // CreateChatRequest is the native http request body for baichuan
-func (c *ChatInstance) CreateChatRequest(props *ChatProps) (string, error) {
+func (c *ChatInstance) CreateChatRequest(props *adaptercommon.ChatProps) (string, error) {
 	res, err := utils.Post(
 		c.GetChatEndpoint(),
 		c.GetHeader(),
@@ -71,7 +64,7 @@ func (c *ChatInstance) CreateChatRequest(props *ChatProps) (string, error) {
 }
 
 // CreateStreamChatRequest is the stream response body for baichuan
-func (c *ChatInstance) CreateStreamChatRequest(props *ChatProps, callback globals.Hook) error {
+func (c *ChatInstance) CreateStreamChatRequest(props *adaptercommon.ChatProps, callback globals.Hook) error {
 	err := utils.EventScanner(&utils.EventScannerProps{
 		Method:  "POST",
 		Uri:     c.GetChatEndpoint(),

adapter/baichuan/struct.go

@@ -1,6 +1,7 @@
 package baichuan
 
 import (
+	factory "chat/adapter/common"
 	"chat/globals"
 	"fmt"
 )
@@ -32,7 +33,7 @@ func NewChatInstance(endpoint, apiKey string) *ChatInstance {
 	}
 }
 
-func NewChatInstanceFromConfig(conf globals.ChannelConfig) *ChatInstance {
+func NewChatInstanceFromConfig(conf globals.ChannelConfig) factory.Factory {
 	return NewChatInstance(
 		conf.GetEndpoint(),
 		conf.GetRandomSecret(),

adapter/bing/chat.go

@@ -1,18 +1,14 @@
 package bing
 
 import (
+	adaptercommon "chat/adapter/common"
 	"chat/globals"
 	"chat/utils"
 	"fmt"
 	"strings"
 )
 
-type ChatProps struct {
+func (c *ChatInstance) CreateStreamChatRequest(props *adaptercommon.ChatProps, hook globals.Hook) error {
-	Message []globals.Message
-	Model   string
-}
-
-func (c *ChatInstance) CreateStreamChatRequest(props *ChatProps, hook globals.Hook) error {
 	var conn *utils.WebSocket
 	if conn = utils.NewWebsocketClient(c.GetEndpoint()); conn == nil {
 		return fmt.Errorf("bing error: websocket connection failed")

adapter/bing/struct.go

@@ -1,6 +1,7 @@
 package bing
 
 import (
+	factory "chat/adapter/common"
 	"chat/globals"
 	"fmt"
 )
@@ -21,7 +22,7 @@ func NewChatInstance(endpoint, secret string) *ChatInstance {
 	}
 }
 
-func NewChatInstanceFromConfig(conf globals.ChannelConfig) *ChatInstance {
+func NewChatInstanceFromConfig(conf globals.ChannelConfig) factory.Factory {
 	return NewChatInstance(
 		conf.GetEndpoint(),
 		conf.GetRandomSecret(),

adapter/chatgpt/test.go

@@ -1,47 +0,0 @@
-package chatgpt
-
-import (
-	"chat/globals"
-	"chat/utils"
-	"fmt"
-	"github.com/spf13/viper"
-	"strings"
-)
-
-func (c *ChatInstance) Test() bool {
-	result, err := c.CreateChatRequest(&ChatProps{
-		Model:   globals.GPT3Turbo,
-		Message: []globals.Message{{Role: globals.User, Content: "hi"}},
-		Token:   utils.ToPtr(1),
-	})
-	if err != nil {
-		fmt.Println(fmt.Sprintf("%s: test failed (%s)", c.GetApiKey(), err.Error()))
-	}
-
-	return err == nil && len(result) > 0
-}
-
-func FilterKeys(v string) []string {
-	endpoint := viper.GetString(fmt.Sprintf("openai.%s.endpoint", v))
-	keys := strings.Split(viper.GetString(fmt.Sprintf("openai.%s.apikey", v)), "|")
-
-	return FilterKeysNative(endpoint, keys)
-}
-
-func FilterKeysNative(endpoint string, keys []string) []string {
-	stack := make(chan string, len(keys))
-	for _, key := range keys {
-		go func(key string) {
-			instance := NewChatInstance(endpoint, key)
-			stack <- utils.Multi[string](instance.Test(), key, "")
-		}(key)
-	}
-
-	var result []string
-	for i := 0; i < len(keys); i++ {
-		if res := <-stack; res != "" {
-			result = append(result, res)
-		}
-	}
-	return result
-}

adapter/claude/chat.go

@@ -1,6 +1,7 @@
 package claude
 
 import (
+	adaptercommon "chat/adapter/common"
 	"chat/globals"
 	"chat/utils"
 	"fmt"
@@ -9,15 +10,6 @@ import (
 
 const defaultTokens = 2500
 
-type ChatProps struct {
-	Model       string
-	Message     []globals.Message
-	Token       *int
-	Temperature *float32
-	TopP        *float32
-	TopK        *int
-}
-
 func (c *ChatInstance) GetChatEndpoint() string {
 	return fmt.Sprintf("%s/v1/complete", c.GetEndpoint())
 }
@@ -52,15 +44,15 @@ func (c *ChatInstance) ConvertMessage(message []globals.Message) string {
 	return fmt.Sprintf("%s\n\nAssistant:", result)
 }
 
-func (c *ChatInstance) GetTokens(props *ChatProps) int {
+func (c *ChatInstance) GetTokens(props *adaptercommon.ChatProps) int {
-	if props.Token == nil || *props.Token <= 0 {
+	if props.MaxTokens == nil || *props.MaxTokens <= 0 {
 		return defaultTokens
 	}
 
-	return *props.Token
+	return *props.MaxTokens
 }
 
-func (c *ChatInstance) GetChatBody(props *ChatProps, stream bool) *ChatBody {
+func (c *ChatInstance) GetChatBody(props *adaptercommon.ChatProps, stream bool) *ChatBody {
 	return &ChatBody{
 		Prompt:            c.ConvertMessage(props.Message),
 		MaxTokensToSample: c.GetTokens(props),
@@ -73,7 +65,7 @@ func (c *ChatInstance) GetChatBody(props *ChatProps, stream bool) *ChatBody {
 }
 
 // CreateChatRequest is the request for anthropic claude
-func (c *ChatInstance) CreateChatRequest(props *ChatProps) (string, error) {
+func (c *ChatInstance) CreateChatRequest(props *adaptercommon.ChatProps) (string, error) {
 	data, err := utils.Post(c.GetChatEndpoint(), c.GetChatHeaders(), c.GetChatBody(props, false))
 	if err != nil {
 		return "", fmt.Errorf("claude error: %s", err.Error())
@@ -115,7 +107,7 @@ func (c *ChatInstance) ProcessLine(buf, data string) (string, error) {
 }
 
 // CreateStreamChatRequest is the stream request for anthropic claude
-func (c *ChatInstance) CreateStreamChatRequest(props *ChatProps, hook globals.Hook) error {
+func (c *ChatInstance) CreateStreamChatRequest(props *adaptercommon.ChatProps, hook globals.Hook) error {
 	buf := ""
 
 	return utils.EventSource(
@@ -124,7 +116,6 @@ func (c *ChatInstance) CreateStreamChatRequest(props *ChatProps, hook globals.Ho
 		c.GetChatHeaders(),
 		c.GetChatBody(props, true),
 		func(data string) error {
-
 			if resp, err := c.ProcessLine(buf, data); err == nil && len(resp) > 0 {
 				buf = ""
 				if err := hook(&globals.Chunk{Content: resp}); err != nil {

adapter/claude/struct.go

@@ -1,6 +1,7 @@
 package claude
 
 import (
+	factory "chat/adapter/common"
 	"chat/globals"
 )
 
@@ -16,7 +17,7 @@ func NewChatInstance(endpoint, apiKey string) *ChatInstance {
 	}
 }
 
-func NewChatInstanceFromConfig(conf globals.ChannelConfig) *ChatInstance {
+func NewChatInstanceFromConfig(conf globals.ChannelConfig) factory.Factory {
 	return NewChatInstance(
 		conf.GetEndpoint(),
 		conf.GetRandomSecret(),

adapter/common/interface.go

@@ -0,0 +1,11 @@
+package adaptercommon
+
+import (
+	"chat/globals"
+)
+
+type Factory interface {
+	CreateStreamChatRequest(props *ChatProps, hook globals.Hook) error
+}
+
+type FactoryCreator func(globals.ChannelConfig) Factory

adapter/common/types.go

@@ -0,0 +1,31 @@
+package adaptercommon
+
+import (
+	"chat/globals"
+	"chat/utils"
+)
+
+type RequestProps struct {
+	MaxRetries *int
+	Current    int
+	Group      string
+}
+
+type ChatProps struct {
+	RequestProps
+
+	Model         string
+	OriginalModel string
+
+	Message           []globals.Message
+	MaxTokens         *int
+	PresencePenalty   *float32
+	FrequencyPenalty  *float32
+	RepetitionPenalty *float32
+	Temperature       *float32
+	TopP              *float32
+	TopK              *int
+	Tools             *globals.FunctionTools
+	ToolChoice        *interface{}
+	Buffer            utils.Buffer
+}

adapter/dashscope/chat.go

@@ -1,6 +1,7 @@
 package dashscope
 
 import (
+	adaptercommon "chat/adapter/common"
 	"chat/globals"
 	"chat/utils"
 	"fmt"
@@ -9,16 +10,6 @@ import (
 
 const defaultMaxTokens = 1500
 
-type ChatProps struct {
-	Model             string
-	Token             *int
-	Temperature       *float32
-	TopP              *float32
-	TopK              *int
-	RepetitionPenalty *float32
-	Message           []globals.Message
-}
-
 func (c *ChatInstance) GetHeader() map[string]string {
 	return map[string]string{
 		"Content-Type":    "application/json",
@@ -43,16 +34,16 @@ func (c *ChatInstance) FormatMessages(message []globals.Message) []Message {
 	return messages
 }
 
-func (c *ChatInstance) GetMaxTokens(props *ChatProps) int {
+func (c *ChatInstance) GetMaxTokens(props *adaptercommon.ChatProps) int {
 	// dashscope has a restriction of 1500 tokens in completion
-	if props.Token == nil || *props.Token <= 0 || *props.Token > 1500 {
+	if props.MaxTokens == nil || *props.MaxTokens <= 0 || *props.MaxTokens > 1500 {
 		return defaultMaxTokens
 	}
 
-	return *props.Token
+	return *props.MaxTokens
 }
 
-func (c *ChatInstance) GetTopP(props *ChatProps) *float32 {
+func (c *ChatInstance) GetTopP(props *adaptercommon.ChatProps) *float32 {
 	// range of top_p should be (0.0, 1.0)
 	if props.TopP == nil {
 		return nil
@@ -67,7 +58,7 @@ func (c *ChatInstance) GetTopP(props *ChatProps) *float32 {
 	return props.TopP
 }
 
-func (c *ChatInstance) GetRepeatPenalty(props *ChatProps) *float32 {
+func (c *ChatInstance) GetRepeatPenalty(props *adaptercommon.ChatProps) *float32 {
 	// range of repetition_penalty should greater than 0.0
 	if props.RepetitionPenalty == nil {
 		return nil
@@ -80,7 +71,7 @@ func (c *ChatInstance) GetRepeatPenalty(props *ChatProps) *float32 {
 	return props.RepetitionPenalty
 }
 
-func (c *ChatInstance) GetChatBody(props *ChatProps) ChatRequest {
+func (c *ChatInstance) GetChatBody(props *adaptercommon.ChatProps) ChatRequest {
 	return ChatRequest{
 		Model: strings.TrimSuffix(props.Model, "-net"),
 		Input: ChatInput{
@@ -102,7 +93,7 @@ func (c *ChatInstance) GetChatEndpoint() string {
 	return fmt.Sprintf("%s/api/v1/services/aigc/text-generation/generation", c.Endpoint)
 }
 
-func (c *ChatInstance) CreateStreamChatRequest(props *ChatProps, callback globals.Hook) error {
+func (c *ChatInstance) CreateStreamChatRequest(props *adaptercommon.ChatProps, callback globals.Hook) error {
 	return utils.EventSource(
 		"POST",
 		c.GetChatEndpoint(),

adapter/dashscope/struct.go

@@ -1,6 +1,7 @@
 package dashscope
 
 import (
+	factory "chat/adapter/common"
 	"chat/globals"
 )
 
@@ -24,7 +25,7 @@ func NewChatInstance(endpoint string, apiKey string) *ChatInstance {
 	}
 }
 
-func NewChatInstanceFromConfig(conf globals.ChannelConfig) *ChatInstance {
+func NewChatInstanceFromConfig(conf globals.ChannelConfig) factory.Factory {
 	return NewChatInstance(
 		conf.GetEndpoint(),
 		conf.GetRandomSecret(),

adapter/hunyuan/chat.go

@@ -1,18 +1,12 @@
 package hunyuan
 
 import (
+	adaptercommon "chat/adapter/common"
 	"chat/globals"
 	"context"
 	"fmt"
 )
 
-type ChatProps struct {
-	Model       string
-	Message     []globals.Message
-	Temperature *float32
-	TopP        *float32
-}
-
 func (c *ChatInstance) FormatMessages(messages []globals.Message) []globals.Message {
 	var result []globals.Message
 	for _, message := range messages {
@@ -36,7 +30,7 @@ func (c *ChatInstance) FormatMessages(messages []globals.Message) []globals.Mess
 	return result
 }
 
-func (c *ChatInstance) CreateStreamChatRequest(props *ChatProps, callback globals.Hook) error {
+func (c *ChatInstance) CreateStreamChatRequest(props *adaptercommon.ChatProps, callback globals.Hook) error {
 	credential := NewCredential(c.GetSecretId(), c.GetSecretKey())
 	client := NewInstance(c.GetAppId(), c.GetEndpoint(), credential)
 	channel, err := client.Chat(context.Background(), NewRequest(Stream, c.FormatMessages(props.Message), props.Temperature, props.TopP))

adapter/hunyuan/struct.go

@@ -1,6 +1,7 @@
 package hunyuan
 
 import (
+	factory "chat/adapter/common"
 	"chat/globals"
 	"chat/utils"
 )
@@ -37,7 +38,7 @@ func NewChatInstance(endpoint, appId, secretId, secretKey string) *ChatInstance
 	}
 }
 
-func NewChatInstanceFromConfig(conf globals.ChannelConfig) *ChatInstance {
+func NewChatInstanceFromConfig(conf globals.ChannelConfig) factory.Factory {
 	params := conf.SplitRandomSecret(3)
 	return NewChatInstance(
 		conf.GetEndpoint(),

adapter/midjourney/chat.go

@@ -1,6 +1,7 @@
 package midjourney
 
 import (
+	adaptercommon "chat/adapter/common"
 	"chat/globals"
 	"chat/utils"
 	"fmt"
@@ -66,16 +67,16 @@ func (c *ChatInstance) GetCleanPrompt(model string, prompt string) string {
 	return target
 }
 
-func (c *ChatInstance) GetPrompt(props *ChatProps) string {
+func (c *ChatInstance) GetPrompt(props *adaptercommon.ChatProps) string {
-	if len(props.Messages) == 0 {
+	if len(props.Message) == 0 {
 		return ""
 	}
 
-	content := props.Messages[len(props.Messages)-1].Content
+	content := props.Message[len(props.Message)-1].Content
 	return c.GetCleanPrompt(props.Model, content)
 }
 
-func (c *ChatInstance) CreateStreamChatRequest(props *ChatProps, callback globals.Hook) error {
+func (c *ChatInstance) CreateStreamChatRequest(props *adaptercommon.ChatProps, callback globals.Hook) error {
 	// partial response like:
 	// ```progress
 	// 0
@@ -95,6 +96,11 @@ func (c *ChatInstance) CreateStreamChatRequest(props *ChatProps, callback global
 	var begin bool
 
 	form, err := c.CreateStreamTask(action, prompt, func(form *StorageForm, progress int) error {
+		if progress == -1 {
+			// ping event
+			return callback(&globals.Chunk{Content: ""})
+		}
+
 		if progress == 0 {
 			begin = true
 			if err := callback(&globals.Chunk{Content: "```progress\n"}); err != nil {

adapter/midjourney/handler.go

@@ -96,6 +96,11 @@ func (c *ChatInstance) CreateStreamTask(action string, prompt string, hook func(
 		utils.Sleep(50)
 		form := getStorage(task)
 		if form == nil {
+			// hook for ping
+			if err := hook(nil, -1); err != nil {
+				return nil, err
+			}
+
 			continue
 		}
 

adapter/midjourney/struct.go

@@ -1,6 +1,7 @@
 package midjourney
 
 import (
+	factory "chat/adapter/common"
 	"chat/globals"
 	"fmt"
 )
@@ -47,7 +48,7 @@ func NewChatInstance(endpoint, apiSecret, whiteList string) *ChatInstance {
 	}
 }
 
-func NewChatInstanceFromConfig(conf globals.ChannelConfig) *ChatInstance {
+func NewChatInstanceFromConfig(conf globals.ChannelConfig) factory.Factory {
 	params := conf.SplitRandomSecret(2)
 
 	return NewChatInstance(

adapter/openai/chat.go

@@ -1,6 +1,7 @@
-package chatgpt
+package openai
 
 import (
+	adaptercommon "chat/adapter/common"
 	"chat/globals"
 	"chat/utils"
 	"errors"
@@ -8,20 +9,7 @@ import (
 	"regexp"
 )
 
-type ChatProps struct {
+func (c *ChatInstance) GetChatEndpoint(props *adaptercommon.ChatProps) string {
-	Model            string
-	Message          []globals.Message
-	Token            *int
-	PresencePenalty  *float32
-	FrequencyPenalty *float32
-	Temperature      *float32
-	TopP             *float32
-	Tools            *globals.FunctionTools
-	ToolChoice       *interface{}
-	Buffer           utils.Buffer
-}
-
-func (c *ChatInstance) GetChatEndpoint(props *ChatProps) string {
 	if props.Model == globals.GPT3TurboInstruct {
 		return fmt.Sprintf("%s/v1/completions", c.GetEndpoint())
 	}
@@ -36,7 +24,7 @@ func (c *ChatInstance) GetCompletionPrompt(messages []globals.Message) string {
 	return result
 }
 
-func (c *ChatInstance) GetLatestPrompt(props *ChatProps) string {
+func (c *ChatInstance) GetLatestPrompt(props *adaptercommon.ChatProps) string {
 	if len(props.Message) == 0 {
 		return ""
 	}
@@ -44,13 +32,13 @@ func (c *ChatInstance) GetLatestPrompt(props *ChatProps) string {
 	return props.Message[len(props.Message)-1].Content
 }
 
-func (c *ChatInstance) GetChatBody(props *ChatProps, stream bool) interface{} {
+func (c *ChatInstance) GetChatBody(props *adaptercommon.ChatProps, stream bool) interface{} {
 	if props.Model == globals.GPT3TurboInstruct {
 		// for completions
 		return CompletionRequest{
 			Model:    props.Model,
 			Prompt:   c.GetCompletionPrompt(props.Message),
-			MaxToken: props.Token,
+			MaxToken: props.MaxTokens,
 			Stream:   stream,
 		}
 	}
@@ -60,7 +48,7 @@ func (c *ChatInstance) GetChatBody(props *ChatProps, stream bool) interface{} {
 	return ChatRequest{
 		Model:            props.Model,
 		Messages:         messages,
-		MaxToken:         props.Token,
+		MaxToken:         props.MaxTokens,
 		Stream:           stream,
 		PresencePenalty:  props.PresencePenalty,
 		FrequencyPenalty: props.FrequencyPenalty,
@@ -71,8 +59,8 @@ func (c *ChatInstance) GetChatBody(props *ChatProps, stream bool) interface{} {
 	}
 }
 
-// CreateChatRequest is the native http request body for chatgpt
+// CreateChatRequest is the native http request body for openai
-func (c *ChatInstance) CreateChatRequest(props *ChatProps) (string, error) {
+func (c *ChatInstance) CreateChatRequest(props *adaptercommon.ChatProps) (string, error) {
 	if globals.IsOpenAIDalleModel(props.Model) {
 		return c.CreateImage(props)
 	}
@@ -84,14 +72,14 @@ func (c *ChatInstance) CreateChatRequest(props *ChatProps) (string, error) {
 	)
 
 	if err != nil || res == nil {
-		return "", fmt.Errorf("chatgpt error: %s", err.Error())
+		return "", fmt.Errorf("openai error: %s", err.Error())
 	}
 
 	data := utils.MapToStruct[ChatResponse](res)
 	if data == nil {
-		return "", fmt.Errorf("chatgpt error: cannot parse response")
+		return "", fmt.Errorf("openai error: cannot parse response")
 	} else if data.Error.Message != "" {
-		return "", fmt.Errorf("chatgpt error: %s", data.Error.Message)
+		return "", fmt.Errorf("openai error: %s", data.Error.Message)
 	}
 	return data.Choices[0].Message.Content, nil
 }
@@ -103,8 +91,8 @@ func hideRequestId(message string) string {
 	return exp.ReplaceAllString(message, "")
 }
 
-// CreateStreamChatRequest is the stream response body for chatgpt
+// CreateStreamChatRequest is the stream response body for openai
-func (c *ChatInstance) CreateStreamChatRequest(props *ChatProps, callback globals.Hook) error {
+func (c *ChatInstance) CreateStreamChatRequest(props *adaptercommon.ChatProps, callback globals.Hook) error {
 	if globals.IsOpenAIDalleModel(props.Model) {
 		if url, err := c.CreateImage(props); err != nil {
 			return err

adapter/openai/image.go

@@ -1,6 +1,7 @@
-package chatgpt
+package openai
 
 import (
+	adaptercommon "chat/adapter/common"
 	"chat/globals"
 	"chat/utils"
 	"fmt"
@@ -37,7 +38,7 @@ func (c *ChatInstance) CreateImageRequest(props ImageProps) (string, error) {
 
 	data := utils.MapToStruct[ImageResponse](res)
 	if data == nil {
-		return "", fmt.Errorf("chatgpt error: cannot parse response")
+		return "", fmt.Errorf("openai error: cannot parse response")
 	} else if data.Error.Message != "" {
 		return "", fmt.Errorf(data.Error.Message)
 	}
@@ -46,7 +47,7 @@ func (c *ChatInstance) CreateImageRequest(props ImageProps) (string, error) {
 }
 
 // CreateImage will create a dalle image from prompt, return markdown of image
-func (c *ChatInstance) CreateImage(props *ChatProps) (string, error) {
+func (c *ChatInstance) CreateImage(props *adaptercommon.ChatProps) (string, error) {
 	url, err := c.CreateImageRequest(ImageProps{
 		Model:  props.Model,
 		Prompt: c.GetLatestPrompt(props),

adapter/openai/processor.go

@@ -1,6 +1,7 @@
-package chatgpt
+package openai
 
 import (
+	adaptercommon "chat/adapter/common"
 	"chat/globals"
 	"chat/utils"
 	"errors"
@@ -8,7 +9,7 @@ import (
 	"regexp"
 )
 
-func formatMessages(props *ChatProps) interface{} {
+func formatMessages(props *adaptercommon.ChatProps) interface{} {
 	if globals.IsVisionModel(props.Model) {
 		return utils.Each[globals.Message, Message](props.Message, func(message globals.Message) Message {
 			if message.Role == globals.User {
@@ -118,7 +119,7 @@ func (c *ChatInstance) ProcessLine(data string, isCompletionType bool) (*globals
 			}, nil
 		}
 
-		globals.Warn(fmt.Sprintf("chatgpt error: cannot parse completion response: %s", data))
+		globals.Warn(fmt.Sprintf("openai error: cannot parse completion response: %s", data))
 		return &globals.Chunk{Content: ""}, errors.New("parser error: cannot parse completion response")
 	}
 
@@ -127,9 +128,9 @@ func (c *ChatInstance) ProcessLine(data string, isCompletionType bool) (*globals
 	}
 
 	if form := processChatErrorResponse(data); form != nil {
-		return &globals.Chunk{Content: ""}, errors.New(fmt.Sprintf("chatgpt error: %s (type: %s)", form.Error.Message, form.Error.Type))
+		return &globals.Chunk{Content: ""}, errors.New(fmt.Sprintf("openai error: %s (type: %s)", form.Error.Message, form.Error.Type))
 	}
 
-	globals.Warn(fmt.Sprintf("chatgpt error: cannot parse chat completion response: %s", data))
+	globals.Warn(fmt.Sprintf("openai error: cannot parse chat completion response: %s", data))
 	return &globals.Chunk{Content: ""}, errors.New("parser error: cannot parse chat completion response")
 }

adapter/openai/struct.go

@@ -1,6 +1,7 @@
-package chatgpt
+package openai
 
 import (
+	factory "chat/adapter/common"
 	"chat/globals"
 	"fmt"
 )
@@ -37,7 +38,7 @@ func NewChatInstance(endpoint, apiKey string) *ChatInstance {
 	}
 }
 
-func NewChatInstanceFromConfig(conf globals.ChannelConfig) *ChatInstance {
+func NewChatInstanceFromConfig(conf globals.ChannelConfig) factory.Factory {
 	return NewChatInstance(
 		conf.GetEndpoint(),
 		conf.GetRandomSecret(),

adapter/openai/types.go

@@ -1,4 +1,4 @@
-package chatgpt
+package openai
 
 import "chat/globals"
 
@@ -24,7 +24,7 @@ type Message struct {
 	ToolCalls    *globals.ToolCalls    `json:"tool_calls,omitempty"`    // only `assistant` role
 }
 
-// ChatRequest is the request body for chatgpt
+// ChatRequest is the request body for openai
 type ChatRequest struct {
 	Model            string                 `json:"model"`
 	Messages         interface{}            `json:"messages"`
@@ -38,7 +38,7 @@ type ChatRequest struct {
 	ToolChoice       *interface{}           `json:"tool_choice,omitempty"` // string or object
 }
 
-// CompletionRequest is the request body for chatgpt completion
+// CompletionRequest is the request body for openai completion
 type CompletionRequest struct {
 	Model    string `json:"model"`
 	Prompt   string `json:"prompt"`
@@ -46,7 +46,7 @@ type CompletionRequest struct {
 	Stream   bool   `json:"stream"`
 }
 
-// ChatResponse is the native http request body for chatgpt
+// ChatResponse is the native http request body for openai
 type ChatResponse struct {
 	ID      string `json:"id"`
 	Object  string `json:"object"`
@@ -62,7 +62,7 @@ type ChatResponse struct {
 	} `json:"error"`
 }
 
-// ChatStreamResponse is the stream response body for chatgpt
+// ChatStreamResponse is the stream response body for openai
 type ChatStreamResponse struct {
 	ID      string `json:"id"`
 	Object  string `json:"object"`
@@ -75,7 +75,7 @@ type ChatStreamResponse struct {
 	} `json:"choices"`
 }
 
-// CompletionResponse is the native http request body / stream response body for chatgpt completion
+// CompletionResponse is the native http request body / stream response body for openai completion
 type CompletionResponse struct {
 	ID      string `json:"id"`
 	Object  string `json:"object"`
@@ -96,7 +96,7 @@ type ChatStreamErrorResponse struct {
 
 type ImageSize string
 
-// ImageRequest is the request body for chatgpt dalle image generation
+// ImageRequest is the request body for openai dalle image generation
 type ImageRequest struct {
 	Model  string    `json:"model"`
 	Prompt string    `json:"prompt"`

adapter/palm2/chat.go

@@ -1,6 +1,7 @@
 package palm2
 
 import (
+	adaptercommon "chat/adapter/common"
 	"chat/globals"
 	"chat/utils"
 	"fmt"
@@ -8,15 +9,6 @@ import (
 
 var geminiMaxImages = 16
 
-type ChatProps struct {
-	Model           string
-	Message         []globals.Message
-	Temperature     *float64
-	TopP            *float64
-	TopK            *int
-	MaxOutputTokens *int
-}
-
 func (c *ChatInstance) GetChatEndpoint(model string) string {
 	if model == globals.ChatBison001 {
 		return fmt.Sprintf("%s/v1beta2/models/%s:generateMessage?key=%s", c.Endpoint, model, c.ApiKey)
@@ -51,7 +43,7 @@ func (c *ChatInstance) ConvertMessage(message []globals.Message) []PalmMessage {
 	return result
 }
 
-func (c *ChatInstance) GetPalm2ChatBody(props *ChatProps) *PalmChatBody {
+func (c *ChatInstance) GetPalm2ChatBody(props *adaptercommon.ChatProps) *PalmChatBody {
 	return &PalmChatBody{
 		Prompt: PalmPrompt{
 			Messages: c.ConvertMessage(props.Message),
@@ -59,12 +51,12 @@ func (c *ChatInstance) GetPalm2ChatBody(props *ChatProps) *PalmChatBody {
 	}
 }
 
-func (c *ChatInstance) GetGeminiChatBody(props *ChatProps) *GeminiChatBody {
+func (c *ChatInstance) GetGeminiChatBody(props *adaptercommon.ChatProps) *GeminiChatBody {
 	return &GeminiChatBody{
 		Contents: c.GetGeminiContents(props.Model, props.Message),
 		GenerationConfig: GeminiConfig{
 			Temperature:     props.Temperature,
-			MaxOutputTokens: props.MaxOutputTokens,
+			MaxOutputTokens: props.MaxTokens,
 			TopP:            props.TopP,
 			TopK:            props.TopK,
 		},
@@ -95,7 +87,7 @@ func (c *ChatInstance) GetGeminiChatResponse(data interface{}) (string, error) {
 	return "", fmt.Errorf("gemini: cannot parse response")
 }
 
-func (c *ChatInstance) CreateChatRequest(props *ChatProps) (string, error) {
+func (c *ChatInstance) CreateChatRequest(props *adaptercommon.ChatProps) (string, error) {
 	uri := c.GetChatEndpoint(props.Model)
 
 	if props.Model == globals.ChatBison001 {
@@ -122,7 +114,7 @@ func (c *ChatInstance) CreateChatRequest(props *ChatProps) (string, error) {
 
 // CreateStreamChatRequest is the mock stream request for palm2
 // tips: palm2 does not support stream request
-func (c *ChatInstance) CreateStreamChatRequest(props *ChatProps, callback globals.Hook) error {
+func (c *ChatInstance) CreateStreamChatRequest(props *adaptercommon.ChatProps, callback globals.Hook) error {
 	response, err := c.CreateChatRequest(props)
 	if err != nil {
 		return err

adapter/palm2/struct.go

@@ -1,6 +1,7 @@
 package palm2
 
 import (
+	factory "chat/adapter/common"
 	"chat/globals"
 )
 
@@ -24,7 +25,7 @@ func NewChatInstance(endpoint string, apiKey string) *ChatInstance {
 	}
 }
 
-func NewChatInstanceFromConfig(conf globals.ChannelConfig) *ChatInstance {
+func NewChatInstanceFromConfig(conf globals.ChannelConfig) factory.Factory {
 	return NewChatInstance(
 		conf.GetEndpoint(),
 		conf.GetRandomSecret(),

adapter/palm2/types.go

@@ -31,9 +31,9 @@ type GeminiChatBody struct {
 }
 
 type GeminiConfig struct {
-	Temperature     *float64 `json:"temperature,omitempty"`
+	Temperature     *float32 `json:"temperature,omitempty"`
 	MaxOutputTokens *int     `json:"maxOutputTokens,omitempty"`
-	TopP            *float64 `json:"topP,omitempty"`
+	TopP            *float32 `json:"topP,omitempty"`
 	TopK            *int     `json:"topK,omitempty"`
 }
 

adapter/request.go

@@ -1,6 +1,7 @@
 package adapter
 
 import (
+	"chat/adapter/common"
 	"chat/globals"
 	"chat/utils"
 	"fmt"
@@ -21,24 +22,24 @@ func isQPSOverLimit(model string, err error) bool {
 	}
 }
 
-func NewChatRequest(conf globals.ChannelConfig, props *ChatProps, hook globals.Hook) error {
+func NewChatRequest(conf globals.ChannelConfig, props *adaptercommon.ChatProps, hook globals.Hook) error {
 	err := createChatRequest(conf, props, hook)
 
 	retries := conf.GetRetry()
 	props.Current++
 
 	if IsAvailableError(err) {
-		if isQPSOverLimit(props.Model, err) {
+		if isQPSOverLimit(props.OriginalModel, err) {
 			// sleep for 0.5s to avoid qps limit
 
-			globals.Info(fmt.Sprintf("qps limit for %s, sleep and retry (times: %d)", props.Model, props.Current))
+			globals.Info(fmt.Sprintf("qps limit for %s, sleep and retry (times: %d)", props.OriginalModel, props.Current))
 			time.Sleep(500 * time.Millisecond)
 			return NewChatRequest(conf, props, hook)
 		}
 
 		if props.Current < retries {
 			content := strings.Replace(err.Error(), "\n", "", -1)
-			globals.Warn(fmt.Sprintf("retrying chat request for %s (attempt %d/%d, error: %s)", props.Model, props.Current+1, retries, content))
+			globals.Warn(fmt.Sprintf("retrying chat request for %s (attempt %d/%d, error: %s)", props.OriginalModel, props.Current+1, retries, content))
 			return NewChatRequest(conf, props, hook)
 		}
 	}

adapter/skylark/chat.go

@@ -1,6 +1,7 @@
 package skylark
 
 import (
+	adaptercommon "chat/adapter/common"
 	"chat/globals"
 	"chat/utils"
 	"fmt"
@@ -10,21 +11,6 @@ import (
 
 const defaultMaxTokens int64 = 1500
 
-type ChatProps struct {
-	Model   string
-	Message []globals.Message
-	Token   *int
-
-	PresencePenalty  *float32
-	FrequencyPenalty *float32
-	RepeatPenalty    *float32
-	Temperature      *float32
-	TopP             *float32
-	TopK             *int
-	Tools            *globals.FunctionTools
-	Buffer           utils.Buffer
-}
-
 func getMessages(messages []globals.Message) []*api.Message {
 	return utils.Each[globals.Message, *api.Message](messages, func(message globals.Message) *api.Message {
 		if message.Role == globals.Tool {
@@ -47,7 +33,7 @@ func (c *ChatInstance) GetMaxTokens(token *int) int64 {
 	return int64(*token)
 }
 
-func (c *ChatInstance) CreateRequest(props *ChatProps) *api.ChatReq {
+func (c *ChatInstance) CreateRequest(props *adaptercommon.ChatProps) *api.ChatReq {
 	return &api.ChatReq{
 		Model: &api.Model{
 			Name: props.Model,
@@ -59,8 +45,8 @@ func (c *ChatInstance) CreateRequest(props *ChatProps) *api.ChatReq {
 			Temperature:       utils.GetPtrVal(props.Temperature, 0.),
 			PresencePenalty:   utils.GetPtrVal(props.PresencePenalty, 0.),
 			FrequencyPenalty:  utils.GetPtrVal(props.FrequencyPenalty, 0.),
-			RepetitionPenalty: utils.GetPtrVal(props.RepeatPenalty, 0.),
+			RepetitionPenalty: utils.GetPtrVal(props.RepetitionPenalty, 0.),
-			MaxTokens:         c.GetMaxTokens(props.Token),
+			MaxTokens:         c.GetMaxTokens(props.MaxTokens),
 		},
 		Functions: getFunctions(props.Tools),
 	}
@@ -96,7 +82,7 @@ func getChoice(choice *api.ChatResp) *globals.Chunk {
 	}
 }
 
-func (c *ChatInstance) CreateStreamChatRequest(props *ChatProps, callback globals.Hook) error {
+func (c *ChatInstance) CreateStreamChatRequest(props *adaptercommon.ChatProps, callback globals.Hook) error {
 	req := c.CreateRequest(props)
 	channel, err := c.Instance.StreamChat(req)
 	if err != nil {

adapter/skylark/struct.go

@@ -1,6 +1,7 @@
 package skylark
 
 import (
+	factory "chat/adapter/common"
 	"chat/globals"
 	"github.com/volcengine/volc-sdk-golang/service/maas"
 	"strings"
@@ -43,7 +44,7 @@ func NewChatInstance(endpoint, accessKey, secretKey string) *ChatInstance {
 	}
 }
 
-func NewChatInstanceFromConfig(conf globals.ChannelConfig) *ChatInstance {
+func NewChatInstanceFromConfig(conf globals.ChannelConfig) factory.Factory {
 	params := conf.SplitRandomSecret(2)
 
 	return NewChatInstance(

adapter/slack/chat.go

@@ -1,15 +1,12 @@
 package slack
 
 import (
+	adaptercommon "chat/adapter/common"
 	"chat/globals"
 	"context"
 )
 
-type ChatProps struct {
+func (c *ChatInstance) CreateStreamChatRequest(props *adaptercommon.ChatProps, hook globals.Hook) error {
-	Message []globals.Message
-}
-
-func (c *ChatInstance) CreateStreamChatRequest(props *ChatProps, hook globals.Hook) error {
 	if err := c.Instance.NewChannel(c.GetChannel()); err != nil {
 		return err
 	}

adapter/slack/struct.go

@@ -1,6 +1,7 @@
 package slack
 
 import (
+	factory "chat/adapter/common"
 	"chat/globals"
 	"fmt"
 	"github.com/bincooo/claude-api"
@@ -46,7 +47,7 @@ func NewChatInstance(botId, token, channel string) *ChatInstance {
 	}
 }
 
-func NewChatInstanceFromConfig(conf globals.ChannelConfig) *ChatInstance {
+func NewChatInstanceFromConfig(conf globals.ChannelConfig) factory.Factory {
 	params := conf.SplitRandomSecret(2)
 	return NewChatInstance(
 		params[0], params[1],

adapter/sparkdesk/chat.go

@@ -1,42 +1,33 @@
 package sparkdesk
 
 import (
+	adaptercommon "chat/adapter/common"
 	"chat/globals"
 	"chat/utils"
 	"fmt"
 	"strings"
 )
 
-type ChatProps struct {
+func GetToken(props *adaptercommon.ChatProps) *int {
-	Model       string
+	if props.MaxTokens == nil {
-	Message     []globals.Message
-	Token       *int
-	Temperature *float32
-	TopK        *int
-	Tools       *globals.FunctionTools
-	Buffer      utils.Buffer
-}
-
-func GetToken(props *ChatProps) *int {
-	if props.Token == nil {
 		return nil
 	}
 
 	switch props.Model {
 	case globals.SparkDeskV2, globals.SparkDeskV3:
-		if *props.Token > 8192 {
+		if *props.MaxTokens > 8192 {
 			return utils.ToPtr(8192)
 		}
 	case globals.SparkDesk:
-		if *props.Token > 4096 {
+		if *props.MaxTokens > 4096 {
 			return utils.ToPtr(4096)
 		}
 	}
 
-	return props.Token
+	return props.MaxTokens
 }
 
-func (c *ChatInstance) GetMessages(props *ChatProps) []Message {
+func (c *ChatInstance) GetMessages(props *adaptercommon.ChatProps) []Message {
 	var messages []Message
 	for _, message := range props.Message {
 		if message.Role == globals.Tool {
@@ -54,7 +45,7 @@ func (c *ChatInstance) GetMessages(props *ChatProps) []Message {
 	return messages
 }
 
-func (c *ChatInstance) GetFunctionCalling(props *ChatProps) *FunctionsPayload {
+func (c *ChatInstance) GetFunctionCalling(props *adaptercommon.ChatProps) *FunctionsPayload {
 	if props.Model != globals.SparkDeskV3 || props.Tools == nil {
 		return nil
 	}
@@ -102,9 +93,11 @@ func getChoice(form *ChatResponse) *globals.Chunk {
 	}
 }
 
-func (c *ChatInstance) CreateStreamChatRequest(props *ChatProps, hook globals.Hook) error {
+func (c *ChatInstance) CreateStreamChatRequest(props *adaptercommon.ChatProps, hook globals.Hook) error {
+	endpoint := fmt.Sprintf("%s/%s/chat", c.Endpoint, TransformAddr(props.Model))
+
 	var conn *utils.WebSocket
-	if conn = utils.NewWebsocketClient(c.GenerateUrl()); conn == nil {
+	if conn = utils.NewWebsocketClient(c.GenerateUrl(endpoint)); conn == nil {
 		return fmt.Errorf("sparkdesk error: websocket connection failed")
 	}
 	defer conn.DeferClose()
@@ -121,7 +114,7 @@ func (c *ChatInstance) CreateStreamChatRequest(props *ChatProps, hook globals.Ho
 		},
 		Parameter: RequestParameter{
 			Chat: ChatParameter{
-				Domain:   c.Model,
+				Domain:   TransformModel(props.Model),
 				MaxToken: GetToken(props),
 			},
 		},

adapter/sparkdesk/struct.go

@@ -1,6 +1,7 @@
 package sparkdesk
 
 import (
+	factory "chat/adapter/common"
 	"chat/globals"
 	"crypto/hmac"
 	"crypto/sha256"
@@ -15,7 +16,6 @@ type ChatInstance struct {
 	AppId     string
 	ApiSecret string
 	ApiKey    string
-	Model     string
 	Endpoint  string
 }
 
@@ -45,24 +45,23 @@ func TransformModel(model string) string {
 	}
 }
 
-func NewChatInstance(conf globals.ChannelConfig, model string) *ChatInstance {
+func NewChatInstanceFromConfig(conf globals.ChannelConfig) factory.Factory {
 	params := conf.SplitRandomSecret(3)
 
 	return &ChatInstance{
 		AppId:     params[0],
 		ApiSecret: params[1],
 		ApiKey:    params[2],
-		Model:     TransformModel(model),
+		Endpoint:  conf.GetEndpoint(),
-		Endpoint:  fmt.Sprintf("%s/%s/chat", conf.GetEndpoint(), TransformAddr(model)),
 	}
 }
 
-func (c *ChatInstance) CreateUrl(host, date, auth string) string {
+func (c *ChatInstance) CreateUrl(endpoint, host, date, auth string) string {
 	v := make(url.Values)
 	v.Add("host", host)
 	v.Add("date", date)
 	v.Add("authorization", auth)
-	return fmt.Sprintf("%s?%s", c.Endpoint, v.Encode())
+	return fmt.Sprintf("%s?%s", endpoint, v.Encode())
 }
 
 func (c *ChatInstance) Sign(data, key string) string {
@@ -72,8 +71,8 @@ func (c *ChatInstance) Sign(data, key string) string {
 }
 
 // GenerateUrl will generate the signed url for sparkdesk api
-func (c *ChatInstance) GenerateUrl() string {
+func (c *ChatInstance) GenerateUrl(endpoint string) string {
-	uri, err := url.Parse(c.Endpoint)
+	uri, err := url.Parse(endpoint)
 	if err != nil {
 		return ""
 	}
@@ -96,5 +95,5 @@ func (c *ChatInstance) GenerateUrl() string {
 		),
 	))
 
-	return c.CreateUrl(uri.Host, date, authorization)
+	return c.CreateUrl(endpoint, uri.Host, date, authorization)
 }

adapter/zhinao/chat.go

@@ -1,22 +1,13 @@
 package zhinao
 
 import (
+	adaptercommon "chat/adapter/common"
 	"chat/globals"
 	"chat/utils"
 	"fmt"
 	"strings"
 )
 
-type ChatProps struct {
-	Model             string
-	Message           []globals.Message
-	Token             *int
-	TopP              *float32
-	TopK              *int
-	Temperature       *float32
-	RepetitionPenalty *float32
-}
-
 func (c *ChatInstance) GetChatEndpoint() string {
 	return fmt.Sprintf("%s/v1/chat/completions", c.GetEndpoint())
 }
@@ -30,10 +21,10 @@ func (c *ChatInstance) GetModel(model string) string {
 	}
 }
 
-func (c *ChatInstance) GetChatBody(props *ChatProps, stream bool) interface{} {
+func (c *ChatInstance) GetChatBody(props *adaptercommon.ChatProps, stream bool) interface{} {
 	// 2048 is the max token for 360GPT
-	if props.Token != nil && *props.Token > 2048 {
+	if props.MaxTokens != nil && *props.MaxTokens > 2048 {
-		props.Token = utils.ToPtr(2048)
+		props.MaxTokens = utils.ToPtr(2048)
 	}
 
 	return ChatRequest{
@@ -45,7 +36,7 @@ func (c *ChatInstance) GetChatBody(props *ChatProps, stream bool) interface{} {
 
 			return &message
 		}),
-		MaxToken:          props.Token,
+		MaxToken:          props.MaxTokens,
 		Stream:            stream,
 		Temperature:       props.Temperature,
 		TopP:              props.TopP,
@@ -55,7 +46,7 @@ func (c *ChatInstance) GetChatBody(props *ChatProps, stream bool) interface{} {
 }
 
 // CreateChatRequest is the native http request body for zhinao
-func (c *ChatInstance) CreateChatRequest(props *ChatProps) (string, error) {
+func (c *ChatInstance) CreateChatRequest(props *adaptercommon.ChatProps) (string, error) {
 	res, err := utils.Post(
 		c.GetChatEndpoint(),
 		c.GetHeader(),
@@ -76,7 +67,7 @@ func (c *ChatInstance) CreateChatRequest(props *ChatProps) (string, error) {
 }
 
 // CreateStreamChatRequest is the stream response body for zhinao
-func (c *ChatInstance) CreateStreamChatRequest(props *ChatProps, callback globals.Hook) error {
+func (c *ChatInstance) CreateStreamChatRequest(props *adaptercommon.ChatProps, callback globals.Hook) error {
 	buf := ""
 	cursor := 0
 	chunk := ""

adapter/zhinao/struct.go

@@ -1,6 +1,7 @@
 package zhinao
 
 import (
+	factory "chat/adapter/common"
 	"chat/globals"
 	"fmt"
 )
@@ -32,7 +33,7 @@ func NewChatInstance(endpoint, apiKey string) *ChatInstance {
 	}
 }
 
-func NewChatInstanceFromConfig(conf globals.ChannelConfig) *ChatInstance {
+func NewChatInstanceFromConfig(conf globals.ChannelConfig) factory.Factory {
 	return NewChatInstance(
 		conf.GetEndpoint(),
 		conf.GetRandomSecret(),

adapter/zhipuai/chat.go

@@ -1,19 +1,13 @@
 package zhipuai
 
 import (
+	adaptercommon "chat/adapter/common"
 	"chat/globals"
 	"chat/utils"
 	"fmt"
 	"strings"
 )
 
-type ChatProps struct {
-	Model       string
-	Message     []globals.Message
-	Temperature *float32 `json:"temperature,omitempty"`
-	TopP        *float32 `json:"top_p,omitempty"`
-}
-
 func (c *ChatInstance) GetChatEndpoint(model string) string {
 	return fmt.Sprintf("%s/api/paas/v3/model-api/%s/sse-invoke", c.GetEndpoint(), c.GetModel(model))
 }
@@ -47,7 +41,7 @@ func (c *ChatInstance) FormatMessages(messages []globals.Message) []globals.Mess
 	return messages
 }
 
-func (c *ChatInstance) GetBody(props *ChatProps) ChatRequest {
+func (c *ChatInstance) GetBody(props *adaptercommon.ChatProps) ChatRequest {
 	return ChatRequest{
 		Prompt:      c.FormatMessages(props.Message),
 		TopP:        props.TopP,
@@ -55,7 +49,7 @@ func (c *ChatInstance) GetBody(props *ChatProps) ChatRequest {
 	}
 }
 
-func (c *ChatInstance) CreateStreamChatRequest(props *ChatProps, hook globals.Hook) error {
+func (c *ChatInstance) CreateStreamChatRequest(props *adaptercommon.ChatProps, hook globals.Hook) error {
 	return utils.EventSource(
 		"POST",
 		c.GetChatEndpoint(props.Model),

adapter/zhipuai/struct.go

@@ -1,6 +1,7 @@
 package zhipuai
 
 import (
+	factory "chat/adapter/common"
 	"chat/globals"
 	"chat/utils"
 	"github.com/dgrijalva/jwt-go"
@@ -47,6 +48,6 @@ func NewChatInstance(endpoint, apikey string) *ChatInstance {
 	}
 }
 
-func NewChatInstanceFromConfig(conf globals.ChannelConfig) *ChatInstance {
+func NewChatInstanceFromConfig(conf globals.ChannelConfig) factory.Factory {
 	return NewChatInstance(conf.GetEndpoint(), conf.GetRandomSecret())
 }

addition/generation/prompt.go

@@ -1,7 +1,7 @@
 package generation
 
 import (
-	"chat/adapter"
+	"chat/adapter/common"
 	"chat/admin"
 	"chat/channel"
 	"chat/globals"
@@ -17,10 +17,10 @@ func CreateGeneration(group, model, prompt, path string, hook func(buffer *utils
 	message := GenerateMessage(prompt)
 	buffer := utils.NewBuffer(model, message, channel.ChargeInstance.GetCharge(model))
 
-	err := channel.NewChatRequest(group, &adapter.ChatProps{
+	err := channel.NewChatRequest(group, &adaptercommon.ChatProps{
-		Model:   model,
+		OriginalModel: model,
-		Message: message,
+		Message:       message,
-		Buffer:  *buffer,
+		Buffer:        *buffer,
 	}, func(data *globals.Chunk) error {
 		buffer.WriteChunk(data)
 		hook(buffer, data.Content)

app/src/admin/channel.ts

@@ -32,18 +32,18 @@ export type ChannelInfo = {
 export const ChannelTypes: Record<string, string> = {
   openai: "OpenAI",
   azure: "Azure OpenAI",
-  claude: "Claude",
+  claude: "Anthropic Claude",
-  slack: "Slack",
-  sparkdesk: "讯飞星火",
-  chatglm: "智谱 ChatGLM",
-  qwen: "通义千问",
-  hunyuan: "腾讯混元",
-  zhinao: "360 智脑",
-  baichuan: "百川 AI",
-  skylark: "火山方舟",
-  bing: "New Bing",
   palm: "Google Gemini",
-  midjourney: "Midjourney",
+  midjourney: "Midjourney Proxy",
+  sparkdesk: "讯飞星火 SparkDesk",
+  chatglm: "智谱 ChatGLM",
+  qwen: "通义千问 TongYi",
+  hunyuan: "腾讯混元 Hunyuan",
+  zhinao: "360智脑 360GLM",
+  baichuan: "百川大模型 BaichuanAI",
+  skylark: "云雀大模型 SkylarkLLM",
+  bing: "New Bing",
+  slack: "Slack Claude",
 };
 
 export const ChannelInfos: Record<string, ChannelInfo> = {

app/src/store/chat.ts

@@ -169,7 +169,7 @@ const chatSlice = createSlice({
         });
 
       const instance = conversation.messages[conversation.messages.length - 1];
-      instance.content += message.message;
+      if (message.message.length > 0) instance.content += message.message;
       if (message.keyword) instance.keyword = message.keyword;
       if (message.quota) instance.quota = message.quota;
       if (message.end) instance.end = message.end;

channel/worker.go

@@ -2,6 +2,7 @@ package channel
 
 import (
 	"chat/adapter"
+	"chat/adapter/common"
 	"chat/globals"
 	"chat/utils"
 	"fmt"
@@ -9,10 +10,10 @@ import (
 	"time"
 )
 
-func NewChatRequest(group string, props *adapter.ChatProps, hook globals.Hook) error {
+func NewChatRequest(group string, props *adaptercommon.ChatProps, hook globals.Hook) error {
-	ticker := ConduitInstance.GetTicker(props.Model, group)
+	ticker := ConduitInstance.GetTicker(props.OriginalModel, group)
 	if ticker == nil || ticker.IsEmpty() {
-		return fmt.Errorf("cannot find channel for model %s", props.Model)
+		return fmt.Errorf("cannot find channel for model %s", props.OriginalModel)
 	}
 
 	var err error
@@ -23,14 +24,14 @@ func NewChatRequest(group string, props *adapter.ChatProps, hook globals.Hook) e
 				return nil
 			}
 
-			globals.Warn(fmt.Sprintf("[channel] caught error %s for model %s at channel %s", err.Error(), props.Model, channel.GetName()))
+			globals.Warn(fmt.Sprintf("[channel] caught error %s for model %s at channel %s", err.Error(), props.OriginalModel, channel.GetName()))
 		}
 	}
 
-	globals.Info(fmt.Sprintf("[channel] channels are exhausted for model %s", props.Model))
+	globals.Info(fmt.Sprintf("[channel] channels are exhausted for model %s", props.OriginalModel))
 
 	if err == nil {
-		err = fmt.Errorf("channels are exhausted for model %s", props.Model)
+		err = fmt.Errorf("channels are exhausted for model %s", props.OriginalModel)
 	}
 
 	return err
@@ -78,9 +79,14 @@ func StoreCache(cache *redis.Client, hash string, index int64, buffer *utils.Buf
 	cache.Set(cache.Context(), key, raw, expire)
 }
 
-func NewChatRequestWithCache(cache *redis.Client, buffer *utils.Buffer, group string, props *adapter.ChatProps, hook globals.Hook) (bool, error) {
+func NewChatRequestWithCache(cache *redis.Client, buffer *utils.Buffer, group string, props *adaptercommon.ChatProps, hook globals.Hook) (bool, error) {
 	hash := utils.Md5Encrypt(utils.Marshal(props))
-	idx, hit, err := PreflightCache(cache, props.Model, hash, buffer, hook)
+
+	if len(props.OriginalModel) == 0 {
+		props.OriginalModel = props.Model
+	}
+
+	idx, hit, err := PreflightCache(cache, props.OriginalModel, hash, buffer, hook)
 	if hit {
 		return true, err
 	}

cli/exec.go

@@ -12,8 +12,6 @@ func Run() bool {
 		Help()
 	case "invite":
 		CreateInvitationCommand(param)
-	case "filter":
-		FilterApiKeyCommand(param)
 	case "token":
 		CreateTokenCommand(param)
 	case "root":

cli/filter.go

@@ -1,18 +0,0 @@
-package cli
-
-import (
-	"chat/adapter/chatgpt"
-	"fmt"
-	"strings"
-)
-
-func FilterApiKeyCommand(args []string) {
-	data := strings.Trim(strings.TrimSpace(GetArgString(args, 0)), "\"")
-	endpoint := "https://api.openai.com"
-	keys := strings.Split(data, "|")
-
-	available := chatgpt.FilterKeysNative(endpoint, keys)
-
-	outputInfo("filter", fmt.Sprintf("filtered %d keys, %d available, %d unavailable", len(keys), len(available), len(keys)-len(available)))
-	fmt.Println(strings.Join(available, "|"))
-}

globals/interface.go

@@ -10,6 +10,7 @@ type ChannelConfig interface {
 	SplitRandomSecret(num int) []string
 	GetEndpoint() string
 	ProcessError(err error) error
+	GetId() int
 }
 
 type AuthLike interface {

manager/chat.go

@@ -2,6 +2,7 @@ package manager
 
 import (
 	"chat/adapter"
+	"chat/adapter/common"
 	"chat/addition/web"
 	"chat/admin"
 	"chat/auth"
@@ -90,7 +91,7 @@ func ChatHandler(conn *Connection, user *auth.User, instance *conversation.Conve
 	hit, err := channel.NewChatRequestWithCache(
 		cache, buffer,
 		auth.GetGroup(db, user),
-		&adapter.ChatProps{
+		&adaptercommon.ChatProps{
 			Model:             model,
 			Message:           segment,
 			Buffer:            *buffer,

manager/chat_completions.go

@@ -1,7 +1,7 @@
 package manager
 
 import (
-	"chat/adapter"
+	"chat/adapter/common"
 	"chat/addition/web"
 	"chat/admin"
 	"chat/auth"
@@ -76,8 +76,8 @@ func ChatRelayAPI(c *gin.Context) {
 	}
 }
 
-func getChatProps(form RelayForm, messages []globals.Message, buffer *utils.Buffer, plan bool) *adapter.ChatProps {
+func getChatProps(form RelayForm, messages []globals.Message, buffer *utils.Buffer, plan bool) *adaptercommon.ChatProps {
-	return &adapter.ChatProps{
+	return &adaptercommon.ChatProps{
 		Model:             form.Model,
 		Message:           messages,
 		MaxTokens:         form.MaxTokens,

manager/completions.go

@@ -1,7 +1,7 @@
 package manager
 
 import (
-	"chat/adapter"
+	"chat/adapter/common"
 	"chat/addition/web"
 	"chat/admin"
 	"chat/auth"
@@ -37,7 +37,7 @@ func NativeChatHandler(c *gin.Context, user *auth.User, model string, message []
 	hit, err := channel.NewChatRequestWithCache(
 		cache, buffer,
 		auth.GetGroup(db, user),
-		&adapter.ChatProps{
+		&adaptercommon.ChatProps{
 			Model:   model,
 			Message: segment,
 			Buffer:  *buffer,

manager/images.go

@@ -1,7 +1,7 @@
 package manager
 
 import (
-	"chat/adapter"
+	"chat/adapter/common"
 	"chat/admin"
 	"chat/auth"
 	"chat/channel"
@@ -57,8 +57,8 @@ func ImagesRelayAPI(c *gin.Context) {
 	createRelayImageObject(c, form, prompt, created, user, supportRelayPlan())
 }
 
-func getImageProps(form RelayImageForm, messages []globals.Message, buffer *utils.Buffer) *adapter.ChatProps {
+func getImageProps(form RelayImageForm, messages []globals.Message, buffer *utils.Buffer) *adaptercommon.ChatProps {
-	return &adapter.ChatProps{
+	return &adaptercommon.ChatProps{
 		Model:     form.Model,
 		Message:   messages,
 		MaxTokens: utils.ToPtr(-1),