Merge pull request #645 from MartialBE/gemini_res_format

支持gemini结构化输出

dto/openai_request.go

@@ -4,6 +4,14 @@ import "encoding/json"
 
 type ResponseFormat struct {
 	Type       string            `json:"type,omitempty"`
+	JsonSchema *FormatJsonSchema `json:"json_schema,omitempty"`
+}
+
+type FormatJsonSchema struct {
+	Description string `json:"description,omitempty"`
+	Name        string `json:"name"`
+	Schema      any    `json:"schema,omitempty"`
+	Strict      any    `json:"strict,omitempty"`
 }
 
 type GeneralOpenAIRequest struct {
@@ -25,7 +33,7 @@ type GeneralOpenAIRequest struct {
 	Functions           any             `json:"functions,omitempty"`
 	FrequencyPenalty    float64         `json:"frequency_penalty,omitempty"`
 	PresencePenalty     float64         `json:"presence_penalty,omitempty"`
-	ResponseFormat      any            `json:"response_format,omitempty"`
+	ResponseFormat      *ResponseFormat `json:"response_format,omitempty"`
 	EncodingFormat      any             `json:"encoding_format,omitempty"`
 	Seed                float64         `json:"seed,omitempty"`
 	Tools               []ToolCall      `json:"tools,omitempty"`

relay/channel/gemini/dto.go

@@ -46,6 +46,8 @@ type GeminiChatGenerationConfig struct {
 	MaxOutputTokens  uint     `json:"maxOutputTokens,omitempty"`
 	CandidateCount   int      `json:"candidateCount,omitempty"`
 	StopSequences    []string `json:"stopSequences,omitempty"`
+	ResponseMimeType string   `json:"responseMimeType,omitempty"`
+	ResponseSchema   any      `json:"responseSchema,omitempty"`
 }
 
 type GeminiChatCandidate struct {

relay/channel/gemini/relay-gemini.go

@@ -77,6 +77,16 @@ func CovertGemini2OpenAI(textRequest dto.GeneralOpenAIRequest) (*GeminiChatReque
 			},
 		}
 	}
+
+	if textRequest.ResponseFormat != nil && (textRequest.ResponseFormat.Type == "json_schema" || textRequest.ResponseFormat.Type == "json_object") {
+		geminiRequest.GenerationConfig.ResponseMimeType = "application/json"
+
+		if textRequest.ResponseFormat.JsonSchema != nil && textRequest.ResponseFormat.JsonSchema.Schema != nil {
+			cleanedSchema := removeAdditionalPropertiesWithDepth(textRequest.ResponseFormat.JsonSchema.Schema, 0)
+			geminiRequest.GenerationConfig.ResponseSchema = cleanedSchema
+		}
+	}
+
 	//shouldAddDummyModelMessage := false
 	for _, message := range textRequest.Messages {
 
@@ -165,6 +175,46 @@ func CovertGemini2OpenAI(textRequest dto.GeneralOpenAIRequest) (*GeminiChatReque
 	return &geminiRequest, nil
 }
 
+func removeAdditionalPropertiesWithDepth(schema interface{}, depth int) interface{} {
+	if depth >= 5 {
+		return schema
+	}
+
+	v, ok := schema.(map[string]interface{})
+	if !ok || len(v) == 0 {
+		return schema
+	}
+
+	// 如果type不为object和array，则直接返回
+	if typeVal, exists := v["type"]; !exists || (typeVal != "object" && typeVal != "array") {
+		return schema
+	}
+
+	switch v["type"] {
+	case "object":
+		delete(v, "additionalProperties")
+		// 处理 properties
+		if properties, ok := v["properties"].(map[string]interface{}); ok {
+			for key, value := range properties {
+				properties[key] = removeAdditionalPropertiesWithDepth(value, depth+1)
+			}
+		}
+		for _, field := range []string{"allOf", "anyOf", "oneOf"} {
+			if nested, ok := v[field].([]interface{}); ok {
+				for i, item := range nested {
+					nested[i] = removeAdditionalPropertiesWithDepth(item, depth+1)
+				}
+			}
+		}
+	case "array":
+		if items, ok := v["items"].(map[string]interface{}); ok {
+			v["items"] = removeAdditionalPropertiesWithDepth(items, depth+1)
+		}
+	}
+
+	return v
+}
+
 func (g *GeminiChatResponse) GetResponseText() string {
 	if g == nil {
 		return ""