diff --git a/common/json.go b/common/json.go index 69aa952e9..13e23a460 100644 --- a/common/json.go +++ b/common/json.go @@ -20,3 +20,25 @@ func DecodeJson(reader *bytes.Reader, v any) error { func Marshal(v any) ([]byte, error) { return json.Marshal(v) } + +func GetJsonType(data json.RawMessage) string { + data = bytes.TrimSpace(data) + if len(data) == 0 { + return "unknown" + } + firstChar := bytes.TrimSpace(data)[0] + switch firstChar { + case '{': + return "object" + case '[': + return "array" + case '"': + return "string" + case 't', 'f': + return "boolean" + case 'n': + return "null" + default: + return "number" + } +} diff --git a/dto/openai_request.go b/dto/openai_request.go index 881ec2241..ca503b259 100644 --- a/dto/openai_request.go +++ b/dto/openai_request.go @@ -760,27 +760,27 @@ type WebSearchOptions struct { // https://platform.openai.com/docs/api-reference/responses/create type OpenAIResponsesRequest struct { - Model string `json:"model"` - Input any `json:"input,omitempty"` - Include json.RawMessage `json:"include,omitempty"` - Instructions json.RawMessage `json:"instructions,omitempty"` - MaxOutputTokens uint `json:"max_output_tokens,omitempty"` - Metadata json.RawMessage `json:"metadata,omitempty"` - ParallelToolCalls bool `json:"parallel_tool_calls,omitempty"` - PreviousResponseID string `json:"previous_response_id,omitempty"` - Reasoning *Reasoning `json:"reasoning,omitempty"` - ServiceTier string `json:"service_tier,omitempty"` - Store bool `json:"store,omitempty"` - Stream bool `json:"stream,omitempty"` - Temperature float64 `json:"temperature,omitempty"` - Text json.RawMessage `json:"text,omitempty"` - ToolChoice json.RawMessage `json:"tool_choice,omitempty"` - Tools []map[string]any `json:"tools,omitempty"` // 需要处理的参数很少，MCP 参数太多不确定，所以用 map - TopP float64 `json:"top_p,omitempty"` - Truncation string `json:"truncation,omitempty"` - User string `json:"user,omitempty"` - MaxToolCalls uint `json:"max_tool_calls,omitempty"` - Prompt json.RawMessage `json:"prompt,omitempty"` + Model string `json:"model"` + Input json.RawMessage `json:"input,omitempty"` + Include json.RawMessage `json:"include,omitempty"` + Instructions json.RawMessage `json:"instructions,omitempty"` + MaxOutputTokens uint `json:"max_output_tokens,omitempty"` + Metadata json.RawMessage `json:"metadata,omitempty"` + ParallelToolCalls bool `json:"parallel_tool_calls,omitempty"` + PreviousResponseID string `json:"previous_response_id,omitempty"` + Reasoning *Reasoning `json:"reasoning,omitempty"` + ServiceTier string `json:"service_tier,omitempty"` + Store bool `json:"store,omitempty"` + Stream bool `json:"stream,omitempty"` + Temperature float64 `json:"temperature,omitempty"` + Text json.RawMessage `json:"text,omitempty"` + ToolChoice json.RawMessage `json:"tool_choice,omitempty"` + Tools json.RawMessage `json:"tools,omitempty"` // 需要处理的参数很少，MCP 参数太多不确定，所以用 map + TopP float64 `json:"top_p,omitempty"` + Truncation string `json:"truncation,omitempty"` + User string `json:"user,omitempty"` + MaxToolCalls uint `json:"max_tool_calls,omitempty"` + Prompt json.RawMessage `json:"prompt,omitempty"` } func (r *OpenAIResponsesRequest) GetTokenCountMeta() *types.TokenCountMeta { @@ -832,8 +832,7 @@ func (r *OpenAIResponsesRequest) GetTokenCountMeta() *types.TokenCountMeta { } if len(r.Tools) > 0 { - toolStr, _ := common.Marshal(r.Tools) - texts = append(texts, string(toolStr)) + texts = append(texts, string(r.Tools)) } return &types.TokenCountMeta{ @@ -853,6 +852,14 @@ func (r *OpenAIResponsesRequest) SetModelName(modelName string) { } } +func (r *OpenAIResponsesRequest) GetToolsMap() []map[string]any { + var toolsMap []map[string]any + if len(r.Tools) > 0 { + _ = common.Unmarshal(r.Tools, &toolsMap) + } + return toolsMap +} + type Reasoning struct { Effort string `json:"effort,omitempty"` Summary string `json:"summary,omitempty"` @@ -879,13 +886,21 @@ func (r *OpenAIResponsesRequest) ParseInput() []MediaInput { var inputs []MediaInput // Try string first - if str, ok := r.Input.(string); ok { + // if str, ok := common.GetJsonType(r.Input); ok { + // inputs = append(inputs, MediaInput{Type: "input_text", Text: str}) + // return inputs + // } + if common.GetJsonType(r.Input) == "string" { + var str string + _ = common.Unmarshal(r.Input, &str) inputs = append(inputs, MediaInput{Type: "input_text", Text: str}) return inputs } // Try array of parts - if array, ok := r.Input.([]any); ok { + if common.GetJsonType(r.Input) == "array" { + var array []any + _ = common.Unmarshal(r.Input, &array) for _, itemAny := range array { // Already parsed MediaInput if media, ok := itemAny.(MediaInput); ok { diff --git a/relay/channel/openai/adaptor.go b/relay/channel/openai/adaptor.go index 3756e677b..1d8286a43 100644 --- a/relay/channel/openai/adaptor.go +++ b/relay/channel/openai/adaptor.go @@ -537,8 +537,14 @@ func detectImageMimeType(filename string) string { func (a *Adaptor) ConvertOpenAIResponsesRequest(c *gin.Context, info *relaycommon.RelayInfo, request dto.OpenAIResponsesRequest) (any, error) { // 转换模型推理力度后缀 effort, originModel := parseReasoningEffortFromModelSuffix(request.Model) - if effort != "" && request.Reasoning != nil { - request.Reasoning.Effort = effort + if effort != "" { + if request.Reasoning == nil { + request.Reasoning = &dto.Reasoning{ + Effort: effort, + } + } else { + request.Reasoning.Effort = effort + } request.Model = originModel } return request, nil diff --git a/relay/channel/openai/relay_responses.go b/relay/channel/openai/relay_responses.go index ab2aa8a4a..444a42cc4 100644 --- a/relay/channel/openai/relay_responses.go +++ b/relay/channel/openai/relay_responses.go @@ -92,6 +92,8 @@ func OaiResponsesStreamHandler(c *gin.Context, info *relaycommon.RelayInfo, resp } } } + } else { + logger.LogError(c, "failed to unmarshal stream response: "+err.Error()) } return true }) diff --git a/relay/common/relay_info.go b/relay/common/relay_info.go index caf8b452e..83314bd50 100644 --- a/relay/common/relay_info.go +++ b/relay/common/relay_info.go @@ -313,7 +313,7 @@ func GenRelayInfoResponses(c *gin.Context, request *dto.OpenAIResponsesRequest) BuiltInTools: make(map[string]*BuildInToolInfo), } if len(request.Tools) > 0 { - for _, tool := range request.Tools { + for _, tool := range request.GetToolsMap() { toolType := common.Interface2String(tool["type"]) info.ResponsesUsageInfo.BuiltInTools[toolType] = &BuildInToolInfo{ ToolName: toolType,