feat: implement Gemini to OpenAI chat stream conversion with state management and terminal handling

This commit is contained in:
CaIon
2026-07-27 20:17:20 +08:00
parent f3ab2cff36
commit a043eef559
10 changed files with 894 additions and 37 deletions
@@ -282,6 +282,107 @@ func StreamResponseGeminiChat2OpenAI(geminiResponse *dto.GeminiChatResponse) (*d
return &response, isStop
}
type GeminiToChatStreamState struct {
id string
created int64
sawToolCall bool
finishEmitted bool
latestUsage *dto.Usage
}
func NewGeminiToChatStreamState(id string, created int64) *GeminiToChatStreamState {
id = strings.TrimSpace(id)
if id == "" {
id = fmt.Sprintf("chatcmpl-%s", kitutil.GetUUID())
}
if created == 0 {
created = kitutil.GetTimestamp()
}
return &GeminiToChatStreamState{id: id, created: created}
}
func (s *GeminiToChatStreamState) ConvertChunk(geminiResponse *dto.GeminiChatResponse, model string, usage *dto.Usage) []*dto.ChatCompletionsStreamResponse {
if s == nil || geminiResponse == nil {
return nil
}
hasNonStopFinish := false
for _, candidate := range geminiResponse.Candidates {
if candidate.FinishReason != nil && *candidate.FinishReason != "" && *candidate.FinishReason != "STOP" {
hasNonStopFinish = true
break
}
}
response, isStop := StreamResponseGeminiChat2OpenAI(geminiResponse)
if response == nil {
return nil
}
response.Id = s.id
response.Created = s.created
response.Model = model
response.Usage = usage
if response.IsToolCall() {
s.sawToolCall = true
if !hasNonStopFinish {
for i := range response.Choices {
if response.Choices[i].FinishReason != nil && *response.Choices[i].FinishReason == types.FinishReasonToolCalls {
response.Choices[i].FinishReason = nil
}
}
}
}
if usage != nil {
s.latestUsage = usage
}
for _, choice := range response.Choices {
if choice.FinishReason != nil && *choice.FinishReason != "" {
s.finishEmitted = true
break
}
}
responses := []*dto.ChatCompletionsStreamResponse{response}
if isStop && !s.finishEmitted {
responses = append(responses, s.terminalChunk(model))
}
return responses
}
func (s *GeminiToChatStreamState) Finalize(model string) []*dto.ChatCompletionsStreamResponse {
if s == nil || s.finishEmitted {
return nil
}
return []*dto.ChatCompletionsStreamResponse{s.terminalChunk(model)}
}
func (s *GeminiToChatStreamState) Usage() *dto.Usage {
if s == nil {
return nil
}
return s.latestUsage
}
func (s *GeminiToChatStreamState) terminalChunk(model string) *dto.ChatCompletionsStreamResponse {
finishReason := types.FinishReasonStop
if s.sawToolCall {
finishReason = types.FinishReasonToolCalls
}
s.finishEmitted = true
return &dto.ChatCompletionsStreamResponse{
Id: s.id,
Object: "chat.completion.chunk",
Created: s.created,
Model: model,
Choices: []dto.ChatCompletionsStreamResponseChoice{
{
Delta: dto.ChatCompletionsStreamResponseChoiceDelta{},
FinishReason: &finishReason,
},
},
Usage: s.latestUsage,
}
}
func geminiResponseToolCall(item *dto.GeminiPart) *dto.ToolCallResponse {
argsBytes, err := kitutil.Marshal(item.FunctionCall.Arguments)
if err != nil {
@@ -17,6 +17,24 @@ func generateStopBlock(index int) *dto.ClaudeResponse {
}
}
func stopOpenBlocks(state *convmeta.ClaudeConvertInfo) []*dto.ClaudeResponse {
if state == nil {
return nil
}
switch state.LastMessagesType {
case convmeta.LastMessageTypeText, convmeta.LastMessageTypeThinking:
return []*dto.ClaudeResponse{generateStopBlock(state.Index)}
case convmeta.LastMessageTypeTools:
responses := make([]*dto.ClaudeResponse, 0, state.ToolCallMaxIndexOffset+1)
for offset := 0; offset <= state.ToolCallMaxIndexOffset; offset++ {
responses = append(responses, generateStopBlock(state.ToolCallBaseIndex+offset))
}
return responses
default:
return nil
}
}
func buildClaudeUsageFromOpenAIUsage(oaiUsage *dto.Usage) *dto.ClaudeUsage {
if oaiUsage == nil {
return nil
@@ -89,16 +107,8 @@ func StreamResponseOpenAI2Claude(openAIResponse *dto.ChatCompletionsStreamRespon
// For text/thinking, there is at most one open block at state.Index.
// For tools, OpenAI tool_calls can stream multiple parallel tool_use blocks (indexed from 0),
// so we may have multiple open blocks and must stop each one explicitly.
stopOpenBlocks := func() {
switch state.LastMessagesType {
case convmeta.LastMessageTypeText, convmeta.LastMessageTypeThinking:
claudeResponses = append(claudeResponses, generateStopBlock(state.Index))
case convmeta.LastMessageTypeTools:
base := state.ToolCallBaseIndex
for offset := 0; offset <= state.ToolCallMaxIndexOffset; offset++ {
claudeResponses = append(claudeResponses, generateStopBlock(base+offset))
}
}
appendStopOpenBlocks := func() {
claudeResponses = append(claudeResponses, stopOpenBlocks(state)...)
}
// stopOpenBlocksAndAdvance closes the currently open block(s) and advances the content block index
// to the next available slot for subsequent content_block_start events.
@@ -109,7 +119,7 @@ func StreamResponseOpenAI2Claude(openAIResponse *dto.ChatCompletionsStreamRespon
if state.LastMessagesType == convmeta.LastMessageTypeNone {
return
}
stopOpenBlocks()
appendStopOpenBlocks()
switch state.LastMessagesType {
case convmeta.LastMessageTypeTools:
state.Index = state.ToolCallBaseIndex + state.ToolCallMaxIndexOffset + 1
@@ -234,23 +244,24 @@ func StreamResponseOpenAI2Claude(openAIResponse *dto.ChatCompletionsStreamRespon
}
}
// 如果首块就带 finish_reason,需要立即发送停止块
// A first chunk can carry finish_reason before usage; defer terminal events until usage arrives.
if len(openAIResponse.Choices) > 0 && openAIResponse.Choices[0].FinishReason != nil && *openAIResponse.Choices[0].FinishReason != "" {
state.FinishReason = *openAIResponse.Choices[0].FinishReason
stopOpenBlocks()
oaiUsage := openAIResponse.Usage
if oaiUsage == nil {
oaiUsage = state.Usage
}
if oaiUsage != nil {
claudeResponses = append(claudeResponses, &dto.ClaudeResponse{
Type: "message_delta",
Usage: buildClaudeUsageFromOpenAIUsage(oaiUsage),
Delta: &dto.ClaudeMediaMessage{
StopReason: kitutil.GetPointer[string](stopReasonOpenAI2Claude(state.FinishReason)),
},
})
if oaiUsage == nil {
return claudeResponses
}
appendStopOpenBlocks()
claudeResponses = append(claudeResponses, &dto.ClaudeResponse{
Type: "message_delta",
Usage: buildClaudeUsageFromOpenAIUsage(oaiUsage),
Delta: &dto.ClaudeMediaMessage{
StopReason: kitutil.GetPointer[string](stopReasonOpenAI2Claude(state.FinishReason)),
},
})
claudeResponses = append(claudeResponses, &dto.ClaudeResponse{
Type: "message_stop",
})
@@ -266,7 +277,7 @@ func StreamResponseOpenAI2Claude(openAIResponse *dto.ChatCompletionsStreamRespon
oaiUsage = state.Usage
}
if oaiUsage != nil {
stopOpenBlocks()
appendStopOpenBlocks()
stopReason := stopReasonOpenAI2Claude(state.FinishReason)
if stopReason == "" {
stopReason = "end_turn"
@@ -403,7 +414,7 @@ func StreamResponseOpenAI2Claude(openAIResponse *dto.ChatCompletionsStreamRespon
}
if doneChunk || state.Done {
stopOpenBlocks()
appendStopOpenBlocks()
oaiUsage := openAIResponse.Usage
if oaiUsage == nil {
oaiUsage = state.Usage
@@ -428,6 +439,34 @@ func StreamResponseOpenAI2Claude(openAIResponse *dto.ChatCompletionsStreamRespon
return claudeResponses
}
func FinalizeStreamResponseOpenAI2Claude(info convmeta.Meta) []*dto.ClaudeResponse {
if info == nil {
info = &convmeta.Values{}
}
state := info.EnsureClaudeConvertInfo()
if state.Done {
return nil
}
stopReason := stopReasonOpenAI2Claude(state.FinishReason)
if stopReason == "" {
stopReason = "end_turn"
}
responses := stopOpenBlocks(state)
responses = append(responses,
&dto.ClaudeResponse{
Type: "message_delta",
Usage: buildClaudeUsageFromOpenAIUsage(state.Usage),
Delta: &dto.ClaudeMediaMessage{
StopReason: kitutil.GetPointer[string](stopReason),
},
},
&dto.ClaudeResponse{Type: "message_stop"},
)
state.Done = true
return responses
}
func ResponseOpenAI2Claude(openAIResponse *dto.OpenAITextResponse, info convmeta.Meta) *dto.ClaudeResponse {
var stopReason string
contents := make([]dto.ClaudeMediaMessage, 0)