Files
new-api/relay/chat_completions_via_responses.go
T
Calcium-Ion 0ed497f066 feat(relay): hosted-tool conversion fidelity, reasoning normalization, and billing usage integrity (#7137)
* feat(relaykit): preserve hosted tools across conversions

- add protocol-neutral hosted-tool DTOs, conversion metadata, and loss policies
- bridge citations, grounding metadata, and hosted-tool stream lifecycles
- document the public conversion behavior and channel policy controls

* refactor(relaykit): normalize reasoning and thinking intent

- centralize provider-neutral reasoning intent, effort, and budget mappings
- parse model suffixes at the host entry boundary while preserving provider-owned tails
- keep adaptive Claude thinking and explicit zero-token compatibility consistent

* fix(billing): preserve authoritative usage across relay hops

- carry native BillingUsage sidecars through direct and streamed protocol bridges
- merge partial and terminal usage monotonically with safe fallback settlement
- retain cache metadata, penultimate usage, and per-call Gemini tool surcharges

* feat(relay): bridge Responses with Claude and Gemini protocols

- add direct request, response, and stream converters across supported relay formats
- expose Claude count_tokens and Chat-to-Responses compatibility endpoints
- carry conversion diagnostics through the host while retaining the curated public goldens

* fix(relay): wire relaykit conversions into host channels

- connect handlers, adaptors, and channel settings to the standalone conversion layer
- keep model mapping, pricing identity, retries, and provider-specific suffix behavior aligned
- ignore local audit artifacts and retain focused public regression coverage
2026-09-01 21:53:35 +08:00

203 lines
6.7 KiB
Go

package relay
import (
"fmt"
"io"
"net/http"
"strings"
"github.com/QuantumNous/new-api/common"
"github.com/QuantumNous/new-api/constant"
"github.com/QuantumNous/new-api/relay/channel"
openaichannel "github.com/QuantumNous/new-api/relay/channel/openai"
relaycommon "github.com/QuantumNous/new-api/relay/common"
relayconstant "github.com/QuantumNous/new-api/relay/constant"
"github.com/QuantumNous/new-api/relaykit/dto"
"github.com/QuantumNous/new-api/relaykit/types"
"github.com/QuantumNous/new-api/service"
"github.com/gin-gonic/gin"
)
func applySystemPromptIfNeeded(c *gin.Context, info *relaycommon.RelayInfo, request *dto.GeneralOpenAIRequest) {
if info == nil || request == nil {
return
}
if info.ChannelSetting.SystemPrompt == "" {
return
}
systemRole := request.GetSystemRoleName()
containSystemPrompt := false
for _, message := range request.Messages {
if message.Role == systemRole {
containSystemPrompt = true
break
}
}
if !containSystemPrompt {
systemMessage := dto.Message{
Role: systemRole,
Content: info.ChannelSetting.SystemPrompt,
}
request.Messages = append([]dto.Message{systemMessage}, request.Messages...)
return
}
if !info.ChannelSetting.SystemPromptOverride {
return
}
common.SetContextKey(c, constant.ContextKeySystemPromptOverride, true)
for i, message := range request.Messages {
if message.Role != systemRole {
continue
}
if message.IsStringContent() {
request.Messages[i].SetStringContent(info.ChannelSetting.SystemPrompt + "\n" + message.StringContent())
return
}
contents := message.ParseContent()
contents = append([]dto.MediaContent{
{
Type: dto.ContentTypeText,
Text: info.ChannelSetting.SystemPrompt,
},
}, contents...)
request.Messages[i].Content = contents
return
}
}
func textRequestViaResponses(c *gin.Context, info *relaycommon.RelayInfo, adaptor channel.Adaptor, request any) (*dto.Usage, *types.NewAPIError) {
paramOverrideApplied := false
if chatRequest, ok := request.(*dto.GeneralOpenAIRequest); ok {
chatJSON, err := common.Marshal(chatRequest)
if err != nil {
return nil, types.NewError(err, types.ErrorCodeConvertRequestFailed, types.ErrOptionWithSkipRetry())
}
chatJSON, err = relaycommon.RemoveDisabledFields(chatJSON, info.ChannelOtherSettings, info.ChannelSetting.PassThroughBodyEnabled)
if err != nil {
return nil, types.NewError(err, types.ErrorCodeConvertRequestFailed, types.ErrOptionWithSkipRetry())
}
if len(info.ParamOverride) > 0 {
chatJSON, err = relaycommon.ApplyParamOverrideWithRelayInfo(chatJSON, info)
if err != nil {
return nil, newAPIErrorFromParamOverride(err)
}
paramOverrideApplied = true
}
var overriddenChatReq dto.GeneralOpenAIRequest
if err := common.Unmarshal(chatJSON, &overriddenChatReq); err != nil {
return nil, types.NewError(err, types.ErrorCodeChannelParamOverrideInvalid, types.ErrOptionWithSkipRetry())
}
request = &overriddenChatReq
}
result, err := service.ConvertRequest(c, info, types.RelayFormatOpenAIResponses, request)
if err != nil {
return nil, types.NewErrorWithStatusCode(err, types.ErrorCodeInvalidRequest, http.StatusBadRequest, types.ErrOptionWithSkipRetry())
}
responsesReq, ok := result.Value.(*dto.OpenAIResponsesRequest)
if !ok {
return nil, types.NewError(fmt.Errorf("expected OpenAI responses request, got %T", result.Value), types.ErrorCodeConvertRequestFailed, types.ErrOptionWithSkipRetry())
}
return relayResponsesRequest(c, info, adaptor, responsesReq, paramOverrideApplied)
}
func relayResponsesRequest(c *gin.Context, info *relaycommon.RelayInfo, adaptor channel.Adaptor, responsesReq *dto.OpenAIResponsesRequest, paramOverrideApplied bool) (*dto.Usage, *types.NewAPIError) {
savedRelayMode := info.RelayMode
savedRequestURLPath := info.RequestURLPath
defer func() {
info.RelayMode = savedRelayMode
info.RequestURLPath = savedRequestURLPath
}()
info.RelayMode = relayconstant.RelayModeResponses
info.RequestURLPath = "/v1/responses"
convertedRequest, err := adaptor.ConvertOpenAIResponsesRequest(c, info, *responsesReq)
if err != nil {
return nil, newConvertRequestFailedError(c, info, err)
}
relaycommon.AppendRequestConversionFromRequest(info, convertedRequest)
jsonData, err := common.Marshal(convertedRequest)
if err != nil {
return nil, types.NewError(err, types.ErrorCodeConvertRequestFailed, types.ErrOptionWithSkipRetry())
}
jsonData, err = relaycommon.RemoveDisabledFields(jsonData, info.ChannelOtherSettings, info.ChannelSetting.PassThroughBodyEnabled)
if err != nil {
return nil, types.NewError(err, types.ErrorCodeConvertRequestFailed, types.ErrOptionWithSkipRetry())
}
if !paramOverrideApplied && len(info.ParamOverride) > 0 {
jsonData, err = relaycommon.ApplyParamOverrideWithRelayInfo(jsonData, info)
if err != nil {
return nil, newAPIErrorFromParamOverride(err)
}
}
body, closer, err := relaycommon.NewOutboundJSONBody(jsonData)
if err != nil {
return nil, types.NewError(err, types.ErrorCodeConvertRequestFailed, types.ErrOptionWithSkipRetry())
}
defer closer.Close()
jsonData = nil
var requestBody io.Reader = body
var httpResp *http.Response
resp, err := adaptor.DoRequest(c, info, requestBody)
if err != nil {
return nil, types.NewOpenAIError(err, types.ErrorCodeDoRequestFailed, http.StatusInternalServerError)
}
if resp == nil {
return nil, types.NewOpenAIError(nil, types.ErrorCodeBadResponse, http.StatusInternalServerError)
}
statusCodeMappingStr := c.GetString("status_code_mapping")
httpResp = resp.(*http.Response)
clientStream := info.IsStream
upstreamStream := isResponsesEventStreamContentType(httpResp.Header.Get("Content-Type"))
info.IsStream = clientStream || upstreamStream
if httpResp.StatusCode != http.StatusOK {
newApiErr := service.RelayErrorHandler(c.Request.Context(), httpResp, false)
service.ResetStatusCode(newApiErr, statusCodeMappingStr)
return nil, newApiErr
}
if upstreamStream && clientStream {
usage, newApiErr := openaichannel.OaiResponsesToChatStreamHandler(c, info, httpResp)
if newApiErr != nil {
service.ResetStatusCode(newApiErr, statusCodeMappingStr)
return nil, newApiErr
}
return usage, nil
}
if upstreamStream {
info.IsStream = false
usage, newApiErr := openaichannel.OaiResponsesToChatBufferedStreamHandler(c, info, httpResp)
if newApiErr != nil {
service.ResetStatusCode(newApiErr, statusCodeMappingStr)
return nil, newApiErr
}
return usage, nil
}
usage, newApiErr := openaichannel.OaiResponsesToChatHandler(c, info, httpResp)
if newApiErr != nil {
service.ResetStatusCode(newApiErr, statusCodeMappingStr)
return nil, newApiErr
}
return usage, nil
}
func isResponsesEventStreamContentType(contentType string) bool {
return strings.Contains(strings.ToLower(contentType), "text/event-stream")
}