mirror of
https://github.com/linux-do/new-api.git
synced 2025-09-17 16:06:38 +08:00
105 lines
3.6 KiB
Go
105 lines
3.6 KiB
Go
package relay
|
|
|
|
import (
|
|
"bytes"
|
|
"encoding/json"
|
|
"fmt"
|
|
"github.com/gin-gonic/gin"
|
|
"net/http"
|
|
"one-api/common"
|
|
"one-api/dto"
|
|
relaycommon "one-api/relay/common"
|
|
"one-api/service"
|
|
)
|
|
|
|
func getRerankPromptToken(rerankRequest dto.RerankRequest) int {
|
|
token, _ := service.CountTokenInput(rerankRequest.Query, rerankRequest.Model)
|
|
for _, document := range rerankRequest.Documents {
|
|
tkm, err := service.CountTokenInput(document, rerankRequest.Model)
|
|
if err == nil {
|
|
token += tkm
|
|
}
|
|
}
|
|
return token
|
|
}
|
|
|
|
func RerankHelper(c *gin.Context, relayMode int) *dto.OpenAIErrorWithStatusCode {
|
|
relayInfo := relaycommon.GenRelayInfo(c)
|
|
|
|
var rerankRequest *dto.RerankRequest
|
|
err := common.UnmarshalBodyReusable(c, &rerankRequest)
|
|
if err != nil {
|
|
common.LogError(c, fmt.Sprintf("getAndValidateTextRequest failed: %s", err.Error()))
|
|
return service.OpenAIErrorWrapperLocal(err, "invalid_text_request", http.StatusBadRequest)
|
|
}
|
|
if rerankRequest.Query == "" {
|
|
return service.OpenAIErrorWrapperLocal(fmt.Errorf("query is empty"), "invalid_query", http.StatusBadRequest)
|
|
}
|
|
if len(rerankRequest.Documents) == 0 {
|
|
return service.OpenAIErrorWrapperLocal(fmt.Errorf("documents is empty"), "invalid_documents", http.StatusBadRequest)
|
|
}
|
|
relayInfo.UpstreamModelName = rerankRequest.Model
|
|
modelPrice, success := common.GetModelPrice(rerankRequest.Model, false)
|
|
groupRatio := common.GetGroupRatio(relayInfo.Group)
|
|
|
|
var preConsumedQuota int
|
|
var ratio float64
|
|
var modelRatio float64
|
|
|
|
promptToken := getRerankPromptToken(*rerankRequest)
|
|
if !success {
|
|
preConsumedTokens := promptToken
|
|
modelRatio = common.GetModelRatio(rerankRequest.Model)
|
|
ratio = modelRatio * groupRatio
|
|
preConsumedQuota = int(float64(preConsumedTokens) * ratio)
|
|
} else {
|
|
preConsumedQuota = int(modelPrice * common.QuotaPerUnit * groupRatio)
|
|
}
|
|
relayInfo.PromptTokens = promptToken
|
|
|
|
// pre-consume quota 预消耗配额
|
|
preConsumedQuota, userQuota, openaiErr := preConsumeQuota(c, preConsumedQuota, relayInfo)
|
|
if openaiErr != nil {
|
|
return openaiErr
|
|
}
|
|
adaptor := GetAdaptor(relayInfo.ApiType)
|
|
if adaptor == nil {
|
|
return service.OpenAIErrorWrapperLocal(fmt.Errorf("invalid api type: %d", relayInfo.ApiType), "invalid_api_type", http.StatusBadRequest)
|
|
}
|
|
adaptor.InitRerank(relayInfo, *rerankRequest)
|
|
|
|
convertedRequest, err := adaptor.ConvertRerankRequest(c, relayInfo.RelayMode, *rerankRequest)
|
|
if err != nil {
|
|
return service.OpenAIErrorWrapperLocal(err, "convert_request_failed", http.StatusInternalServerError)
|
|
}
|
|
jsonData, err := json.Marshal(convertedRequest)
|
|
if err != nil {
|
|
return service.OpenAIErrorWrapperLocal(err, "json_marshal_failed", http.StatusInternalServerError)
|
|
}
|
|
requestBody := bytes.NewBuffer(jsonData)
|
|
statusCodeMappingStr := c.GetString("status_code_mapping")
|
|
resp, err := adaptor.DoRequest(c, relayInfo, requestBody)
|
|
if err != nil {
|
|
return service.OpenAIErrorWrapper(err, "do_request_failed", http.StatusInternalServerError)
|
|
}
|
|
if resp != nil {
|
|
if resp.StatusCode != http.StatusOK {
|
|
returnPreConsumedQuota(c, relayInfo.TokenId, userQuota, preConsumedQuota)
|
|
openaiErr := service.RelayErrorHandler(resp)
|
|
// reset status code 重置状态码
|
|
service.ResetStatusCode(openaiErr, statusCodeMappingStr)
|
|
return openaiErr
|
|
}
|
|
}
|
|
|
|
usage, openaiErr := adaptor.DoResponse(c, resp, relayInfo)
|
|
if openaiErr != nil {
|
|
returnPreConsumedQuota(c, relayInfo.TokenId, userQuota, preConsumedQuota)
|
|
// reset status code 重置状态码
|
|
service.ResetStatusCode(openaiErr, statusCodeMappingStr)
|
|
return openaiErr
|
|
}
|
|
postConsumeQuota(c, relayInfo, rerankRequest.Model, usage, ratio, preConsumedQuota, userQuota, modelRatio, groupRatio, modelPrice, success)
|
|
return nil
|
|
}
|