Files
Calcium-Ion 86ac0f7745 refactor: extract protocol conversion layer into standalone relaykit module (#6369)
* test(relayconvert): add golden snapshot matrix and relaykit boundary guard

Phase 0 of the relaykit extraction plan: pin byte-level output of every
registered (from,to) request/response/stream conversion route, and
forbid kit-bound packages from growing host-only imports.

* wip(relayconvert): drop gin.Context from converter signatures; add convmeta draft

Phase 1 in progress: relayconvert now takes context.Context; host media
resolver adapts gin.Context back at the service boundary.

* refactor(relayconvert): decouple converters from RelayInfo, gin, and settings

Phase 1 of the relaykit extraction plan:
- converters now depend on convmeta.Meta (implemented by RelayInfo) instead
  of *relaycommon.RelayInfo; ClaudeConvertInfo and the format guesser move
  to convmeta with aliases left behind
- host settings reach converters via a convmeta.Options snapshot built in
  RelayInfo.ConvOptions; no more model_setting/reasoning global reads inside
  the conversion layer
- effort-suffix helpers move to service/relayconvert/reasoning (old package
  forwards); chat-to-responses upgrade policy moves to service (host routing
  logic, not conversion)
- golden conversion matrix unchanged

* test(relayconvert): tighten boundary — kit packages now free of gin/setting imports

* refactor(dto): drop gin and logger dependencies

Phase 2 (part 1): dto.Request.IsStream now takes *http.Request instead of
*gin.Context (Gemini's impl reads query/path off the std request); dto's
three logger calls become common.SysError. Boundary test allowlist is now
empty — kit-bound packages import no gin/setting/logger/model.

* refactor(kit): extract dependency-free kitutil; dto/types/relayconvert stop importing common

Phase 2 of the relaykit extraction plan:
- new service/relayconvert/kitutil holds the pure helpers the kit needs
  (JSON wrappers, pointer/string/uuid/timestamp utils, MaskSensitiveInfo,
  pluggable LogInfo/LogError hooks, Debug flag)
- dto, types, and all relayconvert packages now use kitutil; their only
  remaining internal deps are dto/types/constant
- common keeps every original symbol (MaskSensitiveInfo delegates to
  kitutil) so host code is untouched; main.go routes kit logging into
  common.SysLog/SysError and mirrors DebugEnabled
- golden conversion matrix unchanged

* refactor(kit): move EndpointType/FinishReason to types; OpenRouter dialect via Options

Kit packages (dto/types/relayconvert/reasonmap) no longer import constant:
- EndpointType and finish-reason values live in types; constant re-exports
- the OpenRouter special-case in claude->openai request conversion reads
  Options.OpenRouterDialect, set by the host from the channel type;
  InitChannelMeta invalidates the cached snapshot on channel switch

* refactor: extract relaykit submodule (dto/types/relayconvert/reasonmap)

Phase 3 of the relaykit extraction plan:
- new go module github.com/QuantumNous/new-api/relaykit containing dto
  (minus task family), types, relayconvert (with convmeta/kitutil/reasoning),
  and reasonmap; host consumes it via require + replace, go.work for dev
- task-family dto (task/suno/midjourney/video) stays in the host dto
  package; dual-consumer host files alias it as taskdto
- relaykit builds and tests standalone (GOWORK=off): no host imports,
  no gin, no DB, no settings
- golden conversion matrix unchanged

* build(docker): copy relaykit/go.mod before go mod download

The local-replace submodule's go.mod must exist inside the build context
for the main module graph to resolve.

* fix: address relaykit extraction regressions

* fix: address relaykit review regressions

* docs: document Meta nil receiver contract

* fix(relaykit): fail OpenAI→Claude conversion without max_tokens; reject negative default_max_tokens

The Claude Messages API requires max_tokens (omitting it is a 400
"Field required"), but with a nil Options.Claude.DefaultMaxTokens hook
the converters silently emitted a request the upstream is guaranteed to
reject. Both OpenAI Chat and Responses → Claude conversions now return
sharedclaude.ErrMissingMaxTokens when no path (client value, default
hook, thinking-adapter floor) supplied one. Unreachable in the host,
which always configures the hook.

Host side, claude.default_max_tokens now rejects negative values at the
option API before persisting — they would wrap into huge unsigned values
during conversion. Zero stays allowed: the current API treats
max_tokens: 0 as cache pre-warming.

* fix: make Gemini safety settings read path race-free
2026-07-27 15:56:21 +08:00

195 lines
5.2 KiB
Go

package common
import (
"crypto/sha256"
"encoding/hex"
"fmt"
"strings"
"github.com/QuantumNous/new-api/common"
"github.com/QuantumNous/new-api/relaykit/dto"
"github.com/QuantumNous/new-api/setting/operation_setting"
)
var reservedBillableToolNames = map[string]struct{}{
dto.BuildInToolWebSearchPreview: {},
dto.BuildInToolWebSearch: {},
dto.BuildInToolFileSearch: {},
dto.BuildInToolGoogleSearch: {},
dto.BuildInToolImageGeneration: {},
}
// CountBillableToolCall is the single entry point for per-call tool billing counts.
// Built-in call types always count; custom function/tool_use names only count when priced.
func (info *RelayInfo) CountBillableToolCall(itemType string, functionName string) {
if info == nil {
return
}
if info.ResponsesUsageInfo == nil {
info.ResponsesUsageInfo = &ResponsesUsageInfo{
BuiltInTools: make(map[string]*BuildInToolInfo),
}
}
if info.ResponsesUsageInfo.BuiltInTools == nil {
info.ResponsesUsageInfo.BuiltInTools = make(map[string]*BuildInToolInfo)
}
switch itemType {
case dto.BuildInCallWebSearchCall:
info.incrementBillableToolCall(resolveWebSearchToolName(info.ResponsesUsageInfo.BuiltInTools))
case dto.BuildInCallFileSearchCall:
info.incrementBillableToolCall(dto.BuildInToolFileSearch)
case dto.BuildInCallFunctionCall, dto.BuildInCallToolUse:
if functionName == "" {
return
}
if _, reserved := reservedBillableToolNames[functionName]; reserved {
return
}
if operation_setting.GetToolPriceForModel(functionName, info.OriginModelName) <= 0 {
return
}
info.incrementBillableToolCall(functionName)
}
}
func resolveWebSearchToolName(tools map[string]*BuildInToolInfo) string {
if _, ok := tools[dto.BuildInToolWebSearchPreview]; ok {
return dto.BuildInToolWebSearchPreview
}
if _, ok := tools[dto.BuildInToolWebSearch]; ok {
return dto.BuildInToolWebSearch
}
return dto.BuildInToolWebSearchPreview
}
func (info *RelayInfo) incrementBillableToolCall(name string) {
if existing, ok := info.ResponsesUsageInfo.BuiltInTools[name]; ok && existing != nil {
existing.CallCount++
return
}
info.ResponsesUsageInfo.BuiltInTools[name] = &BuildInToolInfo{
ToolName: name,
CallCount: 1,
}
}
// ImageGenerationCallCounter counts completed Responses image_generation_call
// outputs with stream-safe identity deduplication.
type ImageGenerationCallCounter struct {
seen map[string]struct{}
count int
}
// Observe records one completed final image output when billable.
// outputIndex may be nil; when set and nonnegative it participates in dedup.
func (c *ImageGenerationCallCounter) Observe(item *dto.ResponsesOutput, outputIndex *int) {
if c == nil || item == nil {
return
}
if item.Type != dto.ResponsesOutputTypeImageGenerationCall {
return
}
if strings.TrimSpace(item.Result) == "" {
return
}
switch strings.ToLower(strings.TrimSpace(item.Status)) {
case "failed", "cancelled", "canceled", "incomplete", "partial":
return
}
aliases := make([]string, 0, 4)
if item.ID != "" {
aliases = append(aliases, "id:"+item.ID)
}
if item.CallId != "" {
aliases = append(aliases, "call:"+item.CallId)
}
if outputIndex != nil && *outputIndex >= 0 {
aliases = append(aliases, fmt.Sprintf("index:%d", *outputIndex))
}
sum := sha256.Sum256([]byte(item.Result))
aliases = append(aliases, "result:"+hex.EncodeToString(sum[:]))
if c.seen == nil {
c.seen = make(map[string]struct{})
}
for _, alias := range aliases {
if _, ok := c.seen[alias]; ok {
return
}
}
for _, alias := range aliases {
c.seen[alias] = struct{}{}
}
c.count++
}
// Reset clears pending observations (used when a terminal response fails).
func (c *ImageGenerationCallCounter) Reset() {
if c == nil {
return
}
c.seen = nil
c.count = 0
}
// Count returns the deduplicated completed image output count before commit capping.
func (c *ImageGenerationCallCounter) Count() int {
if c == nil {
return 0
}
return c.count
}
// Commit writes the capped completed-output count into RelayInfo once.
// Request tool declarations alone must not become billable calls.
func (c *ImageGenerationCallCounter) Commit(info *RelayInfo) {
if info == nil {
return
}
if info.ResponsesUsageInfo == nil {
info.ResponsesUsageInfo = &ResponsesUsageInfo{
BuiltInTools: make(map[string]*BuildInToolInfo),
}
}
if info.ResponsesUsageInfo.BuiltInTools == nil {
info.ResponsesUsageInfo.BuiltInTools = make(map[string]*BuildInToolInfo)
}
count := 0
if c != nil {
count = c.count
}
if count > dto.MaxImageN {
count = dto.MaxImageN
}
if existing, ok := info.ResponsesUsageInfo.BuiltInTools[dto.BuildInToolImageGeneration]; ok && existing != nil {
existing.CallCount = count
return
}
info.ResponsesUsageInfo.BuiltInTools[dto.BuildInToolImageGeneration] = &BuildInToolInfo{
ToolName: dto.BuildInToolImageGeneration,
CallCount: count,
}
}
// IsNonBillableResponsesStatus reports terminal response statuses that must not
// bill pending image_generation observations.
func IsNonBillableResponsesStatus(status []byte) bool {
if len(status) == 0 {
return false
}
var s string
if err := common.Unmarshal(status, &s); err != nil {
return false
}
switch strings.ToLower(strings.TrimSpace(s)) {
case "failed", "cancelled", "canceled", "incomplete":
return true
default:
return false
}
}