feat: add tri-state support for disable-image-generation configuration

- Introduced `DisableImageGenerationMode` with support for `false`, `true`, and `chat` values.
- Updated payload handling to preserve `image_generation` on images endpoints when `chat` mode is enabled.
- Modified OpenAI image handlers (`ImagesGenerations`, `ImagesEdits`) to respect tri-state logic.
- Added unit tests for `DisableImageGenerationMode` behavior and endpoint-specific handling.
- Enhanced configuration diff logging to support `DisableImageGenerationMode`.
This commit is contained in:
Luis Pater
2026-04-30 11:59:50 +08:00
parent 46018417ad
commit f56a19e5b8
24 changed files with 398 additions and 54 deletions

View File

@@ -1014,7 +1014,7 @@ func (s *Server) UpdateClients(cfg *config.Config) {
}
if oldCfg != nil && oldCfg.DisableImageGeneration != cfg.DisableImageGeneration {
log.Infof("disable-image-generation updated: %t -> %t", oldCfg.DisableImageGeneration, cfg.DisableImageGeneration)
log.Infof("disable-image-generation updated: %v -> %v", oldCfg.DisableImageGeneration, cfg.DisableImageGeneration)
}
applySignatureCacheConfig(oldCfg, cfg)

View File

@@ -610,7 +610,7 @@ func LoadConfigOptional(configFile string, optional bool) (*Config, error) {
cfg.ErrorLogsMaxFiles = 10
cfg.UsageStatisticsEnabled = false
cfg.DisableCooling = false
cfg.DisableImageGeneration = false
cfg.DisableImageGeneration = DisableImageGenerationOff
cfg.Pprof.Enable = false
cfg.Pprof.Addr = DefaultPprofAddr
cfg.AmpCode.RestrictManagementToLocalhost = false // Default to false: API key auth is sufficient

View File

@@ -0,0 +1,136 @@
package config
import (
"bytes"
"encoding/json"
"fmt"
"strings"
"gopkg.in/yaml.v3"
)
// DisableImageGenerationMode is a tri-state config value for disable-image-generation.
//
// It supports:
// - false: enabled
// - true: disabled everywhere (including /v1/images/* endpoints)
// - "chat": disabled for all non-images endpoints, but enabled for /v1/images/generations and /v1/images/edits
type DisableImageGenerationMode int
const (
DisableImageGenerationOff DisableImageGenerationMode = iota
DisableImageGenerationAll
DisableImageGenerationChat
)
func (m DisableImageGenerationMode) String() string {
switch m {
case DisableImageGenerationOff:
return "false"
case DisableImageGenerationAll:
return "true"
case DisableImageGenerationChat:
return "chat"
default:
return "false"
}
}
func (m DisableImageGenerationMode) MarshalYAML() (any, error) {
switch m {
case DisableImageGenerationAll:
return true, nil
case DisableImageGenerationChat:
return "chat", nil
default:
return false, nil
}
}
func (m *DisableImageGenerationMode) UnmarshalYAML(value *yaml.Node) error {
mode, err := parseDisableImageGenerationNode(value)
if err != nil {
return err
}
*m = mode
return nil
}
func (m DisableImageGenerationMode) MarshalJSON() ([]byte, error) {
switch m {
case DisableImageGenerationAll:
return []byte("true"), nil
case DisableImageGenerationChat:
return json.Marshal("chat")
default:
return []byte("false"), nil
}
}
func (m *DisableImageGenerationMode) UnmarshalJSON(data []byte) error {
mode, err := parseDisableImageGenerationJSON(data)
if err != nil {
return err
}
*m = mode
return nil
}
func parseDisableImageGenerationNode(value *yaml.Node) (DisableImageGenerationMode, error) {
if value == nil {
return DisableImageGenerationOff, nil
}
// First try a typed bool decode (covers unquoted true/false and YAML 1.1 bools).
var b bool
if err := value.Decode(&b); err == nil && value.Kind == yaml.ScalarNode && value.ShortTag() == "!!bool" {
if b {
return DisableImageGenerationAll, nil
}
return DisableImageGenerationOff, nil
}
// Fall back to string decoding (covers quoted "true"/"false" and "chat").
var s string
if err := value.Decode(&s); err != nil {
return DisableImageGenerationOff, fmt.Errorf("invalid disable-image-generation value")
}
return parseDisableImageGenerationString(s)
}
func parseDisableImageGenerationJSON(data []byte) (DisableImageGenerationMode, error) {
trimmed := bytes.TrimSpace(data)
if len(trimmed) == 0 || bytes.Equal(trimmed, []byte("null")) {
return DisableImageGenerationOff, nil
}
// bool
var b bool
if err := json.Unmarshal(trimmed, &b); err == nil {
if b {
return DisableImageGenerationAll, nil
}
return DisableImageGenerationOff, nil
}
// string
var s string
if err := json.Unmarshal(trimmed, &s); err != nil {
return DisableImageGenerationOff, fmt.Errorf("invalid disable-image-generation value")
}
return parseDisableImageGenerationString(s)
}
func parseDisableImageGenerationString(s string) (DisableImageGenerationMode, error) {
s = strings.TrimSpace(strings.ToLower(s))
switch s {
case "", "false", "0", "off", "no":
return DisableImageGenerationOff, nil
case "true", "1", "on", "yes":
return DisableImageGenerationAll, nil
case "chat":
return DisableImageGenerationChat, nil
default:
return DisableImageGenerationOff, fmt.Errorf("invalid disable-image-generation value %q (allowed: true, false, chat)", s)
}
}

View File

@@ -0,0 +1,76 @@
package config
import (
"encoding/json"
"testing"
"gopkg.in/yaml.v3"
)
func TestDisableImageGenerationMode_UnmarshalYAML(t *testing.T) {
type wrapper struct {
V DisableImageGenerationMode `yaml:"disable-image-generation"`
}
{
var w wrapper
if err := yaml.Unmarshal([]byte("disable-image-generation: false\n"), &w); err != nil {
t.Fatalf("unmarshal false: %v", err)
}
if w.V != DisableImageGenerationOff {
t.Fatalf("false => %v, want %v", w.V, DisableImageGenerationOff)
}
}
{
var w wrapper
if err := yaml.Unmarshal([]byte("disable-image-generation: true\n"), &w); err != nil {
t.Fatalf("unmarshal true: %v", err)
}
if w.V != DisableImageGenerationAll {
t.Fatalf("true => %v, want %v", w.V, DisableImageGenerationAll)
}
}
{
var w wrapper
if err := yaml.Unmarshal([]byte("disable-image-generation: chat\n"), &w); err != nil {
t.Fatalf("unmarshal chat: %v", err)
}
if w.V != DisableImageGenerationChat {
t.Fatalf("chat => %v, want %v", w.V, DisableImageGenerationChat)
}
}
}
func TestDisableImageGenerationMode_UnmarshalJSON(t *testing.T) {
{
var v DisableImageGenerationMode
if err := json.Unmarshal([]byte("false"), &v); err != nil {
t.Fatalf("unmarshal false: %v", err)
}
if v != DisableImageGenerationOff {
t.Fatalf("false => %v, want %v", v, DisableImageGenerationOff)
}
}
{
var v DisableImageGenerationMode
if err := json.Unmarshal([]byte("true"), &v); err != nil {
t.Fatalf("unmarshal true: %v", err)
}
if v != DisableImageGenerationAll {
t.Fatalf("true => %v, want %v", v, DisableImageGenerationAll)
}
}
{
var v DisableImageGenerationMode
if err := json.Unmarshal([]byte(`"chat"`), &v); err != nil {
t.Fatalf("unmarshal chat: %v", err)
}
if v != DisableImageGenerationChat {
t.Fatalf("chat => %v, want %v", v, DisableImageGenerationChat)
}
}
}

View File

@@ -9,11 +9,15 @@ type SDKConfig struct {
// ProxyURL is the URL of an optional proxy server to use for outbound requests.
ProxyURL string `yaml:"proxy-url" json:"proxy-url"`
// DisableImageGeneration disables the built-in image_generation tool when true.
// When enabled, the server will avoid injecting image_generation into request payloads,
// will remove any existing image_generation tool entries from tools arrays, and will
// return 404 for /v1/images/generations and /v1/images/edits.
DisableImageGeneration bool `yaml:"disable-image-generation" json:"disable-image-generation"`
// DisableImageGeneration controls whether the built-in image_generation tool is injected/allowed.
//
// Supported values:
// - false (default): image_generation is enabled everywhere (normal behavior).
// - true: image_generation is disabled everywhere. The server stops injecting it, removes it from request payloads,
// and returns 404 for /v1/images/generations and /v1/images/edits.
// - "chat": disable image_generation injection for all non-images endpoints (e.g. /v1/responses, /v1/chat/completions),
// while keeping /v1/images/generations and /v1/images/edits enabled and preserving image_generation there.
DisableImageGeneration DisableImageGenerationMode `yaml:"disable-image-generation" json:"disable-image-generation"`
// EnableGeminiCLIEndpoint controls whether Gemini CLI internal endpoints (/v1internal:*) are enabled.
// Default is false for safety; when false, /v1internal:* requests are rejected.

View File

@@ -428,7 +428,8 @@ func (e *AIStudioExecutor) translateRequest(req cliproxyexecutor.Request, opts c
}
payload = fixGeminiImageAspectRatio(baseModel, payload)
requestedModel := helps.PayloadRequestedModel(opts, req.Model)
payload = helps.ApplyPayloadConfigWithRoot(e.cfg, baseModel, to.String(), "", payload, originalTranslated, requestedModel)
requestPath := helps.PayloadRequestPath(opts)
payload = helps.ApplyPayloadConfigWithRoot(e.cfg, baseModel, to.String(), "", payload, originalTranslated, requestedModel, requestPath)
payload, _ = sjson.DeleteBytes(payload, "generationConfig.maxOutputTokens")
payload, _ = sjson.DeleteBytes(payload, "generationConfig.responseMimeType")
payload, _ = sjson.DeleteBytes(payload, "generationConfig.responseJsonSchema")

View File

@@ -521,7 +521,8 @@ func (e *AntigravityExecutor) Execute(ctx context.Context, auth *cliproxyauth.Au
}
requestedModel := helps.PayloadRequestedModel(opts, req.Model)
translated = helps.ApplyPayloadConfigWithRoot(e.cfg, baseModel, "antigravity", "request", translated, originalTranslated, requestedModel)
requestPath := helps.PayloadRequestPath(opts)
translated = helps.ApplyPayloadConfigWithRoot(e.cfg, baseModel, "antigravity", "request", translated, originalTranslated, requestedModel, requestPath)
useCredits := cliproxyauth.AntigravityCreditsRequested(ctx) && antigravityCreditsRetryEnabled(e.cfg)
@@ -718,7 +719,8 @@ func (e *AntigravityExecutor) executeClaudeNonStream(ctx context.Context, auth *
}
requestedModel := helps.PayloadRequestedModel(opts, req.Model)
translated = helps.ApplyPayloadConfigWithRoot(e.cfg, baseModel, "antigravity", "request", translated, originalTranslated, requestedModel)
requestPath := helps.PayloadRequestPath(opts)
translated = helps.ApplyPayloadConfigWithRoot(e.cfg, baseModel, "antigravity", "request", translated, originalTranslated, requestedModel, requestPath)
useCredits := cliproxyauth.AntigravityCreditsRequested(ctx) && antigravityCreditsRetryEnabled(e.cfg)
@@ -1178,7 +1180,8 @@ func (e *AntigravityExecutor) ExecuteStream(ctx context.Context, auth *cliproxya
}
requestedModel := helps.PayloadRequestedModel(opts, req.Model)
translated = helps.ApplyPayloadConfigWithRoot(e.cfg, baseModel, "antigravity", "request", translated, originalTranslated, requestedModel)
requestPath := helps.PayloadRequestPath(opts)
translated = helps.ApplyPayloadConfigWithRoot(e.cfg, baseModel, "antigravity", "request", translated, originalTranslated, requestedModel, requestPath)
useCredits := cliproxyauth.AntigravityCreditsRequested(ctx) && antigravityCreditsRetryEnabled(e.cfg)

View File

@@ -164,7 +164,8 @@ func (e *ClaudeExecutor) Execute(ctx context.Context, auth *cliproxyauth.Auth, r
body = applyCloaking(ctx, e.cfg, auth, body, baseModel, apiKey)
requestedModel := helps.PayloadRequestedModel(opts, req.Model)
body = helps.ApplyPayloadConfigWithRoot(e.cfg, baseModel, to.String(), "", body, originalTranslated, requestedModel)
requestPath := helps.PayloadRequestPath(opts)
body = helps.ApplyPayloadConfigWithRoot(e.cfg, baseModel, to.String(), "", body, originalTranslated, requestedModel, requestPath)
body = ensureModelMaxTokens(body, baseModel)
// Disable thinking if tool_choice forces tool use (Anthropic API constraint)
@@ -349,7 +350,8 @@ func (e *ClaudeExecutor) ExecuteStream(ctx context.Context, auth *cliproxyauth.A
body = applyCloaking(ctx, e.cfg, auth, body, baseModel, apiKey)
requestedModel := helps.PayloadRequestedModel(opts, req.Model)
body = helps.ApplyPayloadConfigWithRoot(e.cfg, baseModel, to.String(), "", body, originalTranslated, requestedModel)
requestPath := helps.PayloadRequestPath(opts)
body = helps.ApplyPayloadConfigWithRoot(e.cfg, baseModel, to.String(), "", body, originalTranslated, requestedModel, requestPath)
body = ensureModelMaxTokens(body, baseModel)
// Disable thinking if tool_choice forces tool use (Anthropic API constraint)

View File

@@ -173,7 +173,8 @@ func (e *CodexExecutor) Execute(ctx context.Context, auth *cliproxyauth.Auth, re
}
requestedModel := helps.PayloadRequestedModel(opts, req.Model)
body = helps.ApplyPayloadConfigWithRoot(e.cfg, baseModel, to.String(), "", body, originalTranslated, requestedModel)
requestPath := helps.PayloadRequestPath(opts)
body = helps.ApplyPayloadConfigWithRoot(e.cfg, baseModel, to.String(), "", body, originalTranslated, requestedModel, requestPath)
body, _ = sjson.SetBytes(body, "model", baseModel)
body, _ = sjson.SetBytes(body, "stream", true)
body, _ = sjson.DeleteBytes(body, "previous_response_id")
@@ -181,7 +182,7 @@ func (e *CodexExecutor) Execute(ctx context.Context, auth *cliproxyauth.Auth, re
body, _ = sjson.DeleteBytes(body, "safety_identifier")
body, _ = sjson.DeleteBytes(body, "stream_options")
body = normalizeCodexInstructions(body)
if e.cfg == nil || !e.cfg.DisableImageGeneration {
if e.cfg == nil || e.cfg.DisableImageGeneration == config.DisableImageGenerationOff {
body = ensureImageGenerationTool(body, baseModel, auth)
}
@@ -327,11 +328,12 @@ func (e *CodexExecutor) executeCompact(ctx context.Context, auth *cliproxyauth.A
}
requestedModel := helps.PayloadRequestedModel(opts, req.Model)
body = helps.ApplyPayloadConfigWithRoot(e.cfg, baseModel, to.String(), "", body, originalTranslated, requestedModel)
requestPath := helps.PayloadRequestPath(opts)
body = helps.ApplyPayloadConfigWithRoot(e.cfg, baseModel, to.String(), "", body, originalTranslated, requestedModel, requestPath)
body, _ = sjson.SetBytes(body, "model", baseModel)
body, _ = sjson.DeleteBytes(body, "stream")
body = normalizeCodexInstructions(body)
if e.cfg == nil || !e.cfg.DisableImageGeneration {
if e.cfg == nil || e.cfg.DisableImageGeneration == config.DisableImageGenerationOff {
body = ensureImageGenerationTool(body, baseModel, auth)
}
@@ -421,14 +423,15 @@ func (e *CodexExecutor) ExecuteStream(ctx context.Context, auth *cliproxyauth.Au
}
requestedModel := helps.PayloadRequestedModel(opts, req.Model)
body = helps.ApplyPayloadConfigWithRoot(e.cfg, baseModel, to.String(), "", body, originalTranslated, requestedModel)
requestPath := helps.PayloadRequestPath(opts)
body = helps.ApplyPayloadConfigWithRoot(e.cfg, baseModel, to.String(), "", body, originalTranslated, requestedModel, requestPath)
body, _ = sjson.DeleteBytes(body, "previous_response_id")
body, _ = sjson.DeleteBytes(body, "prompt_cache_retention")
body, _ = sjson.DeleteBytes(body, "safety_identifier")
body, _ = sjson.DeleteBytes(body, "stream_options")
body, _ = sjson.SetBytes(body, "model", baseModel)
body = normalizeCodexInstructions(body)
if e.cfg == nil || !e.cfg.DisableImageGeneration {
if e.cfg == nil || e.cfg.DisableImageGeneration == config.DisableImageGenerationOff {
body = ensureImageGenerationTool(body, baseModel, auth)
}

View File

@@ -184,14 +184,16 @@ func (e *CodexWebsocketsExecutor) Execute(ctx context.Context, auth *cliproxyaut
}
requestedModel := helps.PayloadRequestedModel(opts, req.Model)
body = helps.ApplyPayloadConfigWithRoot(e.cfg, baseModel, to.String(), "", body, originalTranslated, requestedModel)
requestPath := helps.PayloadRequestPath(opts)
body = helps.ApplyPayloadConfigWithRoot(e.cfg, baseModel, to.String(), "", body, originalTranslated, requestedModel, requestPath)
body, _ = sjson.SetBytes(body, "model", baseModel)
body, _ = sjson.SetBytes(body, "stream", true)
body, _ = sjson.DeleteBytes(body, "previous_response_id")
body, _ = sjson.DeleteBytes(body, "prompt_cache_retention")
body, _ = sjson.DeleteBytes(body, "safety_identifier")
if !gjson.GetBytes(body, "instructions").Exists() {
body, _ = sjson.SetBytes(body, "instructions", "")
body = normalizeCodexInstructions(body)
if e.cfg == nil || e.cfg.DisableImageGeneration == config.DisableImageGenerationOff {
body = ensureImageGenerationTool(body, baseModel, auth)
}
httpURL := strings.TrimSuffix(baseURL, "/") + "/responses"
@@ -387,7 +389,12 @@ func (e *CodexWebsocketsExecutor) ExecuteStream(ctx context.Context, auth *clipr
}
requestedModel := helps.PayloadRequestedModel(opts, req.Model)
body = helps.ApplyPayloadConfigWithRoot(e.cfg, baseModel, to.String(), "", body, body, requestedModel)
requestPath := helps.PayloadRequestPath(opts)
body = helps.ApplyPayloadConfigWithRoot(e.cfg, baseModel, to.String(), "", body, body, requestedModel, requestPath)
body = normalizeCodexInstructions(body)
if e.cfg == nil || e.cfg.DisableImageGeneration == config.DisableImageGenerationOff {
body = ensureImageGenerationTool(body, baseModel, auth)
}
httpURL := strings.TrimSuffix(baseURL, "/") + "/responses"
wsURL, err := buildCodexResponsesWebsocketURL(httpURL)

View File

@@ -139,7 +139,8 @@ func (e *GeminiCLIExecutor) Execute(ctx context.Context, auth *cliproxyauth.Auth
basePayload = fixGeminiCLIImageAspectRatio(baseModel, basePayload)
requestedModel := helps.PayloadRequestedModel(opts, req.Model)
basePayload = helps.ApplyPayloadConfigWithRoot(e.cfg, baseModel, "gemini", "request", basePayload, originalTranslated, requestedModel)
requestPath := helps.PayloadRequestPath(opts)
basePayload = helps.ApplyPayloadConfigWithRoot(e.cfg, baseModel, "gemini", "request", basePayload, originalTranslated, requestedModel, requestPath)
action := "generateContent"
if req.Metadata != nil {
@@ -294,7 +295,8 @@ func (e *GeminiCLIExecutor) ExecuteStream(ctx context.Context, auth *cliproxyaut
basePayload = fixGeminiCLIImageAspectRatio(baseModel, basePayload)
requestedModel := helps.PayloadRequestedModel(opts, req.Model)
basePayload = helps.ApplyPayloadConfigWithRoot(e.cfg, baseModel, "gemini", "request", basePayload, originalTranslated, requestedModel)
requestPath := helps.PayloadRequestPath(opts)
basePayload = helps.ApplyPayloadConfigWithRoot(e.cfg, baseModel, "gemini", "request", basePayload, originalTranslated, requestedModel, requestPath)
projectID := resolveGeminiProjectID(auth)

View File

@@ -132,7 +132,8 @@ func (e *GeminiExecutor) Execute(ctx context.Context, auth *cliproxyauth.Auth, r
body = fixGeminiImageAspectRatio(baseModel, body)
requestedModel := helps.PayloadRequestedModel(opts, req.Model)
body = helps.ApplyPayloadConfigWithRoot(e.cfg, baseModel, to.String(), "", body, originalTranslated, requestedModel)
requestPath := helps.PayloadRequestPath(opts)
body = helps.ApplyPayloadConfigWithRoot(e.cfg, baseModel, to.String(), "", body, originalTranslated, requestedModel, requestPath)
body, _ = sjson.SetBytes(body, "model", baseModel)
action := "generateContent"
@@ -239,7 +240,8 @@ func (e *GeminiExecutor) ExecuteStream(ctx context.Context, auth *cliproxyauth.A
body = fixGeminiImageAspectRatio(baseModel, body)
requestedModel := helps.PayloadRequestedModel(opts, req.Model)
body = helps.ApplyPayloadConfigWithRoot(e.cfg, baseModel, to.String(), "", body, originalTranslated, requestedModel)
requestPath := helps.PayloadRequestPath(opts)
body = helps.ApplyPayloadConfigWithRoot(e.cfg, baseModel, to.String(), "", body, originalTranslated, requestedModel, requestPath)
body, _ = sjson.SetBytes(body, "model", baseModel)
baseURL := resolveGeminiBaseURL(auth)

View File

@@ -335,7 +335,8 @@ func (e *GeminiVertexExecutor) executeWithServiceAccount(ctx context.Context, au
body = fixGeminiImageAspectRatio(baseModel, body)
requestedModel := helps.PayloadRequestedModel(opts, req.Model)
body = helps.ApplyPayloadConfigWithRoot(e.cfg, baseModel, to.String(), "", body, originalTranslated, requestedModel)
requestPath := helps.PayloadRequestPath(opts)
body = helps.ApplyPayloadConfigWithRoot(e.cfg, baseModel, to.String(), "", body, originalTranslated, requestedModel, requestPath)
body, _ = sjson.SetBytes(body, "model", baseModel)
}
@@ -455,7 +456,8 @@ func (e *GeminiVertexExecutor) executeWithAPIKey(ctx context.Context, auth *clip
body = fixGeminiImageAspectRatio(baseModel, body)
requestedModel := helps.PayloadRequestedModel(opts, req.Model)
body = helps.ApplyPayloadConfigWithRoot(e.cfg, baseModel, to.String(), "", body, originalTranslated, requestedModel)
requestPath := helps.PayloadRequestPath(opts)
body = helps.ApplyPayloadConfigWithRoot(e.cfg, baseModel, to.String(), "", body, originalTranslated, requestedModel, requestPath)
body, _ = sjson.SetBytes(body, "model", baseModel)
action := getVertexAction(baseModel, false)
@@ -565,7 +567,8 @@ func (e *GeminiVertexExecutor) executeStreamWithServiceAccount(ctx context.Conte
body = fixGeminiImageAspectRatio(baseModel, body)
requestedModel := helps.PayloadRequestedModel(opts, req.Model)
body = helps.ApplyPayloadConfigWithRoot(e.cfg, baseModel, to.String(), "", body, originalTranslated, requestedModel)
requestPath := helps.PayloadRequestPath(opts)
body = helps.ApplyPayloadConfigWithRoot(e.cfg, baseModel, to.String(), "", body, originalTranslated, requestedModel, requestPath)
body, _ = sjson.SetBytes(body, "model", baseModel)
action := getVertexAction(baseModel, true)
@@ -694,7 +697,8 @@ func (e *GeminiVertexExecutor) executeStreamWithAPIKey(ctx context.Context, auth
body = fixGeminiImageAspectRatio(baseModel, body)
requestedModel := helps.PayloadRequestedModel(opts, req.Model)
body = helps.ApplyPayloadConfigWithRoot(e.cfg, baseModel, to.String(), "", body, originalTranslated, requestedModel)
requestPath := helps.PayloadRequestPath(opts)
body = helps.ApplyPayloadConfigWithRoot(e.cfg, baseModel, to.String(), "", body, originalTranslated, requestedModel, requestPath)
body, _ = sjson.SetBytes(body, "model", baseModel)
action := getVertexAction(baseModel, true)

View File

@@ -16,7 +16,8 @@ import (
// and restricts matches to the given protocol when supplied. Defaults are checked
// against the original payload when provided. requestedModel carries the client-visible
// model name before alias resolution so payload rules can target aliases precisely.
func ApplyPayloadConfigWithRoot(cfg *config.Config, model, protocol, root string, payload, original []byte, requestedModel string) []byte {
// requestPath is the inbound HTTP request path (when available) used for endpoint-scoped gates.
func ApplyPayloadConfigWithRoot(cfg *config.Config, model, protocol, root string, payload, original []byte, requestedModel string, requestPath string) []byte {
if cfg == nil || len(payload) == 0 {
return payload
}
@@ -149,13 +150,34 @@ func ApplyPayloadConfigWithRoot(cfg *config.Config, model, protocol, root string
}
}
if cfg.DisableImageGeneration {
if cfg.DisableImageGeneration != config.DisableImageGenerationOff {
if cfg.DisableImageGeneration == config.DisableImageGenerationChat && isImagesEndpointRequestPath(requestPath) {
return out
}
out = removeToolTypeFromPayloadWithRoot(out, root, "image_generation")
out = removeToolChoiceFromPayloadWithRoot(out, root, "image_generation")
}
return out
}
func isImagesEndpointRequestPath(path string) bool {
path = strings.TrimSpace(path)
if path == "" {
return false
}
if path == "/v1/images/generations" || path == "/v1/images/edits" {
return true
}
// Be tolerant of prefix routers that may report a longer matched route.
if strings.HasSuffix(path, "/v1/images/generations") || strings.HasSuffix(path, "/v1/images/edits") {
return true
}
if strings.HasSuffix(path, "/images/generations") || strings.HasSuffix(path, "/images/edits") {
return true
}
return false
}
func payloadModelRulesMatch(rules []config.PayloadModelRule, protocol string, models []string) bool {
if len(rules) == 0 || len(models) == 0 {
return false
@@ -367,6 +389,24 @@ func PayloadRequestedModel(opts cliproxyexecutor.Options, fallback string) strin
}
}
func PayloadRequestPath(opts cliproxyexecutor.Options) string {
if len(opts.Metadata) == 0 {
return ""
}
raw, ok := opts.Metadata[cliproxyexecutor.RequestPathMetadataKey]
if !ok || raw == nil {
return ""
}
switch v := raw.(type) {
case string:
return strings.TrimSpace(v)
case []byte:
return strings.TrimSpace(string(v))
default:
return ""
}
}
// matchModelPattern performs simple wildcard matching where '*' matches zero or more characters.
// Examples:
//

View File

@@ -9,11 +9,11 @@ import (
func TestApplyPayloadConfigWithRoot_DisableImageGeneration_RemovesToolsEntry(t *testing.T) {
cfg := &config.Config{
SDKConfig: config.SDKConfig{DisableImageGeneration: true},
SDKConfig: config.SDKConfig{DisableImageGeneration: config.DisableImageGenerationAll},
}
payload := []byte(`{"tools":[{"type":"image_generation","output_format":"png"},{"type":"function","name":"f1"}]}`)
out := ApplyPayloadConfigWithRoot(cfg, "gpt-5.4", "openai-response", "", payload, nil, "")
out := ApplyPayloadConfigWithRoot(cfg, "gpt-5.4", "openai-response", "", payload, nil, "", "")
tools := gjson.GetBytes(out, "tools")
if !tools.Exists() || !tools.IsArray() {
@@ -30,11 +30,11 @@ func TestApplyPayloadConfigWithRoot_DisableImageGeneration_RemovesToolsEntry(t *
func TestApplyPayloadConfigWithRoot_DisableImageGeneration_RemovesToolsEntryWithRoot(t *testing.T) {
cfg := &config.Config{
SDKConfig: config.SDKConfig{DisableImageGeneration: true},
SDKConfig: config.SDKConfig{DisableImageGeneration: config.DisableImageGenerationAll},
}
payload := []byte(`{"request":{"tools":[{"type":"image_generation"},{"type":"web_search"}]}}`)
out := ApplyPayloadConfigWithRoot(cfg, "gpt-5.4", "gemini-cli", "request", payload, nil, "")
out := ApplyPayloadConfigWithRoot(cfg, "gpt-5.4", "gemini-cli", "request", payload, nil, "", "")
tools := gjson.GetBytes(out, "request.tools")
if !tools.Exists() || !tools.IsArray() {
@@ -51,11 +51,11 @@ func TestApplyPayloadConfigWithRoot_DisableImageGeneration_RemovesToolsEntryWith
func TestApplyPayloadConfigWithRoot_DisableImageGeneration_RemovesToolChoiceByType(t *testing.T) {
cfg := &config.Config{
SDKConfig: config.SDKConfig{DisableImageGeneration: true},
SDKConfig: config.SDKConfig{DisableImageGeneration: config.DisableImageGenerationAll},
}
payload := []byte(`{"tools":[{"type":"image_generation"},{"type":"function","name":"f1"}],"tool_choice":{"type":"image_generation"}}`)
out := ApplyPayloadConfigWithRoot(cfg, "gpt-5.4", "openai-response", "", payload, nil, "")
out := ApplyPayloadConfigWithRoot(cfg, "gpt-5.4", "openai-response", "", payload, nil, "", "")
if gjson.GetBytes(out, "tool_choice").Exists() {
t.Fatalf("expected tool_choice to be removed")
@@ -64,13 +64,34 @@ func TestApplyPayloadConfigWithRoot_DisableImageGeneration_RemovesToolChoiceByTy
func TestApplyPayloadConfigWithRoot_DisableImageGeneration_RemovesToolChoiceByNameWithRoot(t *testing.T) {
cfg := &config.Config{
SDKConfig: config.SDKConfig{DisableImageGeneration: true},
SDKConfig: config.SDKConfig{DisableImageGeneration: config.DisableImageGenerationAll},
}
payload := []byte(`{"request":{"tools":[{"type":"image_generation"},{"type":"web_search"}],"tool_choice":{"type":"tool","name":"image_generation"}}}`)
out := ApplyPayloadConfigWithRoot(cfg, "gpt-5.4", "gemini-cli", "request", payload, nil, "")
out := ApplyPayloadConfigWithRoot(cfg, "gpt-5.4", "gemini-cli", "request", payload, nil, "", "")
if gjson.GetBytes(out, "request.tool_choice").Exists() {
t.Fatalf("expected request.tool_choice to be removed")
}
}
func TestApplyPayloadConfigWithRoot_DisableImageGenerationChat_KeepsImageGenerationOnImagesEndpoints(t *testing.T) {
cfg := &config.Config{
SDKConfig: config.SDKConfig{DisableImageGeneration: config.DisableImageGenerationChat},
}
payload := []byte(`{"tools":[{"type":"image_generation"},{"type":"function","name":"f1"}],"tool_choice":{"type":"image_generation"}}`)
out := ApplyPayloadConfigWithRoot(cfg, "gpt-5.4", "openai-response", "", payload, nil, "", "/v1/images/generations")
tools := gjson.GetBytes(out, "tools")
if !tools.Exists() || !tools.IsArray() {
t.Fatalf("expected tools array, got %v", tools.Type)
}
arr := tools.Array()
if len(arr) != 2 {
t.Fatalf("expected 2 tools (no removal), got %d", len(arr))
}
if !gjson.GetBytes(out, "tool_choice").Exists() {
t.Fatalf("expected tool_choice to be kept on images endpoint")
}
}

View File

@@ -108,7 +108,8 @@ func (e *KimiExecutor) Execute(ctx context.Context, auth *cliproxyauth.Auth, req
}
requestedModel := helps.PayloadRequestedModel(opts, req.Model)
body = helps.ApplyPayloadConfigWithRoot(e.cfg, baseModel, to.String(), "", body, originalTranslated, requestedModel)
requestPath := helps.PayloadRequestPath(opts)
body = helps.ApplyPayloadConfigWithRoot(e.cfg, baseModel, to.String(), "", body, originalTranslated, requestedModel, requestPath)
body, err = normalizeKimiToolMessageLinks(body)
if err != nil {
return resp, err
@@ -217,7 +218,8 @@ func (e *KimiExecutor) ExecuteStream(ctx context.Context, auth *cliproxyauth.Aut
return nil, fmt.Errorf("kimi executor: failed to set stream_options in payload: %w", err)
}
requestedModel := helps.PayloadRequestedModel(opts, req.Model)
body = helps.ApplyPayloadConfigWithRoot(e.cfg, baseModel, to.String(), "", body, originalTranslated, requestedModel)
requestPath := helps.PayloadRequestPath(opts)
body = helps.ApplyPayloadConfigWithRoot(e.cfg, baseModel, to.String(), "", body, originalTranslated, requestedModel, requestPath)
body, err = normalizeKimiToolMessageLinks(body)
if err != nil {
return nil, err

View File

@@ -97,7 +97,8 @@ func (e *OpenAICompatExecutor) Execute(ctx context.Context, auth *cliproxyauth.A
originalTranslated := sdktranslator.TranslateRequest(from, to, baseModel, originalPayload, opts.Stream)
translated := sdktranslator.TranslateRequest(from, to, baseModel, req.Payload, opts.Stream)
requestedModel := helps.PayloadRequestedModel(opts, req.Model)
translated = helps.ApplyPayloadConfigWithRoot(e.cfg, baseModel, to.String(), "", translated, originalTranslated, requestedModel)
requestPath := helps.PayloadRequestPath(opts)
translated = helps.ApplyPayloadConfigWithRoot(e.cfg, baseModel, to.String(), "", translated, originalTranslated, requestedModel, requestPath)
if opts.Alt == "responses/compact" {
if updated, errDelete := sjson.DeleteBytes(translated, "stream"); errDelete == nil {
translated = updated
@@ -199,7 +200,8 @@ func (e *OpenAICompatExecutor) ExecuteStream(ctx context.Context, auth *cliproxy
originalTranslated := sdktranslator.TranslateRequest(from, to, baseModel, originalPayload, true)
translated := sdktranslator.TranslateRequest(from, to, baseModel, req.Payload, true)
requestedModel := helps.PayloadRequestedModel(opts, req.Model)
translated = helps.ApplyPayloadConfigWithRoot(e.cfg, baseModel, to.String(), "", translated, originalTranslated, requestedModel)
requestPath := helps.PayloadRequestPath(opts)
translated = helps.ApplyPayloadConfigWithRoot(e.cfg, baseModel, to.String(), "", translated, originalTranslated, requestedModel, requestPath)
translated, err = thinking.ApplyThinking(translated, req.Model, from.String(), to.String(), e.Identifier())
if err != nil {

View File

@@ -43,7 +43,7 @@ func BuildConfigChangeDetails(oldCfg, newCfg *config.Config) []string {
changes = append(changes, fmt.Sprintf("disable-cooling: %t -> %t", oldCfg.DisableCooling, newCfg.DisableCooling))
}
if oldCfg.DisableImageGeneration != newCfg.DisableImageGeneration {
changes = append(changes, fmt.Sprintf("disable-image-generation: %t -> %t", oldCfg.DisableImageGeneration, newCfg.DisableImageGeneration))
changes = append(changes, fmt.Sprintf("disable-image-generation: %v -> %v", oldCfg.DisableImageGeneration, newCfg.DisableImageGeneration))
}
if oldCfg.RequestLog != newCfg.RequestLog {
changes = append(changes, fmt.Sprintf("request-log: %t -> %t", oldCfg.RequestLog, newCfg.RequestLog))

View File

@@ -279,7 +279,7 @@ func TestBuildConfigChangeDetails_FlagsAndKeys(t *testing.T) {
APIKeys: []string{" key-1 ", "key-2"},
ForceModelPrefix: true,
NonStreamKeepAliveInterval: 5,
DisableImageGeneration: true,
DisableImageGeneration: config.DisableImageGenerationAll,
},
}
@@ -408,7 +408,7 @@ func TestBuildConfigChangeDetails_AllBranches(t *testing.T) {
RequestLog: true,
ProxyURL: "http://new-proxy",
APIKeys: []string{"keyB"},
DisableImageGeneration: true,
DisableImageGeneration: config.DisableImageGenerationAll,
},
OAuthExcludedModels: map[string][]string{"p1": {"b", "c"}, "p2": {"d"}},
OpenAICompatibility: []config.OpenAICompatibility{