sub2api/backend/internal/service/openai_codex_transform.go

package service

import (
	"fmt"
	"strings"
)

var codexModelMap = map[string]string{
	"gpt-5.5":                    "gpt-5.5",
	"gpt-5.4":                    "gpt-5.4",
	"gpt-5.4-mini":               "gpt-5.4-mini",
	"gpt-5.4-none":               "gpt-5.4",
	"gpt-5.4-low":                "gpt-5.4",
	"gpt-5.4-medium":             "gpt-5.4",
	"gpt-5.4-high":               "gpt-5.4",
	"gpt-5.4-xhigh":              "gpt-5.4",
	"gpt-5.4-chat-latest":        "gpt-5.4",
	"gpt-5.3":                    "gpt-5.3-codex",
	"gpt-5.3-none":               "gpt-5.3-codex",
	"gpt-5.3-low":                "gpt-5.3-codex",
	"gpt-5.3-medium":             "gpt-5.3-codex",
	"gpt-5.3-high":               "gpt-5.3-codex",
	"gpt-5.3-xhigh":              "gpt-5.3-codex",
	"gpt-5.3-codex":              "gpt-5.3-codex",
	"gpt-5.3-codex-spark":        "gpt-5.3-codex-spark",
	"gpt-5.3-codex-spark-low":    "gpt-5.3-codex-spark",
	"gpt-5.3-codex-spark-medium": "gpt-5.3-codex-spark",
	"gpt-5.3-codex-spark-high":   "gpt-5.3-codex-spark",
	"gpt-5.3-codex-spark-xhigh":  "gpt-5.3-codex-spark",
	"gpt-5.3-codex-low":          "gpt-5.3-codex",
	"gpt-5.3-codex-medium":       "gpt-5.3-codex",
	"gpt-5.3-codex-high":         "gpt-5.3-codex",
	"gpt-5.3-codex-xhigh":        "gpt-5.3-codex",
	"gpt-5.2":                    "gpt-5.2",
	"gpt-5.2-none":               "gpt-5.2",
	"gpt-5.2-low":                "gpt-5.2",
	"gpt-5.2-medium":             "gpt-5.2",
	"gpt-5.2-high":               "gpt-5.2",
	"gpt-5.2-xhigh":              "gpt-5.2",
}

type codexTransformResult struct {
	Modified        bool
	NormalizedModel string
	PromptCacheKey  string
}

const (
	codexImageGenerationBridgeMarker = "<sub2api-codex-image-generation>"
	codexImageGenerationBridgeText   = codexImageGenerationBridgeMarker + "\nWhen the user asks for raster image generation or editing, use the OpenAI Responses native `image_generation` tool attached to this request. The local Codex client may not expose an `image_gen` namespace, but that does not mean image generation is unavailable. Do not ask the user to switch to CLI fallback solely because `image_gen` is absent.\n</sub2api-codex-image-generation>"
)

func applyCodexOAuthTransform(reqBody map[string]any, isCodexCLI bool, isCompact bool) codexTransformResult {
	result := codexTransformResult{}
	// 工具续链需求会影响存储策略与 input 过滤逻辑。
	needsToolContinuation := NeedsToolContinuation(reqBody)

	model := ""
	if v, ok := reqBody["model"].(string); ok {
		model = v
	}
	normalizedModel := strings.TrimSpace(model)
	if normalizedModel != "" {
		if model != normalizedModel {
			reqBody["model"] = normalizedModel
			result.Modified = true
		}
		result.NormalizedModel = normalizedModel
	}

	if isCompact {
		if _, ok := reqBody["store"]; ok {
			delete(reqBody, "store")
			result.Modified = true
		}
		if _, ok := reqBody["stream"]; ok {
			delete(reqBody, "stream")
			result.Modified = true
		}
	} else {
		// OAuth 走 ChatGPT internal API 时，store 必须为 false；显式 true 也会强制覆盖。
		// 避免上游返回 "Store must be set to false"。
		if v, ok := reqBody["store"].(bool); !ok || v {
			reqBody["store"] = false
			result.Modified = true
		}
		if v, ok := reqBody["stream"].(bool); !ok || !v {
			reqBody["stream"] = true
			result.Modified = true
		}
	}

	// Strip parameters unsupported by codex models via the Responses API.
	for _, key := range []string{
		"max_output_tokens",
		"max_completion_tokens",
		"temperature",
		"top_p",
		"frequency_penalty",
		"presence_penalty",
		// prompt_cache_retention is a newer Responses API parameter (cache TTL).
		// The ChatGPT internal Codex endpoint rejects it with
		// "Unsupported parameter: prompt_cache_retention". Defense-in-depth
		// for any OAuth path that reaches this transform — the Cursor
		// Responses-shape short-circuit in ForwardAsChatCompletions strips
		// it earlier too, but we keep this line so other OAuth callers are
		// equally protected.
		"prompt_cache_retention",
	} {
		if _, ok := reqBody[key]; ok {
			delete(reqBody, key)
			result.Modified = true
		}
	}

	// 兼容遗留的 functions 和 function_call，转换为 tools 和 tool_choice
	if functionsRaw, ok := reqBody["functions"]; ok {
		if functions, k := functionsRaw.([]any); k {
			tools := make([]any, 0, len(functions))
			for _, f := range functions {
				tools = append(tools, map[string]any{
					"type":     "function",
					"function": f,
				})
			}
			reqBody["tools"] = tools
		}
		delete(reqBody, "functions")
		result.Modified = true
	}

	if fcRaw, ok := reqBody["function_call"]; ok {
		if fcStr, ok := fcRaw.(string); ok {
			// e.g. "auto", "none"
			reqBody["tool_choice"] = fcStr
		} else if fcObj, ok := fcRaw.(map[string]any); ok {
			// e.g. {"name": "my_func"}
			if name, ok := fcObj["name"].(string); ok && strings.TrimSpace(name) != "" {
				reqBody["tool_choice"] = map[string]any{
					"type": "function",
					"function": map[string]any{
						"name": name,
					},
				}
			}
		}
		delete(reqBody, "function_call")
		result.Modified = true
	}

	if normalizeCodexTools(reqBody) {
		result.Modified = true
	}

	if v, ok := reqBody["prompt_cache_key"].(string); ok {
		result.PromptCacheKey = strings.TrimSpace(v)
	}

	// 提取 input 中 role:"system" 消息至 instructions（OAuth 上游不支持 system role）。
	if extractSystemMessagesFromInput(reqBody) {
		result.Modified = true
	}

	// instructions 处理逻辑：根据是否是 Codex CLI 分别调用不同方法
	if applyInstructions(reqBody, isCodexCLI) {
		result.Modified = true
	}

	// 续链场景保留 item_reference 与 id，避免 call_id 上下文丢失。
	if input, ok := reqBody["input"].([]any); ok {
		input = filterCodexInput(input, needsToolContinuation)
		reqBody["input"] = input
		result.Modified = true
	} else if inputStr, ok := reqBody["input"].(string); ok {
		// ChatGPT codex endpoint requires input to be a list, not a string.
		// Convert string input to the expected message array format.
		trimmed := strings.TrimSpace(inputStr)
		if trimmed != "" {
			reqBody["input"] = []any{
				map[string]any{
					"type":    "message",
					"role":    "user",
					"content": inputStr,
				},
			}
		} else {
			reqBody["input"] = []any{}
		}
		result.Modified = true
	}

	return result
}

func normalizeCodexModel(model string) string {
	model = strings.TrimSpace(model)
	if model == "" {
		return "gpt-5.4"
	}
	if isOpenAIImageGenerationModel(model) {
		return model
	}

	modelID := model
	if strings.Contains(modelID, "/") {
		parts := strings.Split(modelID, "/")
		modelID = parts[len(parts)-1]
	}

	if mapped := getNormalizedCodexModel(modelID); mapped != "" {
		return mapped
	}

	normalized := strings.ToLower(modelID)

	if strings.Contains(normalized, "gpt-5.5") || strings.Contains(normalized, "gpt 5.5") {
		return "gpt-5.5"
	}
	if strings.Contains(normalized, "gpt-5.4-mini") || strings.Contains(normalized, "gpt 5.4 mini") {
		return "gpt-5.4-mini"
	}
	if strings.Contains(normalized, "gpt-5.4") || strings.Contains(normalized, "gpt 5.4") {
		return "gpt-5.4"
	}
	if strings.Contains(normalized, "gpt-5.2") || strings.Contains(normalized, "gpt 5.2") {
		return "gpt-5.2"
	}
	if strings.Contains(normalized, "gpt-5.3-codex-spark") || strings.Contains(normalized, "gpt 5.3 codex spark") {
		return "gpt-5.3-codex-spark"
	}
	if strings.Contains(normalized, "gpt-5.3-codex") || strings.Contains(normalized, "gpt 5.3 codex") {
		return "gpt-5.3-codex"
	}
	if strings.Contains(normalized, "gpt-5.3") || strings.Contains(normalized, "gpt 5.3") {
		return "gpt-5.3-codex"
	}
	if strings.Contains(normalized, "codex") {
		return "gpt-5.3-codex"
	}
	if strings.Contains(normalized, "gpt-5") || strings.Contains(normalized, "gpt 5") {
		return "gpt-5.4"
	}

	return "gpt-5.4"
}

func hasOpenAIImageGenerationTool(reqBody map[string]any) bool {
	rawTools, ok := reqBody["tools"]
	if !ok || rawTools == nil {
		return false
	}
	tools, ok := rawTools.([]any)
	if !ok {
		return false
	}
	for _, rawTool := range tools {
		toolMap, ok := rawTool.(map[string]any)
		if !ok {
			continue
		}
		if strings.TrimSpace(firstNonEmptyString(toolMap["type"])) == "image_generation" {
			return true
		}
	}
	return false
}

func normalizeOpenAIResponsesImageGenerationTools(reqBody map[string]any) bool {
	rawTools, ok := reqBody["tools"]
	if !ok || rawTools == nil {
		return false
	}
	tools, ok := rawTools.([]any)
	if !ok {
		return false
	}

	modified := false
	for _, rawTool := range tools {
		toolMap, ok := rawTool.(map[string]any)
		if !ok || strings.TrimSpace(firstNonEmptyString(toolMap["type"])) != "image_generation" {
			continue
		}
		if _, ok := toolMap["output_format"]; !ok {
			if value := strings.TrimSpace(firstNonEmptyString(toolMap["format"])); value != "" {
				toolMap["output_format"] = value
				modified = true
			}
		}
		if _, ok := toolMap["output_compression"]; !ok {
			if value, exists := toolMap["compression"]; exists && value != nil {
				toolMap["output_compression"] = value
				modified = true
			}
		}
		if _, ok := toolMap["format"]; ok {
			delete(toolMap, "format")
			modified = true
		}
		if _, ok := toolMap["compression"]; ok {
			delete(toolMap, "compression")
			modified = true
		}
	}
	return modified
}

func ensureOpenAIResponsesImageGenerationTool(reqBody map[string]any) bool {
	if len(reqBody) == 0 {
		return false
	}

	tool := map[string]any{
		"type":          "image_generation",
		"output_format": "png",
	}

	rawTools, ok := reqBody["tools"]
	if !ok || rawTools == nil {
		reqBody["tools"] = []any{tool}
		return true
	}

	tools, ok := rawTools.([]any)
	if !ok {
		reqBody["tools"] = []any{tool}
		return true
	}
	for _, rawTool := range tools {
		toolMap, ok := rawTool.(map[string]any)
		if !ok {
			continue
		}
		if strings.TrimSpace(firstNonEmptyString(toolMap["type"])) == "image_generation" {
			return false
		}
	}

	reqBody["tools"] = append(tools, tool)
	return true
}

func applyCodexImageGenerationBridgeInstructions(reqBody map[string]any) bool {
	if len(reqBody) == 0 || !hasOpenAIImageGenerationTool(reqBody) {
		return false
	}

	existing, _ := reqBody["instructions"].(string)
	if strings.Contains(existing, codexImageGenerationBridgeMarker) {
		return false
	}

	existing = strings.TrimRight(existing, " \t\r\n")
	if strings.TrimSpace(existing) == "" {
		reqBody["instructions"] = codexImageGenerationBridgeText
		return true
	}

	reqBody["instructions"] = existing + "\n\n" + codexImageGenerationBridgeText
	return true
}

func validateOpenAIResponsesImageModel(reqBody map[string]any, model string) error {
	if !hasOpenAIImageGenerationTool(reqBody) {
		return nil
	}
	model = strings.TrimSpace(model)
	if !isOpenAIImageGenerationModel(model) {
		return nil
	}
	return fmt.Errorf("/v1/responses image_generation requests require a Responses-capable text model; image-only model %q is not allowed", model)
}

func normalizeOpenAIResponsesImageOnlyModel(reqBody map[string]any) bool {
	if len(reqBody) == 0 {
		return false
	}
	imageModel := strings.TrimSpace(firstNonEmptyString(reqBody["model"]))
	if !isOpenAIImageGenerationModel(imageModel) {
		return false
	}

	modified := false
	tools, _ := reqBody["tools"].([]any)
	imageToolIndex := -1
	for i, rawTool := range tools {
		toolMap, ok := rawTool.(map[string]any)
		if !ok {
			continue
		}
		if strings.TrimSpace(firstNonEmptyString(toolMap["type"])) == "image_generation" {
			imageToolIndex = i
			break
		}
	}
	if imageToolIndex < 0 {
		tools = append(tools, map[string]any{
			"type":  "image_generation",
			"model": imageModel,
		})
		imageToolIndex = len(tools) - 1
		reqBody["tools"] = tools
		modified = true
	}

	if toolMap, ok := tools[imageToolIndex].(map[string]any); ok {
		if strings.TrimSpace(firstNonEmptyString(toolMap["model"])) == "" {
			toolMap["model"] = imageModel
			modified = true
		}
		for _, key := range []string{
			"size",
			"quality",
			"background",
			"output_format",
			"output_compression",
			"moderation",
			"style",
			"partial_images",
		} {
			if value, exists := reqBody[key]; exists && value != nil {
				if _, toolHas := toolMap[key]; !toolHas {
					toolMap[key] = value
				}
				delete(reqBody, key)
				modified = true
			}
		}
	}

	if prompt := strings.TrimSpace(firstNonEmptyString(reqBody["prompt"])); prompt != "" {
		if _, hasInput := reqBody["input"]; !hasInput {
			reqBody["input"] = prompt
		}
		delete(reqBody, "prompt")
		modified = true
	}

	if _, ok := reqBody["tool_choice"]; !ok {
		reqBody["tool_choice"] = map[string]any{"type": "image_generation"}
		modified = true
	}
	if imageModel != openAIImagesResponsesMainModel {
		modified = true
	}
	reqBody["model"] = openAIImagesResponsesMainModel
	return modified
}

func normalizeOpenAIModelForUpstream(account *Account, model string) string {
	if account == nil || account.Type == AccountTypeOAuth {
		return normalizeCodexModel(model)
	}
	return strings.TrimSpace(model)
}

func SupportsVerbosity(model string) bool {
	if !strings.HasPrefix(model, "gpt-") {
		return true
	}

	var major, minor int
	n, _ := fmt.Sscanf(model, "gpt-%d.%d", &major, &minor)

	if major > 5 {
		return true
	}
	if major < 5 {
		return false
	}

	// gpt-5
	if n == 1 {
		return true
	}

	return minor >= 3
}

func getNormalizedCodexModel(modelID string) string {
	if modelID == "" {
		return ""
	}
	if mapped, ok := codexModelMap[modelID]; ok {
		return mapped
	}
	lower := strings.ToLower(modelID)
	for key, value := range codexModelMap {
		if strings.ToLower(key) == lower {
			return value
		}
	}
	return ""
}

// extractTextFromContent extracts plain text from a content value that is either
// a Go string or a []any of content-part maps with type:"text".
func extractTextFromContent(content any) string {
	switch v := content.(type) {
	case string:
		return v
	case []any:
		var parts []string
		for _, part := range v {
			m, ok := part.(map[string]any)
			if !ok {
				continue
			}
			if t, _ := m["type"].(string); t == "text" {
				if text, ok := m["text"].(string); ok {
					parts = append(parts, text)
				}
			}
		}
		return strings.Join(parts, "")
	default:
		return ""
	}
}

// extractSystemMessagesFromInput scans the input array for items with role=="system",
// removes them, and merges their content into reqBody["instructions"].
// If instructions is already non-empty, extracted content is prepended with "\n\n".
// Returns true if any system messages were extracted.
func extractSystemMessagesFromInput(reqBody map[string]any) bool {
	input, ok := reqBody["input"].([]any)
	if !ok || len(input) == 0 {
		return false
	}

	var systemTexts []string
	remaining := make([]any, 0, len(input))

	for _, item := range input {
		m, ok := item.(map[string]any)
		if !ok {
			remaining = append(remaining, item)
			continue
		}
		if role, _ := m["role"].(string); role != "system" {
			remaining = append(remaining, item)
			continue
		}
		if text := extractTextFromContent(m["content"]); text != "" {
			systemTexts = append(systemTexts, text)
		}
	}

	if len(systemTexts) == 0 {
		return false
	}

	extracted := strings.Join(systemTexts, "\n\n")
	if existing, ok := reqBody["instructions"].(string); ok && strings.TrimSpace(existing) != "" {
		reqBody["instructions"] = extracted + "\n\n" + existing
	} else {
		reqBody["instructions"] = extracted
	}
	reqBody["input"] = remaining
	return true
}

// applyInstructions 处理 instructions 字段：仅在 instructions 为空时填充默认值。
func applyInstructions(reqBody map[string]any, isCodexCLI bool) bool {
	if !isInstructionsEmpty(reqBody) {
		return false
	}
	reqBody["instructions"] = "You are a helpful coding assistant."
	return true
}

// isInstructionsEmpty 检查 instructions 字段是否为空
// 处理以下情况：字段不存在、nil、空字符串、纯空白字符串
func isInstructionsEmpty(reqBody map[string]any) bool {
	val, exists := reqBody["instructions"]
	if !exists {
		return true
	}
	if val == nil {
		return true
	}
	str, ok := val.(string)
	if !ok {
		return true
	}
	return strings.TrimSpace(str) == ""
}

// filterCodexInput 按需过滤 item_reference 与 id。
// preserveReferences 为 true 时保持引用与 id，以满足续链请求对上下文的依赖。
func filterCodexInput(input []any, preserveReferences bool) []any {
	filtered := make([]any, 0, len(input))
	for _, item := range input {
		m, ok := item.(map[string]any)
		if !ok {
			filtered = append(filtered, item)
			continue
		}
		typ, _ := m["type"].(string)

		// 仅修正真正的 tool/function call 标识，避免误改普通 message/reasoning id；
		// 若 item_reference 指向 legacy call_* 标识，则仅修正该引用本身。
		fixCallIDPrefix := func(id string) string {
			if id == "" || strings.HasPrefix(id, "fc") {
				return id
			}
			if strings.HasPrefix(id, "call_") {
				return "fc" + strings.TrimPrefix(id, "call_")
			}
			return "fc_" + id
		}

		if typ == "item_reference" {
			if !preserveReferences {
				continue
			}
			newItem := make(map[string]any, len(m))
			for key, value := range m {
				newItem[key] = value
			}
			if id, ok := newItem["id"].(string); ok && strings.HasPrefix(id, "call_") {
				newItem["id"] = fixCallIDPrefix(id)
			}
			filtered = append(filtered, newItem)
			continue
		}

		newItem := m
		copied := false
		// 仅在需要修改字段时创建副本，避免直接改写原始输入。
		ensureCopy := func() {
			if copied {
				return
			}
			newItem = make(map[string]any, len(m))
			for key, value := range m {
				newItem[key] = value
			}
			copied = true
		}

		if isCodexToolCallItemType(typ) {
			callID, ok := m["call_id"].(string)
			if !ok || strings.TrimSpace(callID) == "" {
				if id, ok := m["id"].(string); ok && strings.TrimSpace(id) != "" {
					callID = id
					ensureCopy()
					newItem["call_id"] = callID
				}
			}

			if callID != "" {
				fixedCallID := fixCallIDPrefix(callID)
				if fixedCallID != callID {
					ensureCopy()
					newItem["call_id"] = fixedCallID
				}
			}
		}

		if !isCodexToolCallItemType(typ) {
			ensureCopy()
			delete(newItem, "call_id")
		}

		if !preserveReferences {
			ensureCopy()
			delete(newItem, "id")
		}

		filtered = append(filtered, newItem)
	}
	return filtered
}

func isCodexToolCallItemType(typ string) bool {
	switch typ {
	case "function_call",
		"tool_call",
		"local_shell_call",
		"tool_search_call",
		"custom_tool_call",
		"function_call_output",
		"mcp_tool_call_output",
		"custom_tool_call_output",
		"tool_search_output":
		return true
	default:
		return false
	}
}

func normalizeCodexTools(reqBody map[string]any) bool {
	rawTools, ok := reqBody["tools"]
	if !ok || rawTools == nil {
		return false
	}
	tools, ok := rawTools.([]any)
	if !ok {
		return false
	}

	modified := false
	validTools := make([]any, 0, len(tools))

	for _, tool := range tools {
		toolMap, ok := tool.(map[string]any)
		if !ok {
			// Keep unknown structure as-is to avoid breaking upstream behavior.
			validTools = append(validTools, tool)
			continue
		}

		toolType, _ := toolMap["type"].(string)
		toolType = strings.TrimSpace(toolType)
		if toolType != "function" {
			validTools = append(validTools, toolMap)
			continue
		}

		// OpenAI Responses-style tools use top-level name/parameters.
		if name, ok := toolMap["name"].(string); ok && strings.TrimSpace(name) != "" {
			validTools = append(validTools, toolMap)
			continue
		}

		// ChatCompletions-style tools use {type:"function", function:{...}}.
		functionValue, hasFunction := toolMap["function"]
		function, ok := functionValue.(map[string]any)
		if !hasFunction || functionValue == nil || !ok || function == nil {
			// Drop invalid function tools.
			modified = true
			continue
		}

		if _, ok := toolMap["name"]; !ok {
			if name, ok := function["name"].(string); ok && strings.TrimSpace(name) != "" {
				toolMap["name"] = name
				modified = true
			}
		}
		if _, ok := toolMap["description"]; !ok {
			if desc, ok := function["description"].(string); ok && strings.TrimSpace(desc) != "" {
				toolMap["description"] = desc
				modified = true
			}
		}
		if _, ok := toolMap["parameters"]; !ok {
			if params, ok := function["parameters"]; ok {
				toolMap["parameters"] = params
				modified = true
			}
		}
		if _, ok := toolMap["strict"]; !ok {
			if strict, ok := function["strict"]; ok {
				toolMap["strict"] = strict
				modified = true
			}
		}

		validTools = append(validTools, toolMap)
	}

	if modified {
		reqBody["tools"] = validTools
	}

	return modified
}