feat: 修复 Codex 缓存 token 计费并优化代码结构

- 修复 Codex API 缓存 token 重复计费问题 (input_tokens 包含 cached_tokens)
- 提取共享代码到 src/providers/shared/usage-utils.ts，消除重复
- 前端定价页面增加缓存倍率配置列
- 修复 var 为 const，favicon 改为云图标

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>

src/admin/token_utils.ts

@@ -31,23 +31,28 @@ export const TokenUtils = {
     async processUsage(c: Context<HonoCustomType>, apiKey: string, model: string, targetChannelKey: string, targetChannelConfig: ChannelConfig, usage: Usage): Promise<void> {
         console.log("Usage data:", usage);
 
-        // Get pricing and calculate cost
         const pricing = await this.getPricing(c, model, targetChannelConfig);
         const hasTokens = usage.prompt_tokens != null && usage.completion_tokens != null;
 
         if (pricing && hasTokens) {
             const inputCost = usage.prompt_tokens! * pricing.input;
             const outputCost = usage.completion_tokens! * pricing.output;
-            const totalCost = inputCost + outputCost;
-            // Update token usage and log data
+
+            let cacheCost = 0;
+            if (usage.cached_tokens && usage.cached_tokens > 0 && pricing.cache) {
+                cacheCost = usage.cached_tokens * pricing.cache;
+            }
+
+            const totalCost = inputCost + outputCost + cacheCost;
+
             await this.updateUsage(c, apiKey, totalCost);
-            // 长度不够直接显示 *
-            var maskedApiKey = apiKey.length < 3 ? '*'.repeat(apiKey.length) : (
+
+            const maskedApiKey = apiKey.length < 3 ? '*'.repeat(apiKey.length) : (
                 apiKey.slice(0, apiKey.length / 3)
                 + '*'.repeat(apiKey.length / 3)
                 + apiKey.slice((2 * apiKey.length) / 3)
             );
-            console.log(`Model: ${model}, Channel: ${targetChannelKey}, apiKey: ${maskedApiKey}, Cost: ${totalCost}`);
+            console.log(`Model: ${model}, Channel: ${targetChannelKey}, apiKey: ${maskedApiKey}, Cost: ${totalCost} (input: ${inputCost}, cache: ${cacheCost}, output: ${outputCost})`);
         } else {
             console.warn(`No pricing found for model: ${model} in channel: ${targetChannelKey}`);
         }

src/providers/azure-openai-responses-proxy.ts

@@ -1,28 +1,10 @@
 import { Context } from "hono"
-
-type OpenAIResponsesUsage = {
-    input_tokens?: number;
-    output_tokens?: number;
-    total_tokens?: number;
-    prompt_tokens?: number;
-    completion_tokens?: number;
-}
-
-type OpenAIResponsesResponse = {
-    usage?: OpenAIResponsesUsage;
-    input_tokens?: number;
-    output_tokens?: number;
-    total_tokens?: number;
-    prompt_tokens?: number;
-    completion_tokens?: number;
-}
-
-type OpenAIResponsesStreamEvent = {
-    type?: string;
-    response?: OpenAIResponsesResponse;
-    delta?: string;
-    text?: string;
-}
+import {
+    OpenAIResponsesResponse,
+    OpenAIResponsesStreamEvent,
+    extractUsageFromResponse,
+    estimateUsageFromBodies,
+} from "./shared/usage-utils"
 
 const buildProxyRequest = (
     request: Request,
@@ -52,69 +34,6 @@ const buildProxyRequest = (
     })
 }
 
-const normalizeUsage = (raw?: OpenAIResponsesUsage): Usage | null => {
-    if (!raw) return null
-    const promptTokens = raw.prompt_tokens ?? raw.input_tokens
-    const completionTokens = raw.completion_tokens ?? raw.output_tokens
-    const totalTokens = raw.total_tokens ?? (
-        (promptTokens ?? 0) + (completionTokens ?? 0)
-    )
-    if (promptTokens == null && completionTokens == null && totalTokens == null) {
-        return null
-    }
-    return {
-        prompt_tokens: promptTokens,
-        completion_tokens: completionTokens,
-        total_tokens: totalTokens,
-    }
-}
-
-const extractUsageFromResponse = (res?: OpenAIResponsesResponse): Usage | null => {
-    if (!res) return null
-    return normalizeUsage(res.usage || res)
-}
-
-const estimateTokensFromText = (text: string): number => {
-    if (!text) return 0
-    const bytes = new TextEncoder().encode(text).length
-    return Math.ceil(bytes / 4)
-}
-
-const estimateUsageFromBodies = (
-    requestBody: any,
-    responseBody?: any,
-    streamedOutputText?: string
-): Usage | null => {
-    const promptText = requestBody ? JSON.stringify(requestBody) : ""
-    const completionText = streamedOutputText
-        ? streamedOutputText
-        : (responseBody ? JSON.stringify(responseBody) : "")
-    const promptTokens = estimateTokensFromText(promptText)
-    const completionTokens = estimateTokensFromText(completionText)
-    if (!promptTokens && !completionTokens) return null
-    return {
-        prompt_tokens: promptTokens,
-        completion_tokens: completionTokens,
-        total_tokens: promptTokens + completionTokens,
-    }
-}
-
-const checkoutUsageData = async (
-    saveUsage: (usage: Usage) => Promise<void>,
-    response: Response | OpenAIResponsesResponse
-): Promise<void> => {
-    try {
-        const res = response instanceof Response
-            ? await response.clone().json<OpenAIResponsesResponse>()
-            : response
-        const usage = extractUsageFromResponse(res)
-        if (!usage) return
-        await saveUsage(usage)
-    } catch (error) {
-        console.error("Error logging usage data:", error)
-    }
-}
-
 const processStreamData = async (
     lines: string[],
     usageSaved: { value: boolean },

src/providers/openai-responses-proxy.ts

@@ -1,28 +1,10 @@
 import { Context } from "hono"
-
-type OpenAIResponsesUsage = {
-    input_tokens?: number;
-    output_tokens?: number;
-    total_tokens?: number;
-    prompt_tokens?: number;
-    completion_tokens?: number;
-}
-
-type OpenAIResponsesResponse = {
-    usage?: OpenAIResponsesUsage;
-    input_tokens?: number;
-    output_tokens?: number;
-    total_tokens?: number;
-    prompt_tokens?: number;
-    completion_tokens?: number;
-}
-
-type OpenAIResponsesStreamEvent = {
-    type?: string;
-    response?: OpenAIResponsesResponse;
-    delta?: string;
-    text?: string;
-}
+import {
+    OpenAIResponsesResponse,
+    OpenAIResponsesStreamEvent,
+    extractUsageFromResponse,
+    estimateUsageFromBodies,
+} from "./shared/usage-utils"
 
 const buildProxyRequest = (
     request: Request,
@@ -44,69 +26,6 @@ const buildProxyRequest = (
     })
 }
 
-const normalizeUsage = (raw?: OpenAIResponsesUsage): Usage | null => {
-    if (!raw) return null
-    const promptTokens = raw.prompt_tokens ?? raw.input_tokens
-    const completionTokens = raw.completion_tokens ?? raw.output_tokens
-    const totalTokens = raw.total_tokens ?? (
-        (promptTokens ?? 0) + (completionTokens ?? 0)
-    )
-    if (promptTokens == null && completionTokens == null && totalTokens == null) {
-        return null
-    }
-    return {
-        prompt_tokens: promptTokens,
-        completion_tokens: completionTokens,
-        total_tokens: totalTokens,
-    }
-}
-
-const extractUsageFromResponse = (res?: OpenAIResponsesResponse): Usage | null => {
-    if (!res) return null
-    return normalizeUsage(res.usage || res)
-}
-
-const estimateTokensFromText = (text: string): number => {
-    if (!text) return 0
-    const bytes = new TextEncoder().encode(text).length
-    return Math.ceil(bytes / 4)
-}
-
-const estimateUsageFromBodies = (
-    requestBody: any,
-    responseBody?: any,
-    streamedOutputText?: string
-): Usage | null => {
-    const promptText = requestBody ? JSON.stringify(requestBody) : ""
-    const completionText = streamedOutputText
-        ? streamedOutputText
-        : (responseBody ? JSON.stringify(responseBody) : "")
-    const promptTokens = estimateTokensFromText(promptText)
-    const completionTokens = estimateTokensFromText(completionText)
-    if (!promptTokens && !completionTokens) return null
-    return {
-        prompt_tokens: promptTokens,
-        completion_tokens: completionTokens,
-        total_tokens: promptTokens + completionTokens,
-    }
-}
-
-const checkoutUsageData = async (
-    saveUsage: (usage: Usage) => Promise<void>,
-    response: Response | OpenAIResponsesResponse
-): Promise<void> => {
-    try {
-        const res = response instanceof Response
-            ? await response.clone().json<OpenAIResponsesResponse>()
-            : response
-        const usage = extractUsageFromResponse(res)
-        if (!usage) return
-        await saveUsage(usage)
-    } catch (error) {
-        console.error("Error logging usage data:", error)
-    }
-}
-
 const processStreamData = async (
     lines: string[],
     usageSaved: { value: boolean },

src/providers/shared/usage-utils.ts

@@ -0,0 +1,103 @@
+export type InputTokensDetails = {
+    cached_tokens?: number;
+}
+
+export type OpenAIResponsesUsage = {
+    input_tokens?: number;
+    output_tokens?: number;
+    total_tokens?: number;
+    prompt_tokens?: number;
+    completion_tokens?: number;
+    input_tokens_details?: InputTokensDetails;
+    prompt_tokens_details?: InputTokensDetails;
+}
+
+export type OpenAIResponsesResponse = {
+    usage?: OpenAIResponsesUsage;
+    input_tokens?: number;
+    output_tokens?: number;
+    total_tokens?: number;
+    prompt_tokens?: number;
+    completion_tokens?: number;
+    input_tokens_details?: InputTokensDetails;
+    prompt_tokens_details?: InputTokensDetails;
+}
+
+export type OpenAIResponsesStreamEvent = {
+    type?: string;
+    response?: OpenAIResponsesResponse;
+    delta?: string;
+    text?: string;
+}
+
+export const normalizeUsage = (raw?: OpenAIResponsesUsage): Usage | null => {
+    if (!raw) return null
+
+    const cachedTokens =
+        raw.input_tokens_details?.cached_tokens ??
+        raw.prompt_tokens_details?.cached_tokens ??
+        0
+
+    const rawInputTokens = raw.prompt_tokens ?? raw.input_tokens ?? 0
+
+    const actualInputTokens = cachedTokens > 0
+        ? Math.max(0, rawInputTokens - cachedTokens)
+        : rawInputTokens
+
+    const completionTokens = raw.completion_tokens ?? raw.output_tokens
+    const totalTokens = raw.total_tokens ?? (
+        (actualInputTokens ?? 0) + (cachedTokens ?? 0) + (completionTokens ?? 0)
+    )
+
+    if (actualInputTokens == null && completionTokens == null && totalTokens == null) {
+        return null
+    }
+
+    return {
+        prompt_tokens: actualInputTokens,
+        completion_tokens: completionTokens,
+        total_tokens: totalTokens,
+        cached_tokens: cachedTokens > 0 ? cachedTokens : undefined,
+    }
+}
+
+export const extractUsageFromResponse = (res?: OpenAIResponsesResponse): Usage | null => {
+    if (!res) return null
+
+    const usageData: OpenAIResponsesUsage = {
+        input_tokens: res.usage?.input_tokens ?? res.input_tokens,
+        output_tokens: res.usage?.output_tokens ?? res.output_tokens,
+        total_tokens: res.usage?.total_tokens ?? res.total_tokens,
+        prompt_tokens: res.usage?.prompt_tokens ?? res.prompt_tokens,
+        completion_tokens: res.usage?.completion_tokens ?? res.completion_tokens,
+        input_tokens_details: res.usage?.input_tokens_details ?? res.input_tokens_details,
+        prompt_tokens_details: res.usage?.prompt_tokens_details ?? res.prompt_tokens_details,
+    }
+
+    return normalizeUsage(usageData)
+}
+
+export const estimateTokensFromText = (text: string): number => {
+    if (!text) return 0
+    const bytes = new TextEncoder().encode(text).length
+    return Math.ceil(bytes / 4)
+}
+
+export const estimateUsageFromBodies = (
+    requestBody: any,
+    responseBody?: any,
+    streamedOutputText?: string
+): Usage | null => {
+    const promptText = requestBody ? JSON.stringify(requestBody) : ""
+    const completionText = streamedOutputText
+        ? streamedOutputText
+        : (responseBody ? JSON.stringify(responseBody) : "")
+    const promptTokens = estimateTokensFromText(promptText)
+    const completionTokens = estimateTokensFromText(completionText)
+    if (!promptTokens && !completionTokens) return null
+    return {
+        prompt_tokens: promptTokens,
+        completion_tokens: completionTokens,
+        total_tokens: promptTokens + completionTokens,
+    }
+}