oadmin/sub2api
1
0
Fork 0
Code Issues Pull Requests 1 Actions 2 Packages Projects Releases Wiki Activity
1,468 Commits 2 Branches 0 Tags
8da5fac69e2b70d2e928764cac77ce358c63693b
Commit Graph

736 Commits

Author SHA1 Message Date
yangjianbo
836ba14b70 fix: 修复函数签名变更后的调用参数不匹配 
- handleUpstreamError 补齐新增的三个参数 (0, "", false)
- handleStreamingResponse 移除已删除的 nil 参数

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
 2026-02-07 20:05:29 +08:00
yangjianbo
a14dfb769a Merge branch 'dev-release' 2026-02-07 19:58:00 +08:00
yangjianbo
2588fa6a8f fix(audit): 第二批审计修复 — P0 生产 Bug、安全加固、性能优化、缓存一致性、代码质量 
基于 backend-code-audit 审计报告，修复剩余 P0/P1/P2 共 34 项问题：

P0 生产 Bug：
- 修复 time.Since(time.Now()) 计时逻辑错误 (P0-03)
- generateRandomID 改用 crypto/rand 替代固定索引 (P0-04)
- IncrementQuotaUsed 重写为 Ent 原子操作消除 TOCTOU 竞态 (P0-05)

安全加固：
- gateway/openai handler 错误响应替换为泛化消息，防止内部信息泄露 (P1-14)
- usage_log_repo dateFormat 参数改用白名单映射，防止 SQL 注入 (P1-16)
- 默认配置安全加固：sslmode=prefer、response_headers=true、mode=release (P1-18/19, P2-15)

性能优化：
- gateway handler 循环内 defer 替换为显式 releaseWait 闭包 (P1-02)
- group_repo/promo_code_repo Count 前 Clone 查询避免状态污染 (P1-03)
- usage_log_repo 四个查询添加 LIMIT 10000 防止 OOM (P1-07)
- GetBatchUsageStats 添加时间范围参数，默认最近 30 天 (P1-10)
- ip.go CIDR 预编译为包级变量 (P1-11)
- BatchUpdateCredentials 重构为先验证后更新 (P1-13)

缓存一致性：
- billing_cache 添加 jitteredTTL 防止缓存雪崩 (P2-10)
- DeductUserBalance/UpdateSubscriptionUsage 错误传播修复 (P2-12)
- UserService.UpdateBalance 成功后异步失效 billingCache (P2-13)

代码质量：
- search 截断改为按 rune 处理，支持多字节字符 (P2-01)
- TLS Handshake 改为 HandshakeContext 支持 context 取消 (P2-07)
- CORS 预检添加 Access-Control-Max-Age: 86400 (P2-16)

测试覆盖：
- 新增 user_service_test.go（UpdateBalance 缓存失效 6 个用例）
- 新增 batch_update_credentials_test.go（fail-fast + 类型验证 7 个用例）
- 新增 response_transformer_test.go、ip_test.go、usage_log_repo_unit_test.go、search_truncate_test.go
- 集成测试：IncrementQuotaUsed 并发测试、billing_cache 错误传播测试
- config_test.go 补充 server.mode/sslmode 默认值断言

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
 2026-02-07 19:46:42 +08:00
erio
3077fd279d feat: smart retry max 1 attempt + clear sticky session on failure 
- Change antigravitySmartRetryMaxAttempts from 3 to 1 to prevent
  repeated rate limiting and long waits
- Clear sticky session binding (DeleteSessionAccountID) after smart
  retry exhaustion, so subsequent requests don't hit the same
  rate-limited account
- Add flow diagrams to Forward/ForwardGemini doc comments
- Add comprehensive unit tests covering:
  - Sticky session cleared on retry failure (429, 503, network error)
  - Sticky session NOT cleared on retry success
  - Sticky session NOT cleared for non-sticky requests (empty hash)
  - Sticky session NOT cleared on long delay path (handled by handler)
  - Nil cache safety (no panic)
  - MaxAttempts constant verification
  - End-to-end retryLoop → switchError propagation with session clear
 2026-02-07 19:30:58 +08:00
shaw
6aaa4aee6a fix: 收敛 Claude Code 探测拦截并补齐回归测试 2026-02-07 19:04:08 +08:00
erio
e3748da860 fix(lint): handle errcheck for strings.Builder.WriteString 2026-02-07 18:18:15 +08:00
erio
50a783ff01 feat: add Anthropic sticky session digest chain matching via Trie 
The previous fallback (step 3) in GenerateSessionHash hashed system +
all messages together, producing a different hash each round as the
conversation grew ([a] -> [a,b] -> [a,b,c]). This made fallback sticky
sessions ineffective for multi-turn conversations.

Implement per-message Trie digest chain matching (reusing Gemini's Trie
infrastructure) so that the previous round's chain is always a prefix
of the current round's chain, enabling reliable session affinity.
 2026-02-07 18:00:56 +08:00
shaw
1439eb39a9 fix(gateway): harden digest logging and align antigravity ops 
- avoid panic by using safe UUID prefix truncation in Gemini digest fallback logs\n- remove unconditional Antigravity 429 full-body debug logs and honor log truncation config\n- align Antigravity quick preset mappings to opus 4.6-thinking targets only\n- restore scope rate-limit aggregation/output in ops availability stats
 2026-02-07 17:12:15 +08:00
yangjianbo
8226a4ce4d perf(service): 优化 model 替换函数，用 gjson/sjson 替代全量 JSON 序列化 
SSE 热路径中 replaceModelInSSELine 和 replaceModelInResponseBody 原来
使用 json.Unmarshal/Marshal 对每个事件做全量反序列化再序列化，现改为
gjson.Get/sjson.Set 精确字段操作，消除 O(n) 中间 map 分配，保持 JSON
字段顺序不变。涉及 OpenAIGatewayService 和 GatewayService 两个服务。

新增 23 个单元测试覆盖：顶层/嵌套 model 替换、不匹配跳过、空行/[DONE]/
非法 JSON 等边界情况。

Fixes: P1-08

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
 2026-02-07 17:09:55 +08:00
erio
e1a68497d6 refactor: simplify sticky session rate limit handling — switch immediately on any rate limit 
Remove threshold-based waiting in both sticky session and antigravity
pre-check paths. When a model is rate-limited, immediately clear the
sticky session and switch accounts instead of waiting for short durations.
 2026-02-07 17:06:49 +08:00
yangjianbo
a9e256ce8c fix(openai): 修复 usage 为空导致 panic（P0-02） 2026-02-07 16:15:30 +08:00
erio
fa28dcbf32 fix(test): update test calls to match method receivers on handleSmartRetry and antigravityRetryLoop 2026-02-07 16:05:09 +08:00
erio
2656320d04 fix(antigravity): fetch default mapping from API and sync Redis on rate limit 
1. Frontend: replace hardcoded antigravityDefaultMappings with async
   fetch from GET /admin/accounts/antigravity/default-model-mapping,
   eliminating the duplicate data source that caused frontend/backend
   mapping inconsistency.

2. Backend: convert handleSmartRetry and antigravityRetryLoop from
   standalone functions to AntigravityGatewayService methods, enabling
   Redis cache sync (updateAccountModelRateLimitInCache) after both
   rate-limit write paths — long-delay branch and retry-exhausted branch.
 2026-02-07 15:59:27 +08:00
erio
b4f6c4f9d5 style: fix gofmt formatting in gateway_service.go 
Remove extra blank line that caused golangci-lint gofmt check to fail.
 2026-02-07 14:51:20 +08:00
yangjianbo
0e514ed80b perf(middleware): 优化订阅模式认证中间件，5次串行调用降至2步同步+1步异步 
- 为 GetActiveSubscription 添加 ristretto L1 缓存 + singleflight 防击穿
- 合并 ValidateSubscription + CheckUsageLimits 为纯内存 ValidateAndCheckLimits
- 窗口维护操作（激活/重置）异步化，不再阻塞首字节
- 缓存返回浅拷贝，避免并发 data race 和缓存污染
- 所有管理操作（分配/续期/撤销/扩展/窗口重置）同步失效 L1 缓存
- 新增 SubscriptionCacheConfig 可配置 L1 缓存大小/TTL/抖动

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
 2026-02-07 14:43:12 +08:00
erio
14c6c9321a refactor: remove unused IsAntigravityModelSupported function and its tests 2026-02-07 14:42:28 +08:00
erio
386126b1b2 test(antigravity): add missing unit tests for upstream and custom model_mapping 
- Add GetAccessToken upstream branch tests (success/failure/empty/nil)
- Add mapAntigravityModel wildcard-target-equals-request edge case tests
- Add upstream account smart retry test case
- Add GeminiMessagesCompatService custom model_mapping and empty model tests
 2026-02-07 14:39:25 +08:00
erio
de0927289e fix(antigravity): support upstream accounts and custom model_mapping in scheduling 
- GetAccessToken: add upstream branch to read api_key from credentials
- shouldTriggerAntigravitySmartRetry: relax check from IsOAuth to Platform-based
- isModelSupportedByAccount/WithContext: replace IsAntigravityModelSupported
  whitelist with mapAntigravityModel for unified scheduling/forwarding logic
- mapAntigravityModel: fix edge case where wildcard target equals request model
- Update tests for new behavior and add custom model_mapping test cases
 2026-02-07 14:32:08 +08:00
erio
edb0937024 fix: restore non-failover error passthrough from 7b156489 2026-02-07 14:24:55 +08:00
erio
43a4840daf fix: restore error passthrough service improvements from 7b156489 2026-02-07 14:16:19 +08:00
erio
5e98445b22 feat(antigravity): comprehensive enhancements - model mapping, rate limiting, scheduling & ops 
Key changes:
- Upgrade model mapping: Opus 4.5 → Opus 4.6-thinking with precise matching
- Unified rate limiting: scope-level → model-level with Redis snapshot sync
- Load-balanced scheduling by call count with smart retry mechanism
- Force cache billing support
- Model identity injection in prompts with leak prevention
- Thinking mode auto-handling (max_tokens/budget_tokens fix)
- Frontend: whitelist mode toggle, model mapping validation, status indicators
- Gemini session fallback with Redis Trie O(L) matching
- Ops: enhanced concurrency monitoring, account availability, retry logic
- Migration scripts: 049-051 for model mapping unification
 2026-02-07 12:31:10 +08:00
erio
8917afab2a fix(antigravity): reduce 429 fallback cooldown from 5min to 30s 
The default fallback cooldown when rate limit reset time cannot be
parsed was 5 minutes, which is too aggressive and causes accounts
to be unnecessarily locked out. Reduce to 30 seconds for faster
recovery. Config override still works (unit remains minutes).
 2026-02-07 11:54:00 +08:00
shaw
39a5b17d31 fix: 账号测试根据类型使用不同的 beta header 
- OAuth 账号：使用完整的 DefaultBetaHeader 和 Claude Code 客户端 headers
- API Key 账号：使用 APIKeyBetaHeader（不含 oauth beta）
 2026-02-07 11:33:06 +08:00
shaw
5299f3dcf6 fix: ix: antigravity 添加 aude-opus-4-6-thinking 模型支持 2026-02-07 10:38:10 +08:00
shaw
7b1564898b fix: make error passthrough effective for non-failover upstream errors 2026-02-07 10:25:56 +08:00
yangjianbo
4e01126ff2 test(codex): 清理无用的 opencode 缓存测试 
移除不再需要的 setupCodexCache 调用与辅助函数（已不再回源/读写缓存）
 2026-02-07 09:53:01 +08:00
yangjianbo
55b56328da feat(codex): 移除 opencode 指令回源与缓存 
- 不再从 GitHub 拉取 opencode codex_header.txt\n- 删除 ~/.opencode 缓存与异步刷新逻辑\n- 所有 instructions 统一使用内置 codex_cli_instructions.md
 2026-02-07 09:28:32 +08:00
yangjianbo
ce764bf2d9 feat(gateway): 支持强制 Codex CLI 模式并伪装 UA 
- Codex CLI 请求仅使用内置 instructions，不再读取 opencode 缓存/回源\n- 新增 gateway.force_codex_cli（环境变量 GATEWAY_FORCE_CODEX_CLI）\n- ForceCodexCLI=true 时转发上游强制 User-Agent=codex_cli_rs/0.0.0\n- 更新 deploy 示例配置
 2026-02-07 09:21:15 +08:00
yangjianbo
d71537d431 perf(service): SSE Scanner buffer 改用 sync.Pool 复用，减少高并发 GC 压力 
将流式响应中 bufio.Scanner 的 64KB buffer 从每次 make 分配改为
sync.Pool 复用，统一切片表达式为 [:0]、变量命名为 scanBuf，
并补充对应的单元测试。

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
 2026-02-06 22:55:12 +08:00
yangjianbo
ae1ba45350 perf(service): jitterTTL 改用 rand/v2 并移除锁 2026-02-06 21:22:38 +08:00
yangjianbo
c4182f8c33 perf(service): 移除 jitter 随机数全局锁 2026-02-06 21:20:25 +08:00
yangjianbo
2d4bbbf49d feat: 优化codex冷启动, 还有连接池数据库配置信息 2026-02-06 20:31:42 +08:00
shaw
9f4c1ef9f9 fix(ops): 添加 token 相关字段白名单避免误脱敏 
在敏感字段检测中添加白名单，排除 API 参数和用量统计字段：
- max_tokens, max_completion_tokens, max_output_tokens
- completion_tokens, prompt_tokens, total_tokens
- input_tokens, output_tokens
- cache_creation_input_tokens, cache_read_input_tokens

这些字段名虽然包含 "token" 但只是数值参数，不应被脱敏处理。
 2026-02-06 19:47:14 +08:00
Wesley Liddick
a381910e86 Merge pull request #489 from LLLLLLiulei/feat/import-export-bundle 
feat: implement account & proxy import/export with migration-ready JSON bundles
 2026-02-06 16:29:52 +08:00
shaw
d182ef0391 fix(gateway): 移除 PR #316 引入的工具名转换逻辑 
移除响应阶段的工具名/schema/description 转换逻辑，修复第三方工具调用时
工具名被错误转换的问题（如 Task → task）。

移除内容：
- 工具名相关正则变量（toolPrefixRe, toolNameBoundaryRe 等）
- openCodeToolOverrides 和 claudeToolNameOverrides 映射表
- 工具名转换函数（normalizeToolNameForClaude, normalizeToolNameForOpenCode 等）
- 响应体工具名替换函数（replaceToolNamesInText, replaceToolNamesInResponseBody 等）
- 参数名转换函数（normalizeParamNameForOpenCode, rewriteParamKeysInValue）
- 工具描述清理函数（sanitizeToolDescription）
- 输入 schema 转换函数（normalizeToolInputSchema）
- 模型 ID 正则替换函数（replaceModelIDInText）

保留内容：
- 系统提示词清理（sanitizeSystemText）
- Claude Code 指纹 headers 处理
- 模型 ID 映射（通过 JSON 对象操作）
 2026-02-06 16:09:58 +08:00
LLLLLLiulei
7319122e92 merge upstream/main 2026-02-06 11:33:45 +08:00
yangjianbo
ee01f80dc1 test(backend): 修复 usage 类型断言未检查 2026-02-06 09:54:29 +08:00
yangjianbo
98671a73f4 Merge branch 'main' of https://github.com/mt21625457/aicodex2api 
# Conflicts:
#	backend/internal/service/gateway_cached_tokens_test.go
 2026-02-06 09:35:46 +08:00
yangjianbo
f33a950103 fix(兼容): 将 Kimi cached_tokens 映射到 Claude 标准 cache_read_input_tokens 
Kimi 等 Claude 兼容 API 返回缓存信息使用 OpenAI 风格的 cached_tokens 字段，
而非 Claude 标准的 cache_read_input_tokens，导致客户端收不到缓存命中信息且
内部计费缓存折扣为 0。

新增 reconcileCachedTokens 辅助函数，在 cache_read_input_tokens == 0 且
cached_tokens > 0 时自动填充，覆盖流式（message_start/message_delta）和
非流式两种响应路径。对 Claude 原生上游无影响。

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
 2026-02-06 09:27:42 +08:00
程序猿MT
132bf34b69 Merge branch 'Wei-Shaw:main' into main 2026-02-06 08:53:52 +08:00
shaw
01b08e1e43 chore: 前端增加opus4.6模型映射 2026-02-06 08:50:45 +08:00
yangjianbo
c6a456c7c7 fix(兼容): 将 Kimi cached_tokens 映射到 Claude 标准 cache_read_input_tokens 
Kimi 等 Claude 兼容 API 返回缓存信息使用 OpenAI 风格的 cached_tokens 字段，
而非 Claude 标准的 cache_read_input_tokens，导致客户端收不到缓存命中信息且
内部计费缓存折扣为 0。

新增 reconcileCachedTokens 辅助函数，在 cache_read_input_tokens == 0 且
cached_tokens > 0 时自动填充，覆盖流式（message_start/message_delta）和
非流式两种响应路径。对 Claude 原生上游无影响。

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
 2026-02-06 08:42:55 +08:00
yangjianbo
a38bd413ab fix(计费): gpt-5.3-codex 定价回退到 gpt-5.2-codex 2026-02-06 07:40:38 +08:00
yangjianbo
9e1535e203 feat(模型): 添加 gpt-5.3 Codex 映射与价格配置 2026-02-06 07:14:46 +08:00
Wesley Liddick
571d1479a4 Merge pull request #490 from IanShaw027/fix/gemini-oauth-registered-user 
fix(gemini): 修复已注册用户 OAuth 授权问题并增强错误提示
 2026-02-05 22:37:20 +08:00
shaw
39e05a2dad feat: 新增全局错误透传规则功能 
支持管理员配置上游错误如何返回给客户端：
- 新增 ErrorPassthroughRule 数据模型和 Ent Schema
- 实现规则的 CRUD API（/admin/error-passthrough-rules）
- 支持按错误码、关键词匹配，支持 any/all 匹配模式
- 支持按平台过滤（anthropic/openai/gemini/antigravity）
- 支持透传或自定义响应状态码和错误消息
- 实现两级缓存（Redis + 本地内存）和多实例同步
- 集成到 gateway_handler 的错误处理流程
- 新增前端管理界面组件
- 新增单元测试覆盖核心匹配逻辑

优化：
- 移除 refreshLocalCache 中的冗余排序（数据库已排序）
- 后端 Validate() 增加匹配条件非空校验
 2026-02-05 21:52:54 +08:00
LLLLLLiulei
0b45d48e85 perf: batch fetch proxies for account export 2026-02-05 18:40:49 +08:00
LLLLLLiulei
b4bd46d067 feat: add data import/export bundle 2026-02-05 17:46:08 +08:00
shaw
2b192f7dca feat: 支持用户专属分组倍率配置 2026-02-05 16:05:42 +08:00
IanShaw027
979114db45 fix(gemini): 修复已注册用户 OAuth 授权时错误调用 onboardUser 的问题 
问题：Google One Ultra 等已注册用户在 OAuth 授权时，如果 LoadCodeAssist
返回了 currentTier/paidTier 但没有返回 cloudaicompanionProject，之前的
逻辑会继续调用 onboardUser，导致 INVALID_ARGUMENT 错误。

修复：对齐 Gemini CLI 的处理逻辑：
- 当检测到用户已注册（有 currentTier/paidTier）时，不再调用 onboardUser
- 先尝试从 Cloud Resource Manager 获取可用项目
- 如果仍无法获取，返回友好的错误提示，引导用户手动填写 Project ID

这个修复解决了 Google One 订阅用户无法正常授权的问题。
 2026-02-05 13:57:02 +08:00