GitLab

当上游在 SSE 流中途返回 event:error 时，handleStreamingResponse 已将
部分 SSE 事件写入客户端，但原先的 failover 逻辑仍会切换到下一个账号
并写入完整流，导致客户端收到两个 message_start 进而产生 400 错误。

修复方案：在每次 Forward 调用前记录 c.Writer.Size()，若 Forward 返回
UpstreamFailoverError 后 writer 字节数增加，说明 SSE 内容已不可撤销地
发送给客户端，此时直接调用 handleFailoverExhausted 发送 SSE error 事件
终止流，而非继续 failover。

Ping-only 场景不受影响：slot 等待期的 ping 字节在 Forward 前后相等，
正常 failover 流程照常进行。
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

- quota_limited 模式：返回 Key 级别的总额度、速率限制窗口用量和过期时间
- unrestricted 模式：返回订阅限额或钱包余额信息（向后兼容）
- 新增 model_stats 字段，支持 start_date/end_date 参数查询按模型用量统计
- 提取 buildUsageData/parseUsageDateRange 等辅助方法，减少主函数复杂度
- 新增 APIKeyService.GetRateLimitData 和 UsageService.GetAPIKeyModelStats

新增 UMQ (User Message Queue) 双模式支持:
- serialize: 账号级分布式串行锁 + RPM 自适应延迟（严格限流）
- throttle: 仅 RPM 自适应前置延迟，不阻塞并发（软性限速）

后端:
- config: 新增 Mode 字段，保留 Enabled 向后兼容
- service: 新增 UserMessageQueueService（Lua 锁/延迟算法/清理 worker）
- repository: 新增 UserMsgQueueCache（Redis Lua acquire/release/force-release）
- handler: 新增 UserMsgQueueHelper（SSE ping + 等待循环 + throttle）
- gateway: 按 mode 分支集成 serialize/throttle 逻辑
- lint: 修复 gofmt rewrite rules、errcheck 类型断言、staticcheck QF1012

前端:
- 三态选择器 UI（关闭/软性限速/串行队列）替代 toggle 开关
- BulkEdit 支持 null 语义（不修改）
- i18n 中英文文案

通过 6 轮专家评审（42 次 review）、golangci-lint、单元测试、集成测试。

- 通过 User-Agent 识别 Claude Code 客户端并提取版本号
- 在网关层验证客户端版本是否满足管理员配置的最低要求
- 在管理后台提供版本要求配置选项（英文/中文双语）
- 实现原子缓存 + singleflight 防止并发问题和 thundering herd
- 使用 context.WithoutCancel 隔离 DB 查询，避免客户端断连影响缓存
- 双 TTL 策略：60s 正常、5s 错误恢复，保证性能与可用性
- 仅检查 Claude Code 客户端，其他客户端不受影响
- 添加完整单元测试覆盖版本提取、比对、上下文操作

- Move IncrementRPM after Forward success to prevent phantom RPM
  consumption during account switch retries
- Add base_rpm input sanitization (clamp to 0-10000) in Create/Update
- Add WindowCost scheduling checks to legacy path sticky sessions
  (4 check sites + 4 prefetch sites), fixing pre-existing gap
- Clean up rpm_strategy/rpm_sticky_buffer when disabling RPM in
  BulkEditModal (JSONB merge cannot delete keys, use empty values)
- Add json.Number test cases to TestGetBaseRPM/TestGetRPMStickyBuffer
- Document TOCTOU race as accepted soft-limit design trade-off

- Use TxPipeline (MULTI/EXEC) instead of Pipeline for atomic INCR+EXPIRE
- Filter negative values in GetBaseRPM(), update test expectation
- Add RPM batch query (GetRPMBatch) to account List API
- Add warn logs for RPM increment failures in gateway handler
- Reset enableRpmLimit on BulkEditAccountModal close
- Use union type 'tiered' | 'sticky_exempt' for rpmStrategy refs
- Add design decision comments for rdb.Time() RTT trade-off

- Remove Gemini platform exclusion from model restriction UI in
  Create/Edit account modals (Gemini now supports model_mapping)
- Remove outdated Gemini model passthrough info cards
- Add model_mapping field to GeminiCredentials type
- Extend warmup request interception toggle to Antigravity platform
- Remove redundant try/catch in API key account creation
- Remove noisy gateway.request_completed debug log
- Reorganize Gemini model mapping sections in constants.go

- Extract duplicated failover logic from gateway_handler.go (3 places)
  and gemini_v1beta_handler.go into shared failover_loop.go
- Introduce FailoverState with HandleFailoverError and HandleSelectionExhausted
- Move helper functions (needForceCacheBilling, sleepWithContext) into failover_loop.go
- Add comprehensive unit tests (32+ test cases)
- Delete redundant gateway_handler_single_account_retry_test.go

- 新增 UsageRecordWorkerPool，支持有界队列、溢出降级策略与自动扩缩容
- 将 Gateway/OpenAI/Sora/Gemini 使用量记录改为提交到统一任务池执行
- 增加 usage_record 配置默认值与校验规则，并补充配置与任务提交相关测试
- 注入并托管 worker 池生命周期，服务退出时统一 StopAndWait
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

- 将高密度服务与处理器日志迁移到新日志系统（LegacyPrintf/结构化日志）
- 增加 stdlog bridge 与兼容测试，保留旧日志捕获能力
- 将 OpenAI 断流告警改为结构化 Warn 并改造对应测试为 sink 捕获
- 补齐后端相关文件 logger 引用并通过全量 go test

- ops_error_logger: status < 400 分支增加 OpsSkipPassthroughKey 检查
- ops_upstream_context: 新增 checkSkipMonitoringForUpstreamEvent，中间重试/故障转移事件也能触发跳过标记
- gateway_handler/openai_gateway_handler/gemini_v1beta_handler: handleFailoverExhausted 匹配规则后设置 OpsSkipPassthroughKey
- antigravity_gateway_service: writeMappedClaudeError 增加 applyErrorPassthroughRule 调用

For retryable transient errors (Google 400 "invalid project resource name"
and empty stream responses), retry on the same account up to 2 times
(with 500ms delay) before switching to another account.

- Add RetryableOnSameAccount field to UpstreamFailoverError
- Add same-account retry loop in both Gemini and Claude/OpenAI handler paths
- Move temp-unschedule from service layer to handler layer (only after
  all same-account retries exhausted)
- Reduce temp-unschedule cooldown from 30 minutes to 1 minute

单账号 antigravity 分组收到 503 (MODEL_CAPACITY_EXHAUSTED) 时，
原逻辑会设置 ~29s 模型限流标记。由于只有一个账号无法切换，
后续所有新请求在预检查时命中限流 → 几毫秒内直接返回 503，
导致约 30 秒的雪崩窗口。

修复：在 Handler 入口处检查分组是否只有单个 antigravity 账号，
如果是则提前设置 SingleAccountRetry context 标记，让 Service 层
首次 503 就走原地重试逻辑（不设限流标记），避免污染后续请求。

当分组内只有一个可用账号且上游返回 503 (MODEL_CAPACITY_EXHAUSTED) 时，
不再设置模型限流+切换账号（因为切换回来还是同一个账号），而是在 Service 层
原地等待+重试，避免双重等待问题。

主要变更：
- Handler 层：检测单账号 503 场景，清除排除列表并设置 SingleAccountRetry 标记
- Service 层：新增 handleSingleAccountRetryInPlace 原地重试逻辑
- Service 层：预检查跳过单账号模式下的限流检查
- 新增 ctxkey.SingleAccountRetry 上下文标记

ParseGatewayRequest only parsed Anthropic format (system/messages),
ignoring Gemini native format (systemInstruction/contents). This caused
GenerateSessionHash to produce identical hashes for all Gemini sessions.

Add protocol parameter to ParseGatewayRequest to branch between
Anthropic and Gemini parsing. Update GenerateSessionHash message
traversal to extract text from both formats.

Mix SessionContext (ClientIP, UserAgent, APIKeyID) into
GenerateSessionHash 3rd-level fallback to differentiate requests
from different users sending identical content.

Also switch hashContent from SHA256-truncated to XXHash64 for
better performance, and optimize Trie Lua script to match from
longest prefix first.

Upstream accounts now use the standard APIKey type instead of a dedicated
upstream type. GetBaseURL() and new GetGeminiBaseURL() automatically append
/antigravity for Antigravity platform APIKey accounts, eliminating the need
for separate upstream forwarding methods.

- Remove ForwardUpstream, ForwardUpstreamGemini, testUpstreamConnection
- Remove upstream branch guards in Forward/ForwardGemini/TestConnection
- Add migration 052 to convert existing upstream accounts to apikey
- Update frontend CreateAccountModal to create apikey type
- Add unit tests for GetBaseURL and GetGeminiBaseURL

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>

The digest chain fallback is only needed for Gemini endpoints, not
for the Anthropic Messages API path. Remove the handler integration
while keeping the reusable service/repository layer for future use.

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.

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

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

Previously the /v1/usage endpoint aggregated usage stats (today/total
tokens, cost, RPM/TPM) across all API Keys belonging to the user.
This made it impossible to distinguish usage from different API Keys
(e.g. balance vs subscription keys).

Now the usage stats are filtered by the current request's API Key ID,
so each key only sees its own usage data. The balance/remaining fields
are unaffected and still reflect the user-level wallet balance.

Changes:
- Add GetAPIKeyDashboardStats to repository interface and implementation
- Add getPerformanceStatsByAPIKey helper (also fixes TPM to include
  cache_creation_tokens and cache_read_tokens)
- Add GetAPIKeyDashboardStats to UsageService
- Update Usage handler to call GetAPIKeyDashboardStats(apiKey.ID)
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>