Compare commits

..

40 Commits

Author SHA1 Message Date
xiaoxia b4b142180c Merge branch 'develop' into fix/render-engine-security-p0-p1
CI/CD Pipeline / Frontend Lint (pull_request) Successful in 1m56s
CI/CD Pipeline / Validate Code Quality And Tests (pull_request) Successful in 1m59s
CI/CD Pipeline / Build Staging API Image (pull_request) Has been skipped
CI/CD Pipeline / Build Staging Worker Image (pull_request) Has been skipped
CI/CD Pipeline / Build Staging Web Image (pull_request) Has been skipped
CI/CD Pipeline / Build Production API Image (pull_request) Has been skipped
CI/CD Pipeline / Build Production Worker Image (pull_request) Has been skipped
CI/CD Pipeline / Build Production Web Image (pull_request) Has been skipped
CI/CD Pipeline / Deploy Staging (Watchtower auto-deploy) (pull_request) Has been skipped
CI/CD Pipeline / Deploy Production (pull_request) Has been skipped
CI/CD Pipeline / Staging E2E Tests (pull_request) Has been skipped
CI/CD Pipeline / Staging API Integration Tests (pull_request) Has been skipped
CI/CD Pipeline / Production Browser E2E (pull_request) Has been skipped
CI/CD Pipeline / Unit Tests (pull_request) Successful in 2m9s
CI/CD Pipeline / Integration Tests (pull_request) Successful in 1m14s
# Conflicts:
#	apps/worker/video_processing/path_security.py
2026-07-14 18:15:00 +08:00
xiaoxia 7f767e2dd1 feat(ci): Runner物理分层 - 构建任务专属旧服务器
CI/CD Pipeline / Unit Tests (push) Successful in 1m52s
CI/CD Pipeline / Frontend Lint (push) Successful in 2m5s
CI/CD Pipeline / Validate Code Quality And Tests (push) Successful in 2m17s
CI/CD Pipeline / Build Production API Image (push) Has been skipped
CI/CD Pipeline / Build Production Worker Image (push) Has been skipped
CI/CD Pipeline / Build Production Web Image (push) Has been skipped
CI/CD Pipeline / Deploy Production (push) Has been skipped
CI/CD Pipeline / Production Browser E2E (push) Has been skipped
CI/CD Pipeline / Integration Tests (push) Successful in 1m15s
CI/CD Pipeline / Build Staging Web Image (push) Successful in 1m33s
CI/CD Pipeline / Build Staging API Image (push) Has been cancelled
CI/CD Pipeline / Build Staging Worker Image (push) Has been cancelled
CI/CD Pipeline / Deploy Staging (Watchtower auto-deploy) (push) Has been cancelled
CI/CD Pipeline / Staging E2E Tests (push) Has been cancelled
CI/CD Pipeline / Staging API Integration Tests (push) Has been cancelled
2026-07-14 18:11:01 +08:00
xiaoxia de42b1960e style: fix isort import ordering (first-party grouping)
CI/CD Pipeline / Validate Code Quality And Tests (pull_request) Successful in 45s
CI/CD Pipeline / Unit Tests (pull_request) Successful in 1m16s
CI/CD Pipeline / Frontend Lint (pull_request) Successful in 2m7s
CI/CD Pipeline / Build & Push Staging (Watchtower auto-deploy) (pull_request) Has been skipped
CI/CD Pipeline / Build Production Runtime Images (pull_request) Has been skipped
CI/CD Pipeline / Staging E2E Tests (pull_request) Has been skipped
CI/CD Pipeline / Staging API Integration Tests (pull_request) Has been skipped
CI/CD Pipeline / Deploy Production (pull_request) Has been skipped
CI/CD Pipeline / Production Browser E2E (pull_request) Has been skipped
CI/CD Pipeline / Integration Tests (pull_request) Successful in 1m25s
2026-07-14 18:03:19 +08:00
xiaoxia f377670076 style: fix isort import ordering (first-party grouping)
CI/CD Pipeline / Validate Code Quality And Tests (pull_request) Has been cancelled
CI/CD Pipeline / Unit Tests (pull_request) Has been cancelled
CI/CD Pipeline / Integration Tests (pull_request) Has been cancelled
CI/CD Pipeline / Frontend Lint (pull_request) Has been cancelled
CI/CD Pipeline / Build & Push Staging (Watchtower auto-deploy) (pull_request) Has been cancelled
CI/CD Pipeline / Staging E2E Tests (pull_request) Has been cancelled
CI/CD Pipeline / Staging API Integration Tests (pull_request) Has been cancelled
CI/CD Pipeline / Build Production Runtime Images (pull_request) Has been cancelled
CI/CD Pipeline / Deploy Production (pull_request) Has been cancelled
CI/CD Pipeline / Production Browser E2E (pull_request) Has been cancelled
2026-07-14 18:03:17 +08:00
xiaoxia ea93387f98 feat(ci): 代码质量深度加固 - mypy/ruff/vulture告警接入
CI/CD Pipeline / Unit Tests (push) Successful in 1m36s
CI/CD Pipeline / Validate Code Quality And Tests (push) Successful in 1m38s
CI/CD Pipeline / Frontend Lint (push) Successful in 1m56s
CI/CD Pipeline / Build Production Worker Image (push) Has been skipped
CI/CD Pipeline / Build Production Web Image (push) Has been skipped
CI/CD Pipeline / Build Production API Image (push) Has been skipped
CI/CD Pipeline / Deploy Production (push) Has been skipped
CI/CD Pipeline / Production Browser E2E (push) Has been skipped
CI/CD Pipeline / Integration Tests (push) Successful in 1m7s
CI/CD Pipeline / Build Staging Web Image (push) Successful in 3m9s
CI/CD Pipeline / Build Staging API Image (push) Successful in 7m46s
CI/CD Pipeline / Build Staging Worker Image (push) Has been cancelled
CI/CD Pipeline / Deploy Staging (Watchtower auto-deploy) (push) Has been cancelled
CI/CD Pipeline / Staging E2E Tests (push) Has been cancelled
CI/CD Pipeline / Staging API Integration Tests (push) Has been cancelled
feat(ci): 代码质量深度加固 - mypy/ruff/vulture 告警模式接入Validate

- ruff lint 告警模式接入:规则集与原flake8对齐 + bugbear摸底
- mypy 类型检查告警模式接入:检查核心业务代码
- vulture死代码扫描升级:置信度阈值70% + 按置信度排序
- 新增pyproject.toml ruff配置(保留原有black/isort/coverage配置)
- 所有新检查均为告警模式,不阻断CI
2026-07-14 17:56:15 +08:00
xiaoxia 674fc6763d style: fix isort for CI compatibility (5.x format)
CI/CD Pipeline / Validate Code Quality And Tests (pull_request) Failing after 44s
CI/CD Pipeline / Unit Tests (pull_request) Successful in 1m34s
CI/CD Pipeline / Frontend Lint (pull_request) Successful in 1m50s
CI/CD Pipeline / Build Production Runtime Images (pull_request) Has been skipped
CI/CD Pipeline / Build & Push Staging (Watchtower auto-deploy) (pull_request) Has been skipped
CI/CD Pipeline / Staging E2E Tests (pull_request) Has been skipped
CI/CD Pipeline / Staging API Integration Tests (pull_request) Has been skipped
CI/CD Pipeline / Deploy Production (pull_request) Has been skipped
CI/CD Pipeline / Production Browser E2E (pull_request) Has been skipped
CI/CD Pipeline / Integration Tests (pull_request) Successful in 1m29s
2026-07-14 17:52:00 +08:00
xiaoxia 405fb4c8d3 style: fix isort for CI compatibility (5.x format) 2026-07-14 17:51:59 +08:00
xiaoxia 7e172c0907 fix: PR#312安全债务 4个P1修复(路径安全+数量上限) (#322)
CI/CD Pipeline / Unit Tests (push) Successful in 1m33s
CI/CD Pipeline / Frontend Lint (push) Successful in 1m50s
CI/CD Pipeline / Validate Code Quality And Tests (push) Successful in 2m0s
CI/CD Pipeline / Build Production API Image (push) Has been skipped
CI/CD Pipeline / Build Production Worker Image (push) Has been skipped
CI/CD Pipeline / Build Production Web Image (push) Has been skipped
CI/CD Pipeline / Deploy Production (push) Has been skipped
CI/CD Pipeline / Production Browser E2E (push) Has been skipped
CI/CD Pipeline / Integration Tests (push) Successful in 1m12s
CI/CD Pipeline / Build Staging Web Image (push) Successful in 1m28s
CI/CD Pipeline / Build Staging API Image (push) Successful in 4m25s
CI/CD Pipeline / Build Staging Worker Image (push) Has been cancelled
CI/CD Pipeline / Deploy Staging (Watchtower auto-deploy) (push) Has been cancelled
CI/CD Pipeline / Staging E2E Tests (push) Has been cancelled
CI/CD Pipeline / Staging API Integration Tests (push) Has been cancelled
2026-07-14 17:43:57 +08:00
xiaoxia de1816bbd9 style: fix isort import ordering
CI/CD Pipeline / Validate Code Quality And Tests (pull_request) Failing after 20s
CI/CD Pipeline / Unit Tests (pull_request) Successful in 1m9s
CI/CD Pipeline / Integration Tests (pull_request) Successful in 1m24s
CI/CD Pipeline / Frontend Lint (pull_request) Successful in 1m58s
CI/CD Pipeline / Build & Push Staging (Watchtower auto-deploy) (pull_request) Has been skipped
CI/CD Pipeline / Build Production Runtime Images (pull_request) Has been skipped
CI/CD Pipeline / Staging E2E Tests (pull_request) Has been skipped
CI/CD Pipeline / Staging API Integration Tests (pull_request) Has been skipped
CI/CD Pipeline / Deploy Production (pull_request) Has been skipped
CI/CD Pipeline / Production Browser E2E (pull_request) Has been skipped
2026-07-14 17:42:25 +08:00
xiaoxia bbcc8a12bd style: fix isort import ordering
CI/CD Pipeline / Validate Code Quality And Tests (pull_request) Has been cancelled
CI/CD Pipeline / Unit Tests (pull_request) Has been cancelled
CI/CD Pipeline / Integration Tests (pull_request) Has been cancelled
CI/CD Pipeline / Frontend Lint (pull_request) Has been cancelled
CI/CD Pipeline / Build & Push Staging (Watchtower auto-deploy) (pull_request) Has been cancelled
CI/CD Pipeline / Staging E2E Tests (pull_request) Has been cancelled
CI/CD Pipeline / Staging API Integration Tests (pull_request) Has been cancelled
CI/CD Pipeline / Build Production Runtime Images (pull_request) Has been cancelled
CI/CD Pipeline / Deploy Production (pull_request) Has been cancelled
CI/CD Pipeline / Production Browser E2E (pull_request) Has been cancelled
2026-07-14 17:42:24 +08:00
xiaoxia 12862b4e23 style: fix isort import ordering
CI/CD Pipeline / Validate Code Quality And Tests (pull_request) Failing after 25s
CI/CD Pipeline / Unit Tests (pull_request) Successful in 1m19s
CI/CD Pipeline / Frontend Lint (pull_request) Successful in 1m56s
CI/CD Pipeline / Integration Tests (pull_request) Successful in 1m30s
CI/CD Pipeline / Build & Push Staging (Watchtower auto-deploy) (pull_request) Has been skipped
CI/CD Pipeline / Build Production Runtime Images (pull_request) Has been skipped
CI/CD Pipeline / Staging E2E Tests (pull_request) Has been skipped
CI/CD Pipeline / Staging API Integration Tests (pull_request) Has been skipped
CI/CD Pipeline / Deploy Production (pull_request) Has been skipped
CI/CD Pipeline / Production Browser E2E (pull_request) Has been skipped
2026-07-14 17:29:14 +08:00
xiaoxia 3b2bf414eb style: fix isort import ordering
CI/CD Pipeline / Validate Code Quality And Tests (pull_request) Has been cancelled
CI/CD Pipeline / Unit Tests (pull_request) Has been cancelled
CI/CD Pipeline / Integration Tests (pull_request) Has been cancelled
CI/CD Pipeline / Frontend Lint (pull_request) Has been cancelled
CI/CD Pipeline / Build & Push Staging (Watchtower auto-deploy) (pull_request) Has been cancelled
CI/CD Pipeline / Staging E2E Tests (pull_request) Has been cancelled
CI/CD Pipeline / Staging API Integration Tests (pull_request) Has been cancelled
CI/CD Pipeline / Build Production Runtime Images (pull_request) Has been cancelled
CI/CD Pipeline / Deploy Production (pull_request) Has been cancelled
CI/CD Pipeline / Production Browser E2E (pull_request) Has been cancelled
2026-07-14 17:29:13 +08:00
xiaoxia f681904f61 style: fix isort import ordering
CI/CD Pipeline / Validate Code Quality And Tests (pull_request) Has been cancelled
CI/CD Pipeline / Unit Tests (pull_request) Has been cancelled
CI/CD Pipeline / Integration Tests (pull_request) Has been cancelled
CI/CD Pipeline / Frontend Lint (pull_request) Has been cancelled
CI/CD Pipeline / Build & Push Staging (Watchtower auto-deploy) (pull_request) Has been cancelled
CI/CD Pipeline / Staging E2E Tests (pull_request) Has been cancelled
CI/CD Pipeline / Staging API Integration Tests (pull_request) Has been cancelled
CI/CD Pipeline / Build Production Runtime Images (pull_request) Has been cancelled
CI/CD Pipeline / Deploy Production (pull_request) Has been cancelled
CI/CD Pipeline / Production Browser E2E (pull_request) Has been cancelled
2026-07-14 17:29:12 +08:00
xiaoxia 71f7b6795b style: black + isort format
CI/CD Pipeline / Validate Code Quality And Tests (pull_request) Failing after 34s
CI/CD Pipeline / Unit Tests (pull_request) Successful in 1m11s
CI/CD Pipeline / Integration Tests (pull_request) Successful in 1m23s
CI/CD Pipeline / Frontend Lint (pull_request) Successful in 2m8s
CI/CD Pipeline / Build & Push Staging (Watchtower auto-deploy) (pull_request) Has been skipped
CI/CD Pipeline / Build Production Runtime Images (pull_request) Has been skipped
CI/CD Pipeline / Staging E2E Tests (pull_request) Has been skipped
CI/CD Pipeline / Staging API Integration Tests (pull_request) Has been skipped
CI/CD Pipeline / Deploy Production (pull_request) Has been skipped
CI/CD Pipeline / Production Browser E2E (pull_request) Has been skipped
2026-07-14 17:18:23 +08:00
xiaoxia c6c234757a style: black + isort format
CI/CD Pipeline / Validate Code Quality And Tests (pull_request) Has been cancelled
CI/CD Pipeline / Unit Tests (pull_request) Has been cancelled
CI/CD Pipeline / Integration Tests (pull_request) Has been cancelled
CI/CD Pipeline / Frontend Lint (pull_request) Has been cancelled
CI/CD Pipeline / Build & Push Staging (Watchtower auto-deploy) (pull_request) Has been cancelled
CI/CD Pipeline / Staging E2E Tests (pull_request) Has been cancelled
CI/CD Pipeline / Staging API Integration Tests (pull_request) Has been cancelled
CI/CD Pipeline / Build Production Runtime Images (pull_request) Has been cancelled
CI/CD Pipeline / Deploy Production (pull_request) Has been cancelled
CI/CD Pipeline / Production Browser E2E (pull_request) Has been cancelled
2026-07-14 17:18:20 +08:00
xiaoxia 99a6b4d104 style: black + isort format
CI/CD Pipeline / Validate Code Quality And Tests (pull_request) Has been cancelled
CI/CD Pipeline / Unit Tests (pull_request) Has been cancelled
CI/CD Pipeline / Integration Tests (pull_request) Has been cancelled
CI/CD Pipeline / Frontend Lint (pull_request) Has been cancelled
CI/CD Pipeline / Build & Push Staging (Watchtower auto-deploy) (pull_request) Has been cancelled
CI/CD Pipeline / Staging E2E Tests (pull_request) Has been cancelled
CI/CD Pipeline / Staging API Integration Tests (pull_request) Has been cancelled
CI/CD Pipeline / Build Production Runtime Images (pull_request) Has been cancelled
CI/CD Pipeline / Deploy Production (pull_request) Has been cancelled
CI/CD Pipeline / Production Browser E2E (pull_request) Has been cancelled
2026-07-14 17:18:17 +08:00
CI Bot 745ea06659 fix: #317 CI修复 - asset_analyzer缩进 + TTS单测SSRF兼容
CI/CD Pipeline / Validate Code Quality And Tests (pull_request) Failing after 29s
CI/CD Pipeline / Unit Tests (pull_request) Successful in 1m10s
CI/CD Pipeline / Frontend Lint (pull_request) Successful in 1m47s
CI/CD Pipeline / Integration Tests (pull_request) Successful in 1m17s
CI/CD Pipeline / Build & Push Staging (Watchtower auto-deploy) (pull_request) Has been skipped
CI/CD Pipeline / Build Production Runtime Images (pull_request) Has been skipped
CI/CD Pipeline / Staging E2E Tests (pull_request) Has been skipped
CI/CD Pipeline / Staging API Integration Tests (pull_request) Has been skipped
CI/CD Pipeline / Deploy Production (pull_request) Has been skipped
CI/CD Pipeline / Production Browser E2E (pull_request) Has been skipped
P0-1: asset_analyzer.py 缩进错误
- 第147行起23行缩进多了一层(改subprocess时弄乱)
- 修正 streams/format_info/for循环 等缩进

P0-2: TTS单测因SSRF接入挂了8个
- 根因:测试mock httpx,但safe_download_*在发请求前做DNS解析
- 修复:将httpx mock替换为safe_download_bytes/safe_download_file mock
- test_tts_oss_transfer.py: 4个失败修复,mock safe_download_bytes
- test_tts_segment_synthesis.py: 3个失败修复,mock safe_download_file
- test_tts_streaming.py: 1个失败修复,mock safe_download_bytes
- 安全层单测保持独立,33个test_url_security.py全绿

76个相关测试全绿(9+24+10+33)
2026-07-14 17:11:42 +08:00
xiaoxia 6db705a05d feat(ci): Runner物理分层 - 构建任务只调度到旧服务器
CI/CD Pipeline / Validate Code Quality And Tests (pull_request) Successful in 1m16s
CI/CD Pipeline / Unit Tests (pull_request) Successful in 1m26s
CI/CD Pipeline / Integration Tests (pull_request) Successful in 1m8s
CI/CD Pipeline / Frontend Lint (pull_request) Successful in 2m56s
CI/CD Pipeline / Build Staging API Image (pull_request) Has been skipped
CI/CD Pipeline / Build Staging Worker Image (pull_request) Has been skipped
CI/CD Pipeline / Build Staging Web Image (pull_request) Has been skipped
CI/CD Pipeline / Build Production API Image (pull_request) Has been skipped
CI/CD Pipeline / Build Production Worker Image (pull_request) Has been skipped
CI/CD Pipeline / Build Production Web Image (pull_request) Has been skipped
CI/CD Pipeline / Deploy Staging (Watchtower auto-deploy) (pull_request) Has been skipped
CI/CD Pipeline / Staging API Integration Tests (pull_request) Has been skipped
CI/CD Pipeline / Staging E2E Tests (pull_request) Has been skipped
CI/CD Pipeline / Deploy Production (pull_request) Has been skipped
CI/CD Pipeline / Production Browser E2E (pull_request) Has been skipped
将6个构建Job(Staging+Production)的runs-on从saas改为[saas, build-farm],
只有带build-farm标签的旧服务器Runner能接构建任务,新服务器6个Runner
专注跑CI门禁检查,实现真正的物理分层,避免构建任务挤占CI检查资源。
2026-07-14 17:02:16 +08:00
xiaoxia 1c8cb20373 feat(ci): P2债务清理 + Production部署门禁
CI/CD Pipeline / Validate Code Quality And Tests (push) Successful in 1m32s
CI/CD Pipeline / Unit Tests (push) Successful in 1m37s
CI/CD Pipeline / Frontend Lint (push) Successful in 2m7s
CI/CD Pipeline / Build Production API Image (push) Has been skipped
CI/CD Pipeline / Build Production Web Image (push) Has been skipped
CI/CD Pipeline / Build Production Worker Image (push) Has been skipped
CI/CD Pipeline / Deploy Production (push) Has been skipped
CI/CD Pipeline / Production Browser E2E (push) Has been skipped
CI/CD Pipeline / Integration Tests (push) Successful in 1m9s
CI/CD Pipeline / Build Staging Web Image (push) Successful in 2m10s
CI/CD Pipeline / Build Staging API Image (push) Successful in 4m17s
CI/CD Pipeline / Build Staging Worker Image (push) Failing after 20m0s
CI/CD Pipeline / Deploy Staging (Watchtower auto-deploy) (push) Has been skipped
CI/CD Pipeline / Staging E2E Tests (push) Has been skipped
CI/CD Pipeline / Staging API Integration Tests (push) Has been skipped
feat(ci): P2债务清理 + Production部署门禁

- 升级PyJWT 2.9.0 -> 2.13.0,修复4个CVE
- 清理5处死代码(未使用import和参数)
- Production部署加健康检查门禁(API health + 登录接口 + Web前端)
- deploy-production加checkout,修复通知脚本路径问题
2026-07-14 16:57:57 +08:00
CI Bot 2f03e13928 fix: batch_download增加视频MIME类型白名单校验
CI/CD Pipeline / Validate Code Quality And Tests (pull_request) Failing after 14s
CI/CD Pipeline / Unit Tests (pull_request) Failing after 1m9s
CI/CD Pipeline / Integration Tests (pull_request) Successful in 1m9s
CI/CD Pipeline / Frontend Lint (pull_request) Successful in 2m31s
CI/CD Pipeline / Build & Push Staging (Watchtower auto-deploy) (pull_request) Has been skipped
CI/CD Pipeline / Build Production Runtime Images (pull_request) Has been skipped
CI/CD Pipeline / Staging E2E Tests (pull_request) Has been skipped
CI/CD Pipeline / Staging API Integration Tests (pull_request) Has been skipped
CI/CD Pipeline / Deploy Production (pull_request) Has been skipped
CI/CD Pipeline / Production Browser E2E (pull_request) Has been skipped
- HTTP回退下载增加ALLOWED_VIDEO_MIME_TYPES白名单
- 包含application/octet-stream作为兜底兼容
- 6处外部下载全部接入大小+类型双重限制
2026-07-14 16:43:24 +08:00
xiaoxia 74c458e370 fix(ci): deploy-staging阶段添加checkout,修复通知脚本找不到的问题
CI/CD Pipeline / Unit Tests (push) Successful in 1m43s
CI/CD Pipeline / Validate Code Quality And Tests (push) Successful in 1m46s
CI/CD Pipeline / Frontend Lint (push) Successful in 1m52s
CI/CD Pipeline / Build Production API Image (push) Has been skipped
CI/CD Pipeline / Build Production Worker Image (push) Has been skipped
CI/CD Pipeline / Build Production Web Image (push) Has been skipped
CI/CD Pipeline / Integration Tests (push) Successful in 1m25s
CI/CD Pipeline / Build Staging Web Image (push) Successful in 1m32s
CI/CD Pipeline / Deploy Production (push) Has been skipped
CI/CD Pipeline / Production Browser E2E (push) Has been skipped
CI/CD Pipeline / Build Staging API Image (push) Successful in 5m47s
CI/CD Pipeline / Build Staging Worker Image (push) Successful in 22m36s
CI/CD Pipeline / Deploy Staging (Watchtower auto-deploy) (push) Has been cancelled
CI/CD Pipeline / Staging E2E Tests (push) Has been cancelled
CI/CD Pipeline / Staging API Integration Tests (push) Has been cancelled
fix(ci): deploy-staging阶段添加checkout,修复通知脚本找不到的问题
2026-07-14 16:31:25 +08:00
xiaoxia a7d942f705 feat(ci): 接入安全扫描(detect-secrets + pip-audit + vulture)
CI/CD Pipeline / Unit Tests (push) Successful in 1m40s
CI/CD Pipeline / Frontend Lint (push) Successful in 1m55s
CI/CD Pipeline / Validate Code Quality And Tests (push) Successful in 3m2s
CI/CD Pipeline / Build Production API Image (push) Has been skipped
CI/CD Pipeline / Build Production Worker Image (push) Has been skipped
CI/CD Pipeline / Build Production Web Image (push) Has been skipped
CI/CD Pipeline / Deploy Production (push) Has been skipped
CI/CD Pipeline / Production Browser E2E (push) Has been skipped
CI/CD Pipeline / Integration Tests (push) Successful in 1m26s
CI/CD Pipeline / Build Staging Web Image (push) Successful in 1m31s
CI/CD Pipeline / Build Staging API Image (push) Has been cancelled
CI/CD Pipeline / Build Staging Worker Image (push) Has been cancelled
CI/CD Pipeline / Deploy Staging (Watchtower auto-deploy) (push) Has been cancelled
CI/CD Pipeline / Staging E2E Tests (push) Has been cancelled
CI/CD Pipeline / Staging API Integration Tests (push) Has been cancelled
feat(ci): 接入安全扫描(detect-secrets + pip-audit + vulture)
2026-07-14 16:25:07 +08:00
CI Bot 334e2b1fc2 fix: 渲染引擎安全加固P0+P1 补修第二轮
CI/CD Pipeline / Validate Code Quality And Tests (pull_request) Failing after 16s
CI/CD Pipeline / Unit Tests (pull_request) Failing after 1m11s
CI/CD Pipeline / Integration Tests (pull_request) Successful in 1m11s
CI/CD Pipeline / Frontend Lint (pull_request) Successful in 1m46s
CI/CD Pipeline / Build Production Runtime Images (pull_request) Has been skipped
CI/CD Pipeline / Build & Push Staging (Watchtower auto-deploy) (pull_request) Has been skipped
CI/CD Pipeline / Staging E2E Tests (pull_request) Has been skipped
CI/CD Pipeline / Staging API Integration Tests (pull_request) Has been skipped
CI/CD Pipeline / Deploy Production (pull_request) Has been skipped
CI/CD Pipeline / Production Browser E2E (pull_request) Has been skipped
- SSRF漏接点补全:TTS 3个下载点 + BGM 3个下载点 全部接入 url_security.py 校验
  - TTS: workflow.py (_transfer_audio_to_oss, _download_and_merge_segments) + streaming_service.py (_download_audio)
  - BGM: generation.py (外部直链, 预设库) + batch_download.py (HTTP回退)
- 重定向防护:手动跟随重定向,每次跳转前重新校验目标 URL(禁用默认自动跟随)
- 文件大小/类型限制:新增 safe_download_file/safe_download_bytes,流式下载 + 大小上限 + MIME 白名单
- 裸 subprocess 补齐:5 处全部改走统一 run_ffmpeg/run_ffprobe
  - asset_analyzer.py: 3处 (ffprobe + 2个ffmpeg)
  - ingest.py: 1处 (ffprobe)
  - voice_extraction.py: 1处 (ffmpeg)
- URL安全模块迁移到 packages/shared/ 作为单一来源,worker 端保留向后兼容 re-export
- 新增 run_ffprobe 统一工具函数到 ffmpeg_utils
- 新增 7 个下载安全单测,累计 33 个 URL 安全测试
2026-07-14 16:19:39 +08:00
xiaoxia f867897348 fix(ci): 配置docker-container buildx builder
CI/CD Pipeline / Validate Code Quality And Tests (push) Successful in 32s
CI/CD Pipeline / Unit Tests (push) Successful in 1m42s
CI/CD Pipeline / Frontend Lint (push) Successful in 1m53s
CI/CD Pipeline / Integration Tests (push) Successful in 1m29s
CI/CD Pipeline / Build Production API Image (push) Has been skipped
CI/CD Pipeline / Build Production Worker Image (push) Has been skipped
CI/CD Pipeline / Build Production Web Image (push) Has been skipped
CI/CD Pipeline / Deploy Production (push) Has been skipped
CI/CD Pipeline / Production Browser E2E (push) Has been skipped
CI/CD Pipeline / Build Staging Web Image (push) Successful in 1m40s
CI/CD Pipeline / Build Staging API Image (push) Successful in 6m58s
CI/CD Pipeline / Build Staging Worker Image (push) Successful in 23m44s
CI/CD Pipeline / Deploy Staging (Watchtower auto-deploy) (push) Failing after 12m19s
CI/CD Pipeline / Staging E2E Tests (push) Has been skipped
CI/CD Pipeline / Staging API Integration Tests (push) Has been skipped
合并buildx builder修复
2026-07-14 15:37:57 +08:00
xiaoxia 3d1b739e7f fix(ci): 配置docker-container buildx builder以支持cache export
CI/CD Pipeline / Validate Code Quality And Tests (pull_request) Successful in 31s
CI/CD Pipeline / Unit Tests (pull_request) Successful in 1m26s
CI/CD Pipeline / Integration Tests (pull_request) Successful in 1m13s
CI/CD Pipeline / Frontend Lint (pull_request) Successful in 2m11s
CI/CD Pipeline / Build Staging API Image (pull_request) Has been skipped
CI/CD Pipeline / Build Staging Worker Image (pull_request) Has been skipped
CI/CD Pipeline / Build Staging Web Image (pull_request) Has been skipped
CI/CD Pipeline / Build Production API Image (pull_request) Has been skipped
CI/CD Pipeline / Build Production Worker Image (pull_request) Has been skipped
CI/CD Pipeline / Build Production Web Image (pull_request) Has been skipped
CI/CD Pipeline / Deploy Staging (Watchtower auto-deploy) (pull_request) Has been skipped
CI/CD Pipeline / Staging E2E Tests (pull_request) Has been skipped
CI/CD Pipeline / Staging API Integration Tests (pull_request) Has been skipped
CI/CD Pipeline / Deploy Production (pull_request) Has been skipped
CI/CD Pipeline / Production Browser E2E (pull_request) Has been skipped
新CI服务器默认docker driver不支持buildx cache export功能,
导致Staging/Production构建直接失败:
ERROR: Cache export is not supported for the docker driver.

在6个镜像构建Job中统一添加builder初始化步骤,
自动创建/使用docker-container类型的builder实例。
2026-07-14 15:29:43 +08:00
xiaoxia f268e208de feat: 多轨道混音 + 字幕渲染引擎 + 视频拼接(后端) (#312)
CI/CD Pipeline / Validate Code Quality And Tests (push) Successful in 33s
CI/CD Pipeline / Unit Tests (push) Successful in 1m35s
CI/CD Pipeline / Integration Tests (push) Successful in 1m18s
CI/CD Pipeline / Frontend Lint (push) Successful in 2m27s
CI/CD Pipeline / Build Production API Image (push) Has been skipped
CI/CD Pipeline / Build Production Worker Image (push) Has been skipped
CI/CD Pipeline / Build Production Web Image (push) Has been skipped
CI/CD Pipeline / Deploy Production (push) Has been skipped
CI/CD Pipeline / Production Browser E2E (push) Has been skipped
CI/CD Pipeline / Build Staging API Image (push) Failing after 8s
CI/CD Pipeline / Build Staging Worker Image (push) Failing after 11s
CI/CD Pipeline / Build Staging Web Image (push) Failing after 1m50s
CI/CD Pipeline / Deploy Staging (Watchtower auto-deploy) (push) Has been skipped
CI/CD Pipeline / Staging E2E Tests (push) Has been skipped
CI/CD Pipeline / Staging API Integration Tests (push) Has been skipped
2026-07-14 15:20:47 +08:00
xiaoxia b05966ff48 feat(ci): 新增CHANGELOG自动生成脚本 (#307)
CI/CD Pipeline / Validate Code Quality And Tests (push) Has been cancelled
CI/CD Pipeline / Unit Tests (push) Has been cancelled
CI/CD Pipeline / Integration Tests (push) Has been cancelled
CI/CD Pipeline / Frontend Lint (push) Has been cancelled
CI/CD Pipeline / Build Staging API Image (push) Has been cancelled
CI/CD Pipeline / Build Staging Worker Image (push) Has been cancelled
CI/CD Pipeline / Build Staging Web Image (push) Has been cancelled
CI/CD Pipeline / Deploy Staging (Watchtower auto-deploy) (push) Has been cancelled
CI/CD Pipeline / Staging E2E Tests (push) Has been cancelled
CI/CD Pipeline / Staging API Integration Tests (push) Has been cancelled
CI/CD Pipeline / Build Production API Image (push) Has been cancelled
CI/CD Pipeline / Build Production Worker Image (push) Has been cancelled
CI/CD Pipeline / Build Production Web Image (push) Has been cancelled
CI/CD Pipeline / Deploy Production (push) Has been cancelled
CI/CD Pipeline / Production Browser E2E (push) Has been cancelled
2026-07-14 15:20:31 +08:00
xiaoxia 3b80edd8c5 style: add nosec comment for bandit B108 false positive
CI/CD Pipeline / Validate Code Quality And Tests (pull_request) Successful in 38s
CI/CD Pipeline / Unit Tests (pull_request) Successful in 1m13s
CI/CD Pipeline / Integration Tests (pull_request) Successful in 1m22s
CI/CD Pipeline / Frontend Lint (pull_request) Successful in 2m22s
CI/CD Pipeline / Build & Push Staging (Watchtower auto-deploy) (pull_request) Has been skipped
CI/CD Pipeline / Build Production Runtime Images (pull_request) Has been skipped
CI/CD Pipeline / Staging E2E Tests (pull_request) Has been skipped
CI/CD Pipeline / Staging API Integration Tests (pull_request) Has been skipped
CI/CD Pipeline / Deploy Production (pull_request) Has been skipped
CI/CD Pipeline / Production Browser E2E (pull_request) Has been skipped
2026-07-14 15:20:20 +08:00
xiaoxia 01daa72f2a style: black + isort format
CI/CD Pipeline / Validate Code Quality And Tests (pull_request) Failing after 39s
CI/CD Pipeline / Unit Tests (pull_request) Successful in 1m24s
CI/CD Pipeline / Integration Tests (pull_request) Successful in 1m24s
CI/CD Pipeline / Frontend Lint (pull_request) Successful in 1m45s
CI/CD Pipeline / Build & Push Staging (Watchtower auto-deploy) (pull_request) Has been skipped
CI/CD Pipeline / Build Production Runtime Images (pull_request) Has been skipped
CI/CD Pipeline / Staging E2E Tests (pull_request) Has been skipped
CI/CD Pipeline / Staging API Integration Tests (pull_request) Has been skipped
CI/CD Pipeline / Deploy Production (pull_request) Has been skipped
CI/CD Pipeline / Production Browser E2E (pull_request) Has been skipped
2026-07-14 15:12:43 +08:00
xiaoxia e0f841fcee style: black + isort format
CI/CD Pipeline / Validate Code Quality And Tests (pull_request) Has been cancelled
CI/CD Pipeline / Unit Tests (pull_request) Has been cancelled
CI/CD Pipeline / Integration Tests (pull_request) Has been cancelled
CI/CD Pipeline / Frontend Lint (pull_request) Has been cancelled
CI/CD Pipeline / Build & Push Staging (Watchtower auto-deploy) (pull_request) Has been cancelled
CI/CD Pipeline / Staging E2E Tests (pull_request) Has been cancelled
CI/CD Pipeline / Staging API Integration Tests (pull_request) Has been cancelled
CI/CD Pipeline / Build Production Runtime Images (pull_request) Has been cancelled
CI/CD Pipeline / Deploy Production (pull_request) Has been cancelled
CI/CD Pipeline / Production Browser E2E (pull_request) Has been cancelled
2026-07-14 15:12:41 +08:00
xiaoxia 70a85b1463 style: black + isort format
CI/CD Pipeline / Validate Code Quality And Tests (pull_request) Has been cancelled
CI/CD Pipeline / Unit Tests (pull_request) Has been cancelled
CI/CD Pipeline / Integration Tests (pull_request) Has been cancelled
CI/CD Pipeline / Frontend Lint (pull_request) Has been cancelled
CI/CD Pipeline / Build & Push Staging (Watchtower auto-deploy) (pull_request) Has been cancelled
CI/CD Pipeline / Staging E2E Tests (pull_request) Has been cancelled
CI/CD Pipeline / Staging API Integration Tests (pull_request) Has been cancelled
CI/CD Pipeline / Build Production Runtime Images (pull_request) Has been cancelled
CI/CD Pipeline / Deploy Production (pull_request) Has been cancelled
CI/CD Pipeline / Production Browser E2E (pull_request) Has been cancelled
2026-07-14 15:12:39 +08:00
xiaoxia 79b82978d8 feat(ci): 镜像并行构建 - 三镜像并行构建,构建时间从17min压缩到8-10min (#314)
CI/CD Pipeline / Validate Code Quality And Tests (push) Successful in 35s
CI/CD Pipeline / Unit Tests (push) Successful in 1m8s
CI/CD Pipeline / Integration Tests (push) Successful in 1m15s
CI/CD Pipeline / Frontend Lint (push) Successful in 2m18s
CI/CD Pipeline / Build Production API Image (push) Has been skipped
CI/CD Pipeline / Build Staging API Image (push) Failing after 8s
CI/CD Pipeline / Build Production Worker Image (push) Has been skipped
CI/CD Pipeline / Build Production Web Image (push) Has been skipped
CI/CD Pipeline / Deploy Production (push) Has been skipped
CI/CD Pipeline / Build Staging Worker Image (push) Failing after 10s
CI/CD Pipeline / Production Browser E2E (push) Has been skipped
CI/CD Pipeline / Build Staging Web Image (push) Failing after 2m59s
CI/CD Pipeline / Deploy Staging (Watchtower auto-deploy) (push) Has been skipped
CI/CD Pipeline / Staging E2E Tests (push) Has been skipped
CI/CD Pipeline / Staging API Integration Tests (push) Has been skipped
2026-07-14 15:08:43 +08:00
xiaoxia 08b51ffa1d feat(ci): Runner标签分层路由 - 测试/构建分机运行 (#315)
CI/CD Pipeline / Unit Tests (push) Successful in 1m17s
CI/CD Pipeline / Frontend Lint (push) Successful in 2m11s
CI/CD Pipeline / Validate Code Quality And Tests (push) Successful in 2m21s
CI/CD Pipeline / Build Production Runtime Images (push) Has been skipped
CI/CD Pipeline / Deploy Production (push) Has been skipped
CI/CD Pipeline / Production Browser E2E (push) Has been skipped
CI/CD Pipeline / Integration Tests (push) Successful in 1m35s
CI/CD Pipeline / Build & Push Staging (Watchtower auto-deploy) (push) Has been cancelled
CI/CD Pipeline / Staging E2E Tests (push) Has been cancelled
CI/CD Pipeline / Staging API Integration Tests (push) Has been cancelled
2026-07-14 14:56:08 +08:00
xiaoxia 23d2406c27 feat(ci): 集成gitleaks + pip-audit + vulture安全扫描到Validate阶段 (#311)
CI/CD Pipeline / Validate Code Quality And Tests (push) Has been cancelled
CI/CD Pipeline / Unit Tests (push) Has been cancelled
CI/CD Pipeline / Integration Tests (push) Has been cancelled
CI/CD Pipeline / Frontend Lint (push) Has been cancelled
CI/CD Pipeline / Build & Push Staging (Watchtower auto-deploy) (push) Has been cancelled
CI/CD Pipeline / Staging E2E Tests (push) Has been cancelled
CI/CD Pipeline / Staging API Integration Tests (push) Has been cancelled
CI/CD Pipeline / Build Production Runtime Images (push) Has been cancelled
CI/CD Pipeline / Deploy Production (push) Has been cancelled
CI/CD Pipeline / Production Browser E2E (push) Has been cancelled
2026-07-14 14:55:58 +08:00
CI Bot 9e10e31a29 fix: 渲染引擎全链路安全加固 P0+P1
CI/CD Pipeline / Validate Code Quality And Tests (pull_request) Failing after 23s
CI/CD Pipeline / Integration Tests (pull_request) Successful in 1m17s
CI/CD Pipeline / Frontend Lint (pull_request) Failing after 15m19s
CI/CD Pipeline / Unit Tests (pull_request) Failing after 15m19s
CI/CD Pipeline / Build & Push Staging (Watchtower auto-deploy) (pull_request) Has been skipped
CI/CD Pipeline / Build Production Runtime Images (pull_request) Has been skipped
CI/CD Pipeline / Staging E2E Tests (pull_request) Has been skipped
CI/CD Pipeline / Deploy Production (pull_request) Has been skipped
CI/CD Pipeline / Staging API Integration Tests (pull_request) Has been skipped
CI/CD Pipeline / Production Browser E2E (pull_request) Has been skipped
P0-1 SSRF防护:统一URL安全校验方案
- 新增 url_security.py:SSRF防护工具(scheme白名单、端口白名单、内网IP拦截、内网主机名拦截、DNS解析校验)
- generation.py URL校验函数增加SSRF前置校验
- PiP引擎URL类型增加安全校验
- 贴纸引擎URL类型增加安全校验

P0-2 路径遍历防护:输入路径校验强化
- 新增 path_security.py:路径安全工具(路径遍历防护、local:// schema校验、文件名清理、允许目录白名单)
- resolve_asset_path 加固:本地绝对路径白名单、空字节检测、存储键路径遍历拦截
- PiP引擎local_path类型:增加允许目录校验
- 贴纸引擎:区分本地路径/URL,分别增加安全校验

P1 裸subprocess统一:3处FFmpeg调用走run_ffmpeg
- unified_render_service.py _extract_audio:改用 run_ffmpeg
- compose_video.py:改用 run_ffmpeg
- edit_plan_generation.py:改用 run_ffmpeg

新增测试:
- test_url_security.py:29个SSRF防护单测
- test_path_security.py:27个路径遍历防护单测
- 原有171+105=276个单测零回归
2026-07-14 13:18:16 +08:00
xiaoxia def6ee2363 feat(ci): 新增发布与灰度部署脚本
CI/CD Pipeline / Validate Code Quality And Tests (push) Successful in 46s
CI/CD Pipeline / Unit Tests (push) Successful in 1m40s
CI/CD Pipeline / Frontend Lint (push) Successful in 2m52s
CI/CD Pipeline / Build Production Runtime Images (push) Has been skipped
CI/CD Pipeline / Deploy Production (push) Has been skipped
CI/CD Pipeline / Production Browser E2E (push) Has been skipped
CI/CD Pipeline / Integration Tests (push) Successful in 1m42s
CI/CD Pipeline / Build & Push Staging (Watchtower auto-deploy) (push) Successful in 16m34s
CI/CD Pipeline / Staging E2E Tests (push) Successful in 2m52s
CI/CD Pipeline / Staging API Integration Tests (push) Successful in 3m55s
合并PR #316:新增release.sh / gray_deploy.sh / rollback_gray.sh三个脚本,适配生产环境架构(nginx upstream权重 + canary容器)
2026-07-14 13:12:41 +08:00
xiaoxia 2b5b650b9e fix(ci): 修复生产部署SSH密钥查找失败 + 通知脚本路径问题 (#313)
CI/CD Pipeline / Staging E2E Tests (push) Failing after 29m8s
CI/CD Pipeline / Validate Code Quality And Tests (push) Successful in 38s
CI/CD Pipeline / Unit Tests (push) Successful in 1m50s
CI/CD Pipeline / Integration Tests (push) Successful in 1m40s
CI/CD Pipeline / Frontend Lint (push) Successful in 3m25s
CI/CD Pipeline / Build & Push Staging (Watchtower auto-deploy) (push) Successful in 29m31s
CI/CD Pipeline / Staging API Integration Tests (push) Successful in 4m13s
CI/CD Pipeline / Build Production Runtime Images (push) Has been skipped
CI/CD Pipeline / Deploy Production (push) Has been skipped
CI/CD Pipeline / Production Browser E2E (push) Has been skipped
2026-07-14 11:53:02 +08:00
xiaoxia 2b8326987e Merge pull request 'fix(e2e): 5步向导测试末尾加unrouteAll兜底,避免页面关闭时报错' (#308) from fix/e2e-unroute-all-stability into develop
CI/CD Pipeline / Validate Code Quality And Tests (push) Successful in 28s
CI/CD Pipeline / Unit Tests (push) Successful in 1m3s
CI/CD Pipeline / Frontend Lint (push) Successful in 1m36s
CI/CD Pipeline / Build Production Runtime Images (push) Has been skipped
CI/CD Pipeline / Integration Tests (push) Successful in 1m8s
CI/CD Pipeline / Deploy Production (push) Has been skipped
CI/CD Pipeline / Production Browser E2E (push) Has been skipped
CI/CD Pipeline / Build & Push Staging (Watchtower auto-deploy) (push) Has been cancelled
CI/CD Pipeline / Staging E2E Tests (push) Has been cancelled
CI/CD Pipeline / Staging API Integration Tests (push) Has been cancelled
2026-07-14 11:47:12 +08:00
xiaoxia 2081c72be6 feat: 视频调速引擎(快进/慢放) (#294)
CI/CD Pipeline / Unit Tests (push) Successful in 1m8s
CI/CD Pipeline / Frontend Lint (push) Successful in 1m18s
CI/CD Pipeline / Validate Code Quality And Tests (push) Successful in 2m3s
CI/CD Pipeline / Integration Tests (push) Successful in 1m12s
CI/CD Pipeline / Build Production Runtime Images (push) Successful in 5m31s
CI/CD Pipeline / Deploy Production (push) Failing after 11s
CI/CD Pipeline / Production Browser E2E (push) Has been skipped
CI/CD Pipeline / Build & Push Staging (Watchtower auto-deploy) (push) Has been cancelled
CI/CD Pipeline / Staging E2E Tests (push) Has been cancelled
CI/CD Pipeline / Staging API Integration Tests (push) Has been cancelled
2026-07-14 11:36:24 +08:00
CI Bot a681844a44 fix(e2e): 5步向导测试末尾加unrouteAll兜底,避免页面关闭时报错
CI/CD Pipeline / Validate Code Quality And Tests (pull_request) Successful in 43s
CI/CD Pipeline / Unit Tests (pull_request) Successful in 1m20s
CI/CD Pipeline / Frontend Lint (pull_request) Successful in 1m25s
CI/CD Pipeline / Build & Push Staging (Watchtower auto-deploy) (pull_request) Has been skipped
CI/CD Pipeline / Build Production Runtime Images (pull_request) Has been skipped
CI/CD Pipeline / Staging E2E Tests (pull_request) Has been skipped
CI/CD Pipeline / Staging API Integration Tests (pull_request) Has been skipped
CI/CD Pipeline / Deploy Production (pull_request) Has been skipped
CI/CD Pipeline / Production Browser E2E (pull_request) Has been skipped
CI/CD Pipeline / Integration Tests (pull_request) Successful in 1m17s
2026-07-14 11:32:03 +08:00
38 changed files with 6420 additions and 393 deletions
+1 -1
View File
@@ -1 +1 @@
# CI trigger Fri Jun 26 09:53:28 PM CST 2026
trigger: 1784009947
+868 -31
View File
@@ -101,6 +101,62 @@ jobs:
bandit --version
pytest --version
- name: Secret detection (detect-secrets)
shell: sh
run: |
set -eu
echo "=== Installing detect-secrets ==="
python3 -m pip install -q detect-secrets
detect-secrets --version
echo ""
echo "=== Running secret scan ==="
detect-secrets scan \
--all-files \
--exclude-files '(^|/)(tests|test|e2e|__tests__|spec|docs|node_modules|site-packages|migrations|alembic|.gitea|.git|.pytest_cache|.next|dist|build)/' \
--exclude-files '\.(md|rst|txt|lock|example|sample|min\.js|min\.css|spec\.ts|test\.ts|test\.py)$' \
--exclude-files '(package-lock|yarn\.lock|poetry\.lock|Pipfile\.lock)$' \
--disable-plugin Base64HighEntropyString \
--disable-plugin HexHighEntropyString \
--disable-plugin BasicAuthDetector \
--disable-plugin KeywordDetector \
--disable-plugin IPPublicDetector \
2>&1 | tee /tmp/secrets-scan.json
FOUND=$(python3 -c "
import json
try:
with open('/tmp/secrets-scan.json') as f:
data = json.load(f)
results = data.get('results', {})
total = sum(len(v) for v in results.values())
print(total)
except Exception:
print('error')
")
echo ""
echo "Secrets detected: $FOUND"
if [ "$FOUND" != "0" ] && [ "$FOUND" != "error" ]; then
echo ""
echo "=== Secret details ==="
python3 -c "
import json
with open('/tmp/secrets-scan.json') as f:
data = json.load(f)
for fpath, items in data.get('results', {}).items():
for item in items:
line = item.get('line_number', '?')
stype = item.get('type', '?')
hashed = item.get('hashed_secret', '')[:16]
print(f' {fpath}:{line} [{stype}] {hashed}...')
"
echo ""
echo "ERROR: Potential secrets detected in code!"
echo "If these are false positives, add exclusions in the CI workflow."
exit 1
fi
echo "Secret scan completed - no secrets detected"
- name: Run code quality checks
shell: sh
run: |
@@ -110,12 +166,110 @@ jobs:
python3 -m isort --check-only alembic apps packages tests scripts
python3 -m flake8 apps packages tests --count --statistics
- name: Run security scan
- name: Ruff lint (advisory mode - 摸底阶段)
if: always()
shell: sh
run: |
set +e
echo "=== Installing ruff ==="
python3 -m pip install -q ruff
ruff --version
echo ""
echo "=== Running ruff lint (advisory mode) ==="
echo "告警模式,不阻断CI。用于摸底问题数量,后续分批修复后正式替换flake8。"
echo ""
ruff check apps packages tests scripts --statistics --output-format concise 2>&1 | tail -30
EXIT_CODE=$?
echo ""
if [ "$EXIT_CODE" != "0" ]; then
echo "ruff 发现 lint 问题(告警模式,不阻断)"
echo "问题分类统计见上方,后续将分批修复"
else
echo "ruff 检查全部通过 ✅"
fi
exit 0
- name: Type check (mypy, advisory mode)
if: always()
shell: sh
run: |
set +e
echo "=== Installing mypy ==="
python3 -m pip install -q mypy
mypy --version
echo ""
echo "=== Running mypy type check (advisory mode) ==="
echo "告警模式,不阻断CI"
echo ""
# 只检查核心业务代码,跳过测试和迁移
EXIT_CODE=0
mypy apps/api/app packages --ignore-missing-imports --no-site-packages --no-strict-optional --explicit-package-bases --exclude 'tests/|test_|migrations/|alembic/' --no-error-summary 2>&1 | head -60 || EXIT_CODE=$?
echo ""
if [ "$EXIT_CODE" != "0" ]; then
echo "mypy 发现类型问题(告警模式,不阻断)"
echo "建议后续逐步修复"
else
echo "mypy 类型检查通过 ✅"
fi
exit 0
- name: Run security scan (bandit)
shell: sh
run: |
set -eu
bandit -r apps packages -q -ll
- name: Python dependency vulnerability scan (pip-audit)
shell: sh
run: |
set -eu
echo "=== Installing pip-audit ==="
python3 -m pip install -q pip-audit
pip-audit --version
echo ""
echo "=== Scanning Python dependencies ==="
EXIT_CODE=0
for req_file in requirements.txt requirements-base.txt requirements-dev.txt; do
if [ -f "$req_file" ]; then
echo "--- Scanning $req_file ---"
pip-audit -r "$req_file" --desc on 2>&1 | head -40 || EXIT_CODE=$?
echo ""
fi
done
echo "pip-audit scan completed (advisory mode - warnings only, not blocking CI)"
if [ "$EXIT_CODE" != "0" ]; then
echo "WARNING: Potential vulnerabilities found in dependencies."
fi
exit 0
- name: Dead code detection (vulture)
if: always()
shell: sh
run: |
set +e
echo "=== Installing vulture ==="
python3 -m pip install -q vulture
vulture --version
echo ""
echo "=== Running vulture dead code scan (confidence >= 70%) ==="
echo "告警模式,不阻断CI。置信度>=90%建议尽快确认。"
echo ""
# 按置信度从高到低输出,便于优先查看高价值条目
vulture apps packages scripts \
--exclude "tests,test,migrations,.gitea,docs,node_modules,site-packages,*/test_*.py,*/conftest.py" \
--min-confidence 70 \
2>&1 | sort -t'(' -k2 -rn | head -80
EXIT_CODE=$?
echo ""
echo "=== vulture scan summary ==="
if [ "$EXIT_CODE" != "0" ]; then
echo "发现潜在死代码(可能包含框架装饰器注册的函数,为误报)"
echo "建议:定期人工审查高置信度(>=90%)条目"
else
echo "未发现明显死代码 ✅"
fi
exit 0
- name: Validate release scripts syntax
shell: sh
run: |
@@ -651,10 +805,10 @@ jobs:
echo "=== CI 失败通知 ==="
FAILED_JOB="Frontend Lint" python3 scripts/ci_notify_failure.py
deploy-staging:
name: Build & Push Staging (Watchtower auto-deploy)
runs-on: saas
timeout-minutes: 30
build-staging-api:
name: Build Staging API Image
runs-on: [saas, build-farm]
timeout-minutes: 20
needs: [validate, frontend-lint]
if: github.event_name == 'push' && (github.ref_name == 'main' || github.ref_name == 'develop')
@@ -704,30 +858,391 @@ jobs:
if member.name:
tar.extract(member, '.')
INNERPY
- name: Build and push all images to Gitea Registry
- name: Docker login to Registry
shell: sh
env:
REGISTRY_TOKEN: ${{ secrets.REGISTRY_TOKEN }}
run: |
set -eu
chmod +x scripts/build_release_images.sh
ALLOW_SHARED_PRODUCTION_BUILD_HOST=true REGISTRY_TOKEN="${REGISTRY_TOKEN}" \
scripts/build_release_images.sh "${GITHUB_SHA}" staging
printf '%s' "${REGISTRY_TOKEN}" | docker login git.xiaoxiajianji.com -u xiaoxia --password-stdin
echo "Docker login successful"
- name: Setup cache strategy
shell: sh
run: |
set -eu
# develop/main 分支写回缓存,其他分支只读
if [ "${GITHUB_REF_NAME}" = "develop" ] || [ "${GITHUB_REF_NAME}" = "main" ]; then
echo "CACHE_MODE=read-write" >> $GITHUB_ENV
echo "Cache mode: read-write (will push cache)"
else
echo "CACHE_MODE=read-only" >> $GITHUB_ENV
echo "Cache mode: read-only"
fi
- name: Tag and push :staging images (Watchtower auto-update)
- name: Setup buildx builder (docker-container driver)
shell: sh
run: |
set -eu
# 确保使用 docker-container driver 以支持 cache export 功能
if ! docker buildx inspect ci-builder > /dev/null 2>&1; then
docker buildx create --use --name ci-builder --driver docker-container
echo "Created ci-builder (docker-container driver)"
else
docker buildx use ci-builder
echo "Using existing ci-builder"
fi
docker buildx inspect --bootstrap
- name: Build and push API image (buildx cache)
shell: sh
env:
REGISTRY_TOKEN: ${{ secrets.REGISTRY_TOKEN }}
run: |
set -eu
REGISTRY="git.xiaoxiajianji.com/xiaoxia/xiaoxia-saas"
if [ -n "${REGISTRY_TOKEN:-}" ]; then
printf '%s' "${REGISTRY_TOKEN}" | docker login git.xiaoxiajianji.com -u xiaoxia --password-stdin 2>/dev/null
IMAGE_NAME="xiaoxia-saas-api"
CACHE_REF="${REGISTRY}/api-cache:develop"
CACHE_FROM="type=registry,ref=${CACHE_REF},ignore-error=true"
if [ "${CACHE_MODE}" = "read-write" ]; then
CACHE_TO="type=registry,ref=${CACHE_REF},mode=max"
echo "Building API image with read-write cache..."
docker buildx build --build-arg APP_VERSION="${GITHUB_SHA}" --cache-from "${CACHE_FROM}" --cache-to "${CACHE_TO}" -f infra/docker/api.Dockerfile -t "${REGISTRY}/${IMAGE_NAME}:${GITHUB_SHA}" --push .
else
echo "Building API image with read-only cache..."
docker buildx build --build-arg APP_VERSION="${GITHUB_SHA}" --cache-from "${CACHE_FROM}" -f infra/docker/api.Dockerfile -t "${REGISTRY}/${IMAGE_NAME}:${GITHUB_SHA}" --push .
fi
echo "API image pushed: ${REGISTRY}/${IMAGE_NAME}:${GITHUB_SHA}"
- name: Notify CI failure
if: failure()
shell: sh
run: |
set +e
echo "=== CI 失败通知 ==="
FAILED_JOB="Build Staging API Image" python3 scripts/ci_notify_failure.py
build-staging-worker:
name: Build Staging Worker Image
runs-on: [saas, build-farm]
timeout-minutes: 20
needs: [validate, frontend-lint]
if: github.event_name == 'push' && (github.ref_name == 'main' || github.ref_name == 'develop')
steps:
- name: Checkout code
shell: sh
env:
GITHUB_TOKEN: ${{ github.token }}
run: |
set -eu
python3 - <<'INNERPY'
import io, os, tarfile, time, urllib.request, urllib.error
url = f"{os.environ['GITHUB_API_URL']}/repos/{os.environ['GITHUB_REPOSITORY']}/archive/{os.environ['GITHUB_SHA']}.tar.gz"
request = urllib.request.Request(url, headers={"Authorization": f"token {os.environ['GITHUB_TOKEN']}"})
last_err = None
for attempt in range(5):
try:
with urllib.request.urlopen(request, timeout=120) as response:
archive = response.read()
break
except urllib.error.HTTPError as e:
last_err = e
if e.code >= 500 and attempt < 4:
wait = 2 ** attempt
print(f"Checkout HTTP {e.code}, retrying in {wait}s (attempt {attempt+1}/5)...")
time.sleep(wait)
continue
raise
except Exception as e:
last_err = e
if attempt < 4:
wait = 2 ** attempt
print(f"Checkout error: {e}, retrying in {wait}s (attempt {attempt+1}/5)...")
time.sleep(wait)
continue
else:
raise last_err
with tarfile.open(fileobj=io.BytesIO(archive), mode='r:gz') as tar:
root_prefix = tar.getmembers()[0].name.split('/', 1)[0] + '/'
for member in tar.getmembers():
name = member.name
if name == root_prefix[:-1]:
continue
if name.startswith(root_prefix):
member.name = name[len(root_prefix):]
if member.name:
tar.extract(member, '.')
INNERPY
- name: Docker login to Registry
shell: sh
env:
REGISTRY_TOKEN: ${{ secrets.REGISTRY_TOKEN }}
run: |
set -eu
printf '%s' "${REGISTRY_TOKEN}" | docker login git.xiaoxiajianji.com -u xiaoxia --password-stdin
echo "Docker login successful"
- name: Setup cache strategy
shell: sh
run: |
set -eu
# develop/main 分支写回缓存,其他分支只读
if [ "${GITHUB_REF_NAME}" = "develop" ] || [ "${GITHUB_REF_NAME}" = "main" ]; then
echo "CACHE_MODE=read-write" >> $GITHUB_ENV
echo "Cache mode: read-write (will push cache)"
else
echo "CACHE_MODE=read-only" >> $GITHUB_ENV
echo "Cache mode: read-only"
fi
- name: Setup buildx builder (docker-container driver)
shell: sh
run: |
set -eu
# 确保使用 docker-container driver 以支持 cache export 功能
if ! docker buildx inspect ci-builder > /dev/null 2>&1; then
docker buildx create --use --name ci-builder --driver docker-container
echo "Created ci-builder (docker-container driver)"
else
docker buildx use ci-builder
echo "Using existing ci-builder"
fi
docker buildx inspect --bootstrap
- name: Build and push Worker image (buildx cache)
shell: sh
run: |
set -eu
REGISTRY="git.xiaoxiajianji.com/xiaoxia/xiaoxia-saas"
IMAGE_NAME="xiaoxia-saas-worker"
CACHE_REF="${REGISTRY}/worker-cache:develop"
CACHE_FROM="type=registry,ref=${CACHE_REF},ignore-error=true"
if [ "${CACHE_MODE}" = "read-write" ]; then
CACHE_TO="type=registry,ref=${CACHE_REF},mode=max"
echo "Building Worker image with read-write cache..."
docker buildx build --build-arg APP_VERSION="${GITHUB_SHA}" --cache-from "${CACHE_FROM}" --cache-to "${CACHE_TO}" -f infra/docker/worker.Dockerfile -t "${REGISTRY}/${IMAGE_NAME}:${GITHUB_SHA}" --push .
else
echo "Building Worker image with read-only cache..."
docker buildx build --build-arg APP_VERSION="${GITHUB_SHA}" --cache-from "${CACHE_FROM}" -f infra/docker/worker.Dockerfile -t "${REGISTRY}/${IMAGE_NAME}:${GITHUB_SHA}" --push .
fi
echo "Worker image pushed: ${REGISTRY}/${IMAGE_NAME}:${GITHUB_SHA}"
- name: Notify CI failure
if: failure()
shell: sh
run: |
set +e
echo "=== CI 失败通知 ==="
FAILED_JOB="Build Staging Worker Image" python3 scripts/ci_notify_failure.py
build-staging-web:
name: Build Staging Web Image
runs-on: [saas, build-farm]
timeout-minutes: 20
needs: [validate, frontend-lint]
if: github.event_name == 'push' && (github.ref_name == 'main' || github.ref_name == 'develop')
steps:
- name: Checkout code
shell: sh
env:
GITHUB_TOKEN: ${{ github.token }}
run: |
set -eu
python3 - <<'INNERPY'
import io, os, tarfile, time, urllib.request, urllib.error
url = f"{os.environ['GITHUB_API_URL']}/repos/{os.environ['GITHUB_REPOSITORY']}/archive/{os.environ['GITHUB_SHA']}.tar.gz"
request = urllib.request.Request(url, headers={"Authorization": f"token {os.environ['GITHUB_TOKEN']}"})
last_err = None
for attempt in range(5):
try:
with urllib.request.urlopen(request, timeout=120) as response:
archive = response.read()
break
except urllib.error.HTTPError as e:
last_err = e
if e.code >= 500 and attempt < 4:
wait = 2 ** attempt
print(f"Checkout HTTP {e.code}, retrying in {wait}s (attempt {attempt+1}/5)...")
time.sleep(wait)
continue
raise
except Exception as e:
last_err = e
if attempt < 4:
wait = 2 ** attempt
print(f"Checkout error: {e}, retrying in {wait}s (attempt {attempt+1}/5)...")
time.sleep(wait)
continue
else:
raise last_err
with tarfile.open(fileobj=io.BytesIO(archive), mode='r:gz') as tar:
root_prefix = tar.getmembers()[0].name.split('/', 1)[0] + '/'
for member in tar.getmembers():
name = member.name
if name == root_prefix[:-1]:
continue
if name.startswith(root_prefix):
member.name = name[len(root_prefix):]
if member.name:
tar.extract(member, '.')
INNERPY
- name: Docker login to Registry
shell: sh
env:
REGISTRY_TOKEN: ${{ secrets.REGISTRY_TOKEN }}
run: |
set -eu
printf '%s' "${REGISTRY_TOKEN}" | docker login git.xiaoxiajianji.com -u xiaoxia --password-stdin
echo "Docker login successful"
- name: Setup cache strategy
shell: sh
run: |
set -eu
# develop/main 分支写回缓存,其他分支只读
if [ "${GITHUB_REF_NAME}" = "develop" ] || [ "${GITHUB_REF_NAME}" = "main" ]; then
echo "CACHE_MODE=read-write" >> $GITHUB_ENV
echo "Cache mode: read-write (will push cache)"
else
echo "CACHE_MODE=read-only" >> $GITHUB_ENV
echo "Cache mode: read-only"
fi
- name: Build frontend assets (npm build)
shell: sh
run: |
set -eu
NPM_CACHE_VOLUME="xiaoxia-npm-cache"
if ! docker volume inspect "$NPM_CACHE_VOLUME" >/dev/null 2>&1; then
docker volume create "$NPM_CACHE_VOLUME" >/dev/null
echo "Created npm cache volume: $NPM_CACHE_VOLUME"
fi
docker run --rm -v "$PWD:/workspace" -v "$NPM_CACHE_VOLUME:/workspace/apps/web/node_modules" -w /workspace/apps/web docker.m.daocloud.io/library/node:20 sh -lc "npm ci && npm run build"
test -f apps/web/dist/index.html
echo "Frontend build complete: $(ls apps/web/dist/ | head -5)"
- name: Setup buildx builder (docker-container driver)
shell: sh
run: |
set -eu
# 确保使用 docker-container driver 以支持 cache export 功能
if ! docker buildx inspect ci-builder > /dev/null 2>&1; then
docker buildx create --use --name ci-builder --driver docker-container
echo "Created ci-builder (docker-container driver)"
else
docker buildx use ci-builder
echo "Using existing ci-builder"
fi
docker buildx inspect --bootstrap
- name: Build and push Web image (buildx cache)
shell: sh
run: |
set -eu
REGISTRY="git.xiaoxiajianji.com/xiaoxia/xiaoxia-saas"
IMAGE_NAME="xiaoxia-saas-web"
CACHE_REF="${REGISTRY}/web-cache:develop"
NGINX_CONF="infra/docker/nginx-staging.conf"
CACHE_FROM="type=registry,ref=${CACHE_REF},ignore-error=true"
if [ "${CACHE_MODE}" = "read-write" ]; then
CACHE_TO="type=registry,ref=${CACHE_REF},mode=max"
echo "Building Web image with read-write cache..."
docker buildx build --cache-from "${CACHE_FROM}" --cache-to "${CACHE_TO}" -f infra/docker/web-artifact.Dockerfile --build-arg "NGINX_CONF=${NGINX_CONF}" -t "${REGISTRY}/${IMAGE_NAME}:${GITHUB_SHA}" --push .
else
echo "Building Web image with read-only cache..."
docker buildx build --cache-from "${CACHE_FROM}" -f infra/docker/web-artifact.Dockerfile --build-arg "NGINX_CONF=${NGINX_CONF}" -t "${REGISTRY}/${IMAGE_NAME}:${GITHUB_SHA}" --push .
fi
echo "Web image pushed: ${REGISTRY}/${IMAGE_NAME}:${GITHUB_SHA}"
- name: Notify CI failure
if: failure()
shell: sh
run: |
set +e
echo "=== CI 失败通知 ==="
FAILED_JOB="Build Staging Web Image" python3 scripts/ci_notify_failure.py
deploy-staging:
name: Deploy Staging (Watchtower auto-deploy)
runs-on: saas
timeout-minutes: 15
needs: [build-staging-api, build-staging-worker, build-staging-web]
if: github.event_name == 'push' && (github.ref_name == 'main' || github.ref_name == 'develop')
steps:
- name: Checkout code
shell: sh
env:
GITHUB_TOKEN: ${{ github.token }}
run: |
set -eu
python3 - <<'INNERPY'
import io, os, tarfile, time, urllib.request, urllib.error
url = f"{os.environ['GITHUB_API_URL']}/repos/{os.environ['GITHUB_REPOSITORY']}/archive/{os.environ['GITHUB_SHA']}.tar.gz"
request = urllib.request.Request(url, headers={"Authorization": f"token {os.environ['GITHUB_TOKEN']}"})
last_err = None
for attempt in range(5):
try:
with urllib.request.urlopen(request, timeout=120) as response:
archive = response.read()
break
except urllib.error.HTTPError as e:
last_err = e
if e.code >= 500 and attempt < 4:
wait = 2 ** attempt
print(f"Checkout HTTP {e.code}, retrying in {wait}s (attempt {attempt+1}/5)...")
time.sleep(wait)
continue
raise
except Exception as e:
last_err = e
if attempt < 4:
wait = 2 ** attempt
print(f"Checkout error: {e}, retrying in {wait}s (attempt {attempt+1}/5)...")
time.sleep(wait)
continue
else:
raise last_err
with tarfile.open(fileobj=io.BytesIO(archive), mode='r:gz') as tar:
root_prefix = tar.getmembers()[0].name.split('/', 1)[0] + '/'
for member in tar.getmembers():
name = member.name
if name == root_prefix[:-1]:
continue
if name.startswith(root_prefix):
member.name = name[len(root_prefix):]
if member.name:
tar.extract(member, '.')
INNERPY
- name: Docker login to Registry
shell: sh
env:
REGISTRY_TOKEN: ${{ secrets.REGISTRY_TOKEN }}
run: |
set -eu
printf '%s' "${REGISTRY_TOKEN}" | docker login git.xiaoxiajianji.com -u xiaoxia --password-stdin
echo "Docker login successful"
- name: Tag and push :staging images (Watchtower auto-update)
shell: sh
run: |
set -eu
REGISTRY="git.xiaoxiajianji.com/xiaoxia/xiaoxia-saas"
for svc in api worker web; do
echo "Pulling ${REGISTRY}/xiaoxia-saas-${svc}:${GITHUB_SHA} ..."
docker pull "${REGISTRY}/xiaoxia-saas-${svc}:${GITHUB_SHA}"
docker tag "${REGISTRY}/xiaoxia-saas-${svc}:${GITHUB_SHA}" "${REGISTRY}/xiaoxia-saas-${svc}:staging"
docker push "${REGISTRY}/xiaoxia-saas-${svc}:staging"
echo "$svc :staging tagged and pushed"
done
echo "All :staging images pushed. Watchtower will auto-deploy within 60s."
@@ -796,8 +1311,7 @@ jobs:
run: |
set +e
echo "=== CI 失败通知 ==="
FAILED_JOB="Build & Push Staging (Watchtower auto-deploy)" python3 scripts/ci_notify_failure.py
FAILED_JOB="Deploy Staging" python3 scripts/ci_notify_failure.py
staging-e2e:
@@ -951,10 +1465,10 @@ jobs:
build-production-runtime-images:
name: Build Production Runtime Images
runs-on: saas
timeout-minutes: 30
build-production-api:
name: Build Production API Image
runs-on: [saas, build-farm]
timeout-minutes: 20
needs: [validate, frontend-lint]
if: startsWith(github.ref, 'refs/tags/v')
@@ -966,7 +1480,7 @@ jobs:
GITHUB_TOKEN: ${{ github.token }}
run: |
set -eu
python3 - <<'PY'
python3 - <<'INNERPY'
import io, os, tarfile, time, urllib.request, urllib.error
url = f"{os.environ['GITHUB_API_URL']}/repos/{os.environ['GITHUB_REPOSITORY']}/archive/{os.environ['GITHUB_SHA']}.tar.gz"
request = urllib.request.Request(url, headers={"Authorization": f"token {os.environ['GITHUB_TOKEN']}"})
@@ -991,7 +1505,6 @@ jobs:
print(f"Checkout error: {e}, retrying in {wait}s (attempt {attempt+1}/5)...")
time.sleep(wait)
continue
raise
else:
raise last_err
with tarfile.open(fileobj=io.BytesIO(archive), mode='r:gz') as tar:
@@ -1004,16 +1517,251 @@ jobs:
member.name = name[len(root_prefix):]
if member.name:
tar.extract(member, '.')
PY
- name: Build and push all images (api + worker + web, with buildx cache)
INNERPY
- name: Docker login to Registry
shell: sh
env:
REGISTRY_TOKEN: ${{ secrets.REGISTRY_TOKEN }}
run: |
set -eu
chmod +x scripts/build_release_images.sh
REGISTRY_TOKEN="${REGISTRY_TOKEN}" scripts/build_release_images.sh "${GITHUB_REF_NAME}"
printf '%s' "${REGISTRY_TOKEN}" | docker login git.xiaoxiajianji.com -u xiaoxia --password-stdin
echo "Docker login successful"
- name: Setup buildx builder (docker-container driver)
shell: sh
run: |
set -eu
# 确保使用 docker-container driver 以支持 cache export 功能
if ! docker buildx inspect ci-builder > /dev/null 2>&1; then
docker buildx create --use --name ci-builder --driver docker-container
echo "Created ci-builder (docker-container driver)"
else
docker buildx use ci-builder
echo "Using existing ci-builder"
fi
docker buildx inspect --bootstrap
- name: Build and push API image (buildx cache)
shell: sh
run: |
set -eu
REGISTRY="git.xiaoxiajianji.com/xiaoxia/xiaoxia-saas"
IMAGE_NAME="xiaoxia-saas-api"
VERSION="${GITHUB_REF_NAME}"
CACHE_REF="${REGISTRY}/api-cache:main"
echo "Building Production API image: ${VERSION}"
docker buildx build --build-arg APP_VERSION="${VERSION}" --cache-from "type=registry,ref=${CACHE_REF},ignore-error=true" --cache-to "type=registry,ref=${CACHE_REF},mode=max" -f infra/docker/api.Dockerfile -t "${REGISTRY}/${IMAGE_NAME}:${VERSION}" --push .
echo "Production API image pushed: ${REGISTRY}/${IMAGE_NAME}:${VERSION}"
- name: Notify CI failure
if: failure()
shell: sh
run: |
set +e
echo "=== CI 失败通知 ==="
FAILED_JOB="Build Production API Image" python3 scripts/ci_notify_failure.py
build-production-worker:
name: Build Production Worker Image
runs-on: [saas, build-farm]
timeout-minutes: 20
needs: [validate, frontend-lint]
if: startsWith(github.ref, 'refs/tags/v')
steps:
- name: Checkout code
shell: sh
env:
GITHUB_TOKEN: ${{ github.token }}
run: |
set -eu
python3 - <<'INNERPY'
import io, os, tarfile, time, urllib.request, urllib.error
url = f"{os.environ['GITHUB_API_URL']}/repos/{os.environ['GITHUB_REPOSITORY']}/archive/{os.environ['GITHUB_SHA']}.tar.gz"
request = urllib.request.Request(url, headers={"Authorization": f"token {os.environ['GITHUB_TOKEN']}"})
last_err = None
for attempt in range(5):
try:
with urllib.request.urlopen(request, timeout=120) as response:
archive = response.read()
break
except urllib.error.HTTPError as e:
last_err = e
if e.code >= 500 and attempt < 4:
wait = 2 ** attempt
print(f"Checkout HTTP {e.code}, retrying in {wait}s (attempt {attempt+1}/5)...")
time.sleep(wait)
continue
raise
except Exception as e:
last_err = e
if attempt < 4:
wait = 2 ** attempt
print(f"Checkout error: {e}, retrying in {wait}s (attempt {attempt+1}/5)...")
time.sleep(wait)
continue
else:
raise last_err
with tarfile.open(fileobj=io.BytesIO(archive), mode='r:gz') as tar:
root_prefix = tar.getmembers()[0].name.split('/', 1)[0] + '/'
for member in tar.getmembers():
name = member.name
if name == root_prefix[:-1]:
continue
if name.startswith(root_prefix):
member.name = name[len(root_prefix):]
if member.name:
tar.extract(member, '.')
INNERPY
- name: Docker login to Registry
shell: sh
env:
REGISTRY_TOKEN: ${{ secrets.REGISTRY_TOKEN }}
run: |
set -eu
printf '%s' "${REGISTRY_TOKEN}" | docker login git.xiaoxiajianji.com -u xiaoxia --password-stdin
echo "Docker login successful"
- name: Setup buildx builder (docker-container driver)
shell: sh
run: |
set -eu
# 确保使用 docker-container driver 以支持 cache export 功能
if ! docker buildx inspect ci-builder > /dev/null 2>&1; then
docker buildx create --use --name ci-builder --driver docker-container
echo "Created ci-builder (docker-container driver)"
else
docker buildx use ci-builder
echo "Using existing ci-builder"
fi
docker buildx inspect --bootstrap
- name: Build and push Worker image (buildx cache)
shell: sh
run: |
set -eu
REGISTRY="git.xiaoxiajianji.com/xiaoxia/xiaoxia-saas"
IMAGE_NAME="xiaoxia-saas-worker"
VERSION="${GITHUB_REF_NAME}"
CACHE_REF="${REGISTRY}/worker-cache:main"
echo "Building Production Worker image: ${VERSION}"
docker buildx build --build-arg APP_VERSION="${VERSION}" --cache-from "type=registry,ref=${CACHE_REF},ignore-error=true" --cache-to "type=registry,ref=${CACHE_REF},mode=max" -f infra/docker/worker.Dockerfile -t "${REGISTRY}/${IMAGE_NAME}:${VERSION}" --push .
echo "Production Worker image pushed: ${REGISTRY}/${IMAGE_NAME}:${VERSION}"
- name: Notify CI failure
if: failure()
shell: sh
run: |
set +e
echo "=== CI 失败通知 ==="
FAILED_JOB="Build Production Worker Image" python3 scripts/ci_notify_failure.py
build-production-web:
name: Build Production Web Image
runs-on: [saas, build-farm]
timeout-minutes: 20
needs: [validate, frontend-lint]
if: startsWith(github.ref, 'refs/tags/v')
steps:
- name: Checkout code
shell: sh
env:
GITHUB_TOKEN: ${{ github.token }}
run: |
set -eu
python3 - <<'INNERPY'
import io, os, tarfile, time, urllib.request, urllib.error
url = f"{os.environ['GITHUB_API_URL']}/repos/{os.environ['GITHUB_REPOSITORY']}/archive/{os.environ['GITHUB_SHA']}.tar.gz"
request = urllib.request.Request(url, headers={"Authorization": f"token {os.environ['GITHUB_TOKEN']}"})
last_err = None
for attempt in range(5):
try:
with urllib.request.urlopen(request, timeout=120) as response:
archive = response.read()
break
except urllib.error.HTTPError as e:
last_err = e
if e.code >= 500 and attempt < 4:
wait = 2 ** attempt
print(f"Checkout HTTP {e.code}, retrying in {wait}s (attempt {attempt+1}/5)...")
time.sleep(wait)
continue
raise
except Exception as e:
last_err = e
if attempt < 4:
wait = 2 ** attempt
print(f"Checkout error: {e}, retrying in {wait}s (attempt {attempt+1}/5)...")
time.sleep(wait)
continue
else:
raise last_err
with tarfile.open(fileobj=io.BytesIO(archive), mode='r:gz') as tar:
root_prefix = tar.getmembers()[0].name.split('/', 1)[0] + '/'
for member in tar.getmembers():
name = member.name
if name == root_prefix[:-1]:
continue
if name.startswith(root_prefix):
member.name = name[len(root_prefix):]
if member.name:
tar.extract(member, '.')
INNERPY
- name: Docker login to Registry
shell: sh
env:
REGISTRY_TOKEN: ${{ secrets.REGISTRY_TOKEN }}
run: |
set -eu
printf '%s' "${REGISTRY_TOKEN}" | docker login git.xiaoxiajianji.com -u xiaoxia --password-stdin
echo "Docker login successful"
- name: Build frontend assets (npm build)
shell: sh
run: |
set -eu
NPM_CACHE_VOLUME="xiaoxia-npm-cache"
if ! docker volume inspect "$NPM_CACHE_VOLUME" >/dev/null 2>&1; then
docker volume create "$NPM_CACHE_VOLUME" >/dev/null
fi
docker run --rm -v "$PWD:/workspace" -v "$NPM_CACHE_VOLUME:/workspace/apps/web/node_modules" -w /workspace/apps/web docker.m.daocloud.io/library/node:20 sh -lc "npm ci && npm run build"
test -f apps/web/dist/index.html
echo "Frontend build complete"
- name: Setup buildx builder (docker-container driver)
shell: sh
run: |
set -eu
# 确保使用 docker-container driver 以支持 cache export 功能
if ! docker buildx inspect ci-builder > /dev/null 2>&1; then
docker buildx create --use --name ci-builder --driver docker-container
echo "Created ci-builder (docker-container driver)"
else
docker buildx use ci-builder
echo "Using existing ci-builder"
fi
docker buildx inspect --bootstrap
- name: Build and push Web image (buildx cache)
shell: sh
run: |
set -eu
REGISTRY="git.xiaoxiajianji.com/xiaoxia/xiaoxia-saas"
IMAGE_NAME="xiaoxia-saas-web"
VERSION="${GITHUB_REF_NAME}"
CACHE_REF="${REGISTRY}/web-cache:main"
NGINX_CONF="infra/docker/nginx-production.conf"
echo "Building Production Web image: ${VERSION}"
docker buildx build --cache-from "type=registry,ref=${CACHE_REF},ignore-error=true" --cache-to "type=registry,ref=${CACHE_REF},mode=max" -f infra/docker/web-artifact.Dockerfile --build-arg "NGINX_CONF=${NGINX_CONF}" -t "${REGISTRY}/${IMAGE_NAME}:${VERSION}" --push .
echo "Production Web image pushed: ${REGISTRY}/${IMAGE_NAME}:${VERSION}"
- name: Cleanup old Docker images
if: always()
@@ -1036,17 +1784,60 @@ jobs:
run: |
set +e
echo "=== CI 失败通知 ==="
FAILED_JOB="Build Production Runtime Images" python3 scripts/ci_notify_failure.py
FAILED_JOB="Build Production Web Image" python3 scripts/ci_notify_failure.py
deploy-production:
name: Deploy Production
runs-on: saas
timeout-minutes: 20
if: startsWith(github.ref, 'refs/tags/v')
needs: build-production-runtime-images
needs: [build-production-api, build-production-worker, build-production-web]
steps:
- name: Checkout code
shell: sh
env:
GITHUB_TOKEN: ${{ github.token }}
run: |
set -eu
python3 - <<'INNERPY'
import io, os, tarfile, time, urllib.request, urllib.error
url = f"{os.environ['GITHUB_API_URL']}/repos/{os.environ['GITHUB_REPOSITORY']}/archive/{os.environ['GITHUB_SHA']}.tar.gz"
request = urllib.request.Request(url, headers={"Authorization": f"token {os.environ['GITHUB_TOKEN']}"})
last_err = None
for attempt in range(5):
try:
with urllib.request.urlopen(request, timeout=120) as response:
archive = response.read()
break
except urllib.error.HTTPError as e:
last_err = e
if e.code >= 500 and attempt < 4:
wait = 2 ** attempt
print(f"Checkout HTTP {e.code}, retrying in {wait}s (attempt {attempt+1}/5)...")
time.sleep(wait)
continue
raise
except Exception as e:
last_err = e
if attempt < 4:
wait = 2 ** attempt
print(f"Checkout error: {e}, retrying in {wait}s (attempt {attempt+1}/5)...")
time.sleep(wait)
continue
else:
raise last_err
with tarfile.open(fileobj=io.BytesIO(archive), mode='r:gz') as tar:
root_prefix = tar.getmembers()[0].name.split('/', 1)[0] + '/'
for member in tar.getmembers():
name = member.name
if name == root_prefix[:-1]:
continue
if name.startswith(root_prefix):
member.name = name[len(root_prefix):]
if member.name:
tar.extract(member, '.')
INNERPY
- name: Install SSH client
shell: sh
run: |
@@ -1104,6 +1895,52 @@ jobs:
echo "$DEPLOY_B64" | base64 -d | ssh -p 22222 -i "$key_path" "$production_user@$production_host" "IMAGE_TAG='${GITHUB_REF_NAME}' REGISTRY_TOKEN='${REGISTRY_TOKEN}' sh"
- name: Production smoke test (健康检查门禁)
if: success()
shell: sh
run: |
set -eu
API_BASE="https://api.xiaoxiajianji.com"
WEB_BASE="https://saas.xiaoxiajianji.com"
echo "=== 生产部署门禁:外部健康检查 ==="
echo "等待服务启动稳定(30s..."
sleep 30
echo "--- Check 1: API health endpoint ---"
for i in $(seq 1 20); do
HEALTH=$(curl -sf --max-time 10 "${API_BASE}/health") && break
echo " Attempt $i/20: not ready yet, waiting 5s..."
sleep 5
done
if [ -z "$HEALTH" ]; then
echo "FAIL: API /health unreachable after 100s"
echo "生产环境健康检查未通过,部署失败!"
echo "(回滚机制待接入,当前需手动回滚)"
exit 1
fi
echo "API health OK: $HEALTH"
echo "--- Check 2: API login endpoint (expect 401/422) ---"
HTTP_CODE=$(curl -s -o /dev/null -w "%{http_code}" --max-time 10 -X POST "${API_BASE}/api/v1/auth/login" -H "Content-Type: application/json" -d '{"email":"smoke@test.com","password":"wrong"}')
if [ "$HTTP_CODE" != "401" ] && [ "$HTTP_CODE" != "422" ]; then
echo "FAIL: login returned HTTP $HTTP_CODE (expected 401 or 422)"
exit 1
fi
echo "Login API OK: HTTP $HTTP_CODE"
echo "--- Check 3: Web frontend ---"
HTTP_CODE=$(curl -s -o /dev/null -w "%{http_code}" --max-time 10 "${WEB_BASE}/")
if [ "$HTTP_CODE" != "200" ]; then
echo "FAIL: web frontend returned HTTP $HTTP_CODE (expected 200)"
exit 1
fi
echo "Web frontend OK: HTTP $HTTP_CODE"
echo ""
echo "=== ✅ 生产环境健康检查全部通过 ==="
echo "Version: ${GITHUB_REF_NAME}"
- name: Notify CI success
if: success()
shell: sh
-1
View File
@@ -45,7 +45,6 @@ from packages.application.template.use_cases import (
CreateTemplateUseCase,
DeleteCategoryUseCase,
DeleteTemplateUseCase,
GetTemplateUsageUseCase,
GetTemplateUseCase,
ListCategoriesUseCase,
ListTagsUseCase,
+3
View File
@@ -251,6 +251,9 @@ test.describe("Core generation flow", () => {
await expect(page.locator(".xx-products-page")).toBeVisible({
timeout: 15_000,
});
// 清理所有路由,避免页面关闭时飞地API请求导致测试报错
await page.unrouteAll({ behavior: "ignoreErrors" });
});
test("generation task API creates and lists tasks", async ({ request }) => {
+697
View File
@@ -0,0 +1,697 @@
"""视频拼接/合并引擎 — 多段视频按顺序拼接成一个成片.
基于 FFmpeg 实现两种拼接模式:
1. **concat demuxerstream copy**:最快,所有视频编码参数必须一致
2. **concat filter(重新编码)**:更灵活,支持不同分辨率/编码/帧率的视频
使用场景:
- 多段素材按顺序合并成一个视频
- 视频分割后重新拼接
- 片头 + 正片 + 片尾拼接
降级策略:
- 优先尝试 stream copy(速度快、无质量损失)
- 参数不一致时自动降级到 concat filter
- 某段视频失败时跳过,不阻断整体拼接
"""
from __future__ import annotations
import logging
import tempfile
from dataclasses import dataclass, field
from pathlib import Path
from typing import Any
from video_processing.ffmpeg_utils import FFMPEG_BIN, probe_duration, probe_video_info, run_ffmpeg
from video_processing.path_security import PathSecurityError, is_in_allowed_dirs, safe_resolve_path
logger = logging.getLogger(__name__)
# ── 常量 ──────────────────────────────────────────────────────────────────────
MAX_CONCAT_SEGMENTS = 50 # 最大拼接段数(安全上限,防止OOM)
ALLOWED_VIDEO_EXTENSIONS = {".mp4", ".mov", ".avi", ".mkv", ".webm", ".flv", ".wmv"}
# concat demuxer 要求一致的参数列表
CONCAT_DEMUXER_REQUIRED_PARAMS = [
"codec_name", # 视频编码
"width", # 宽度
"height", # 高度
"r_frame_rate", # 帧率
"pix_fmt", # 像素格式
"sample_rate", # 音频采样率
"channels", # 音频声道数
"audio_codec", # 音频编码
]
# ── 拼接片段配置 ──────────────────────────────────────────────────────────────
@dataclass
class ConcatSegment:
"""单个拼接片段."""
video_path: str # 视频文件路径
start_time: float = 0.0 # 开始时间(秒),从视频的哪个位置开始取
duration: float = 0.0 # 持续时长(秒),0表示取到末尾
has_audio: bool = True # 是否包含音频
@classmethod
def from_dict(cls, seg: dict) -> "ConcatSegment":
"""从字典创建拼接片段,带安全类型转换."""
try:
start_time = max(0.0, float(seg.get("start_time", 0.0)))
except (TypeError, ValueError):
start_time = 0.0
try:
duration = max(0.0, float(seg.get("duration", 0.0)))
except (TypeError, ValueError):
duration = 0.0
return cls(
video_path=str(seg.get("video_path", "")),
start_time=start_time,
duration=duration,
has_audio=bool(seg.get("has_audio", True)),
)
@dataclass
class ConcatConfig:
"""视频拼接配置."""
segments: list[ConcatSegment] = field(default_factory=list)
output_width: int = 0 # 输出宽度(0=自动取第一段)
output_height: int = 0 # 输出高度(0=自动取第一段)
output_fps: float = 0.0 # 输出帧率(0=自动取第一段)
force_reencode: bool = False # 强制重新编码(不用 stream copy
transition: str = "none" # 转场效果(none/crossfade- 预留
transition_duration: float = 0.3 # 转场时长
@classmethod
def from_config_dict(cls, config: dict | None) -> "ConcatConfig":
"""从配置字典创建 ConcatConfig."""
if not config or not isinstance(config, dict):
return cls()
segments_raw = config.get("segments", [])
segments: list[ConcatSegment] = []
if isinstance(segments_raw, list):
for s in segments_raw:
if isinstance(s, dict) and s.get("video_path"):
try:
seg = ConcatSegment.from_dict(s)
if seg.video_path:
segments.append(seg)
except Exception:
logger.warning("[concat] skip invalid segment: %s", s)
continue
try:
output_width = max(0, int(config.get("output_width", 0)))
except (TypeError, ValueError):
output_width = 0
try:
output_height = max(0, int(config.get("output_height", 0)))
except (TypeError, ValueError):
output_height = 0
try:
output_fps = max(0.0, float(config.get("output_fps", 0.0)))
except (TypeError, ValueError):
output_fps = 0.0
return cls(
segments=segments,
output_width=output_width,
output_height=output_height,
output_fps=output_fps,
force_reencode=bool(config.get("force_reencode", False)),
transition=str(config.get("transition", "none")),
transition_duration=max(0.1, float(config.get("transition_duration", 0.3))),
)
@property
def has_effect(self) -> bool:
"""是否有有效片段需要拼接."""
return len([s for s in self.segments if s.video_path]) >= 2
@property
def total_segments(self) -> int:
"""有效片段数量."""
return len([s for s in self.segments if s.video_path])
# ── 路径安全校验 ────────────────────────────────────────────────────────────
def _validate_video_path(video_path: str, work_dir: Path) -> None:
"""校验视频文件路径安全性.
规则:
- local:// schema → 必须在 work_dir 内
- 相对路径 → 必须在 work_dir 内
- 绝对路径 → 必须在允许目录白名单内
- 扩展名必须是视频格式
Raises:
PathSecurityError: 路径不安全
"""
if not video_path or not isinstance(video_path, str):
raise PathSecurityError("视频路径不能为空")
# 本地路径(local:// 或相对路径 / 绝对路径)
if video_path.startswith("local://") or not video_path.startswith(("http://", "https://", "oss://")):
is_abs = video_path.startswith("/") and not video_path.startswith("local://")
resolved_path = safe_resolve_path(
video_path,
work_dir,
allow_outside=is_abs,
allowed_extensions=ALLOWED_VIDEO_EXTENSIONS,
)
# 绝对路径额外检查白名单目录(用realpath规范化后的真实路径比较,防止 ../ 遍历绕过)
if is_abs:
resolved_work_dir = work_dir.resolve()
try:
resolved_path.relative_to(resolved_work_dir)
except ValueError:
if not is_in_allowed_dirs(resolved_path):
raise PathSecurityError(f"视频路径不在允许目录内: {video_path[:80]}")
# URL类型路径不做本地路径校验(由下载阶段的SSRF防护负责)
# 但检查扩展名
else:
path_part = video_path.split("?")[0].split("#")[0]
ext = Path(path_part).suffix.lower()
if ext and ext not in ALLOWED_VIDEO_EXTENSIONS:
raise PathSecurityError(f"不允许的视频文件类型: {ext}")
# ── 视频拼接引擎 ──────────────────────────────────────────────────────────────
class ConcatEngine:
"""视频拼接引擎 — 支持 stream copy 和重新编码两种模式."""
def __init__(self, work_dir: Path):
self.work_dir = work_dir
self.work_dir.mkdir(parents=True, exist_ok=True)
# ── 主入口 ────────────────────────────────────────────────────────
def concat_videos(
self,
config: ConcatConfig,
output_path: Path,
) -> Path:
"""拼接多段视频.
自动选择最优拼接策略:
1. 所有片段参数一致 → concat demuxerstream copy,最快)
2. 参数不一致或有裁剪 → concat filter(重新编码)
Args:
config: 拼接配置
output_path: 输出文件路径
Returns:
输出文件路径
"""
valid_segments = [s for s in config.segments if s.video_path]
if not valid_segments:
raise ValueError("No valid video segments to concat")
# ── 安全校验:段数上限 ──
if len(valid_segments) > MAX_CONCAT_SEGMENTS:
raise ValueError(f"Too many concat segments: {len(valid_segments)} > {MAX_CONCAT_SEGMENTS}")
# ── 安全校验:所有视频路径白名单校验 ──
safe_segments = []
for seg in valid_segments:
try:
_validate_video_path(seg.video_path, self.work_dir)
safe_segments.append(seg)
except PathSecurityError as e:
logger.warning("[concat] skip segment: path security check failed: %s", e)
if len(safe_segments) != len(valid_segments):
valid_segments = safe_segments
config.segments = safe_segments
logger.info("[concat] %d segments passed security check", len(safe_segments))
if not valid_segments:
raise ValueError("No valid video segments after security check")
if len(valid_segments) == 1:
# 只有一段,直接复制
import shutil
logger.info("[concat] single segment, copy directly")
shutil.copy2(valid_segments[0].video_path, output_path)
return output_path
# 判断能否用 stream copy
can_stream_copy = self._can_use_stream_copy(config)
if can_stream_copy and not config.force_reencode:
logger.info("[concat] using concat demuxer (stream copy)")
try:
return self._concat_demuxer(config, output_path)
except Exception as e:
logger.warning("[concat] demuxer failed, fallback to filter: %s", e)
# 降级到 concat filter
logger.info("[concat] using concat filter (re-encode)")
return self._concat_filter(config, output_path)
# ── 模式判断 ──────────────────────────────────────────────────────
def _can_use_stream_copy(self, config: ConcatConfig) -> bool:
"""判断是否可以使用 concat demuxerstream copy.
条件:
1. 所有视频编码参数一致(分辨率、帧率、编码、像素格式)
2. 所有音频参数一致(采样率、声道、编码)
3. 没有设置 start_time 裁剪(或可以通过 concat demuxer 的 inpoint/outpoint 实现)
4. 没有强制重新编码
"""
if config.force_reencode:
return False
# 如果有转场效果,必须重新编码
if config.transition != "none":
return False
# 探测所有视频的参数
video_infos = []
for seg in config.segments:
if not seg.video_path:
continue
try:
info = probe_video_info(seg.video_path)
video_infos.append(info)
except Exception:
logger.warning("[concat] probe failed for %s", seg.video_path[-40:])
return False
if len(video_infos) < 2:
return False
# 检查参数一致性
base_info = video_infos[0]
for info in video_infos[1:]:
for param in CONCAT_DEMUXER_REQUIRED_PARAMS:
base_val = base_info.get(param)
curr_val = info.get(param)
if base_val != curr_val:
logger.debug(
"[concat] param mismatch: %s (%s vs %s)",
param,
base_val,
curr_val,
)
return False
# 检查是否有裁剪需求
# concat demuxer 支持 inpoint/outpoint,所以有裁剪也可以用
# 但为了简单和稳定性,有裁剪时也用 filter 模式
# inpoint/outpoint 不是所有格式都支持得好)
has_trimming = any(seg.start_time > 0 or seg.duration > 0 for seg in config.segments if seg.video_path)
if has_trimming:
return False
return True
# ── 模式1concat demuxerstream copy) ──────────────────────────
def _concat_demuxer(self, config: ConcatConfig, output_path: Path) -> Path:
"""使用 concat demuxer 拼接(stream copy.
优点:速度极快,无质量损失
缺点:要求所有视频参数完全一致
"""
# 生成 concat 文件列表
list_file = self.work_dir / "concat_list.txt"
lines = []
for seg in config.segments:
if not seg.video_path:
continue
# 路径转义:单引号替换为 '\''
safe_path = str(seg.video_path).replace("'", "'\\''")
lines.append(f"file '{safe_path}'")
list_file.write_text("\n".join(lines), encoding="utf-8")
command = [
FFMPEG_BIN,
"-y",
"-f",
"concat",
"-safe",
"0",
"-i",
str(list_file),
"-c",
"copy",
"-copyts",
str(output_path),
]
logger.info("[concat] demuxer: %d segments", config.total_segments)
run_ffmpeg(command)
return output_path
# ── 模式2concat filter(重新编码) ──────────────────────────────
def _concat_filter(self, config: ConcatConfig, output_path: Path) -> Path:
"""使用 concat filter 拼接(重新编码).
优点:支持不同参数的视频,支持裁剪
缺点:需要重新编码,较慢
"""
valid_segments = [s for s in config.segments if s.video_path]
num_segments = len(valid_segments)
# 构建输入参数
input_args: list[str] = []
for seg in valid_segments:
input_args.extend(["-i", seg.video_path])
# 确定输出参数
output_width, output_height, output_fps = self._get_output_params(config)
# 构建 filter_complex
filter_parts: list[str] = []
concat_inputs = ""
for i, seg in enumerate(valid_segments):
vid_label = f"v{i}"
aud_label = f"a{i}"
seg_filters: list[str] = []
# 1. 裁剪(start_time + duration
if seg.start_time > 0 or seg.duration > 0:
start = seg.start_time
if seg.duration > 0:
end = start + seg.duration
seg_filters.append(f"trim=start={start:.3f}:end={end:.3f}")
else:
seg_filters.append(f"trim=start={start:.3f}")
seg_filters.append("setpts=PTS-STARTPTS")
# 音频同步裁剪
if seg.has_audio:
if seg.duration > 0:
filter_parts.append(
f"[{i}:a]atrim=start={start:.3f}:end={end:.3f}," f"asetpts=PTS-STARTPTS[{aud_label}]"
)
else:
filter_parts.append(f"[{i}:a]atrim=start={start:.3f}," f"asetpts=PTS-STARTPTS[{aud_label}]")
else:
# 无音频时生成静音轨
filter_parts.append(
f"[{i}:v]trim=start={start:.3f}," f"setpts=PTS-STARTPTS, " f"aevalsrc=0:d={0.1}[{aud_label}]"
)
else:
# 无裁剪,直接用原始标签
if not seg.has_audio:
# 无音频时需要生成静音
try:
dur = probe_duration(seg.video_path)
except Exception:
dur = 10.0
filter_parts.append(f"aevalsrc=0:d={dur:.3f}:s=44100[{aud_label}]")
# 2. 缩放/帧率统一
vf_parts = []
if not seg_filters:
vf_parts.append(f"[{i}:v]")
else:
vf_parts.append("")
# 分辨率统一
if output_width and output_height:
vf_parts.append(
f"scale={output_width}:{output_height}:force_original_aspect_ratio=decrease,"
f"pad={output_width}:{output_height}:(ow-iw)/2:(oh-ih)/2:black"
)
# 帧率统一
if output_fps > 0:
vf_parts.append(f"fps={output_fps}")
# 像素格式统一
vf_parts.append("format=yuv420p")
if len(vf_parts) > 1 or (seg_filters and vf_parts):
if seg_filters:
# 先裁剪后缩放
crop_str = "".join(seg_filters)
scale_str = "".join(vf_parts[1:]) # 跳过空字符串
if scale_str:
filter_parts.append(f"[{i}:v]{crop_str},{scale_str}[{vid_label}]")
else:
filter_parts.append(f"[{i}:v]{crop_str}[{vid_label}]")
else:
filter_parts.append(f"{vf_parts[0]}{''.join(vf_parts[1:])}[{vid_label}]")
else:
if seg_filters:
filter_parts.append(f"[{i}:v]{''.join(seg_filters)}[{vid_label}]")
else:
# 什么都不需要,直接用输入
pass
# 拼接 concat 的输入标签
if seg_filters or (output_width and output_height) or output_fps > 0:
concat_inputs += f"[{vid_label}]"
else:
concat_inputs += f"[{i}:v]"
# 音频标签
if seg.start_time > 0 or seg.duration > 0:
# 已经生成了 aud_label
pass
elif not seg.has_audio:
# 已经生成了静音 aud_label
pass
else:
# 使用原始音频
pass
# 简化处理:用更直接的方式构建 filter
# 重新整理一下,确保所有输入都有对应的 v_i 和 a_i 标签
filter_parts.clear()
concat_inputs = "" # 按段交织: [v0][a0][v1][a1]...
for i, seg in enumerate(valid_segments):
v_label = f"v{i}_in"
a_label = f"a{i}_in"
# 视频处理链
v_steps: list[str] = [f"[{i}:v]"]
# 裁剪
if seg.start_time > 0 or seg.duration > 0:
start = seg.start_time
if seg.duration > 0:
end = start + seg.duration
v_steps.append(f"trim=start={start:.3f}:end={end:.3f},")
else:
v_steps.append(f"trim=start={start:.3f},")
v_steps.append("setpts=PTS-STARTPTS,")
# 缩放
if output_width and output_height:
v_steps.append(
f"scale={output_width}:{output_height}:force_original_aspect_ratio=decrease,"
f"pad={output_width}:{output_height}:(ow-iw)/2:(oh-ih)/2:black,"
)
# 帧率
if output_fps > 0:
v_steps.append(f"fps={output_fps},")
# 像素格式
v_steps.append("format=yuv420p")
v_filter = "".join(v_steps) + f"[{v_label}]"
filter_parts.append(v_filter)
# 音频处理链
a_steps: list[str] = []
if seg.has_audio:
a_steps.append(f"[{i}:a]")
if seg.start_time > 0 or seg.duration > 0:
start = seg.start_time
if seg.duration > 0:
end = start + seg.duration
a_steps.append(f"atrim=start={start:.3f}:end={end:.3f},")
else:
a_steps.append(f"atrim=start={start:.3f},")
a_steps.append("asetpts=PTS-STARTPTS,")
a_steps.append("aformat=sample_fmts=fltp:sample_rates=44100:channel_layouts=stereo")
else:
# 生成静音音频
try:
dur = probe_duration(seg.video_path)
except Exception:
dur = 10.0
# 减去裁剪
if seg.start_time > 0:
dur = max(0.1, dur - seg.start_time)
if seg.duration > 0 and seg.duration < dur:
dur = seg.duration
a_steps.append(f"aevalsrc=0:d={dur:.3f}:s=44100:c=stereo")
a_filter = "".join(a_steps) + f"[{a_label}]"
filter_parts.append(a_filter)
# 按段交织排列(v_i, a_i),这是 FFmpeg concat filter 要求的顺序
concat_inputs += f"[{v_label}][{a_label}]"
# concat filter: 输入按 [v0][a0][v1][a1]... 顺序
filter_parts.append(f"{concat_inputs}" f"concat=n={num_segments}:v=1:a=1[vout][aout]")
filter_complex = ";".join(filter_parts)
command = [
FFMPEG_BIN,
"-y",
*input_args,
"-filter_complex",
filter_complex,
"-map",
"[vout]",
"-map",
"[aout]",
"-c:v",
"libx264",
"-preset",
"fast",
"-crf",
"23",
"-c:a",
"aac",
"-b:a",
"128k",
"-movflags",
"+faststart",
str(output_path),
]
logger.info(
"[concat] filter: %d segments, %dx%d, %.2f fps",
num_segments,
output_width,
output_height,
output_fps,
)
run_ffmpeg(command)
return output_path
# ── 辅助方法 ──────────────────────────────────────────────────────
def _get_output_params(self, config: ConcatConfig) -> tuple[int, int, float]:
"""获取输出参数(宽、高、帧率).
优先级:
1. config 中显式指定的
2. 第一段视频的参数
"""
valid_segments = [s for s in config.segments if s.video_path]
width = config.output_width
height = config.output_height
fps = config.output_fps
# 如果没有显式指定,用第一段的参数
if (width == 0 or height == 0 or fps == 0) and valid_segments:
try:
info = probe_video_info(valid_segments[0].video_path)
if width == 0:
width = int(info.get("width", 1080))
if height == 0:
height = int(info.get("height", 1920))
if fps == 0:
fps_str = info.get("r_frame_rate", "30/1")
if "/" in str(fps_str):
num, den = str(fps_str).split("/")
try:
fps = float(num) / float(den)
except (ValueError, ZeroDivisionError):
fps = 30.0
else:
fps = float(fps_str) if fps_str else 30.0
except Exception:
# 探测失败,用默认值
if width == 0:
width = 1080
if height == 0:
height = 1920
if fps == 0:
fps = 30.0
return width, height, fps
# ── 便捷函数 ──────────────────────────────────────────────────────────────────
def concat_video_files(
video_paths: list[str],
output_path: Path,
*,
work_dir: Path | None = None,
force_reencode: bool = False,
) -> Path:
"""简单拼接多个视频文件.
Args:
video_paths: 视频文件路径列表
output_path: 输出路径
work_dir: 工作目录(默认输出文件所在目录)
force_reencode: 是否强制重新编码
Returns:
输出文件路径
"""
if work_dir is None:
work_dir = output_path.parent
segments = [ConcatSegment(video_path=p) for p in video_paths if p]
config = ConcatConfig(segments=segments, force_reencode=force_reencode)
engine = ConcatEngine(work_dir)
return engine.concat_videos(config, output_path)
def concat_videos_from_config(
config_dict: dict | None,
output_path: Path,
*,
work_dir: Path,
) -> Path | None:
"""从配置字典执行视频拼接.
降级策略:配置无效或拼接失败时返回 None.
"""
config = ConcatConfig.from_config_dict(config_dict)
if not config.has_effect:
return None
try:
engine = ConcatEngine(work_dir)
return engine.concat_videos(config, output_path)
except Exception as e:
logger.error("[concat] concat failed: %s", e)
return None
@@ -119,6 +119,54 @@ def run_ffmpeg(
raise
def run_ffprobe(
command: list[str],
*,
capture_output: bool = True,
timeout: int = 30,
) -> tuple[str, str]:
"""执行 FFprobe 命令。
Args:
command: 完整的 ffprobe 命令列表(含 "ffprobe" 本身)
capture_output: 是否捕获 stdout/stderr
timeout: 超时时间(秒),默认 30s;None 表示不设超时
Returns:
(stdout, stderr) 元组
Raises:
subprocess.CalledProcessError: 命令执行失败时抛出
subprocess.TimeoutExpired: 超时未完成时抛出
"""
try:
result = subprocess.run( # nosec B603
command,
check=True,
stdout=subprocess.PIPE if capture_output else None,
stderr=subprocess.PIPE if capture_output else None,
text=True,
timeout=timeout,
)
return (result.stdout or "", result.stderr or "")
except subprocess.TimeoutExpired:
logger.error(
"FFprobe 命令超时 (%ds): command=%s",
timeout or -1,
" ".join(str(c) for c in command[:20]),
)
raise
except subprocess.CalledProcessError as e:
stderr_text = (e.stderr or "").strip()
logger.error(
"FFprobe 命令失败: exit_code=%d command=%s\nstderr:\n%s",
e.returncode,
" ".join(str(c) for c in command[:20]),
stderr_text[:5000],
)
raise
def probe_has_audio(local_path: str | Path) -> bool:
"""探测文件是否包含音频流。
+478
View File
@@ -0,0 +1,478 @@
"""多轨道混音引擎 — 支持多路音频独立音量调节与混合.
基于 FFmpeg amix / amerge 实现:
- 支持任意数量音频轨道(原音、BGM、配音、音效等)
- 每轨独立音量调节
- 每轨独立淡入淡出
- 每轨独立时间偏移(delay
- 总输出音量归一化补偿
作为 render_audio.py 的增强模块,在 mix_audio 后处理阶段被调用。
与 bgm_mixer.py 的关系:
- bgm_mixer 专注 BGM 单轨道的复杂处理(循环、人声闪避)
- 本模块专注多路轨道的统一音量调节与混合
"""
from __future__ import annotations
import logging
from dataclasses import dataclass, field
from pathlib import Path
from typing import TYPE_CHECKING
from video_processing.ffmpeg_utils import FFMPEG_BIN, probe_duration, run_ffmpeg
from video_processing.path_security import PathSecurityError, is_in_allowed_dirs, safe_resolve_path
if TYPE_CHECKING:
from video_processing.render_audio import RenderContext
logger = logging.getLogger(__name__)
# ── 常量 ──────────────────────────────────────────────────────────────────────
TRACK_TYPE_MAIN = "main" # 原音(视频原声)
TRACK_TYPE_BGM = "bgm" # 背景音乐
TRACK_TYPE_VOICEOVER = "voiceover" # 配音(TTS/人声)
TRACK_TYPE_SFX = "sfx" # 音效
TRACK_TYPE_AMBIENT = "ambient" # 环境音
MAX_AUDIO_TRACKS = 8 # 最大混音轨道数(安全上限,防止资源耗尽)
# 各轨道默认音量(相对主音频)
DEFAULT_VOLUMES = {
TRACK_TYPE_MAIN: 1.0,
TRACK_TYPE_BGM: 0.3,
TRACK_TYPE_VOICEOVER: 1.0,
TRACK_TYPE_SFX: 0.7,
TRACK_TYPE_AMBIENT: 0.2,
}
@dataclass
class AudioTrack:
"""单条音频轨道配置."""
track_id: str # 轨道唯一标识
track_type: str # 轨道类型(main/bgm/voiceover/sfx/ambient
audio_path: str # 音频文件路径
volume: float = 1.0 # 音量 0.0 ~ 2.0
fade_in: float = 0.0 # 淡入时长(秒)
fade_out: float = 0.0 # 淡出时长(秒)
start_time: float = 0.0 # 开始时间(相对于视频起点,秒)
duration: float = 0.0 # 持续时长(0表示到文件末尾)
enabled: bool = True # 是否启用
@classmethod
def from_dict(cls, track: dict) -> "AudioTrack":
"""从字典创建 AudioTrack,带安全类型转换."""
track_type = str(track.get("track_type", TRACK_TYPE_SFX))
default_vol = DEFAULT_VOLUMES.get(track_type, 1.0)
try:
volume = float(track.get("volume", default_vol))
except (TypeError, ValueError):
volume = default_vol
volume = max(0.0, min(2.0, volume))
try:
fade_in = max(0.0, float(track.get("fade_in", 0.0)))
except (TypeError, ValueError):
fade_in = 0.0
try:
fade_out = max(0.0, float(track.get("fade_out", 0.0)))
except (TypeError, ValueError):
fade_out = 0.0
try:
start_time = max(0.0, float(track.get("start_time", 0.0)))
except (TypeError, ValueError):
start_time = 0.0
try:
duration = max(0.0, float(track.get("duration", 0.0)))
except (TypeError, ValueError):
duration = 0.0
return cls(
track_id=str(track.get("track_id", "")),
track_type=track_type,
audio_path=str(track.get("audio_path", "")),
volume=volume,
fade_in=fade_in,
fade_out=fade_out,
start_time=start_time,
duration=duration,
enabled=bool(track.get("enabled", True)),
)
@dataclass
class MultiTrackMixConfig:
"""多轨道混音配置."""
tracks: list[AudioTrack] = field(default_factory=list)
master_volume: float = 1.0 # 主输出音量
normalize: bool = True # 是否自动归一化补偿
max_output_volume: float = 1.5 # 最大输出音量(防止爆音)
@classmethod
def from_config_dict(cls, config: dict | None) -> "MultiTrackMixConfig":
"""从 plan.config.audio_tracks 字典创建配置."""
if not config or not isinstance(config, dict):
return cls()
tracks_raw = config.get("tracks", [])
tracks: list[AudioTrack] = []
if isinstance(tracks_raw, list):
for t in tracks_raw:
if isinstance(t, dict) and t.get("audio_path"):
try:
track = AudioTrack.from_dict(t)
if track.enabled and track.audio_path:
tracks.append(track)
except Exception:
logger.warning("[multi-track] skip invalid track config: %s", t)
continue
try:
master_volume = float(config.get("master_volume", 1.0))
master_volume = max(0.0, min(2.0, master_volume))
except (TypeError, ValueError):
master_volume = 1.0
return cls(
tracks=tracks,
master_volume=master_volume,
normalize=bool(config.get("normalize", True)),
max_output_volume=float(config.get("max_output_volume", 1.5)),
)
@property
def has_effect(self) -> bool:
"""是否有有效轨道需要混音."""
return len([t for t in self.tracks if t.enabled and t.audio_path]) > 0
# ── 路径安全校验 ────────────────────────────────────────────────────────────
ALLOWED_AUDIO_EXTENSIONS = {".mp3", ".wav", ".aac", ".ogg", ".flac", ".m4a", ".wma"}
def _validate_audio_path(audio_path: str, work_dir: Path) -> None:
"""校验音频文件路径安全性.
规则:
- local:// schema → 必须在 work_dir 内
- 相对路径 → 必须在 work_dir 内
- 绝对路径 → 必须在允许目录白名单内
- 扩展名必须是音频格式
Raises:
PathSecurityError: 路径不安全
"""
if not audio_path or not isinstance(audio_path, str):
raise PathSecurityError("音频路径不能为空")
# 本地路径(local:// 或相对路径)
if audio_path.startswith("local://") or not audio_path.startswith(("http://", "https://", "oss://")):
is_abs = audio_path.startswith("/") and not audio_path.startswith("local://")
resolved_path = safe_resolve_path(
audio_path,
work_dir,
allow_outside=is_abs,
allowed_extensions=ALLOWED_AUDIO_EXTENSIONS,
)
# 绝对路径额外检查白名单目录(用realpath规范化后的真实路径比较,防止 ../ 遍历绕过)
if is_abs:
resolved_work_dir = work_dir.resolve()
try:
resolved_path.relative_to(resolved_work_dir)
except ValueError:
if not is_in_allowed_dirs(resolved_path):
raise PathSecurityError(f"音频路径不在允许目录内: {audio_path[:80]}")
# URL类型路径不做本地路径校验(由下载阶段的SSRF防护负责)
# 但检查扩展名
else:
# URL路径,检查扩展名白名单(取 ? 之前的部分)
path_part = audio_path.split("?")[0].split("#")[0]
from pathlib import Path as _P
ext = _P(path_part).suffix.lower()
if ext and ext not in ALLOWED_AUDIO_EXTENSIONS:
raise PathSecurityError(f"不允许的音频文件类型: {ext}")
# ── 单轨道预处理 ────────────────────────────────────────────────────────────
def _prepare_single_track(
ctx: "RenderContext",
track: AudioTrack,
target_duration: float,
output_path: Path,
) -> bool:
"""预处理单条轨道:音量 + 淡入淡出 + 时间偏移 + 截断.
生成一个精确对齐时间轴的音频文件,后续统一 amix 混音。
Returns:
True 表示处理成功,False 表示失败(跳过)
"""
try:
audio_dur = probe_duration(track.audio_path)
except Exception:
logger.warning("[multi-track] probe failed, skip track: %s", track.track_id)
return False
if audio_dur <= 0:
return False
# 计算实际有效时长
effective_start = track.start_time
if track.duration > 0:
effective_dur = min(track.duration, audio_dur)
else:
effective_dur = audio_dur
# 如果轨道完全在视频时长之外,跳过
if effective_start >= target_duration:
return False
if effective_start + effective_dur <= 0:
return False
# 构建滤镜链
filter_parts: list[str] = []
# 1. 先截断到有效范围
trim_start = 0.0 # 从源文件的哪个位置开始取
if effective_start < 0:
trim_start = -effective_start
effective_start = 0.0
# 实际需要的源时长
need_dur = min(effective_dur, target_duration - effective_start)
if need_dur <= 0:
return False
filter_parts.append(f"atrim={trim_start:.3f}:{trim_start + need_dur:.3f}")
filter_parts.append("asetpts=N/SR/TB")
# 2. 音量调节
if abs(track.volume - 1.0) > 0.001:
filter_parts.append(f"volume={track.volume:.3f}")
# 3. 淡入
if track.fade_in > 0 and track.fade_in < need_dur:
filter_parts.append(f"afade=t=in:st=0:d={track.fade_in:.3f}")
# 4. 淡出
if track.fade_out > 0 and track.fade_out < need_dur:
fade_start = need_dur - track.fade_out
if fade_start > 0:
filter_parts.append(f"afade=t=out:st={fade_start:.3f}:d={track.fade_out:.3f}")
# 5. 时间偏移(用 adelay 实现开头静音填充)
if effective_start > 0.01:
delay_ms = int(effective_start * 1000)
filter_parts.append(f"adelay={delay_ms}|{delay_ms}")
# 6. 最终截断到目标总时长
filter_parts.append(f"atrim=0:{target_duration:.3f}")
filter_parts.append("asetpts=N/SR/TB")
filter_str = ",".join(filter_parts)
command = [
FFMPEG_BIN,
"-y",
"-i",
track.audio_path,
"-filter:a",
filter_str,
"-c:a",
"aac",
"-b:a",
"128k",
str(output_path),
]
logger.info(
"[multi-track] prepare track: id=%s type=%s vol=%.2f start=%.2f dur=%.2f",
track.track_id,
track.track_type,
track.volume,
effective_start,
need_dur,
)
try:
run_ffmpeg(command)
return True
except Exception as e:
logger.warning("[multi-track] track prepare failed: %s, error=%s", track.track_id, e)
return False
# ── 多轨道混音主入口 ─────────────────────────────────────────────────────────
def mix_multi_track(
ctx: "RenderContext",
main_audio_path: Path,
config: MultiTrackMixConfig,
target_duration: float,
) -> Path:
"""多轨道混音:主音频 + 多条附加轨道.
Args:
ctx: 渲染上下文
main_audio_path: 主音频文件路径(原音)
config: 多轨道混音配置
target_duration: 目标总时长
Returns:
混音后的音频文件路径
"""
output_path = ctx.work_dir / f"multi_track_mix_{ctx.plan_id}.aac"
if target_duration <= 0:
target_duration = 5.0
# ── 安全校验:轨道数量上限 ──
enabled_tracks = [t for t in config.tracks if t.enabled and t.audio_path]
if len(enabled_tracks) > MAX_AUDIO_TRACKS:
logger.warning(
"[multi-track] too many tracks: %d > %d, truncating to max",
len(enabled_tracks),
MAX_AUDIO_TRACKS,
)
enabled_tracks = enabled_tracks[:MAX_AUDIO_TRACKS]
# 更新 config.tracks 为截断后的列表
config.tracks = enabled_tracks
# ── 安全校验:所有音频路径白名单校验 ──
# 主音频路径
try:
_validate_audio_path(str(main_audio_path), ctx.work_dir)
except PathSecurityError as e:
logger.error("[multi-track] main audio path security check failed: %s", e)
raise
# 各轨道音频路径
valid_tracks = []
for track in enabled_tracks:
try:
_validate_audio_path(track.audio_path, ctx.work_dir)
valid_tracks.append(track)
except PathSecurityError as e:
logger.warning("[multi-track] skip track %s: path security check failed: %s", track.track_id, e)
if len(valid_tracks) != len(enabled_tracks):
config.tracks = valid_tracks
logger.info("[multi-track] %d tracks passed security check", len(valid_tracks))
# 收集所有有效轨道(已预处理好的)
prepared_tracks: list[Path] = []
# 主音频作为第0轨
prepared_tracks.append(main_audio_path)
# 预处理每条附加轨道
for i, track in enumerate(config.tracks):
if not track.enabled or not track.audio_path:
continue
track_out = ctx.work_dir / f"track_{i}_{ctx.plan_id}.aac"
if _prepare_single_track(ctx, track, target_duration, track_out):
prepared_tracks.append(track_out)
# 如果只有主音频,直接返回(无需混音)
if len(prepared_tracks) <= 1:
import shutil
shutil.copy2(main_audio_path, output_path)
return output_path
# 使用 amix 混音
num_inputs = len(prepared_tracks)
# 构建输入参数
input_args: list[str] = []
for tp in prepared_tracks:
input_args.extend(["-i", str(tp)])
# amix 的 duration=first 以第一个输入(主音频)时长为准
# normalize 补偿:amix 会把每路音量除以 N,需要乘回来
# 但如果所有轨道都同时有声,可能会爆音,所以用 master_volume 控制
if config.normalize:
# 经验值:不是所有轨道都同时有声,补偿系数取 N * 0.7
compensate = num_inputs * 0.7
else:
compensate = 1.0
final_volume = compensate * config.master_volume
final_volume = min(final_volume, config.max_output_volume)
# 构建 filter_complex
inputs_label = "".join(f"[{i}:a]" for i in range(num_inputs))
filter_complex = (
f"{inputs_label}amix=inputs={num_inputs}:duration=first:dropout_transition=0[outa];"
f"[outa]volume={final_volume:.3f}[final]"
)
command = [
FFMPEG_BIN,
"-y",
*input_args,
"-filter_complex",
filter_complex,
"-map",
"[final]",
"-c:a",
"aac",
"-b:a",
"128k",
str(output_path),
]
logger.info(
"[multi-track] mix %d tracks, master_vol=%.2f compensate=%.2f final_vol=%.2f",
num_inputs,
config.master_volume,
compensate,
final_volume,
)
try:
run_ffmpeg(command)
except Exception as e:
logger.error("[multi-track] mix failed, fallback to main audio only: %s", e)
import shutil
shutil.copy2(main_audio_path, output_path)
return output_path
# ── 便捷函数:从 plan.config 快速混音 ───────────────────────────────────────
def mix_audio_tracks_from_config(
ctx: "RenderContext",
main_audio_path: Path,
audio_tracks_config: dict | None,
target_duration: float,
) -> Path:
"""从 plan.config.audio_tracks 配置执行多轨道混音.
降级策略:配置无效或混音失败时返回主音频。
"""
config = MultiTrackMixConfig.from_config_dict(audio_tracks_config)
if not config.has_effect:
return main_audio_path
return mix_multi_track(ctx, main_audio_path, config, target_duration)
+48 -9
View File
@@ -220,25 +220,64 @@ def resolve_asset_path(asset_id: str, work_dir: Path) -> Path | None:
"""从 asset_id 解析到本地文件路径。
策略(按优先级):
1. 如果 asset_id 是本地绝对路径(/var/storage/...)→ 直接返回
1. 如果 asset_id 是本地绝对路径(/var/storage/...)→ 安全校验后返回
2. 如果 work_dir 下已有缓存文件 → 返回缓存路径
3. 从 OSS 下载到 work_dir/{hash}.mp4 → 返回下载路径
4. 下载失败 → 返回 None
缓存策略:以 asset_id 的 SHA256 前 16 位为文件名,避免重复下载。
"""
# 1. 本地绝对路径
if asset_id.startswith("/") and os.path.exists(asset_id):
return Path(asset_id)
# 2. 缓存命中
安全:
- 本地绝对路径必须在 ASSET_ALLOWED_DIRS 环境变量指定的目录内
- 文件名经过 sanitize,防止路径遍历
- 禁止空字节、控制字符
"""
from video_processing.path_security import (
PathSecurityError,
get_allowed_local_dirs,
is_in_allowed_dirs,
sanitize_filename,
)
if not asset_id or not isinstance(asset_id, str):
return None
# 空字节检测
if "\x00" in asset_id:
logger.warning("asset_id 包含空字节,拒绝: %s", asset_id[:50])
return None
# 1. 本地绝对路径 — 必须在允许的目录内
if asset_id.startswith("/") and os.path.exists(asset_id):
try:
resolved = Path(asset_id).resolve()
if is_in_allowed_dirs(resolved, get_allowed_local_dirs()):
return resolved
else:
logger.warning(
"本地素材路径不在允许目录内,拒绝: %s (allowed=%s)",
asset_id[:80],
get_allowed_local_dirs(),
)
return None
except (OSError, PathSecurityError):
return None
# 2. 缓存命中(使用 hash 而非原始 ID,防止路径遍历)
cache_hash = hashlib.sha256(asset_id.encode()).hexdigest()[:16]
cached_path = work_dir / f"{cache_hash}.mp4"
safe_name = sanitize_filename(cache_hash)
cached_path = work_dir / f"{safe_name}.mp4"
if cached_path.exists() and cached_path.stat().st_size > 0:
return cached_path
# 3. 从 OSS 下载
if download_asset(asset_id, cached_path):
# 3. 从 OSS 下载(先标准化 key,防止路径遍历注入)
safe_key = normalize_storage_key(asset_id)
# 额外校验:存储键不能包含 ../ 或绝对路径
if ".." in safe_key or safe_key.startswith("/"):
logger.warning("asset_id 包含路径遍历模式,拒绝下载: %s", asset_id[:80])
return None
if download_asset(safe_key, cached_path):
return cached_path
return None
@@ -0,0 +1,303 @@
"""路径安全校验工具 — 路径遍历防护.
统一的文件路径安全校验方案,覆盖所有渲染管线中的路径处理场景:
- 本地素材路径校验
- local:// 路径 schema 校验
- 工作目录内路径安全约束
- 防止路径遍历攻击 (../)
防护要点:
1. 所有用户可控路径必须在允许的目录内
2. 解析符号链接后的真实路径仍需在允许目录内
3. 禁止空路径、相对路径遍历、绝对路径逃逸
4. 路径字符限制与规范化
"""
from __future__ import annotations
import logging
import os
from pathlib import Path
logger = logging.getLogger(__name__)
# 最大路径长度
MAX_PATH_LENGTH = 4096
# 允许的文件扩展名(渲染相关)
ALLOWED_MEDIA_EXTENSIONS = {
".mp4",
".mov",
".avi",
".mkv",
".webm",
".flv",
".wmv", # 视频
".mp3",
".wav",
".aac",
".ogg",
".flac",
".m4a",
".wma", # 音频
".jpg",
".jpeg",
".png",
".gif",
".bmp",
".webp",
".tiff", # 图片
".srt",
".ass",
".vtt",
".sub", # 字幕
".txt",
".json", # 文本/配置
}
# local:// schema 前缀
LOCAL_SCHEMA_PREFIX = "local://"
class PathSecurityError(ValueError):
"""路径安全校验失败."""
pass
def safe_resolve_path(
input_path: str | Path,
base_dir: str | Path,
*,
allow_outside: bool = False,
allowed_extensions: set[str] | None = None,
) -> Path:
"""安全解析路径,确保最终路径在 base_dir 内.
Args:
input_path: 输入路径(相对或绝对)
base_dir: 基路径目录,解析后的路径必须在此目录内
allow_outside: 是否允许路径在 base_dir 外(默认禁止)
allowed_extensions: 允许的文件扩展名集合(None 表示不限制)
Returns:
解析后的绝对路径 Path 对象
Raises:
PathSecurityError: 路径不安全
"""
if input_path is None:
raise PathSecurityError("路径不能为空")
path_str = str(input_path).strip()
if not path_str:
raise PathSecurityError("路径不能为空")
if len(path_str) > MAX_PATH_LENGTH:
raise PathSecurityError(f"路径过长 ({len(path_str)} > {MAX_PATH_LENGTH})")
# 空字节检测(必须在 Path() 之前)
if "\x00" in path_str:
raise PathSecurityError("路径包含空字节")
# 处理 local:// schema
if path_str.startswith(LOCAL_SCHEMA_PREFIX):
path_str = path_str[len(LOCAL_SCHEMA_PREFIX) :]
# local:// 后必须是相对路径(相对于 base_dir),不能是绝对路径
if os.path.isabs(path_str):
raise PathSecurityError("local:// 路径不能是绝对路径")
# 规范化 base_dir
base_dir = Path(base_dir).resolve()
if not base_dir.is_dir():
raise PathSecurityError(f"基路径不是有效目录: {base_dir}")
# 解析输入路径
input_path_obj = Path(path_str)
# 如果是绝对路径且不允许外部路径
if input_path_obj.is_absolute() and not allow_outside:
raise PathSecurityError("禁止使用绝对路径(需在工作目录内)")
# 组合并解析为绝对路径
if input_path_obj.is_absolute():
full_path = input_path_obj.resolve()
else:
full_path = (base_dir / input_path_obj).resolve()
# 检查路径遍历 — 确保最终路径在 base_dir 内
if not allow_outside:
try:
full_path.relative_to(base_dir)
except ValueError:
raise PathSecurityError(f"路径遍历检测:路径 '{path_str}' 超出基路径 '{base_dir}' 范围")
# 扩展名校验
if allowed_extensions is not None:
ext = full_path.suffix.lower()
if ext and ext not in allowed_extensions:
raise PathSecurityError(f"不允许的文件类型: {ext}")
# 检查危险路径模式
_check_dangerous_patterns(full_path)
return full_path
def _check_dangerous_patterns(path: Path) -> None:
"""检查危险路径模式."""
path_str = str(path)
# 检查空字节
if "\x00" in path_str:
raise PathSecurityError("路径包含空字节")
# 检查特殊设备文件(Linux
dangerous_prefixes = [
"/proc/",
"/sys/",
"/dev/",
"/etc/passwd",
"/etc/shadow",
"/root/",
"/boot/",
"/var/run/",
]
for prefix in dangerous_prefixes:
if path_str.startswith(prefix):
raise PathSecurityError(f"禁止访问系统路径: {prefix}")
def is_path_safe(
input_path: str | Path,
base_dir: str | Path,
*,
allow_outside: bool = False,
) -> bool:
"""便捷函数:检查路径是否安全,不抛异常."""
try:
safe_resolve_path(input_path, base_dir, allow_outside=allow_outside)
return True
except PathSecurityError:
return False
def validate_local_schema_path(
schema_path: str,
work_dir: str | Path,
) -> Path:
"""校验 local:// schema 路径,返回安全的本地路径.
local:// 路径规则:
- 必须以 local:// 开头
- 后面必须是相对路径
- 最终解析后必须在 work_dir 内
- 不允许 ../ 遍历
Args:
schema_path: local:// 开头的路径
work_dir: 工作目录
Returns:
解析后的安全路径
Raises:
PathSecurityError: 路径不安全
"""
if not schema_path.startswith(LOCAL_SCHEMA_PREFIX):
raise PathSecurityError(f"路径必须以 {LOCAL_SCHEMA_PREFIX} 开头")
return safe_resolve_path(schema_path, work_dir, allow_outside=False)
def sanitize_filename(filename: str) -> str:
"""清理文件名,移除危险字符.
保留:字母、数字、下划线、连字符、点、中文字符
移除:路径分隔符、控制字符、特殊符号等
"""
import re
if not filename:
return "unnamed"
# 移除路径分隔符和危险字符
# 保留: 字母数字、中文字符、下划线、连字符、点、空格
sanitized = re.sub(r'[\\/\x00-\x1f\x7f<>:"|?*]', "_", filename)
# 移除开头的点和连续的点(防止隐藏文件和路径遍历)
while sanitized.startswith("."):
sanitized = sanitized[1:]
# 限制长度
if len(sanitized) > 255:
name, ext = os.path.splitext(sanitized)
sanitized = name[: 255 - len(ext)] + ext
# 空文件名兜底
if not sanitized or sanitized == ".":
sanitized = "unnamed"
return sanitized
# ── 允许目录配置 ──────────────────────────────────────────────────────────────
def get_allowed_local_dirs() -> list[Path]:
"""获取允许的本地素材目录列表(从环境变量读取).
环境变量 ASSET_ALLOWED_DIRS,多个目录用冒号分隔(Linux)或分号分隔(Windows)。
默认包含 /tmp。
用于:
- resolve_asset_path 本地绝对路径白名单
- PiP local_path 类型白名单
- 贴纸本地路径白名单
"""
env_dirs = os.environ.get("ASSET_ALLOWED_DIRS", "")
dirs: list[Path] = []
if env_dirs:
import re
sep = ";" if os.name == "nt" else ":"
for d in re.split(f"[{sep}]", env_dirs):
d = d.strip()
if d:
try:
dirs.append(Path(d).resolve())
except OSError:
pass
# 默认允许 /tmp
if not dirs:
try:
dirs.append(Path("/tmp").resolve()) # nosec B108
except OSError:
pass
return dirs
def is_in_allowed_dirs(path: str | Path, allowed_dirs: list[Path] | None = None) -> bool:
"""检查路径是否在允许的目录列表内.
Args:
path: 待检查的路径
allowed_dirs: 允许的目录列表,None 则使用默认配置
Returns:
True 表示在允许目录内
"""
if allowed_dirs is None:
allowed_dirs = get_allowed_local_dirs()
try:
resolved = Path(path).resolve()
for allowed in allowed_dirs:
try:
resolved.relative_to(allowed)
return True
except ValueError:
continue
return False
except OSError:
return False
+32 -6
View File
@@ -465,18 +465,44 @@ class PiPEngine:
layer: PiPLayerConfig,
asset_path_map: dict[str, Path],
) -> Path | None:
"""验证图层素材是否可用,返回本地路径或None(降级跳过)."""
"""验证图层素材是否可用,返回本地路径或None(降级跳过).
安全:
- local_path 类型:必须在允许的目录内,防止路径遍历
- url 类型:必须通过 SSRF 安全校验
"""
from video_processing.path_security import is_in_allowed_dirs
from video_processing.url_security import UrlSecurityError, validate_url_safety
try:
if layer.source_type == "local_path":
path = Path(layer.source)
if path.exists():
return path
if not layer.source:
return None
# 路径安全校验:必须在允许目录内
src_path = Path(layer.source)
if not src_path.exists():
return None
if not is_in_allowed_dirs(src_path):
logger.warning(
"PiP local_path 不在允许目录内,拒绝: %s",
layer.source[:80],
)
return None
return src_path.resolve()
elif layer.source_type == "asset_id":
if layer.source in asset_path_map:
return asset_path_map[layer.source]
return None
elif layer.source_type == "url":
# URL类型由调用者负责下载,这里返回标记
return None # 暂时不支持直接URL
# URL类型:先做SSRF安全校验,由调用者负责实际下载
try:
validate_url_safety(layer.source, purpose="pip_source")
logger.info("PiP URL 安全校验通过: %s", layer.source[:80])
except UrlSecurityError as e:
logger.warning("PiP URL 安全校验失败: %s (error=%s)", layer.source[:80], e)
return None
# 暂时不支持直接URL下载,返回None表示降级跳过
return None
except Exception as e:
logger.warning("PiP素材验证失败: %s", e)
+17 -2
View File
@@ -77,6 +77,7 @@ def mix_audio(
*,
bgm_path: str | None = None,
bgm_config: dict | None = None,
audio_tracks_config: dict | None = None,
) -> Path | None:
"""音频后处理混音.
@@ -87,6 +88,8 @@ def mix_audio(
4. 输出时长截断到 video_duration
5. 无音频流的 clip 会被自动跳过,避免 FFmpeg 引用 [i:a] 失败
6. 如果提供了 bgm_path,则额外混入 BGM(支持淡入淡出、循环、人声闪避)
7. 如果配置了 audio_tracks,则混入多轨道音频(配音、音效等)
8. 如果配置了降噪,最后应用降噪
Args:
ctx: 渲染上下文
@@ -94,6 +97,7 @@ def mix_audio(
video_duration: 视频总时长(用于截断音频)
bgm_path: BGM 音频本地路径,为 None 时不混入 BGM
bgm_config: BGM 配置字典(volume/fade_in/fade_out/sidechain 等)
audio_tracks_config: 多轨道音频配置(tracks/master_volume 等)
Returns:
混音后的音频文件路径,无音频时返回 None
@@ -157,10 +161,21 @@ def mix_audio(
try:
# 这里 main_audio 就是 output_path,先有主音频再混 BGM
final_path = mix_bgm_with_main(ctx, output_path, bgm_cfg, video_duration)
return _apply_noise_reduction_if_needed(ctx, final_path)
output_path = final_path
except Exception:
logger.exception("[bgm] BGM 混音失败,回退到无 BGM 音频: plan_id=%s", ctx.plan_id)
return _apply_noise_reduction_if_needed(ctx, output_path)
# ── 多轨道混音(配音/音效等) ──
if audio_tracks_config and audio_tracks_config.get("enabled", False):
from video_processing.multi_track_mixer import mix_audio_tracks_from_config
try:
tracks_config = audio_tracks_config.get("tracks_config") or audio_tracks_config
multi_output = mix_audio_tracks_from_config(ctx, output_path, tracks_config, video_duration)
if multi_output and multi_output != output_path:
output_path = multi_output
except Exception:
logger.exception("[multi-track] 多轨道混音失败,回退: plan_id=%s", ctx.plan_id)
return _apply_noise_reduction_if_needed(ctx, output_path)
+44 -6
View File
@@ -392,10 +392,48 @@ class StickerEngine:
)
parsed_stickers.append((z, config))
else:
# 图片贴纸
image_path = s.get("image_path", "") or s.get("image_url", "")
if not image_path or not Path(image_path).exists():
logger.warning("贴纸素材不存在,跳过: %s", image_path)
# 图片贴纸 — 安全校验:区分本地路径和URL
image_path = s.get("image_path", "")
image_url = s.get("image_url", "")
safe_image_path: Path | None = None
if image_path:
# 本地路径:路径遍历防护
from video_processing.path_security import is_in_allowed_dirs
try:
p = Path(image_path)
if not p.exists():
logger.warning("贴纸素材不存在,跳过: %s", image_path[:80])
continue
if not is_in_allowed_dirs(p):
logger.warning("贴纸路径不在允许目录内,拒绝: %s", image_path[:80])
continue
safe_image_path = p.resolve()
except Exception as e:
logger.warning("贴纸路径校验失败,跳过: %s error=%s", image_path[:80], e)
continue
elif image_url:
# URL:SSRF 安全校验(暂不自动下载,仅校验安全性)
from video_processing.url_security import (
UrlSecurityError,
validate_url_safety,
)
try:
validate_url_safety(image_url, purpose="sticker_image")
except UrlSecurityError as e:
logger.warning("贴纸URL安全校验失败,跳过: %s error=%s", image_url[:80], e)
continue
# URL 类型暂不支持自动下载,跳过
logger.info("贴纸URL类型暂不支持自动下载,跳过: %s", image_url[:80])
continue
else:
logger.warning("贴纸缺少 image_path 和 image_url,跳过")
continue
if safe_image_path is None:
continue
config = ImageStickerConfig(
@@ -414,11 +452,11 @@ class StickerEngine:
fade_in=float(s.get("fade_in", 0)),
fade_out=float(s.get("fade_out", 0)),
z_index=z,
image_url=str(s.get("image_url", "")),
image_url=image_url,
)
parsed_stickers.append((z, config))
image_stickers.append(config)
image_paths.append(image_path)
image_paths.append(str(safe_image_path))
except Exception as e:
logger.warning("贴纸配置解析失败,跳过: %s", e)
+687
View File
@@ -0,0 +1,687 @@
"""字幕渲染引擎 — 统一管理字幕样式配置与视频烧录.
与现有模块的关系:
- render_subtitles.py:生成静态整段标题/字幕的 ASS 文件
- subtitle_generator.py:从 ASR 时间轴生成 ASS 文件
- 本模块:统一的字幕样式配置 + 烧录滤镜生成 + 多源字幕合并
支持的字幕来源:
1. 静态标题/字幕(title_config / subtitle_config
2. ASR 自动字幕(asr_subtitle_timeline
3. 手动字幕(manual_subtitles 时间轴)
支持的样式配置:
- 字体、字号、颜色
- 描边(颜色、宽度)
- 阴影(偏移、模糊、颜色)
- 背景框(颜色、透明度、圆角、边距)
- 位置(9宫格 + 自定义坐标)
- 对齐方式
- 动画(淡入淡出、滑入滑出、打字机)
- 多行/换行规则
"""
from __future__ import annotations
import logging
from dataclasses import dataclass, field
from pathlib import Path
from typing import Any
from video_processing.path_security import PathSecurityError, is_in_allowed_dirs, safe_resolve_path
from video_processing.render_subtitles import generate_ass_subtitles
from video_processing.subtitle_generator import generate_ass_from_timeline
logger = logging.getLogger(__name__)
# ── 常量 ──────────────────────────────────────────────────────────────────────
ALLOWED_SUBTITLE_EXTENSIONS = {".srt", ".ass", ".vtt", ".sub"}
# 9宫格位置映射(ASS alignment 编号)
POSITION_ALIGNMENT = {
"top_left": 7,
"top_center": 8,
"top_right": 9,
"middle_left": 4,
"center": 5,
"middle_right": 6,
"bottom_left": 1,
"bottom_center": 2,
"bottom_right": 3,
}
# 位置简称兼容
POSITION_ALIASES = {
"top": "top_center",
"bottom": "bottom_center",
"middle": "center",
"left": "middle_left",
"right": "middle_right",
}
DEFAULT_FONT = "思源黑体"
DEFAULT_FONT_SIZE = 24
DEFAULT_COLOR = "#FFFFFF"
DEFAULT_STROKE_COLOR = "#000000"
DEFAULT_STROKE_WIDTH = 1.5
DEFAULT_POSITION = "bottom_center"
DEFAULT_MAX_CHARS_PER_LINE = 20
# ── 字幕样式配置 ────────────────────────────────────────────────────────────
@dataclass
class SubtitleStyle:
"""字幕样式配置."""
font_name: str = DEFAULT_FONT
font_size: int = DEFAULT_FONT_SIZE
font_color: str = DEFAULT_COLOR
bold: bool = False
italic: bool = False
# 描边
stroke_enabled: bool = True
stroke_color: str = DEFAULT_STROKE_COLOR
stroke_width: float = DEFAULT_STROKE_WIDTH
# 阴影
shadow_enabled: bool = False
shadow_color: str = "#000000"
shadow_offset_x: int = 2
shadow_offset_y: int = 2
shadow_blur: float = 0.0
# 背景框
background_enabled: bool = False
background_color: str = "#000000"
background_opacity: float = 0.5 # 0.0 ~ 1.0
background_padding: int = 8
background_radius: int = 4
# 位置
position: str = DEFAULT_POSITION # 9宫格位置名
margin_v: int = 60 # 垂直边距
margin_l: int = 40 # 左边距
margin_r: int = 40 # 右边距
# 多行
max_chars_per_line: int = DEFAULT_MAX_CHARS_PER_LINE
line_spacing: int = 0 # 行间距
# 动画
fade_in: float = 0.0 # 淡入时长(秒)
fade_out: float = 0.0 # 淡出时长(秒)
animation_type: str = "none" # none/fade/slide/typewriter
@classmethod
def from_dict(cls, config: dict[str, Any] | None) -> "SubtitleStyle":
"""从字典创建样式配置,带安全类型转换."""
if not config or not isinstance(config, dict):
return cls()
def safe_str(key: str, default: str) -> str:
val = config.get(key, default)
return str(val) if val is not None else default
def safe_int(key: str, default: int) -> int:
try:
return int(config.get(key, default))
except (TypeError, ValueError):
return default
def safe_float(key: str, default: float) -> float:
try:
return float(config.get(key, default))
except (TypeError, ValueError):
return default
def safe_bool(key: str, default: bool) -> bool:
return bool(config.get(key, default))
position = safe_str("position", DEFAULT_POSITION)
position = POSITION_ALIASES.get(position, position)
if position not in POSITION_ALIGNMENT:
position = DEFAULT_POSITION
return cls(
font_name=safe_str("font", DEFAULT_FONT),
font_size=safe_int("size", DEFAULT_FONT_SIZE),
font_color=safe_str("color", DEFAULT_COLOR),
bold=safe_bool("bold", False),
italic=safe_bool("italic", False),
stroke_enabled=safe_bool("stroke_enabled", True),
stroke_color=safe_str("stroke_color", DEFAULT_STROKE_COLOR),
stroke_width=safe_float("stroke_width", DEFAULT_STROKE_WIDTH),
shadow_enabled=safe_bool("shadow_enabled", False),
shadow_color=safe_str("shadow_color", "#000000"),
shadow_offset_x=safe_int("shadow_offset_x", 2),
shadow_offset_y=safe_int("shadow_offset_y", 2),
shadow_blur=safe_float("shadow_blur", 0.0),
background_enabled=safe_bool("background_enabled", False),
background_color=safe_str("background_color", "#000000"),
background_opacity=max(0.0, min(1.0, safe_float("background_opacity", 0.5))),
background_padding=safe_int("background_padding", 8),
background_radius=safe_int("background_radius", 4),
position=position,
margin_v=safe_int("margin_v", 60),
margin_l=safe_int("margin_l", 40),
margin_r=safe_int("margin_r", 40),
max_chars_per_line=safe_int("max_chars_per_line", DEFAULT_MAX_CHARS_PER_LINE),
line_spacing=safe_int("line_spacing", 0),
fade_in=max(0.0, safe_float("fade_in", 0.0)),
fade_out=max(0.0, safe_float("fade_out", 0.0)),
animation_type=safe_str("animation_type", "none"),
)
@property
def alignment(self) -> int:
"""获取 ASS alignment 编号."""
return POSITION_ALIGNMENT.get(self.position, 2)
@property
def ass_font_color(self) -> str:
"""ASS 格式颜色 &HAABBGGRR."""
return _hex_to_ass_color(self.font_color)
@property
def ass_stroke_color(self) -> str:
return _hex_to_ass_color(self.stroke_color)
@property
def ass_shadow_color(self) -> str:
return _hex_to_ass_color(self.shadow_color)
@property
def ass_background_color(self) -> str:
"""背景框颜色(ASS BackColour),带透明度."""
alpha_hex = _opacity_to_ass_alpha(self.background_opacity)
color_bgr = _hex_to_ass_bgr(self.background_color)
return f"&H{alpha_hex}{color_bgr}"
# ── 工具函数 ──────────────────────────────────────────────────────────────────
def _hex_to_ass_color(hex_color: str) -> str:
"""HEX → ASS 颜色 &HAABBGGRR(默认不透明)."""
hex_color = hex_color.lstrip("#")
if len(hex_color) != 6:
return "&H00FFFFFF"
r, g, b = hex_color[0:2], hex_color[2:4], hex_color[4:6]
return f"&H00{b.upper()}{g.upper()}{r.upper()}"
def _hex_to_ass_bgr(hex_color: str) -> str:
"""HEX → ASS BGR 部分(不含 alpha."""
hex_color = hex_color.lstrip("#")
if len(hex_color) != 6:
return "FFFFFF"
r, g, b = hex_color[0:2], hex_color[2:4], hex_color[4:6]
return f"{b.upper()}{g.upper()}{r.upper()}"
def _opacity_to_ass_alpha(opacity: float) -> str:
"""不透明度 → ASS alpha(00=不透明,FF=完全透明)."""
alpha = 255 - int(opacity * 255)
return f"{alpha:02X}"
def _escape_ass_text(text: str) -> str:
"""转义 ASS 文本特殊字符."""
text = text.replace("\r\n", "\\N").replace("\n", "\\N").replace("\r", "\\N")
text = text.replace("{", "(").replace("}", ")")
return text
def _format_ass_time(seconds: float) -> str:
"""秒 → ASS 时间格式 H:MM:SS.cc."""
hours = int(seconds // 3600)
minutes = int((seconds % 3600) // 60)
secs = seconds % 60
return f"{hours}:{minutes:02d}:{secs:05.2f}"
def _wrap_text(text: str, max_chars: int) -> list[str]:
"""按字数换行,优先标点断开."""
if len(text) <= max_chars:
return [text]
lines: list[str] = []
remaining = text
while len(remaining) > max_chars:
break_point = max_chars
punctuations = ",。!?、;:,.;:!?"
for i in range(max_chars, max_chars // 2, -1):
if i < len(remaining) and remaining[i] in punctuations:
break_point = i + 1
break
lines.append(remaining[:break_point])
remaining = remaining[break_point:]
if remaining:
lines.append(remaining)
return lines
# ── 字幕片段 ──────────────────────────────────────────────────────────────────
@dataclass
class SubtitleSegment:
"""单个字幕片段."""
start: float # 开始时间(秒)
end: float # 结束时间(秒)
text: str # 字幕文本
style_name: str = "Default" # 使用的样式名
# ── 字幕渲染引擎 ──────────────────────────────────────────────────────────────
class SubtitleRenderEngine:
"""字幕渲染引擎 — 统一管理多源字幕的 ASS 文件生成.
支持合并多个字幕来源到同一个 ASS 文件:
- 标题(顶部,单独样式)
- 字幕(底部,单独样式)
- ASR 时间轴字幕
- 手动字幕
输出一个统一的 ASS 文件,供 FFmpeg subtitles filter 烧录。
"""
def __init__(
self,
video_width: int = 1080,
video_height: int = 1920,
video_duration: float = 0.0,
):
self.video_width = video_width
self.video_height = video_height
self.video_duration = video_duration
self._styles: dict[str, SubtitleStyle] = {}
self._segments: list[SubtitleSegment] = []
self._style_counter = 0
# ── 样式管理 ──────────────────────────────────────────────────────
def add_style(self, name: str, style: SubtitleStyle) -> str:
"""注册一个样式,返回样式名."""
self._styles[name] = style
return name
def get_or_create_style(self, base_name: str, style: SubtitleStyle) -> str:
"""获取或创建样式(避免重复)."""
if base_name in self._styles:
return base_name
self._styles[base_name] = style
return base_name
# ── 字幕源添加 ────────────────────────────────────────────────────
def add_title(self, text: str, style: SubtitleStyle | None = None) -> None:
"""添加整段标题(显示整个视频时长)."""
if not text or not text.strip():
return
style = style or SubtitleStyle(
position="top_center",
font_size=48,
bold=True,
stroke_enabled=True,
stroke_width=2.0,
)
style_name = self.get_or_create_style("TitleStyle", style)
self._segments.append(
SubtitleSegment(
start=0.0,
end=self.video_duration if self.video_duration > 0 else 9999.0,
text=text.strip(),
style_name=style_name,
)
)
def add_subtitle_text(self, text: str, style: SubtitleStyle | None = None) -> None:
"""添加整段字幕(显示整个视频时长)."""
if not text or not text.strip():
return
style = style or SubtitleStyle()
style_name = self.get_or_create_style("SubtitleStyle", style)
self._segments.append(
SubtitleSegment(
start=0.0,
end=self.video_duration if self.video_duration > 0 else 9999.0,
text=text.strip(),
style_name=style_name,
)
)
def add_timeline_segments(
self,
segments: list[dict] | list[SubtitleSegment],
style: SubtitleStyle | None = None,
) -> None:
"""添加时间轴字幕片段(ASR 或手动字幕).
segments 可以是:
- SubtitleSegment 列表
- dict 列表,每个 dict 含 start/end/text 字段
"""
if not segments:
return
style = style or SubtitleStyle()
style_name = self.get_or_create_style("Default", style)
for seg in segments:
if isinstance(seg, SubtitleSegment):
seg.style_name = style_name
self._segments.append(seg)
elif isinstance(seg, dict):
try:
start = float(seg.get("start", 0))
end = float(seg.get("end", 0))
text = str(seg.get("text", ""))
if end > start and text.strip():
self._segments.append(
SubtitleSegment(
start=start,
end=end,
text=text.strip(),
style_name=style_name,
)
)
except (TypeError, ValueError):
continue
def add_asr_timeline(self, timeline: Any, style: SubtitleStyle | None = None) -> None:
"""从 SubtitleTimeline 对象添加 ASR 字幕."""
if not timeline or not hasattr(timeline, "segments") or not timeline.segments:
return
style = style or SubtitleStyle()
style_name = self.get_or_create_style("ASRStyle", style)
for seg in timeline.segments:
if hasattr(seg, "start") and hasattr(seg, "end") and hasattr(seg, "text"):
if seg.end > seg.start and seg.text.strip():
self._segments.append(
SubtitleSegment(
start=seg.start,
end=seg.end,
text=seg.text.strip(),
style_name=style_name,
)
)
# ── ASS 文件生成 ──────────────────────────────────────────────────
def generate_ass(self, output_path: Path) -> Path:
"""生成 ASS 字幕文件.
Returns:
生成的文件路径;如果没有字幕内容,返回空文件。
"""
if not self._segments:
output_path.parent.mkdir(parents=True, exist_ok=True)
output_path.write_text("", encoding="utf-8")
return output_path
# 确保至少有 Default 样式
if "Default" not in self._styles:
self._styles["Default"] = SubtitleStyle()
# 生成样式行
style_lines = []
for name, style in self._styles.items():
style_lines.append(self._build_ass_style_line(name, style))
# 生成事件行(按时间排序)
self._segments.sort(key=lambda s: s.start)
event_lines = []
for seg in self._segments:
event_lines.append(self._build_ass_event_line(seg))
# 组装文件
ass_content = f"""[Script Info]
ScriptType: v4.00+
PlayResX: {self.video_width}
PlayResY: {self.video_height}
ScaledBorderAndShadow: yes
WrapStyle: 2
Encoding: UTF-8
[V4+ Styles]
Format: Name, Fontname, Fontsize, PrimaryColour, SecondaryColour, OutlineColour, BackColour, Bold, Italic, Underline, StrikeOut, ScaleX, ScaleY, Spacing, Angle, BorderStyle, Outline, Shadow, Alignment, MarginL, MarginR, MarginV, Encoding
{chr(10).join(style_lines)}
[Events]
Format: Layer, Start, End, Style, Name, MarginL, MarginR, MarginV, Effect, Text
{chr(10).join(event_lines)}
"""
output_path.parent.mkdir(parents=True, exist_ok=True)
output_path.write_text(ass_content, encoding="utf-8")
return output_path
def _build_ass_style_line(self, name: str, style: SubtitleStyle) -> str:
"""构建一条 ASS Style 行."""
bold_val = -1 if style.bold else 0
italic_val = -1 if style.italic else 0
# BorderStyle: 1=outline+shadow, 3=opaque box(背景框)
if style.background_enabled:
border_style = 3
back_color = style.ass_background_color
else:
border_style = 1
back_color = style.ass_shadow_color if style.shadow_enabled else style.ass_font_color
outline_val = style.stroke_width if style.stroke_enabled else 0.0
shadow_val = style.shadow_offset_y if style.shadow_enabled else 0
return (
f"Style: {name},{style.font_name},{style.font_size},{style.ass_font_color},"
f"&H000000FF,{style.ass_stroke_color},{back_color},"
f"{bold_val},{italic_val},0,0,100,100,0,0,"
f"{border_style},{outline_val},{shadow_val},{style.alignment},"
f"{style.margin_l},{style.margin_r},{style.margin_v},1"
)
def _build_ass_event_line(self, seg: SubtitleSegment) -> str:
"""构建一条 ASS Dialogue 事件行."""
style = self._styles.get(seg.style_name, SubtitleStyle())
max_chars = style.max_chars_per_line
# 自动换行
lines = _wrap_text(seg.text, max_chars)
display_text = "\\N".join(lines)
# 动画效果(淡入淡出)
effect_tags = ""
if style.fade_in > 0 or style.fade_out > 0:
fade_in_ms = int(style.fade_in * 1000)
fade_out_ms = int(style.fade_out * 1000)
effect_tags = f"{{\\fad({fade_in_ms},{fade_out_ms})}}"
safe_text = _escape_ass_text(display_text)
start_time = _format_ass_time(max(0, seg.start))
end_time = _format_ass_time(max(seg.start + 0.1, seg.end))
return f"Dialogue: 0,{start_time},{end_time},{seg.style_name},,0,0,0,," f"{effect_tags}{safe_text}"
@property
def has_subtitles(self) -> bool:
"""是否有字幕内容."""
return len(self._segments) > 0
# ── 便捷函数:从 plan.config 快速生成 ASS ────────────────────────────────────
def build_subtitles_from_plan(
output_path: Path,
plan_config: dict,
*,
video_width: int,
video_height: int,
video_duration: float,
asr_timeline: Any = None,
) -> Path | None:
"""从 plan.config 构建字幕 ASS 文件.
支持的配置项:
- title_config: 标题配置(含 text/style
- subtitle_config: 字幕配置(含 text/style
- asr_subtitles: ASR 字幕开关 + 样式
- manual_subtitles: 手动字幕片段列表
Returns:
生成的 ASS 文件路径;如果没有任何字幕,返回 None
"""
engine = SubtitleRenderEngine(
video_width=video_width,
video_height=video_height,
video_duration=video_duration,
)
has_any = False
# 1. 标题
title_cfg = plan_config.get("title_config") or {}
if isinstance(title_cfg, dict):
title_text = str(title_cfg.get("text", ""))
title_enabled = title_cfg.get("enabled", True)
if title_enabled and title_text.strip():
style_dict = title_cfg.get("style") or {}
style = SubtitleStyle.from_dict(style_dict)
# 标题默认样式:顶部、大字号、粗体
if style.position == DEFAULT_POSITION and style.font_size == DEFAULT_FONT_SIZE:
style.position = "top_center"
style.font_size = 48
style.bold = True
engine.add_title(title_text, style)
has_any = True
# 2. 静态字幕
sub_cfg = plan_config.get("subtitle_config") or {}
if isinstance(sub_cfg, dict):
sub_text = str(sub_cfg.get("text", ""))
sub_enabled = sub_cfg.get("enabled", True)
if sub_enabled and sub_text.strip():
style_dict = sub_cfg.get("style") or {}
style = SubtitleStyle.from_dict(style_dict)
engine.add_subtitle_text(sub_text, style)
has_any = True
# 3. ASR 自动字幕
asr_cfg = plan_config.get("asr_subtitles") or {}
if isinstance(asr_cfg, dict) and asr_cfg.get("enabled", False):
if asr_timeline is not None:
style_dict = asr_cfg.get("style") or {}
style = SubtitleStyle.from_dict(style_dict)
engine.add_asr_timeline(asr_timeline, style)
has_any = has_any or engine.has_subtitles
# 4. 手动字幕
manual_segs = plan_config.get("manual_subtitles") or []
if isinstance(manual_segs, list) and manual_segs:
style_dict = (plan_config.get("manual_subtitle_style") or {}) or {}
style = SubtitleStyle.from_dict(style_dict)
engine.add_timeline_segments(manual_segs, style)
has_any = has_any or engine.has_subtitles
if not has_any:
return None
return engine.generate_ass(output_path)
# ── FFmpeg 烧录滤镜生成 ───────────────────────────────────────────────────────
def build_subtitle_filter(
ass_path: Path | str,
*,
video_input_label: str = "0:v",
output_label: str = "subtitled",
work_dir: Path | str | None = None,
) -> str:
"""生成 FFmpeg subtitles 滤镜字符串.
Args:
ass_path: ASS 字幕文件路径
video_input_label: 视频输入标签(如 "0:v""[v_out]"
output_label: 输出标签
work_dir: 工作目录(必填,用于路径安全校验,防止路径遍历绕过)
Returns:
filter_complex 片段,如 "[0:v]subtitles=xxx.ass[subtitled]"
Raises:
PathSecurityError: 字幕路径不安全或 work_dir 未提供
"""
# ── 安全校验:字幕文件路径白名单 ──
ass_path_str = str(ass_path)
if work_dir is None or not str(work_dir).strip():
raise PathSecurityError("work_dir 必须提供,不能为 None 或空")
_validate_subtitle_path(ass_path_str, Path(work_dir))
# FFmpeg subtitles filter 的路径需要转义:
# - Windows 路径的 \ → /
# - 冒号 : → \:
# - 单引号 ' → '\''
safe_path = ass_path_str.replace("\\", "/").replace(":", "\\:").replace("'", "'\\''")
return f"{video_input_label}subtitles='{safe_path}'[{output_label}]"
def _validate_subtitle_path(subtitle_path: str, work_dir: Path) -> None:
"""校验字幕文件路径安全性.
规则:
- 必须是本地路径(不支持远程URL字幕)
- local:// schema → 必须在 work_dir 内
- 相对路径 → 必须在 work_dir 内
- 绝对路径 → 必须在允许目录白名单内
- 扩展名必须是字幕格式
Raises:
PathSecurityError: 路径不安全
"""
if not subtitle_path or not isinstance(subtitle_path, str):
raise PathSecurityError("字幕路径不能为空")
# 不允许远程URL字幕(subtitles滤镜不支持远程加载,且有SSRF风险)
if subtitle_path.startswith(("http://", "https://", "oss://")):
raise PathSecurityError("不允许使用远程URL字幕文件")
is_abs = subtitle_path.startswith("/") and not subtitle_path.startswith("local://")
resolved_path = safe_resolve_path(
subtitle_path,
work_dir,
allow_outside=is_abs,
allowed_extensions=ALLOWED_SUBTITLE_EXTENSIONS,
)
# 绝对路径额外检查白名单目录(用realpath规范化后的真实路径比较,防止 ../ 遍历绕过)
if is_abs:
resolved_work_dir = work_dir.resolve()
try:
resolved_path.relative_to(resolved_work_dir)
except ValueError:
if not is_in_allowed_dirs(resolved_path):
raise PathSecurityError(f"字幕路径不在允许目录内: {subtitle_path[:80]}")
+9 -13
View File
@@ -28,7 +28,6 @@ from dataclasses import dataclass, field
from pathlib import Path
from typing import Any
from video_processing.chroma_key_engine import apply_chroma_key_if_needed
from video_processing.color_grade_engine import ColorGradeConfig, ColorGradeEngine
from video_processing.ffmpeg_utils import (
DEFAULT_FPS,
@@ -46,7 +45,7 @@ from video_processing.render_audio import RenderContext, merge_audio_video, mix_
from video_processing.render_subtitles import generate_ass_subtitles
from video_processing.reverse_engine import ReverseConfig, ReverseEngine
from video_processing.speed_engine import SpeedConfig, SpeedEngine
from video_processing.sticker_engine import StickerEngine, parse_stickers_from_config
from video_processing.sticker_engine import StickerEngine
from video_processing.subtitle_generator import generate_ass_from_timeline
from video_processing.transition_engine import TransitionEngine
from video_processing.trim_engine import TrimConfig, TrimEngine, extract_trim_from_clip_config
@@ -343,6 +342,7 @@ class UnifiedRenderService:
else:
config = self.plan.config or {}
bgm_config = config.get("bgm", {}) or {}
audio_tracks_config = config.get("audio_tracks") or {}
noise_reduction_config = config.get("audio_noise_reduction")
ctx = RenderContext(
work_dir=self.work_dir,
@@ -355,6 +355,7 @@ class UnifiedRenderService:
video_duration,
bgm_path=self.bgm_path,
bgm_config=bgm_config,
audio_tracks_config=audio_tracks_config,
)
t_audio_end = time.time()
audio_mix_ms = int((t_audio_end - t_audio_start) * 1000)
@@ -639,8 +640,8 @@ class UnifiedRenderService:
return timeline
def _extract_audio(self, video_path: Path, output_path: Path) -> None:
"""从视频中提取音频为16kHz单声道wav(ASR友好格式)"""
import subprocess
"""从视频中提取音频为16kHz单声道wav(ASR友好格式)."""
from video_processing.ffmpeg_utils import run_ffmpeg
cmd = [
"ffmpeg",
@@ -657,15 +658,10 @@ class UnifiedRenderService:
str(output_path),
]
result = subprocess.run(
cmd,
capture_output=True,
text=True,
timeout=120,
)
if result.returncode != 0:
raise RuntimeError(f"音频提取失败: {result.stderr[:200]}")
try:
run_ffmpeg(cmd, timeout=120)
except Exception as e:
raise RuntimeError(f"音频提取失败: {str(e)[:200]}") from e
def _maybe_add_voiceover_layer(
self,
@@ -0,0 +1,21 @@
"""URL 安全校验工具 — SSRF 防护(向后兼容层).
本模块为向后兼容而保留,实际实现已迁移至 packages.shared.url_security。
所有符号均从该模块重新导出,请新代码直接 import packages.shared.url_security。
"""
from packages.shared.url_security import ( # noqa: F401
ALLOWED_AUDIO_MIME_TYPES,
ALLOWED_IMAGE_MIME_TYPES,
ALLOWED_PORTS,
ALLOWED_SCHEMES,
ALLOWED_VIDEO_MIME_TYPES,
DEFAULT_MAX_DOWNLOAD_SIZE,
MAX_URL_LENGTH,
TRUSTED_DOMAINS,
UrlSecurityError,
is_url_safe,
safe_download_bytes,
safe_download_file,
validate_url_safety,
)
+43 -43
View File
@@ -130,6 +130,8 @@ class AssetAnalyzer:
info = VideoInfo()
try:
from video_processing.ffmpeg_utils import run_ffprobe
cmd = [
"ffprobe",
"-v",
@@ -140,38 +142,31 @@ class AssetAnalyzer:
"-show_streams",
self.video_path,
]
result = subprocess.run(
cmd,
capture_output=True,
text=True,
timeout=30,
)
stdout, _ = run_ffprobe(cmd, timeout=30)
data = json.loads(stdout)
streams = data.get("streams", [])
format_info = data.get("format", {})
if result.returncode == 0:
data = json.loads(result.stdout)
streams = data.get("streams", [])
format_info = data.get("format", {})
for stream in streams:
if stream.get("codec_type") == "video":
info.width = int(stream.get("width", 0))
info.height = int(stream.get("height", 0))
info.codec = stream.get("codec_name", "")
for stream in streams:
if stream.get("codec_type") == "video":
info.width = int(stream.get("width", 0))
info.height = int(stream.get("height", 0))
info.codec = stream.get("codec_name", "")
# 解析帧率
fps_str = stream.get("r_frame_rate", "0/1")
if "/" in fps_str:
num, denom = fps_str.split("/")
info.fps = float(num) / float(denom) if float(denom) != 0 else 0.0
else:
info.fps = float(fps_str)
# 解析帧率
fps_str = stream.get("r_frame_rate", "0/1")
if "/" in fps_str:
num, denom = fps_str.split("/")
info.fps = float(num) / float(denom) if float(denom) != 0 else 0.0
else:
info.fps = float(fps_str)
elif stream.get("codec_type") == "audio":
info.has_audio = True
elif stream.get("codec_type") == "audio":
info.has_audio = True
info.duration = float(format_info.get("duration", 0))
info.bitrate = int(format_info.get("bit_rate", 0))
info.file_size = int(format_info.get("size", 0))
info.duration = float(format_info.get("duration", 0))
info.bitrate = int(format_info.get("bit_rate", 0))
info.file_size = int(format_info.get("size", 0))
except Exception as e:
logger.warning(f"Failed to get video info: {e}")
@@ -179,7 +174,7 @@ class AssetAnalyzer:
self._video_info = info
return info
def extract_frames(self, count: int = 10, max_frames: int = 30) -> list[np.ndarray]:
def extract_frames(self, count: int = 10) -> list[np.ndarray]:
"""
从视频中均匀抽取帧
@@ -224,14 +219,14 @@ class AssetAnalyzer:
output_path,
]
result = subprocess.run(
cmd,
capture_output=True,
text=True,
timeout=10,
)
from video_processing.ffmpeg_utils import run_ffmpeg
if result.returncode == 0 and os.path.exists(output_path):
try:
run_ffmpeg(cmd, timeout=10)
except Exception:
continue
if os.path.exists(output_path):
# 读取帧并转换为 numpy 数组
img = self._load_image_as_array(output_path)
if img is not None:
@@ -397,14 +392,19 @@ class AssetAnalyzer:
audio_path,
]
result_audio = subprocess.run(
cmd,
capture_output=True,
text=True,
timeout=30,
)
from video_processing.ffmpeg_utils import run_ffmpeg
if result_audio.returncode == 0 and os.path.exists(audio_path):
try:
run_ffmpeg(cmd, timeout=30)
except Exception:
# 音频提取失败,返回默认分析结果
return AudioAnalysis(
has_speech=False,
speech_ratio=0.0,
avg_volume=0.0,
)
if os.path.exists(audio_path):
# 读取音频数据
import struct
+12 -3
View File
@@ -106,7 +106,16 @@ def _download_video_to_file(url: str, dest_path: str) -> None:
except Exception:
pass
# 回退到 HTTP 下载
import urllib.request
# 回退到 HTTP 下载(含 SSRF 防护 + 大小限制 + 类型校验)
from video_processing.url_security import (
ALLOWED_VIDEO_MIME_TYPES,
safe_download_file,
)
urllib.request.urlretrieve(url, dest_path) # nosec B310
safe_download_file(
url,
dest_path,
purpose="batch_video_download",
allowed_mime_types=ALLOWED_VIDEO_MIME_TYPES | {"application/octet-stream"},
timeout=300.0,
)
+6 -11
View File
@@ -6,7 +6,6 @@
from __future__ import annotations
import os
import subprocess
import tempfile
from pathlib import Path
@@ -115,16 +114,12 @@ def _compose_with_legacy_engine(task, job_service, job, plan_id: str, db) -> dic
logger.info("Executing FFmpeg for job %s, plan %s", job_id, plan_id)
try:
subprocess.run(
compose_cmd.command,
check=True,
stdout=subprocess.PIPE,
stderr=subprocess.PIPE,
text=True,
timeout=3600,
)
except subprocess.CalledProcessError as e:
job_service.fail_job(job_id, f"FFmpeg 执行失败: {e.stderr[:500]}")
from video_processing.ffmpeg_utils import run_ffmpeg
run_ffmpeg(compose_cmd.command, timeout=3600)
except Exception as e:
error_msg = f"FFmpeg 执行失败: {str(e)[:500]}"
job_service.fail_job(job_id, error_msg)
raise
# 上传结果
@@ -257,7 +257,6 @@ def _render_with_legacy(
) -> dict:
"""旧引擎路径(VideoComposeService + FFmpeg filter_complex)。"""
import os
import subprocess
from apps.api.app.services.video_compose_service import VideoComposeService
@@ -278,16 +277,11 @@ def _render_with_legacy(
logger.info("执行 FFmpeg (legacy): plan_id=%s", plan_id)
try:
subprocess.run(
compose_cmd.command,
check=True,
stdout=subprocess.PIPE,
stderr=subprocess.PIPE,
text=True,
timeout=3600,
)
except subprocess.CalledProcessError as e:
error_msg = f"FFmpeg 执行失败: {e.stderr[:500]}"
from video_processing.ffmpeg_utils import run_ffmpeg
run_ffmpeg(compose_cmd.command, timeout=3600)
except Exception as e:
error_msg = f"FFmpeg 执行失败: {str(e)[:500]}"
logger.error("FFmpeg 执行失败(legacy): %s%s", plan_id, error_msg)
_mark_plan_failed(plan_repo, plan_id, gen_task_repo, generation_task_id, error_msg)
return {"status": "error", "message": error_msg}
+37 -7
View File
@@ -364,10 +364,19 @@ def _prepare_bgm_track(
try:
parsed = urlparse(audio_url)
if parsed.scheme in ("http", "https"):
import urllib.request
from video_processing.url_security import (
ALLOWED_AUDIO_MIME_TYPES,
safe_download_file,
)
logger.info("[task_id=%s] [BGM] 从URL下载: %s", task_id, audio_url[:80])
urllib.request.urlretrieve(audio_url, bgm_file) # nosec B310
safe_download_file(
audio_url,
str(bgm_file),
purpose="bgm_download",
allowed_mime_types=ALLOWED_AUDIO_MIME_TYPES,
timeout=60.0,
)
if bgm_file.exists() and bgm_file.stat().st_size > 0:
return str(bgm_file)
except Exception as e:
@@ -401,10 +410,19 @@ def _prepare_bgm_track(
preset = get_preset_bgm(preset_id)
if preset and preset.audio_url:
import urllib.request
from video_processing.url_security import (
ALLOWED_AUDIO_MIME_TYPES,
safe_download_file,
)
logger.info("[task_id=%s] [BGM] 从预设库下载: preset_id=%s", task_id, preset_id)
urllib.request.urlretrieve(preset.audio_url, bgm_file) # nosec B310
safe_download_file(
preset.audio_url,
str(bgm_file),
purpose="bgm_preset_download",
allowed_mime_types=ALLOWED_AUDIO_MIME_TYPES,
timeout=60.0,
)
if bgm_file.exists() and bgm_file.stat().st_size > 0:
return str(bgm_file)
except Exception as e:
@@ -418,17 +436,31 @@ def _prepare_bgm_track(
def _verify_url_accessible(url: str, timeout: float = 10.0, retries: int = 2) -> bool:
"""HEAD 请求校验 URL 可访问(含重试,防止 OSS 抖动误报)。
安全:
- 请求前先做 SSRF 安全校验(内网IP/回环地址/链路本地地址等)
- scheme 仅允许 http/https
- 端口仅允许 80/443
Args:
url: 待校验的 URL
timeout: 单次请求超时时间(秒)
retries: 最大重试次数(默认 2 次,首次失败后间隔 1s 重试)
Returns:
True 表示 URL 可访问(HTTP 2xx/3xx),False 表示所有尝试均失败。
True 表示 URL 可访问(HTTP 2xx/3xx),False 表示所有尝试均失败或安全校验不通过
"""
import time
import urllib.request
from video_processing.url_security import UrlSecurityError, validate_url_safety
# P0-1 SSRF 防护:请求前先校验 URL 安全性
try:
validate_url_safety(url, purpose="url_verify")
except UrlSecurityError as e:
logger.warning("URL 安全校验失败,拒绝访问: url=%s error=%s", url[:80], e)
return False
last_error: Exception | None = None
for attempt in range(1 + retries):
try:
@@ -461,7 +493,6 @@ def _download_library_assets(
asset_library_id: str = "",
project_id: str = "",
asset_ids: list[str] | None = None,
video_extensions: tuple = (".mp4", ".mov", ".avi", ".mkv", ".webm"),
strict: bool = True,
task_id: str = "",
gen_task=None,
@@ -480,7 +511,6 @@ def _download_library_assets(
asset_library_id: 素材库 ID(可选,与 project_id 二选一)
project_id: 项目 ID(可选,与 asset_library_id 二选一)
asset_ids: 指定素材 ID 列表,为空则下载全部 ready 视频素材
video_extensions: 支持的视频扩展名(保留兼容,当前按 file_type 过滤)
strict: 严格模式(默认 True)。
True — 任何素材下载失败立即抛 RuntimeError
False — 跳过失败素材,返回成功列表(调用方可通过日志感知失败)。
+8 -8
View File
@@ -45,6 +45,8 @@ def extract_media_metadata(file_url: str, media_type: str) -> dict:
try:
if media_type == "video":
# 使用 ffprobe 提取视频元数据
from video_processing.ffmpeg_utils import run_ffprobe
cmd = [
"ffprobe",
"-v",
@@ -55,16 +57,11 @@ def extract_media_metadata(file_url: str, media_type: str) -> dict:
"-show_streams",
file_url,
]
result = subprocess.run(
cmd,
capture_output=True,
text=True,
timeout=30,
)
if result.returncode == 0:
try:
stdout, _ = run_ffprobe(cmd, timeout=30)
import json as json_lib
probe_data = json_lib.loads(result.stdout)
probe_data = json_lib.loads(stdout)
# 提取视频流信息
for stream in probe_data.get("streams", []):
@@ -83,6 +80,9 @@ def extract_media_metadata(file_url: str, media_type: str) -> dict:
metadata["size_bytes"] = int(format_info.get("size", 0))
metadata["bitrate"] = int(format_info.get("bit_rate", 0))
except Exception as e:
logger.warning("视频元数据提取失败: %s", e)
elif media_type == "image":
# 使用 Pillow 提取图片元数据
try:
@@ -19,14 +19,12 @@ class VoiceExtractor:
"""Extract voice tracks and background music from videos using FFmpeg."""
@staticmethod
def _run_ffmpeg(cmd: list[str]) -> subprocess.CompletedProcess:
"""Run FFmpeg command and return result."""
logger.info(f"Running FFmpeg: {chr(39).join(cmd)}")
result = subprocess.run(cmd, capture_output=True, text=True)
if result.returncode != 0:
logger.error(f"FFmpeg error: {result.stderr}")
raise RuntimeError(f"FFmpeg failed: {result.stderr}")
return result
def _run_ffmpeg(cmd: list[str]) -> None:
"""Run FFmpeg command using 统一 run_ffmpeg 工具."""
from video_processing.ffmpeg_utils import run_ffmpeg
logger.info("Running FFmpeg: %s", " ".join(cmd[:10]))
run_ffmpeg(cmd)
def extract_voice(
self,
@@ -15,6 +15,7 @@ import httpx
from packages.application.cosyvoice_service import CosyVoiceError, CosyVoiceService
from packages.application.tts_job.text_splitter import split_text
from packages.shared.url_security import ALLOWED_AUDIO_MIME_TYPES, safe_download_bytes
logger = logging.getLogger(__name__)
@@ -224,10 +225,13 @@ class TTSStreamingService:
# ── 工具方法 ────────────────────────────────────────────
def _download_audio(self, url: str) -> bytes:
"""下载音频数据。"""
resp = httpx.get(url, timeout=60.0, follow_redirects=True)
resp.raise_for_status()
return resp.content
"""下载音频数据(含 SSRF 防护 + 大小限制 + 重定向校验)"""
return safe_download_bytes(
url,
purpose="tts_streaming_download",
allowed_mime_types=ALLOWED_AUDIO_MIME_TYPES,
timeout=60.0,
)
async def _stream_audio_chunks(self, websocket: Any, audio_data: bytes) -> int:
"""将音频数据分块通过 WebSocket 推送。
+23 -15
View File
@@ -19,16 +19,19 @@ from typing import Optional
import httpx
from packages.application.cosyvoice_service import (
CosyVoiceAuthError,
CosyVoiceError,
CosyVoiceService,
)
from packages.application.cosyvoice_service import CosyVoiceAuthError, CosyVoiceError, CosyVoiceService
from packages.application.tts_job.audio_merger import AudioMerger
from packages.application.tts_job.text_splitter import split_text
from packages.domain.tts_job import TTSJob, TTSJobStatus
from packages.ports.tts_job_repository import TTSJobRepository
from packages.shared.storage import SharedStorageService, get_shared_storage_service
from packages.shared.url_security import (
ALLOWED_AUDIO_MIME_TYPES,
UrlSecurityError,
safe_download_bytes,
safe_download_file,
validate_url_safety,
)
logger = logging.getLogger(__name__)
@@ -96,10 +99,13 @@ class TTSWorkflowService:
content_type = content_type_map.get(audio_format, "application/octet-stream")
try:
# 下载临时音频
resp = httpx.get(temp_url, timeout=60.0, follow_redirects=True)
resp.raise_for_status()
audio_data = resp.content
# 安全下载临时音频(SSRF 防护 + 大小限制 + 重定向校验)
audio_data = safe_download_bytes(
temp_url,
purpose="tts_audio_download",
allowed_mime_types=ALLOWED_AUDIO_MIME_TYPES,
timeout=60.0,
)
# 上传到 OSS
file_obj = io.BytesIO(audio_data)
@@ -463,13 +469,15 @@ class TTSWorkflowService:
total_duration += result.get("duration", 0.0)
# 下载分段音频到临时文件
resp = httpx.get(audio_url, timeout=60.0, follow_redirects=True)
resp.raise_for_status()
# 安全下载分段音频到临时文件SSRF 防护 + 大小限制)
seg_path = os.path.join(temp_dir, f"seg_{idx:03d}.{job.format}")
with open(seg_path, "wb") as f:
f.write(resp.content)
safe_download_file(
audio_url,
seg_path,
purpose="tts_segment_download",
allowed_mime_types=ALLOWED_AUDIO_MIME_TYPES,
timeout=60.0,
)
audio_paths.append(seg_path)
# 合并
+402
View File
@@ -0,0 +1,402 @@
"""URL 安全校验工具 — SSRF 防护.
统一的外部 URL 安全校验方案,覆盖所有渲染管线和 TTS 中的外部下载场景。
放在 packages/shared/ 作为单一来源,worker 和 application 层都可引用。
防护要点:
1. Scheme 白名单:仅允许 http/https
2. 主机 SSRF 防护:禁止内网 IP、回环地址、链路本地地址、元数据服务
3. 端口白名单:仅允许 80/443(标准 HTTP/HTTPS
4. 域名校验:禁止 IP 直接访问(除非在白名单中)
5. 重定向防护:手动跟随重定向,每次跳转前重新校验目标 URL
6. 文件大小限制:流式下载,超过上限立即中断
7. MIME 类型白名单:可选的内容类型校验
"""
from __future__ import annotations
import ipaddress
import logging
import os
import socket
import urllib.error
import urllib.request
from urllib.parse import urljoin, urlparse
logger = logging.getLogger(__name__)
# 允许的 URL scheme
ALLOWED_SCHEMES = {"http", "https"}
# 允许的端口(标准 HTTP/HTTPS
ALLOWED_PORTS = {80, 443}
# 可信域名白名单(可根据实际 OSS/CDN 域名配置)
# 从环境变量读取,格式:"oss-cn-hangzhou.aliyuncs.com,cdn.example.com"
# 默认空表示所有公网域名都允许,但仍会做 SSRF 检查
TRUSTED_DOMAINS: set[str] = set()
_env_trusted = os.environ.get("URL_SECURITY_TRUSTED_DOMAINS", "")
if _env_trusted:
TRUSTED_DOMAINS = {d.strip() for d in _env_trusted.split(",") if d.strip()}
# 是否允许 IP 直接访问(默认禁止,防止绕过 DNS 校验)
ALLOW_DIRECT_IP = os.environ.get("URL_SECURITY_ALLOW_DIRECT_IP", "false").lower() == "true"
# 最大 URL 长度
MAX_URL_LENGTH = 2048
# 单次下载最大文件大小(默认 200MB)
DEFAULT_MAX_DOWNLOAD_SIZE = int(os.environ.get("URL_SECURITY_MAX_DOWNLOAD_MB", "200")) * 1024 * 1024
# 允许的音频 MIME 类型白名单
ALLOWED_AUDIO_MIME_TYPES = {
"audio/mpeg",
"audio/mp3",
"audio/wav",
"audio/x-wav",
"audio/pcm",
"audio/ogg",
"audio/opus",
"audio/flac",
"audio/aac",
"audio/m4a",
"audio/x-m4a",
"audio/mp4",
"application/octet-stream", # 兼容一些 CDN 返回通用类型
}
# 允许的视频 MIME 类型白名单
ALLOWED_VIDEO_MIME_TYPES = {
"video/mp4",
"video/quicktime",
"video/x-matroska",
"video/webm",
"video/avi",
"video/x-msvideo",
"video/mpeg",
"application/octet-stream",
}
# 允许的图片 MIME 类型白名单
ALLOWED_IMAGE_MIME_TYPES = {
"image/jpeg",
"image/png",
"image/gif",
"image/webp",
"image/bmp",
}
# 下载块大小
_DOWNLOAD_CHUNK_SIZE = 8192
# 最大重定向次数
_MAX_REDIRECTS = 5
class UrlSecurityError(ValueError):
"""URL 安全校验失败."""
pass
class NoRedirectHandler(urllib.request.HTTPRedirectHandler):
"""禁止自动重定向的 handler,用于手动控制重定向以做安全校验."""
def redirect_request(self, req, fp, code, msg, headers, newurl): # noqa: N802
return None
def validate_url_safety(url: str, *, purpose: str = "download") -> str:
"""校验 URL 安全性,返回标准化后的 URL(供下游使用).
Args:
url: 待校验的 URL
purpose: 用途描述(用于日志),如 "bgm_download""tts_download"
Returns:
标准化后的 URL
Raises:
UrlSecurityError: URL 不安全
"""
if not url:
raise UrlSecurityError("URL 为空")
if len(url) > MAX_URL_LENGTH:
raise UrlSecurityError(f"URL 过长 ({len(url)} > {MAX_URL_LENGTH})")
# 解析 URL
try:
parsed = urlparse(url)
except Exception as e:
raise UrlSecurityError(f"URL 解析失败: {e}") from e
# 1. Scheme 校验
if not parsed.scheme or parsed.scheme.lower() not in ALLOWED_SCHEMES:
raise UrlSecurityError(f"不允许的 URL scheme: {parsed.scheme}")
# 2. 主机名校验
hostname = parsed.hostname
if not hostname:
raise UrlSecurityError("URL 缺少主机名")
# 2.1 常见内网主机名前置拦截(防止 DNS rebinding 绕过)
_check_internal_hostnames(hostname)
# 3. 端口校验
port = parsed.port
if port is not None and port not in ALLOWED_PORTS:
raise UrlSecurityError(f"不允许的端口: {port}")
# 4. SSRF 防护 - 解析 IP 并检查
try:
# 先判断是否是 IP 地址
ip_obj = None
try:
ip_obj = ipaddress.ip_address(hostname)
except ValueError:
pass # 不是 IP,继续走域名解析
if ip_obj is not None:
# 是直接 IP 访问
if not ALLOW_DIRECT_IP and not _is_trusted_ip(ip_obj):
raise UrlSecurityError(f"禁止直接 IP 访问: {hostname}")
_check_ssrf_ip(ip_obj)
else:
# 域名 — 解析 DNS 检查 SSRF
_check_ssrf_domain(hostname)
except UrlSecurityError:
raise
except Exception as e:
logger.warning("URL 安全校验异常: url=%s purpose=%s error=%s", url[:80], purpose, e)
raise UrlSecurityError(f"URL 安全校验异常: {e}") from e
# 5. 可信域名校验(如果配置了白名单)
if TRUSTED_DOMAINS and not _is_trusted_domain(hostname):
raise UrlSecurityError(f"域名不在可信白名单中: {hostname}")
logger.debug("URL 安全校验通过: url=%s purpose=%s", url[:80], purpose)
return url
def _check_internal_hostnames(hostname: str) -> None:
"""前置检查常见内网/敏感主机名,防止 DNS 解析层绕过."""
hostname_lower = hostname.lower()
internal_hostnames = {
"localhost",
"localhost.localdomain",
"ip6-localhost",
"ip6-loopback",
"metadata",
"metadata.google.internal",
"169.254.169.254", # 云元数据服务
}
if hostname_lower in internal_hostnames:
raise UrlSecurityError(f"禁止访问内部主机名: {hostname}")
# 检查以 .local / .internal 结尾的主机名
if hostname_lower.endswith((".local", ".internal", ".localdomain")):
raise UrlSecurityError(f"禁止访问内网域名: {hostname}")
def _check_ssrf_ip(ip_obj: ipaddress.IPv4Address | ipaddress.IPv6Address) -> None:
"""检查 IP 是否属于 SSRF 风险范围."""
# 回环地址
if ip_obj.is_loopback:
raise UrlSecurityError(f"禁止访问回环地址: {ip_obj}")
# 私有地址(内网)
if ip_obj.is_private:
raise UrlSecurityError(f"禁止访问内网地址: {ip_obj}")
# 链路本地地址
if ip_obj.is_link_local:
raise UrlSecurityError(f"禁止访问链路本地地址: {ip_obj}")
# 组播地址
if ip_obj.is_multicast:
raise UrlSecurityError(f"禁止访问组播地址: {ip_obj}")
# 未指定地址(0.0.0.0 / ::
if ip_obj.is_unspecified:
raise UrlSecurityError(f"禁止访问未指定地址: {ip_obj}")
# 保留地址
if ip_obj.is_reserved:
raise UrlSecurityError(f"禁止访问保留地址: {ip_obj}")
def _check_ssrf_domain(hostname: str) -> None:
"""对域名做 DNS 解析并检查所有解析结果的 IP 是否安全.
注意:这不能完全防止 DNS rebinding,但能防御大部分 SSRF 场景。
"""
try:
# 解析所有地址
infos = socket.getaddrinfo(hostname, None)
if not infos:
raise UrlSecurityError(f"域名解析失败: {hostname}")
for info in infos:
ip_str = info[4][0]
try:
ip_obj = ipaddress.ip_address(ip_str)
_check_ssrf_ip(ip_obj)
except ValueError:
# 无法解析为 IP,跳过(不应该发生)
continue
except socket.gaierror as e:
raise UrlSecurityError(f"域名解析失败: {hostname} ({e})") from e
def _is_trusted_ip(ip_obj: ipaddress.IPv4Address | ipaddress.IPv6Address) -> bool:
"""检查 IP 是否在可信列表中(目前通过环境变量配置域名,IP 级信任暂不开放)."""
return False
def _is_trusted_domain(hostname: str) -> bool:
"""检查域名是否在可信白名单中(支持子域名匹配)."""
hostname_lower = hostname.lower()
if hostname_lower in TRUSTED_DOMAINS:
return True
# 检查子域名
for domain in TRUSTED_DOMAINS:
if hostname_lower.endswith("." + domain.lower()):
return True
return False
def is_url_safe(url: str, *, purpose: str = "download") -> bool:
"""便捷函数:检查 URL 是否安全,不抛异常."""
try:
validate_url_safety(url, purpose=purpose)
return True
except UrlSecurityError:
return False
# ── 安全下载 ───────────────────────────────────────────────────────────────────
def safe_download_file(
url: str,
dest_path: str,
*,
purpose: str = "download",
max_size: int = DEFAULT_MAX_DOWNLOAD_SIZE,
allowed_mime_types: set[str] | None = None,
timeout: float = 60.0,
) -> int:
"""安全下载 URL 到本地文件。
包含防护:
- SSRF 校验(初始 URL + 每次重定向后都校验)
- 重定向次数限制 + 手动跟随(避免重定向绕过 SSRF)
- 文件大小限制(流式读取,超过立即中断)
- MIME 类型白名单(可选)
Args:
url: 下载 URL
dest_path: 目标文件路径
purpose: 用途描述(日志用)
max_size: 最大下载字节数,超过则中断并抛出 UrlSecurityError
allowed_mime_types: 允许的 Content-Type 集合,None 表示不校验
timeout: 单次请求超时(秒)
Returns:
实际下载的字节数
Raises:
UrlSecurityError: 安全校验失败
"""
current_url = url
redirect_count = 0
total_bytes = 0
# 使用不自动跟随重定向的 opener
no_redirect_opener = urllib.request.build_opener(NoRedirectHandler())
while True:
# 每次请求前都做 SSRF 校验(重定向目标也会校验)
validate_url_safety(current_url, purpose=purpose)
req = urllib.request.Request(current_url, method="GET")
req.add_header("User-Agent", "xiaoxia-saas-worker/1.0")
try:
resp = no_redirect_opener.open(req, timeout=timeout) # nosec B310
except urllib.error.HTTPError as e:
# 3xx 重定向
if 300 <= e.code < 400 and e.headers.get("Location"):
if redirect_count >= _MAX_REDIRECTS:
raise UrlSecurityError(f"重定向次数超过限制 ({_MAX_REDIRECTS})") from e
redirect_count += 1
current_url = urljoin(current_url, e.headers["Location"])
continue
raise UrlSecurityError(f"HTTP 错误: {e.code} {e.reason}") from e
except urllib.error.URLError as e:
raise UrlSecurityError(f"URL 错误: {e.reason}") from e
try:
# Content-Type 校验
if allowed_mime_types is not None:
content_type = resp.headers.get("Content-Type", "").split(";")[0].strip().lower()
if content_type and content_type not in allowed_mime_types:
raise UrlSecurityError(
f"不允许的 Content-Type: {content_type}, " f"允许: {sorted(allowed_mime_types)}"
)
# Content-Length 预检
content_length = resp.headers.get("Content-Length")
if content_length and int(content_length) > max_size:
raise UrlSecurityError(f"文件过大: {content_length} bytes > {max_size} bytes 上限")
# 流式下载,实时检查大小
with open(dest_path, "wb") as f:
while True:
chunk = resp.read(_DOWNLOAD_CHUNK_SIZE)
if not chunk:
break
total_bytes += len(chunk)
if total_bytes > max_size:
raise UrlSecurityError(f"下载超过大小限制: {total_bytes} bytes > {max_size} bytes")
f.write(chunk)
return total_bytes
finally:
resp.close()
def safe_download_bytes(
url: str,
*,
purpose: str = "download",
max_size: int = DEFAULT_MAX_DOWNLOAD_SIZE,
allowed_mime_types: set[str] | None = None,
timeout: float = 60.0,
) -> bytes:
"""安全下载 URL 并返回字节内容。
防护同 safe_download_file,但结果返回在内存中(适合小文件)。
"""
import tempfile
fd, tmp_path = tempfile.mkstemp()
os.close(fd)
try:
safe_download_file(
url,
tmp_path,
purpose=purpose,
max_size=max_size,
allowed_mime_types=allowed_mime_types,
timeout=timeout,
)
with open(tmp_path, "rb") as f:
return f.read()
finally:
try:
os.unlink(tmp_path)
except OSError:
pass
+50 -18
View File
@@ -48,23 +48,55 @@ omit = [
]
branch = true
[tool.coverage.report]
exclude_lines = [
"pragma: no cover",
"def __repr__",
"if __name__ == .__main__.:",
"raise NotImplementedError",
"pass",
"if TYPE_CHECKING:",
"class .*Protocol",
"@abstractmethod",
"raise AssertionError",
"raise RuntimeError",
"if 0:",
"if __debug__:",
[tool.ruff]
target-version = "py311"
line-length = 120
exclude = [
".git",
".cache",
"__pycache__",
".venv",
"venv",
"node_modules",
"alembic",
".gitea",
".next",
"dist",
"build",
]
show_missing = true
skip_covered = false
[tool.coverage.xml]
output = "coverage.xml"
[tool.ruff.lint]
# 当前阶段:摸底模式,规则集与原flake8对齐
# 后续迭代计划:
# Phase 1: 修完 bugbear 后正式替换 flake8
# Phase 2: 启用 UP(pyupgrade) + SIM(simplify)
# Phase 3: 启用 RET(return) + ARG(unused-args)
select = [
"E", # pycodestyle errors(同flake8
"F", # pyflakes(同flake8
"W", # pycodestyle warnings(同flake8
"B", # flake8-bugbear(新增,摸底用)
]
# 与原 setup.cfg flake8 配置对齐,确保不新增阻断
ignore = [
"E203",
"W503",
"E501", # line-too-longblack管)
"E302",
"E402", # module-import-not-at-top(循环导入多)
"E722", # bare-except
"W291",
"W293",
"F401", # unused-import
"F403",
"F405",
"F841", # unused-variable
"B008", # do-not-perform-callback-from-argfastapi依赖注入)
]
[tool.ruff.lint.per-file-ignores]
"__init__.py" = ["F401", "F403", "F405"]
"tests/*" = ["E402", "F401", "F841"]
"packages/ports/*" = ["E301", "E704"]
"apps/*/migrations/*" = ["ALL"]
"alembic/*" = ["ALL"]
Executable → Regular
+1 -1
View File
@@ -13,7 +13,7 @@ uvicorn[standard]==0.32.0
pydantic==2.9.0
# 认证核心
pyjwt==2.9.0
pyjwt==2.13.0
bcrypt==4.2.0
# Redis
+183
View File
@@ -0,0 +1,183 @@
#!/usr/bin/env python3
"""
自动生成 CHANGELOG 条目。
用法:
python3 scripts/generate_changelog.py v0.1.128 v0.1.129
python3 scripts/generate_changelog.py v0.1.128 HEAD
"""
from __future__ import annotations
import json
import os
import re
import sys
import urllib.error
import urllib.request
from datetime import datetime
GITEA_URL = os.environ.get("GITEA_URL", "https://git.xiaoxiajianji.com")
REPO = os.environ.get("GITEA_REPO", "xiaoxia/xiaoxia-saas")
TOKEN = os.environ.get("GITEA_TOKEN", "")
def gitea_api(path: str) -> dict | list:
url = f"{GITEA_URL}/api/v1{path}"
req = urllib.request.Request(url)
if TOKEN:
req.add_header("Authorization", f"token {TOKEN}")
try:
with urllib.request.urlopen(req, timeout=30) as resp:
return json.loads(resp.read().decode())
except urllib.error.HTTPError as e:
print(f"API Error: {e.code} {e.reason}", file=sys.stderr)
raise
def get_tag_date(tag: str) -> str:
try:
info = gitea_api(f"/repos/{REPO}/git/refs/tags/{tag}")
if isinstance(info, dict):
sha = info.get("object", {}).get("sha", "")
if sha:
commit = gitea_api(f"/repos/{REPO}/git/commits/{sha}")
if isinstance(commit, dict):
return commit.get("committer", {}).get("date", "")[:10]
except Exception:
pass
return ""
def get_merged_prs_between(from_tag: str, to_tag: str) -> list[dict]:
all_prs: list[dict] = []
page = 1
while True:
prs = gitea_api(f"/repos/{REPO}/pulls?state=closed&sort=merged&direction=desc" f"&per_page=50&page={page}")
if not isinstance(prs, list) or not prs:
break
all_prs.extend(prs)
if len(prs) < 50:
break
page += 1
if page > 10:
break
merged = [pr for pr in all_prs if pr.get("merged_at")]
from_date = get_tag_date(from_tag)
to_date = get_tag_date(to_tag) if not to_tag.startswith("HEAD") else datetime.now().strftime("%Y-%m-%d")
if not from_date:
return merged[:50]
result = []
for pr in merged:
merged_at = pr.get("merged_at", "")[:10]
if from_date <= merged_at <= to_date:
result.append(pr)
return result
def categorize_pr(title: str) -> tuple[str, str]:
title = title.strip()
lower = title.lower()
m = re.match(r"^(feat|fix|chore|perf|docs|refactor|test|ci|style|build|security)\s*[:]", title)
if m:
prefix = m.group(1)
clean_title = title[m.end() :].strip()
else:
prefix = ""
clean_title = title
if prefix in ("feat", "feature"):
return "✨ 功能", clean_title
elif prefix == "fix":
return "🐛 Bug 修复", clean_title
elif prefix in ("refactor", "chore", "style"):
return "🔄 重构与清理", clean_title
elif prefix in ("perf", "performance"):
return "⚡ 性能优化", clean_title
elif prefix == "security":
return "🔒 安全修复", clean_title
elif prefix == "docs":
return "📝 文档", clean_title
elif prefix == "test":
return "🧪 测试", clean_title
elif prefix in ("ci", "build"):
return "🚀 CI/CD & 基础设施", clean_title
else:
if any(k in lower for k in ["安全", "security", "cve", "漏洞"]):
return "🔒 安全修复", clean_title
elif any(k in lower for k in ["修复", "bug"]):
return "🐛 Bug 修复", clean_title
elif any(k in lower for k in ["新增", "添加", "feat", "功能"]):
return "✨ 功能", clean_title
elif any(k in lower for k in ["ci", "构建", "workflow", "pipeline"]):
return "🚀 CI/CD & 基础设施", clean_title
elif any(k in lower for k in ["测试", "test", "e2e"]):
return "🧪 测试", clean_title
else:
return "📌 其他", clean_title
def generate_changelog(from_tag: str, to_tag: str, version: str = "") -> str:
if not version:
version = to_tag
prs = get_merged_prs_between(from_tag, to_tag)
categories: dict[str, list[tuple[int, str]]] = {}
for pr in prs:
cat, title = categorize_pr(pr["title"])
pr_num = pr["number"]
categories.setdefault(cat, []).append((pr_num, title))
order = [
"🔒 安全修复",
"✨ 功能",
"🐛 Bug 修复",
"⚡ 性能优化",
"🔄 重构与清理",
"📝 文档",
"🧪 测试",
"🚀 CI/CD & 基础设施",
"📌 其他",
]
date_str = get_tag_date(to_tag) if not to_tag.startswith("HEAD") else datetime.now().strftime("%Y-%m-%d")
lines = [f"## [{version}] - {date_str}", ""]
for cat in order:
items = categories.get(cat, [])
if not items:
continue
lines.append(f"### {cat}")
lines.append("")
for num, title in sorted(items, key=lambda x: x[0]):
short_title = title.split("")[0].split(" - ")[0]
if len(short_title) > 80:
short_title = short_title[:77] + "..."
lines.append(f"- #{num} {short_title}")
lines.append("")
lines.append("---")
lines.append("")
return "\n".join(lines)
def main():
if len(sys.argv) < 3:
print(f"用法: {sys.argv[0]} <from_tag> <to_tag> [version]")
sys.exit(1)
from_tag = sys.argv[1]
to_tag = sys.argv[2]
version = sys.argv[3] if len(sys.argv) > 3 else ""
changelog = generate_changelog(from_tag, to_tag, version)
print(changelog)
if __name__ == "__main__":
main()
+184 -74
View File
@@ -1,9 +1,13 @@
#!/bin/bash
# 灰度发布脚本:通过Nginx权重调整流量比例
# 灰度发布脚本:在生产服务器上启动 canary 版本,通过 Nginx 权重切流
# 用法: ./scripts/gray_deploy.sh <版本号> <灰度百分比>
#
# 需要在目标服务器上执行,或通过SSH执行
# 前提:服务器上运行两个版本的容器(stable + canary),Nginx做加权轮询
# 前提:
# - 在生产服务器上执行(或通过 SSH 管道执行)
# - 当前已有全量运行的 production 容器
# - Nginx 配置在 /etc/nginx/sites-enabled/00-xiaoxia-saas
#
# 灰度范围:API + Web(Worker 暂时全量升级,队列消费无法按比例切流)
set -euo pipefail
@@ -12,106 +16,212 @@ GRAY_PCT="${2:-10}"
if [[ -z "$VERSION" ]]; then
echo "用法: $0 <版本号> [灰度百分比]"
echo "示例: $0 v0.1.129 5"
echo "示例: $0 v0.1.130 5"
exit 1
fi
STABLE_VERSION="${STABLE_VERSION:-current}"
REGISTRY="${REGISTRY:-git.xiaoxiajianji.com/xiaoxia/xiaoxia-saas}"
NGINX_CONF="${NGINX_CONF:-/etc/nginx/sites-enabled/00-xiaoxia-saas}"
ENV_FILE="${ENV_FILE:-/var/lib/xiaoxia-saas-production/.env}"
GENERATED_DIR="${GENERATED_DIR:-/var/lib/xiaoxia-saas-production/generated}"
LEGACY_ASSETS_DIR="${LEGACY_ASSETS_DIR:-/var/lib/xiaoxia-saas-production/legacy-assets}"
echo "=========================================="
# Canary 端口(与 production 错开)
CANARY_API_PORT=18001
CANARY_WEB_PORT=13002
echo "============================================"
echo " 灰度发布"
echo " 新版本: $VERSION"
echo " 灰度比例: ${GRAY_PCT}%"
echo " 稳定版本: $STABLE_VERSION"
echo "=========================================="
echo " Canary API 端口: $CANARY_API_PORT"
echo " Canary Web 端口: $CANARY_WEB_PORT"
echo "============================================"
# 1. 拉取新版本镜像
# 1. 检查环境
if [[ ! -f "$ENV_FILE" ]]; then
echo "错误: 环境文件不存在: $ENV_FILE"
exit 1
fi
if [[ ! -f "$NGINX_CONF" ]]; then
echo "错误: Nginx 配置不存在: $NGINX_CONF"
exit 1
fi
# 2. 拉取新版本镜像
echo ""
echo ">>> 拉取新版本镜像..."
for component in api worker web; do
for component in api web worker; do
echo " 拉取 $component:$VERSION ..."
docker pull "${REGISTRY}-${component}:${VERSION}" 2>&1 | tail -1
done
echo " ✅ 镜像拉取完成"
# 2. 启动灰度版本容器(canary
# 3. 启动 API Canary
echo ""
echo ">>> 启动灰度版本容器..."
# API canary
CANARY_API_NAME="saas-api-canary"
if docker ps -a --format '{{.Names}}' | grep -q "^${CANARY_API_NAME}$"; then
echo " 停止旧 canary 容器..."
docker stop "$CANARY_API_NAME" 2>/dev/null || true
docker rm "$CANARY_API_NAME" 2>/dev/null || true
echo ">>> 启动 API Canary 容器..."
CANARY_API="xiaoxia-api-canary"
if docker ps -a --format '{{.Names}}' | grep -q "^${CANARY_API}$"; then
echo " 停止旧 canary..."
docker rm -f "$CANARY_API" >/dev/null 2>&1
fi
echo " 启动 api canary..."
docker run -d \
--name "$CANARY_API_NAME" \
--network saas-network \
-e DATABASE_URL="${DATABASE_URL}" \
-e REDIS_URL="${REDIS_URL}" \
-e FEATURE_FLAG_PROVIDER=redis \
--name "$CANARY_API" \
--env-file "$ENV_FILE" \
--network xiaoxia-net-production \
-p "127.0.0.1:${CANARY_API_PORT}:8000" \
-e APP_ENV=production \
-e APP_VERSION="$VERSION-canary" \
-e GENERATED_FILES_DIR=/app/generated \
-e GENERATED_FILES_URL_PREFIX=/generated-files \
-e PUBLIC_API_BASE_URL=https://saas-api.xiaoxiajianji.com \
-v "${GENERATED_DIR}:/app/generated" \
--restart unless-stopped \
"${REGISTRY}-api:${VERSION}"
--cpus 1 \
--memory 1g \
--health-cmd "python -c \"import urllib.request; urllib.request.urlopen('http://localhost:8000/health', timeout=5)\"" \
--health-interval 30s \
--health-timeout 10s \
--health-retries 3 \
--health-start-period 40s \
--log-driver json-file \
--log-opt max-size=50m \
--log-opt max-file=3 \
"${REGISTRY}-api:${VERSION}" >/dev/null
# Worker canary
CANARY_WORKER_NAME="saas-worker-canary"
if docker ps -a --format '{{.Names}}' | grep -q "^${CANARY_WORKER_NAME}$"; then
echo " 停止旧 worker canary..."
docker stop "$CANARY_WORKER_NAME" 2>/dev/null || true
docker rm "$CANARY_WORKER_NAME" 2>/dev/null || true
echo " ✅ API Canary 已启动(端口 $CANARY_API_PORT"
# 4. 启动 Web Canary
echo ""
echo ">>> 启动 Web Canary 容器..."
CANARY_WEB="xiaoxia-web-canary"
if docker ps -a --format '{{.Names}}' | grep -q "^${CANARY_WEB}$"; then
echo " 停止旧 canary..."
docker rm -f "$CANARY_WEB" >/dev/null 2>&1
fi
echo " 启动 worker canary..."
docker run -d \
--name "$CANARY_WORKER_NAME" \
--network saas-network \
-e DATABASE_URL="${DATABASE_URL}" \
-e REDIS_URL="${REDIS_URL}" \
--restart unless-stopped \
"${REGISTRY}-worker:${VERSION}"
LEGACY_VOLUME=""
if [[ -d "$LEGACY_ASSETS_DIR" ]] && [[ -n "$(ls -A "$LEGACY_ASSETS_DIR" 2>/dev/null)" ]]; then
LEGACY_VOLUME="-v ${LEGACY_ASSETS_DIR}:/usr/share/nginx/html/assets-legacy/assets:ro"
fi
# 3. 等待容器健康
docker run -d \
--name "$CANARY_WEB" \
--network xiaoxia-net-production \
-p "127.0.0.1:${CANARY_WEB_PORT}:80" \
--restart unless-stopped \
--cpus 0.5 \
--memory 256m \
$LEGACY_VOLUME \
--health-cmd "wget --spider -q http://127.0.0.1:80" \
--health-interval 30s \
--health-timeout 5s \
--health-retries 3 \
--health-start-period 10s \
--log-driver json-file \
--log-opt max-size=50m \
--log-opt max-file=3 \
"${REGISTRY}-web:${VERSION}" >/dev/null
echo " ✅ Web Canary 已启动(端口 $CANARY_WEB_PORT"
# 5. 等待健康检查
echo ""
echo ">>> 等待容器健康..."
sleep 10
if ! docker ps --format '{{.Names}} {{.Status}}' | grep -q "$CANARY_API_NAME"; then
echo "错误: API canary 容器未运行"
docker logs "$CANARY_API_NAME" --tail 20
echo ">>> 等待 Canary 容器健康..."
for i in $(seq 1 40); do
api_healthy=$(docker inspect --format='{{.State.Health.Status}}' "$CANARY_API" 2>/dev/null || echo "starting")
web_healthy=$(docker inspect --format='{{.State.Health.Status}}' "$CANARY_WEB" 2>/dev/null || echo "starting")
if [[ "$api_healthy" == "healthy" && "$web_healthy" == "healthy" ]]; then
echo " ✅ API + Web Canary 均健康(用时 ${i}s"
break
fi
if [[ "$api_healthy" == "unhealthy" ]]; then
echo " ❌ API Canary 健康检查失败"
docker logs --tail 30 "$CANARY_API"
exit 1
fi
if [[ "$web_healthy" == "unhealthy" ]]; then
echo " ❌ Web Canary 健康检查失败"
docker logs --tail 20 "$CANARY_WEB"
exit 1
fi
sleep 3
done
# 6. 更新 Nginx 配置 - 添加 upstream 权重
echo ""
echo ">>> 更新 Nginx 权重(稳定: $((100-GRAY_PCT))% / 灰度: ${GRAY_PCT}%..."
# 备份
BAK_FILE="${NGINX_CONF}.bak.gray.$(date +%Y%m%d%H%M%S)"
cp "$NGINX_CONF" "$BAK_FILE"
echo " 已备份: $BAK_FILE"
# 生成 upstream 块
UPSTREAM_BLOCK="
# Gray release upstreams(自动生成 - gray_deploy.sh
upstream saas_api_backend {
server 127.0.0.1:8001 weight=$((100-GRAY_PCT));
server 127.0.0.1:${CANARY_API_PORT} weight=${GRAY_PCT};
}
upstream saas_web_backend {
server 127.0.0.1:3002 weight=$((100-GRAY_PCT));
server 127.0.0.1:${CANARY_WEB_PORT} weight=${GRAY_PCT};
}
"
# 在文件最前面插入 upstream 块
TMP_CONF=$(mktemp)
{
echo "$UPSTREAM_BLOCK"
cat "$NGINX_CONF"
} > "$TMP_CONF"
# 替换 proxy_pass 指向 upstream
# API: proxy_pass http://127.0.0.1:8001 -> proxy_pass http://saas_api_backend
sed -i 's|proxy_pass http://127\.0\.0\.1:8001|proxy_pass http://saas_api_backend|g' "$TMP_CONF"
# Web: proxy_pass http://127.0.0.1:3002/ -> proxy_pass http://saas_web_backend/
sed -i 's|proxy_pass http://127\.0\.0\.1:3002/|proxy_pass http://saas_web_backend/|g' "$TMP_CONF"
# 测试配置
mv "$TMP_CONF" "$NGINX_CONF"
if ! nginx -t 2>&1; then
echo " ❌ Nginx 配置测试失败,回滚..."
cp "$BAK_FILE" "$NGINX_CONF"
nginx -t
exit 1
fi
echo " ✅ API canary 运行中"
# 4. 更新Nginx权重
nginx -s reload
echo " ✅ Nginx 已 reload,灰度生效"
# 7. 验证灰度流量
echo ""
echo ">>> 更新Nginx权重 (稳定: $((100-GRAY_PCT))% / 灰度: ${GRAY_PCT}%)..."
NGINX_CONF="${NGINX_CONF:-/etc/nginx/conf.d/saas-api.conf}"
if [[ -f "$NGINX_CONF" ]]; then
# 备份
cp "$NGINX_CONF" "${NGINX_CONF}.bak.$(date +%Y%m%d%H%M%S)"
# 更新 upstream 权重(需要根据实际配置调整)
echo " 请手动更新 Nginx upstream 配置中的权重"
echo " 示例配置:"
cat <<EOF
upstream saas_api_backend {
server saas-api:8000 weight=$((100-GRAY_PCT));
server saas-api-canary:8000 weight=${GRAY_PCT};
}
EOF
nginx -t && nginx -s reload
echo " ✅ Nginx 已reload"
else
echo " 警告: Nginx 配置文件不存在 ($NGINX_CONF)"
echo " 请手动配置灰度流量权重"
fi
echo ">>> 验证灰度流量..."
gray_hits=0
total_hits=20
for i in $(seq 1 $total_hits); do
resp=$(curl -s -o /dev/null -w "%{http_code}" -H "X-Gray-Test: 1" http://127.0.0.1:${CANARY_API_PORT}/health 2>/dev/null || echo "000")
if [[ "$resp" == "200" ]]; then
gray_hits=$((gray_hits + 1))
fi
sleep 0.1
done
echo " Canary 健康验证: $gray_hits/$total_hits 请求成功"
echo ""
echo "=========================================="
echo "============================================"
echo " ✅ 灰度发布完成"
echo " 版本: $VERSION (${GRAY_PCT}%流量)"
echo " 监控: Grafana / 日志"
echo "=========================================="
echo " 版本: $VERSION (${GRAY_PCT}%流量)"
echo " API: 127.0.0.1:$CANARY_API_PORT"
echo " Web: 127.0.0.1:$CANARY_WEB_PORT"
echo " Nginx 备份: $BAK_FILE"
echo " 回滚: ./scripts/rollback.sh"
echo " Worker: 暂不灰度(队列消费无法按比例切流)"
echo "============================================"
+37 -50
View File
@@ -1,6 +1,11 @@
#!/bin/bash
# 一键发布脚本:打tag → 触发生产镜像构建 → 部署到灰度
# 用法: ./scripts/release.sh v0.1.129 [--gray 5]
# 一键发布脚本:打 tag → 触发 CI 构建 → 可选灰度发布
# 用法: ./scripts/release.sh v0.1.130 [--gray 5]
#
# 说明:
# - 打 tag 后 CI 会自动构建镜像并全量部署到生产
# - 加 --gray 参数则在构建完成后执行灰度切流(需 SSH 到生产服务器执行)
# - 加 --no-deploy 只打 tag 不触发自动部署
set -euo pipefail
@@ -11,13 +16,12 @@ usage() {
echo "用法: $0 <版本号> [--gray 百分比] [--no-deploy]"
echo ""
echo "示例:"
echo " $0 v0.1.129 # 打tag + 全量发布"
echo " $0 v0.1.129 --gray 5 # 打tag + 5%灰度发布"
echo " $0 v0.1.129 --no-deploy # 只打tag,不部署"
echo " $0 v0.1.130 # 打tag + 全量发布CI自动部署)"
echo " $0 v0.1.130 --gray 5 # 打tag + 5%灰度发布"
echo " $0 v0.1.130 --no-deploy # 只打tag,不部署"
exit 1
}
# 参数解析
VERSION=""
GRAY_PCT=0
DEPLOY=true
@@ -47,80 +51,63 @@ while [[ $# -gt 0 ]]; do
done
if [[ -z "$VERSION" ]]; then
echo "错误: 请指定版本号(如 v0.1.129"
echo "错误: 请指定版本号(如 v0.1.130"
usage
fi
echo "=========================================="
echo "============================================"
echo " 发布版本: $VERSION"
echo " 灰度比例: ${GRAY_PCT}%"
echo " 自动部署: $DEPLOY"
echo "=========================================="
echo "============================================"
cd "$REPO_ROOT"
# 1. 确认在 develop 分支
# 1. 确认分支
CURRENT_BRANCH=$(git rev-parse --abbrev-ref HEAD)
if [[ "$CURRENT_BRANCH" != "develop" ]]; then
echo "错误: 请切换到 develop 分支后再发布"
echo "错误: 请 develop 分支上打 tag"
exit 1
fi
# 2. 拉取最新代码
# 2. 拉取最新
echo ""
echo ">>> 拉取最新代码..."
git pull origin develop
# 3. 生成 CHANGELOG
echo ""
echo ">>> 生成 CHANGELOG..."
if [[ -f "scripts/generate_changelog.py" ]]; then
PREV_TAG=$(git describe --tags --abbrev=0 HEAD^ 2>/dev/null || echo "")
if [[ -n "$PREV_TAG" ]]; then
python3 scripts/generate_changelog.py \
--from-tag "$PREV_TAG" \
--to-tag HEAD \
--gitea-token "${GITEA_TOKEN:-}" \
--output /tmp/changelog_$$.md
echo "CHANGELOG 已生成到 /tmp/changelog_$$.md"
fi
fi
# 4. 打tag
echo ""
echo ">>> 打 tag $VERSION ..."
# 3. 检查 tag 是否已存在
if git rev-parse "$VERSION" >/dev/null 2>&1; then
echo "警告: tag $VERSION 已存在,跳过打tag"
echo "警告: tag $VERSION 已存在,跳过打 tag"
else
echo ""
echo ">>> 打 tag $VERSION ..."
git tag -a "$VERSION" -m "Release $VERSION"
git push origin "$VERSION"
echo "Tag $VERSION 已推送,触发生产镜像构建..."
echo " ✅ Tag 已推送,CI 将自动构建生产镜像"
fi
# 5. 等待镜像构建
# 4. 部署提示
if [[ "$DEPLOY" == "true" ]]; then
echo ""
echo ">>> 等待镜像构建完成(约10-15分钟)..."
echo " 镜像: git.xiaoxiajianji.com/xiaoxia/xiaoxia-saas-{api,worker,web}:$VERSION"
# 这里可以加镜像存在性检查
echo " (镜像构建由CI自动完成,请在Gitea Actions中确认)"
fi
# 6. 灰度部署
if [[ "$DEPLOY" == "true" && "$GRAY_PCT" -gt 0 ]]; then
echo ">>> 构建 & 部署"
echo " CI 会自动执行:"
echo " 1. Build Production Runtime Images(约10-15分钟)"
echo " 2. Deploy ProductionSSH 到生产服务器部署)"
echo ""
echo ">>> 灰度部署: ${GRAY_PCT}% 流量到 $VERSION"
if [[ -f "scripts/gray_deploy.sh" ]]; then
./scripts/gray_deploy.sh "$VERSION" "$GRAY_PCT"
else
echo "警告: gray_deploy.sh 不存在,跳过灰度部署"
echo " 查看进度: Gitea Actions → 对应 tag 的 run"
if [[ "$GRAY_PCT" -gt 0 ]]; then
echo ""
echo ">>> 灰度发布"
echo " 构建部署完成后,在生产服务器上执行:"
echo " cd /var/lib/xiaoxia-saas-production"
echo " ./gray_deploy.sh $VERSION $GRAY_PCT"
fi
fi
echo ""
echo "=========================================="
echo " ✅ 发布流程完成"
echo "============================================"
echo " ✅ 发布流程触发完成"
echo " 版本: $VERSION"
echo " 灰度: ${GRAY_PCT}%"
echo "=========================================="
echo "============================================"
+77
View File
@@ -0,0 +1,77 @@
#!/bin/bash
# 灰度回滚脚本:切回全量稳定版本,停止 canary 容器
# 用法: ./scripts/rollback_gray.sh
set -euo pipefail
NGINX_CONF="${NGINX_CONF:-/etc/nginx/sites-enabled/00-xiaoxia-saas}"
CANARY_API="${CANARY_API:-xiaoxia-api-canary}"
CANARY_WEB="${CANARY_WEB:-xiaoxia-web-canary}"
echo "============================================"
echo " 灰度回滚"
echo " 目标: 全量切回稳定版本"
echo "============================================"
# 1. 找最近的灰度备份
echo ""
echo ">>> 查找最近的灰度备份..."
LATEST_BAK=$(ls -t "${NGINX_CONF}".bak.gray.* 2>/dev/null | head -1 || true)
if [[ -z "$LATEST_BAK" ]]; then
echo " 未找到灰度备份,尝试手动移除 upstream 配置..."
# 手动回滚:移除 upstream 块,把 proxy_pass 改回 127.0.0.1
TMP_CONF=$(mktemp)
# 移除 upstream 块(从 "# Gray release upstreams" 到空行结束)
awk '
/^# Gray release upstreams/ { skip=1; next }
skip && /^$/ && !found_first_empty { found_first_empty=1; next }
skip && found_first_empty && /^$/ { skip=0; found_first_empty=0; next }
skip { next }
{ print }
' "$NGINX_CONF" > "$TMP_CONF"
# 把 upstream 名改回 IP
sed -i 's|proxy_pass http://saas_api_backend|proxy_pass http://127.0.0.1:8001|g' "$TMP_CONF"
sed -i 's|proxy_pass http://saas_web_backend/|proxy_pass http://127.0.0.1:3002/|g' "$TMP_CONF"
mv "$TMP_CONF" "$NGINX_CONF"
else
echo " 从备份恢复: $LATEST_BAK"
cp "$LATEST_BAK" "$NGINX_CONF"
fi
# 2. 测试并 reload nginx
echo ""
echo ">>> Nginx 测试 & reload..."
if ! nginx -t 2>&1; then
echo " ❌ Nginx 配置测试失败!请检查"
exit 1
fi
nginx -s reload
echo " ✅ Nginx 已回滚,全量切回稳定版本"
# 3. 停止 canary 容器(延迟停止,保留30分钟便于排查)
echo ""
echo ">>> Canary 容器将在30分钟后停止(便于排查)"
echo " 立即停止请执行: docker rm -f $CANARY_API $CANARY_WEB"
# 30分钟后停止(后台执行,不阻塞脚本)
(
sleep 1800
for c in "$CANARY_API" "$CANARY_WEB"; do
if docker ps --format '{{.Names}}' | grep -q "^${c}$"; then
docker stop "$c" >/dev/null 2>&1 && docker rm "$c" >/dev/null 2>&1
echo "[$(date)] 已停止 canary 容器: $c"
fi
done
) &
echo ""
echo "============================================"
echo " ✅ 灰度回滚完成"
echo " 流量已全部切回稳定版本"
echo " Canary 容器: 30分钟后自动清理"
echo "============================================"
File diff suppressed because it is too large Load Diff
+243
View File
@@ -0,0 +1,243 @@
"""路径安全校验工具单元测试 — 路径遍历防护."""
from __future__ import annotations
import os
import sys
import tempfile
import unittest
from pathlib import Path
sys.path.insert(0, os.path.join(os.path.dirname(__file__), "..", "..", "apps", "worker"))
from video_processing.path_security import ( # noqa: E402
PathSecurityError,
get_allowed_local_dirs,
is_in_allowed_dirs,
is_path_safe,
safe_resolve_path,
sanitize_filename,
validate_local_schema_path,
)
class TestSafeResolvePath(unittest.TestCase):
"""安全路径解析测试."""
def setUp(self):
self.tmpdir = tempfile.mkdtemp()
def tearDown(self):
import shutil
shutil.rmtree(self.tmpdir, ignore_errors=True)
# ── 正常路径 ─────────────────────────────────────────────────────────
def test_simple_relative_path(self):
"""简单相对路径应该正常解析."""
result = safe_resolve_path("test.mp4", self.tmpdir)
self.assertEqual(result.name, "test.mp4")
self.assertTrue(str(result).startswith(self.tmpdir))
def test_subdirectory_path(self):
"""子目录路径应该正常解析."""
result = safe_resolve_path("sub/dir/file.mp4", self.tmpdir)
self.assertTrue(str(result).startswith(self.tmpdir))
self.assertIn("sub/dir/file.mp4", str(result).replace("\\", "/"))
def test_dot_slash_path(self):
"""./ 开头的路径应该正常解析."""
result = safe_resolve_path("./test.mp4", self.tmpdir)
self.assertEqual(result.name, "test.mp4")
# ── 路径遍历防护 ─────────────────────────────────────────────────────
def test_parent_traversal_rejected(self):
"""../ 路径遍历应该被拒绝."""
with self.assertRaises(PathSecurityError):
safe_resolve_path("../etc/passwd", self.tmpdir)
def test_multiple_parent_traversal_rejected(self):
"""多级 ../ 遍历应该被拒绝."""
with self.assertRaises(PathSecurityError):
safe_resolve_path("../../etc/passwd", self.tmpdir)
def test_mixed_traversal_rejected(self):
"""混合路径遍历应该被拒绝."""
with self.assertRaises(PathSecurityError):
safe_resolve_path("./sub/../../etc/shadow", self.tmpdir)
def test_absolute_path_rejected(self):
"""绝对路径(超出基目录)应该被拒绝."""
with self.assertRaises(PathSecurityError):
safe_resolve_path("/etc/passwd", self.tmpdir)
# ── 空字节注入 ───────────────────────────────────────────────────────
def test_null_byte_rejected(self):
"""空字节注入应该被拒绝."""
with self.assertRaises(PathSecurityError):
safe_resolve_path("test\x00.mp4", self.tmpdir)
# ── 空路径 ──────────────────────────────────────────────────────────
def test_empty_path_rejected(self):
"""空路径应该被拒绝."""
with self.assertRaises(PathSecurityError):
safe_resolve_path("", self.tmpdir)
def test_none_path_rejected(self):
"""None 路径应该被拒绝."""
with self.assertRaises(PathSecurityError):
safe_resolve_path(None, self.tmpdir) # type: ignore
def test_whitespace_path_rejected(self):
"""空白路径应该被拒绝."""
with self.assertRaises(PathSecurityError):
safe_resolve_path(" ", self.tmpdir)
# ── 路径长度 ────────────────────────────────────────────────────────
def test_too_long_path_rejected(self):
"""超长路径应该被拒绝."""
long_path = "a" * 5000 + ".mp4"
with self.assertRaises(PathSecurityError):
safe_resolve_path(long_path, self.tmpdir)
# ── 系统路径防护 ─────────────────────────────────────────────────────
def test_proc_path_rejected_when_absolute(self):
"""/proc/ 路径在绝对路径模式下应该被拒绝(因为超出基目录)."""
with self.assertRaises(PathSecurityError):
safe_resolve_path("/proc/self/environ", self.tmpdir)
# ── 扩展名校验 ───────────────────────────────────────────────────────
def test_extension_whitelist_pass(self):
"""白名单内的扩展名应该通过."""
result = safe_resolve_path(
"test.mp4",
self.tmpdir,
allowed_extensions={".mp4", ".mov"},
)
self.assertEqual(result.suffix.lower(), ".mp4")
def test_extension_whitelist_reject(self):
"""白名单外的扩展名应该被拒绝."""
with self.assertRaises(PathSecurityError):
safe_resolve_path(
"test.exe",
self.tmpdir,
allowed_extensions={".mp4", ".mov"},
)
class TestLocalSchemaPath(unittest.TestCase):
"""local:// schema 路径测试."""
def setUp(self):
self.tmpdir = tempfile.mkdtemp()
def tearDown(self):
import shutil
shutil.rmtree(self.tmpdir, ignore_errors=True)
def test_valid_local_schema(self):
"""有效的 local:// 相对路径应该通过."""
# 创建测试文件
test_file = Path(self.tmpdir) / "test.mp4"
test_file.touch()
result = validate_local_schema_path("local://test.mp4", self.tmpdir)
self.assertTrue(result.exists())
def test_local_schema_absolute_rejected(self):
"""local:// + 绝对路径应该被拒绝."""
with self.assertRaises(PathSecurityError):
validate_local_schema_path("local:///etc/passwd", self.tmpdir)
def test_local_schema_traversal_rejected(self):
"""local:// + 路径遍历应该被拒绝."""
with self.assertRaises(PathSecurityError):
validate_local_schema_path("local://../etc/passwd", self.tmpdir)
def test_non_local_schema_rejected(self):
"""非 local:// 开头的路径应该被拒绝."""
with self.assertRaises(PathSecurityError):
validate_local_schema_path("http://example.com/test", self.tmpdir)
class TestSanitizeFilename(unittest.TestCase):
"""文件名清理测试."""
def test_normal_filename(self):
"""正常文件名应该保持不变."""
self.assertEqual(sanitize_filename("video.mp4"), "video.mp4")
def test_path_separators_removed(self):
"""路径分隔符应该被替换."""
self.assertNotIn("/", sanitize_filename("../path/to/file.mp4"))
self.assertNotIn("\\", sanitize_filename("..\\path\\file.mp4"))
def test_leading_dots_removed(self):
"""开头的点应该被移除."""
result = sanitize_filename(".hidden")
self.assertFalse(result.startswith("."))
self.assertEqual(result, "hidden")
def test_multiple_leading_dots_removed(self):
"""多个开头的点应该全部被移除."""
result = sanitize_filename("...hidden")
self.assertFalse(result.startswith("."))
def test_empty_filename_default(self):
"""空文件名应该返回 unnamed."""
self.assertEqual(sanitize_filename(""), "unnamed")
def test_special_chars_removed(self):
"""特殊字符应该被替换."""
result = sanitize_filename('file<name>:"test|?*.mp4')
self.assertNotIn("<", result)
self.assertNotIn(">", result)
self.assertNotIn(":", result)
self.assertNotIn('"', result)
self.assertNotIn("|", result)
self.assertNotIn("?", result)
self.assertNotIn("*", result)
def test_chinese_filename_preserved(self):
"""中文文件名应该保留."""
result = sanitize_filename("视频素材.mp4")
self.assertIn("视频素材", result)
def test_long_filename_truncated(self):
"""超长文件名应该被截断."""
long_name = "a" * 300 + ".mp4"
result = sanitize_filename(long_name)
self.assertLessEqual(len(result), 255)
self.assertTrue(result.endswith(".mp4"))
class TestAllowedDirs(unittest.TestCase):
"""允许目录配置测试."""
def test_get_allowed_dirs_returns_list(self):
"""get_allowed_local_dirs 应该返回列表."""
dirs = get_allowed_local_dirs()
self.assertIsInstance(dirs, list)
def test_is_in_allowed_dirs_tmp(self):
"""/tmp 应该在默认允许目录内."""
self.assertTrue(is_in_allowed_dirs("/tmp/test.mp4"))
def test_is_path_safe_convenience(self):
"""is_path_safe 便捷函数应该正常工作."""
with tempfile.TemporaryDirectory() as tmpdir:
self.assertTrue(is_path_safe("test.mp4", tmpdir))
self.assertFalse(is_path_safe("../etc/passwd", tmpdir))
if __name__ == "__main__":
unittest.main()
+484
View File
@@ -0,0 +1,484 @@
"""PR #312 安全债务修复 单元测试.
测试4个P1安全修复:
1. 多轨道混音:audio_path 路径安全 + 轨道数量上限
2. 视频拼接:video_path 路径安全 + 段数上限
3. 字幕渲染:字幕文件路径白名单校验
"""
import sys
import tempfile
from pathlib import Path
sys.path.insert(0, str(Path(__file__).resolve().parents[2] / "apps" / "worker"))
sys.path.insert(0, str(Path(__file__).resolve().parents[2] / "apps" / "api"))
import pytest
from video_processing.path_security import PathSecurityError
# ── Fixtures ──────────────────────────────────────────────────────────────────
@pytest.fixture
def work_dir(tmp_path):
return tmp_path
@pytest.fixture
def sample_audio(work_dir):
"""生成一个测试音频文件."""
import subprocess
path = work_dir / "test.aac"
subprocess.run(
[
"ffmpeg",
"-y",
"-f",
"lavfi",
"-i",
"sine=frequency=440:duration=1:sample_rate=44100",
"-c:a",
"aac",
"-b:a",
"128k",
str(path),
],
capture_output=True,
check=True,
timeout=30,
)
return path
@pytest.fixture
def sample_video(work_dir):
"""生成一个测试视频文件."""
import subprocess
path = work_dir / "test.mp4"
subprocess.run(
[
"ffmpeg",
"-y",
"-f",
"lavfi",
"-i",
"testsrc=duration=1:size=320x240:rate=30",
"-f",
"lavfi",
"-i",
"sine=frequency=440:duration=1:sample_rate=44100",
"-c:v",
"libx264",
"-preset",
"ultrafast",
"-c:a",
"aac",
"-b:a",
"128k",
"-shortest",
str(path),
],
capture_output=True,
check=True,
timeout=60,
)
return path
# ══════════════════════════════════════════════════════════════════════════════
# 1. 多轨道混音安全测试
# ══════════════════════════════════════════════════════════════════════════════
class TestMultiTrackSecurity:
"""多轨道混音安全测试."""
def test_track_count_limit_exceeded(self, work_dir, sample_audio):
"""超过最大轨道数时应截断到上限."""
from unittest.mock import MagicMock, patch
from video_processing.multi_track_mixer import (
MAX_AUDIO_TRACKS,
AudioTrack,
MultiTrackMixConfig,
mix_multi_track,
)
# 创建超过上限的轨道数
tracks = []
for i in range(MAX_AUDIO_TRACKS + 5):
tracks.append(
AudioTrack(
track_id=f"track_{i}",
track_type="sfx",
audio_path=str(sample_audio),
volume=0.5,
)
)
config = MultiTrackMixConfig(tracks=tracks)
ctx = MagicMock()
ctx.work_dir = work_dir
ctx.plan_id = "test_plan"
# mock _prepare_single_track 避免实际跑ffmpeg
with patch("video_processing.multi_track_mixer._prepare_single_track", return_value=True):
with patch("video_processing.multi_track_mixer.run_ffmpeg"):
import shutil
with patch("shutil.copy2"):
result = mix_multi_track(ctx, sample_audio, config, 10.0)
# 验证轨道被截断到上限
assert len(config.tracks) == MAX_AUDIO_TRACKS
assert result is not None
def test_track_count_within_limit(self, work_dir, sample_audio):
"""轨道数在限制内时正常处理."""
from unittest.mock import MagicMock, patch
from video_processing.multi_track_mixer import (
MAX_AUDIO_TRACKS,
AudioTrack,
MultiTrackMixConfig,
mix_multi_track,
)
tracks = []
for i in range(3):
tracks.append(
AudioTrack(
track_id=f"track_{i}",
track_type="sfx",
audio_path=str(sample_audio),
volume=0.5,
)
)
config = MultiTrackMixConfig(tracks=tracks)
ctx = MagicMock()
ctx.work_dir = work_dir
ctx.plan_id = "test_plan"
with patch("video_processing.multi_track_mixer._prepare_single_track", return_value=True):
with patch("video_processing.multi_track_mixer.run_ffmpeg"):
result = mix_multi_track(ctx, sample_audio, config, 10.0)
assert len(config.tracks) == 3
assert result is not None
def test_audio_path_traversal_attack(self, work_dir, sample_audio):
"""路径遍历攻击应被拦截."""
from video_processing.multi_track_mixer import _validate_audio_path
# 路径遍历
with pytest.raises(PathSecurityError):
_validate_audio_path("../../../etc/passwd", work_dir)
# local:// 路径遍历
with pytest.raises(PathSecurityError):
_validate_audio_path("local://../../../etc/passwd", work_dir)
def test_audio_path_allowed_extension(self, work_dir, sample_audio):
"""允许的音频扩展名应通过校验."""
from video_processing.multi_track_mixer import _validate_audio_path
# 在work_dir内的音频文件
test_file = work_dir / "test.mp3"
test_file.touch()
_validate_audio_path(str(test_file), work_dir) # 不应抛异常
test_file2 = work_dir / "test.wav"
test_file2.touch()
_validate_audio_path(str(test_file2), work_dir) # 不应抛异常
def test_audio_path_disallowed_extension(self, work_dir):
"""不允许的文件扩展名应被拦截."""
from video_processing.multi_track_mixer import _validate_audio_path
test_file = work_dir / "test.exe"
test_file.touch()
with pytest.raises(PathSecurityError):
_validate_audio_path(str(test_file), work_dir)
test_file2 = work_dir / "test.php"
test_file2.touch()
with pytest.raises(PathSecurityError):
_validate_audio_path(str(test_file2), work_dir)
def test_audio_path_empty(self, work_dir):
"""空路径应被拦截."""
from video_processing.multi_track_mixer import _validate_audio_path
with pytest.raises(PathSecurityError):
_validate_audio_path("", work_dir)
with pytest.raises(PathSecurityError):
_validate_audio_path(None, work_dir)
def test_audio_path_traversal_bypass_startswith(self, work_dir):
"""【P1绕过】用../构造伪work_dir前缀路径,真实路径逃逸,必须被拦截.
漏洞:旧代码用 startswith(str(work_dir)) 比原始字符串,
/tmp/work/../../opt/secret.aac 会通过 startswith 检查,跳过白名单校验。
修复:用 realpath 规范化后再比较。
"""
from video_processing.multi_track_mixer import _validate_audio_path
evil_path = str(work_dir / "../../../../opt/secret.aac")
with pytest.raises(PathSecurityError, match="不在允许目录"):
_validate_audio_path(evil_path, work_dir)
# ══════════════════════════════════════════════════════════════════════════════
# 2. 视频拼接安全测试
# ══════════════════════════════════════════════════════════════════════════════
class TestConcatSecurity:
"""视频拼接安全测试."""
def test_segment_count_limit_exceeded(self, work_dir, sample_video):
"""超过最大段数时应报错."""
from video_processing.concat_engine import (
MAX_CONCAT_SEGMENTS,
ConcatConfig,
ConcatEngine,
ConcatSegment,
)
# 创建超过上限的段数
segments = []
for i in range(MAX_CONCAT_SEGMENTS + 5):
segments.append(ConcatSegment(video_path=str(sample_video)))
config = ConcatConfig(segments=segments)
engine = ConcatEngine(work_dir)
output_path = work_dir / "output.mp4"
with pytest.raises(ValueError, match="Too many concat segments"):
engine.concat_videos(config, output_path)
def test_segment_count_within_limit(self, work_dir, sample_video):
"""段数在限制内时正常处理."""
from unittest.mock import patch
from video_processing.concat_engine import (
MAX_CONCAT_SEGMENTS,
ConcatConfig,
ConcatEngine,
ConcatSegment,
)
segments = [
ConcatSegment(video_path=str(sample_video)),
ConcatSegment(video_path=str(sample_video)),
ConcatSegment(video_path=str(sample_video)),
]
config = ConcatConfig(segments=segments)
engine = ConcatEngine(work_dir)
output_path = work_dir / "output.mp4"
# mock ffmpeg执行
with patch.object(engine, "_concat_filter", return_value=output_path):
with patch.object(engine, "_can_use_stream_copy", return_value=False):
result = engine.concat_videos(config, output_path)
assert result == output_path
def test_video_path_traversal_attack(self, work_dir):
"""路径遍历攻击应被拦截."""
from video_processing.concat_engine import _validate_video_path
with pytest.raises(PathSecurityError):
_validate_video_path("../../../etc/passwd", work_dir)
with pytest.raises(PathSecurityError):
_validate_video_path("local://../../../etc/passwd", work_dir)
def test_video_path_allowed_extension(self, work_dir):
"""允许的视频扩展名应通过校验."""
from video_processing.concat_engine import _validate_video_path
for ext in [".mp4", ".mov", ".avi", ".mkv", ".webm"]:
test_file = work_dir / f"test{ext}"
test_file.touch()
_validate_video_path(str(test_file), work_dir) # 不应抛异常
def test_video_path_disallowed_extension(self, work_dir):
"""不允许的文件扩展名应被拦截."""
from video_processing.concat_engine import _validate_video_path
test_file = work_dir / "test.exe"
test_file.touch()
with pytest.raises(PathSecurityError):
_validate_video_path(str(test_file), work_dir)
test_file2 = work_dir / "test.js"
test_file2.touch()
with pytest.raises(PathSecurityError):
_validate_video_path(str(test_file2), work_dir)
def test_video_path_empty(self, work_dir):
"""空路径应被拦截."""
from video_processing.concat_engine import _validate_video_path
with pytest.raises(PathSecurityError):
_validate_video_path("", work_dir)
with pytest.raises(PathSecurityError):
_validate_video_path(None, work_dir)
def test_video_path_traversal_bypass_startswith(self, work_dir):
"""【P1绕过】视频路径../遍历绕过startswith检查,必须被拦截.
漏洞:旧代码用 startswith(str(work_dir)) 比原始字符串,
/tmp/work/../../opt/secret.mp4 会通过 startswith 检查,跳过白名单校验。
修复:用 realpath 规范化后再比较。
"""
from video_processing.concat_engine import _validate_video_path
evil_path = str(work_dir / "../../../../opt/secret.mp4")
with pytest.raises(PathSecurityError, match="不在允许目录"):
_validate_video_path(evil_path, work_dir)
def test_invalid_segments_skipped(self, work_dir, sample_video):
"""路径不安全的片段应被跳过."""
from unittest.mock import patch
from video_processing.concat_engine import (
ConcatConfig,
ConcatEngine,
ConcatSegment,
)
segments = [
ConcatSegment(video_path=str(sample_video)),
ConcatSegment(video_path="../../../etc/passwd"), # 不安全路径
ConcatSegment(video_path=str(sample_video)),
]
config = ConcatConfig(segments=segments)
engine = ConcatEngine(work_dir)
output_path = work_dir / "output.mp4"
with patch.object(engine, "_concat_filter", return_value=output_path):
with patch.object(engine, "_can_use_stream_copy", return_value=False):
result = engine.concat_videos(config, output_path)
# 验证只有2个安全片段保留
assert len(config.segments) == 2
assert result == output_path
# ══════════════════════════════════════════════════════════════════════════════
# 3. 字幕渲染安全测试
# ══════════════════════════════════════════════════════════════════════════════
class TestSubtitleSecurity:
"""字幕渲染安全测试."""
def test_subtitle_path_traversal_attack(self, work_dir):
"""路径遍历攻击应被拦截."""
from video_processing.subtitle_render_engine import _validate_subtitle_path
with pytest.raises(PathSecurityError):
_validate_subtitle_path("../../../etc/passwd", work_dir)
with pytest.raises(PathSecurityError):
_validate_subtitle_path("local://../../../etc/shadow", work_dir)
def test_subtitle_path_allowed_extension(self, work_dir):
"""允许的字幕扩展名应通过校验."""
from video_processing.subtitle_render_engine import _validate_subtitle_path
for ext in [".srt", ".ass", ".vtt", ".sub"]:
test_file = work_dir / f"test{ext}"
test_file.touch()
_validate_subtitle_path(str(test_file), work_dir) # 不应抛异常
def test_subtitle_path_disallowed_extension(self, work_dir):
"""不允许的文件扩展名应被拦截."""
from video_processing.subtitle_render_engine import _validate_subtitle_path
test_file = work_dir / "test.exe"
test_file.touch()
with pytest.raises(PathSecurityError):
_validate_subtitle_path(str(test_file), work_dir)
test_file2 = work_dir / "test.mp4"
test_file2.touch()
with pytest.raises(PathSecurityError):
_validate_subtitle_path(str(test_file2), work_dir)
def test_subtitle_remote_url_blocked(self, work_dir):
"""远程URL字幕应被拦截."""
from video_processing.subtitle_render_engine import _validate_subtitle_path
with pytest.raises(PathSecurityError, match="远程URL"):
_validate_subtitle_path("http://evil.com/evil.ass", work_dir)
with pytest.raises(PathSecurityError, match="远程URL"):
_validate_subtitle_path("https://evil.com/evil.srt", work_dir)
def test_subtitle_path_empty(self, work_dir):
"""空路径应被拦截."""
from video_processing.subtitle_render_engine import _validate_subtitle_path
with pytest.raises(PathSecurityError):
_validate_subtitle_path("", work_dir)
with pytest.raises(PathSecurityError):
_validate_subtitle_path(None, work_dir)
def test_build_filter_with_safe_path(self, work_dir):
"""安全路径应正常生成滤镜字符串."""
from video_processing.subtitle_render_engine import build_subtitle_filter
ass_file = work_dir / "subtitle.ass"
ass_file.write_text("test", encoding="utf-8")
result = build_subtitle_filter(ass_file, work_dir=work_dir)
assert "subtitles=" in result
assert "subtitle.ass" in result
assert "[subtitled]" in result
def test_build_filter_with_unsafe_path_raises(self, work_dir):
"""不安全路径应抛出异常."""
from video_processing.subtitle_render_engine import build_subtitle_filter
with pytest.raises(PathSecurityError):
build_subtitle_filter("../../../etc/passwd", work_dir=work_dir)
def test_build_filter_work_dir_required(self, work_dir):
"""不传work_dir时必须报错(防止自证清白绕过)."""
from video_processing.subtitle_render_engine import build_subtitle_filter
ass_file = work_dir / "sub.ass"
ass_file.write_text("test", encoding="utf-8")
# 不传 work_dir 必须报错
with pytest.raises(PathSecurityError, match="work_dir"):
build_subtitle_filter(ass_file) # type: ignore[call-arg]
# 传 None 也必须报错
with pytest.raises(PathSecurityError, match="work_dir"):
build_subtitle_filter(ass_file, work_dir=None) # type: ignore[arg-type]
# 传空字符串也必须报错
with pytest.raises(PathSecurityError, match="work_dir"):
build_subtitle_filter(ass_file, work_dir="")
def test_subtitle_path_traversal_bypass_startswith(self, work_dir):
"""【P1绕过】字幕路径../遍历绕过startswith检查,必须被拦截."""
from video_processing.subtitle_render_engine import _validate_subtitle_path
evil_path = str(work_dir / "../../../../opt/secret.srt")
with pytest.raises(PathSecurityError, match="不在允许目录"):
_validate_subtitle_path(evil_path, work_dir)
+28 -41
View File
@@ -6,6 +6,7 @@
from __future__ import annotations
import unittest
from datetime import datetime, timezone
from unittest.mock import MagicMock, patch
@@ -67,13 +68,10 @@ def _make_workflow(
class TestTransferAudioToOSS:
"""测试 _transfer_audio_to_oss 方法。"""
@patch("packages.application.tts_job.workflow.httpx")
def test_success_download_and_upload(self, mock_httpx: MagicMock) -> None:
@patch("packages.application.tts_job.workflow.safe_download_bytes")
def test_success_download_and_upload(self, mock_download: MagicMock) -> None:
"""成功下载音频并上传到 OSS,返回永久 URL 和 storage_key。"""
mock_resp = MagicMock()
mock_resp.content = b"fake audio data"
mock_resp.raise_for_status.return_value = None
mock_httpx.get.return_value = mock_resp
mock_download.return_value = b"fake audio data"
storage = MagicMock()
storage.upload_file.return_value = "https://oss.example.com/tts-outputs/user_001/job_123.mp3"
@@ -89,20 +87,21 @@ class TestTransferAudioToOSS:
assert url == "https://oss.example.com/tts-outputs/user_001/job_123.mp3"
assert key == "tts-outputs/user_001/job_123.mp3"
mock_httpx.get.assert_called_once_with(
mock_download.assert_called_once_with(
"https://cosyvoice-temp.com/audio.mp3",
purpose="tts_audio_download",
allowed_mime_types=unittest.mock.ANY,
timeout=60.0,
follow_redirects=True,
)
storage.upload_file.assert_called_once()
call_args = storage.upload_file.call_args
assert call_args[0][1] == "tts-outputs/user_001/job_123.mp3"
assert call_args[1]["content_type"] == "audio/mpeg"
@patch("packages.application.tts_job.workflow.httpx")
def test_download_failure_fallback(self, mock_httpx: MagicMock) -> None:
@patch("packages.application.tts_job.workflow.safe_download_bytes")
def test_download_failure_fallback(self, mock_download: MagicMock) -> None:
"""下载失败时回退到原始临时 URL,storage_key 为空。"""
mock_httpx.get.side_effect = Exception("Network error")
mock_download.side_effect = Exception("Network error")
workflow = _make_workflow()
url, key = workflow._transfer_audio_to_oss(
@@ -114,13 +113,10 @@ class TestTransferAudioToOSS:
assert url == "https://cosyvoice-temp.com/audio.mp3"
assert key == ""
@patch("packages.application.tts_job.workflow.httpx")
def test_upload_failure_fallback(self, mock_httpx: MagicMock) -> None:
@patch("packages.application.tts_job.workflow.safe_download_bytes")
def test_upload_failure_fallback(self, mock_download: MagicMock) -> None:
"""上传 OSS 失败时回退到原始临时 URL。"""
mock_resp = MagicMock()
mock_resp.content = b"fake audio data"
mock_resp.raise_for_status.return_value = None
mock_httpx.get.return_value = mock_resp
mock_download.return_value = b"fake audio data"
storage = MagicMock()
storage.upload_file.side_effect = Exception("OSS bucket error")
@@ -135,13 +131,10 @@ class TestTransferAudioToOSS:
assert url == "https://cosyvoice-temp.com/audio.mp3"
assert key == ""
@patch("packages.application.tts_job.workflow.httpx")
def test_wav_content_type(self, mock_httpx: MagicMock) -> None:
@patch("packages.application.tts_job.workflow.safe_download_bytes")
def test_wav_content_type(self, mock_download: MagicMock) -> None:
"""wav 格式使用正确的 content_type。"""
mock_resp = MagicMock()
mock_resp.content = b"fake wav data"
mock_resp.raise_for_status.return_value = None
mock_httpx.get.return_value = mock_resp
mock_download.return_value = b"fake wav data"
storage = MagicMock()
storage.upload_file.return_value = "https://oss.example.com/audio.wav"
@@ -162,13 +155,10 @@ class TestTransferAudioToOSS:
class TestProcessSynthesisResultWithOSS:
"""测试 process_synthesis_result 集成 OSS 转存。"""
@patch("packages.application.tts_job.workflow.httpx")
def test_stores_permanent_url_and_key(self, mock_httpx: MagicMock) -> None:
@patch("packages.application.tts_job.workflow.safe_download_bytes")
def test_stores_permanent_url_and_key(self, mock_download: MagicMock) -> None:
"""合成结果存 OSS 永久 URL 和 storage_key。"""
mock_resp = MagicMock()
mock_resp.content = b"audio bytes"
mock_resp.raise_for_status.return_value = None
mock_httpx.get.return_value = mock_resp
mock_download.return_value = b"audio bytes"
storage = MagicMock()
storage.upload_file.return_value = "https://oss.example.com/tts-outputs/user_001/test_job_001.mp3"
@@ -192,10 +182,10 @@ class TestProcessSynthesisResultWithOSS:
assert result.duration == 5.0
assert result.file_size == 50000
@patch("packages.application.tts_job.workflow.httpx")
def test_fallback_to_temp_url_on_oss_failure(self, mock_httpx: MagicMock) -> None:
@patch("packages.application.tts_job.workflow.safe_download_bytes")
def test_fallback_to_temp_url_on_oss_failure(self, mock_download: MagicMock) -> None:
"""OSS 转存失败时,使用 CosyVoice 临时 URL(不阻塞合成流程)。"""
mock_httpx.get.side_effect = Exception("Download failed")
mock_download.side_effect = Exception("Download failed")
repo = MagicMock()
job = _make_job(status=TTSJobStatus.PROCESSING)
@@ -216,13 +206,10 @@ class TestProcessSynthesisResultWithOSS:
class TestStartSynthesisSyncWithOSS:
"""测试 start_synthesis 同步路径的 OSS 转存。"""
@patch("packages.application.tts_job.workflow.httpx")
def test_sync_path_transfers_to_oss(self, mock_httpx: MagicMock) -> None:
@patch("packages.application.tts_job.workflow.safe_download_bytes")
def test_sync_path_transfers_to_oss(self, mock_download: MagicMock) -> None:
"""CosyVoice 同步返回 audio_url 时,也走 OSS 转存。"""
mock_resp = MagicMock()
mock_resp.content = b"sync audio bytes"
mock_resp.raise_for_status.return_value = None
mock_httpx.get.return_value = mock_resp
mock_download.return_value = b"sync audio bytes"
storage = MagicMock()
storage.upload_file.return_value = "https://oss.example.com/tts-outputs/user_001/test_job_001.mp3"
@@ -248,10 +235,10 @@ class TestStartSynthesisSyncWithOSS:
assert job.output_audio_key == "tts-outputs/user_001/test_job_001.mp3"
assert job.duration == 2.0
@patch("packages.application.tts_job.workflow.httpx")
def test_sync_path_oss_failure_stores_temp_url(self, mock_httpx: MagicMock) -> None:
@patch("packages.application.tts_job.workflow.safe_download_bytes")
def test_sync_path_oss_failure_stores_temp_url(self, mock_download: MagicMock) -> None:
"""同步路径 OSS 失败时,降级存储临时 URL。"""
mock_httpx.get.side_effect = Exception("Network error")
mock_download.side_effect = Exception("Network error")
service = MagicMock(spec=CosyVoiceService)
service.submit_synthesize_task.return_value = {
+11 -23
View File
@@ -233,14 +233,11 @@ class TestStartSegmentSynthesis:
# 短文本走普通路径,不调用分段
assert job.status == TTSJobStatus.PROCESSING
@patch("packages.application.tts_job.workflow.httpx")
def test_long_text_sync_segments(self, mock_httpx: MagicMock) -> None:
@patch("packages.application.tts_job.workflow.safe_download_file")
def test_long_text_sync_segments(self, mock_download: MagicMock) -> None:
"""长文本同步分段:所有段立即返回 audio_url,直接合并。"""
# Mock 分段音频下载
mock_resp = MagicMock()
mock_resp.content = b"segment audio"
mock_resp.raise_for_status.return_value = None
mock_httpx.get.return_value = mock_resp
mock_download.return_value = 1024 # 模拟文件大小
service = MagicMock(spec=CosyVoiceService)
# 每个分段都同步返回 audio_url
@@ -355,14 +352,11 @@ class TestHandleSegmentFailure:
class TestPollSegmentTasks:
"""测试 _poll_segment_tasks 分段缺失重新合成(适配同步接口)。"""
@patch("packages.application.tts_job.workflow.httpx")
def test_all_segments_done(self, mock_httpx: MagicMock) -> None:
@patch("packages.application.tts_job.workflow.safe_download_file")
def test_all_segments_done(self, mock_download: MagicMock) -> None:
"""所有分段缺少 audio_url 时重新同步合成,合并后标记完成。"""
# Mock 下载分段音频
mock_resp = MagicMock()
mock_resp.content = b"seg audio"
mock_resp.raise_for_status.return_value = None
mock_httpx.get.return_value = mock_resp
mock_download.return_value = 1024 # 模拟文件大小
service = MagicMock(spec=CosyVoiceService)
service.submit_synthesize_task.side_effect = [
@@ -422,13 +416,10 @@ class TestPollSegmentTasks:
assert result.status == TTSJobStatus.FAILED
@patch("packages.application.tts_job.workflow.httpx")
def test_partial_audio_urls_reuse_existing(self, mock_httpx: MagicMock) -> None:
@patch("packages.application.tts_job.workflow.safe_download_file")
def test_partial_audio_urls_reuse_existing(self, mock_download: MagicMock) -> None:
"""部分分段已有 audio_url 时直接复用,缺失的重新合成。"""
mock_resp = MagicMock()
mock_resp.content = b"seg audio"
mock_resp.raise_for_status.return_value = None
mock_httpx.get.return_value = mock_resp
mock_download.return_value = 1024 # 模拟文件大小
service = MagicMock(spec=CosyVoiceService)
# 只有 1 个分段需要重新合成
@@ -515,11 +506,8 @@ class TestPollAndProcessSynthesisSegmentDetection:
workflow = _make_workflow(cosyvoice_service=service, repo=repo, storage=storage)
with patch("packages.application.tts_job.workflow.httpx") as mock_httpx:
mock_resp = MagicMock()
mock_resp.content = b"audio"
mock_resp.raise_for_status.return_value = None
mock_httpx.get.return_value = mock_resp
with patch("packages.application.tts_job.workflow.safe_download_file") as mock_download:
mock_download.return_value = 1024 # 模拟文件大小
result = workflow.poll_and_process_synthesis("test_job_seg")
Regular → Executable
+4 -7
View File
@@ -221,17 +221,14 @@ class TestTTSStreamingService:
@pytest.mark.asyncio
async def test_download_audio(self):
"""下载音频数据。"""
"""下载音频数据SSRF防护走safe_download_bytesmock掉安全层)"""
cosyvoice = MagicMock(spec=CosyVoiceService)
service = TTSStreamingService(cosyvoice)
with patch("packages.application.tts_job.streaming_service.httpx") as mock_httpx:
mock_resp = MagicMock()
mock_resp.content = b"audio data"
mock_resp.raise_for_status.return_value = None
mock_httpx.get.return_value = mock_resp
with patch("packages.application.tts_job.streaming_service.safe_download_bytes") as mock_download:
mock_download.return_value = b"audio data"
result = service._download_audio("https://example.com/audio.mp3")
assert result == b"audio data"
mock_httpx.get.assert_called_once()
mock_download.assert_called_once()
+296
View File
@@ -0,0 +1,296 @@
"""URL 安全校验工具单元测试 — SSRF 防护."""
from __future__ import annotations
import os
import sys
import unittest
sys.path.insert(0, os.path.join(os.path.dirname(__file__), "..", "..", "apps", "worker"))
import shutil
import tempfile
from video_processing.url_security import ( # noqa: E402
ALLOWED_AUDIO_MIME_TYPES,
UrlSecurityError,
is_url_safe,
safe_download_bytes,
safe_download_file,
validate_url_safety,
)
class TestUrlSecurityValidation(unittest.TestCase):
"""URL 安全校验测试."""
# ── Scheme 白名单 ──────────────────────────────────────────────────────
def test_http_scheme_allowed(self):
"""HTTP scheme 应该被允许."""
result = validate_url_safety("http://example.com/test", purpose="test")
self.assertEqual(result, "http://example.com/test")
def test_https_scheme_allowed(self):
"""HTTPS scheme 应该被允许."""
result = validate_url_safety("https://example.com/test", purpose="test")
self.assertEqual(result, "https://example.com/test")
def test_file_scheme_rejected(self):
"""file:// scheme 应该被拒绝."""
with self.assertRaises(UrlSecurityError):
validate_url_safety("file:///etc/passwd", purpose="test")
def test_ftp_scheme_rejected(self):
"""ftp:// scheme 应该被拒绝."""
with self.assertRaises(UrlSecurityError):
validate_url_safety("ftp://example.com/test", purpose="test")
def test_empty_scheme_rejected(self):
"""空 scheme 应该被拒绝."""
with self.assertRaises(UrlSecurityError):
validate_url_safety("example.com/test", purpose="test")
# ── 端口白名单 ────────────────────────────────────────────────────────
def test_port_80_allowed(self):
"""端口 80 应该被允许."""
# 80端口是默认HTTP端口,不显式指定也可以
result = validate_url_safety("http://example.com:80/test", purpose="test")
self.assertIn("example.com", result)
def test_port_443_allowed(self):
"""端口 443 应该被允许."""
result = validate_url_safety("https://example.com:443/test", purpose="test")
self.assertIn("example.com", result)
def test_port_8080_rejected(self):
"""非标准端口 8080 应该被拒绝."""
with self.assertRaises(UrlSecurityError):
validate_url_safety("http://example.com:8080/test", purpose="test")
def test_port_22_rejected(self):
"""SSH 端口 22 应该被拒绝."""
with self.assertRaises(UrlSecurityError):
validate_url_safety("http://example.com:22/test", purpose="test")
# ── SSRF: 直接 IP 访问 ───────────────────────────────────────────────
def test_loopback_ip_rejected(self):
"""回环地址 127.0.0.1 应该被拒绝."""
with self.assertRaises(UrlSecurityError):
validate_url_safety("http://127.0.0.1/test", purpose="test")
def test_private_ip_192_rejected(self):
"""内网地址 192.168.x.x 应该被拒绝."""
with self.assertRaises(UrlSecurityError):
validate_url_safety("http://192.168.1.1/test", purpose="test")
def test_private_ip_10_rejected(self):
"""内网地址 10.x.x.x 应该被拒绝."""
with self.assertRaises(UrlSecurityError):
validate_url_safety("http://10.0.0.1/test", purpose="test")
def test_private_ip_172_rejected(self):
"""内网地址 172.16.x.x 应该被拒绝."""
with self.assertRaises(UrlSecurityError):
validate_url_safety("http://172.16.0.1/test", purpose="test")
def test_unspecified_ip_rejected(self):
"""未指定地址 0.0.0.0 应该被拒绝."""
with self.assertRaises(UrlSecurityError):
validate_url_safety("http://0.0.0.0/test", purpose="test")
def test_ipv6_loopback_rejected(self):
"""IPv6 回环 ::1 应该被拒绝."""
with self.assertRaises(UrlSecurityError):
validate_url_safety("http://[::1]/test", purpose="test")
def test_ipv6_link_local_rejected(self):
"""IPv6 链路本地地址应该被拒绝."""
with self.assertRaises(UrlSecurityError):
validate_url_safety("http://[fe80::1]/test", purpose="test")
# ── SSRF: 内网主机名 ─────────────────────────────────────────────────
def test_localhost_rejected(self):
"""localhost 应该被拒绝."""
with self.assertRaises(UrlSecurityError):
validate_url_safety("http://localhost/test", purpose="test")
def test_local_domain_rejected(self):
""".local 域名应该被拒绝."""
with self.assertRaises(UrlSecurityError):
validate_url_safety("http://printer.local/test", purpose="test")
def test_internal_domain_rejected(self):
""".internal 域名应该被拒绝."""
with self.assertRaises(UrlSecurityError):
validate_url_safety("http://db.internal/test", purpose="test")
# ── URL 格式校验 ─────────────────────────────────────────────────────
def test_empty_url_rejected(self):
"""空 URL 应该被拒绝."""
with self.assertRaises(UrlSecurityError):
validate_url_safety("", purpose="test")
def test_none_url_rejected(self):
"""None URL 应该被拒绝."""
with self.assertRaises(UrlSecurityError):
validate_url_safety(None, purpose="test") # type: ignore
def test_url_too_long_rejected(self):
"""超长 URL 应该被拒绝."""
long_url = "https://example.com/" + "a" * 3000
with self.assertRaises(UrlSecurityError):
validate_url_safety(long_url, purpose="test")
def test_no_hostname_rejected(self):
"""缺少主机名应该被拒绝."""
with self.assertRaises(UrlSecurityError):
validate_url_safety("http:///test", purpose="test")
# ── is_url_safe 便捷函数 ─────────────────────────────────────────────
def test_is_url_safe_true(self):
"""安全 URL 应该返回 True."""
self.assertTrue(is_url_safe("https://example.com/test", purpose="test"))
def test_is_url_safe_false(self):
"""不安全 URL 应该返回 False."""
self.assertFalse(is_url_safe("http://127.0.0.1/test", purpose="test"))
def test_is_url_safe_empty(self):
"""空 URL 应该返回 False."""
self.assertFalse(is_url_safe("", purpose="test"))
if __name__ == "__main__":
unittest.main()
class TestSafeDownload(unittest.TestCase):
"""安全下载函数测试."""
def setUp(self):
self.temp_dir = tempfile.mkdtemp()
def tearDown(self):
shutil.rmtree(self.temp_dir, ignore_errors=True)
def test_safe_download_file_rejects_ssrf(self):
"""SSRF 风险 URL 应该被拒绝下载."""
dest = os.path.join(self.temp_dir, "test.bin")
with self.assertRaises(UrlSecurityError):
safe_download_file("http://127.0.0.1/test", dest, purpose="test")
def test_safe_download_bytes_rejects_ssrf(self):
"""SSRF 风险 URL 应该被拒绝下载(bytes 版本)."""
with self.assertRaises(UrlSecurityError):
safe_download_bytes("http://localhost/test", purpose="test")
def test_safe_download_file_size_limit(self):
"""超过大小限制应该被拒绝."""
# 用 mock server 测试太大的 content-length
dest = os.path.join(self.temp_dir, "test.bin")
# 直接验证参数:max_size=0 时任何下载都应超限
# (这里用一个可访问的 URL 并设置极小的限制)
# 为避免依赖外部网络,这里只测试函数参数传递
with unittest.mock.patch("urllib.request.build_opener") as mock_opener:
mock_resp = unittest.mock.MagicMock()
mock_resp.headers = {"Content-Length": "1000"}
mock_resp.read.return_value = b""
mock_opener.return_value.open.return_value = mock_resp
# 设置 max_size=500content-length=1000 应被拒绝
with self.assertRaises(UrlSecurityError):
safe_download_file(
"https://example.com/test",
dest,
purpose="test",
max_size=500,
)
def test_safe_download_file_mime_rejected(self):
"""不允许的 MIME 类型应该被拒绝."""
dest = os.path.join(self.temp_dir, "test.bin")
with unittest.mock.patch("urllib.request.build_opener") as mock_opener:
mock_resp = unittest.mock.MagicMock()
mock_resp.headers = {"Content-Type": "text/html"}
mock_resp.read.return_value = b""
mock_opener.return_value.open.return_value = mock_resp
with self.assertRaises(UrlSecurityError):
safe_download_file(
"https://example.com/test.mp3",
dest,
purpose="test",
allowed_mime_types=ALLOWED_AUDIO_MIME_TYPES,
)
def test_safe_download_file_mime_allowed(self):
"""允许的 MIME 类型应该通过."""
dest = os.path.join(self.temp_dir, "test.mp3")
with unittest.mock.patch("urllib.request.build_opener") as mock_opener:
mock_resp = unittest.mock.MagicMock()
mock_resp.headers = {"Content-Type": "audio/mpeg"}
mock_resp.read.side_effect = [b"audio_data", b""]
mock_resp.geturl.return_value = "https://example.com/test.mp3"
mock_opener.return_value.open.return_value = mock_resp
size = safe_download_file(
"https://example.com/test.mp3",
dest,
purpose="test",
allowed_mime_types=ALLOWED_AUDIO_MIME_TYPES,
)
self.assertEqual(size, 10)
self.assertTrue(os.path.exists(dest))
def test_safe_download_file_stream_size_limit(self):
"""流式下载时超过大小限制应该中断."""
dest = os.path.join(self.temp_dir, "test.bin")
with unittest.mock.patch("urllib.request.build_opener") as mock_opener:
mock_resp = unittest.mock.MagicMock()
mock_resp.headers = {}
# 每次返回 100 字节,max_size=500,第 6 次读取就超限
mock_resp.read.side_effect = lambda n: b"x" * n if n < 1000 else b"x" * 100
# 改成返回固定 100 字节,直到第 N 次后返回空
call_count = [0]
def mock_read(size):
call_count[0] += 1
if call_count[0] > 10:
return b""
return b"x" * 100
mock_resp.read = mock_read
mock_opener.return_value.open.return_value = mock_resp
with self.assertRaises(UrlSecurityError):
safe_download_file(
"https://example.com/test",
dest,
purpose="test",
max_size=500, # 500 字节上限
)
def test_safe_download_bytes_returns_content(self):
"""safe_download_bytes 应该返回文件内容."""
test_data = b"hello world test audio"
with unittest.mock.patch("urllib.request.build_opener") as mock_opener:
mock_resp = unittest.mock.MagicMock()
mock_resp.headers = {"Content-Type": "audio/mpeg"}
call_count = [0]
def mock_read(size):
call_count[0] += 1
if call_count[0] > 1:
return b""
return test_data
mock_resp.read = mock_read
mock_opener.return_value.open.return_value = mock_resp
result = safe_download_bytes(
"https://example.com/test.mp3",
purpose="test",
allowed_mime_types=ALLOWED_AUDIO_MIME_TYPES,
)
self.assertEqual(result, test_data)