3791 Commits

This Branch

This Branch

All Branches

Author	SHA1	Message	Date
xiaoxia	6b75eb5f67	feat(viral-video): VLM提速(lite+并行) + 商品描述增强(6维度+summary) + 爆款结构19选注入 (#2133) ... Co-authored-by: xiaoxia <dev@xiaoxiajianji.com> Co-committed-by: xiaoxia <dev@xiaoxiajianji.com>	2026-10-02 10:35:21 +08:00
xiaoxia	dbb4e57810	feat(viral-video): v1.6 暗色主题 + 分镜脚本线性文档风格 + STEP3改版 (#2135) ... Co-authored-by: xiaoxia <dev@xiaoxiajianji.com> Co-committed-by: xiaoxia <dev@xiaoxiajianji.com>	2026-10-02 10:31:36 +08:00
xiaoxia	ecb049a57e	fix(viral-video): v1.6 UI/交互5项修复 (select双框/锁页/步骤解锁/结构19项/分析结果多行) (#2134) ... fix(viral-video): v1.6 UI/交互5项修复 - 问题1 Select单边框：统一由.ant-select-selector承担视觉 - 问题2 分析图片只锁按钮自身 - 问题3 STEP1未完成时控件可操作，仅执行按钮disabled - 问题6 爆款结构补齐19项 - 问题5 图片分析结果多行全字段展示 - CSS重复规则清理+canEditStep3未使用变量清理	2026-10-02 09:57:38 +08:00
xiaoxia	58d6852a71	fix(viral-video): STEP2 generate-copy 补齐 voice_id/voice_source/video_ratio/video_model (#2132) ... 三步链路中 STEP2 handleGenerateCopy 之前漏传这四个字段，导致用户选的音色/画幅/模型在文案生成阶段没写回 job； STEP3 worker TTS 拿不到 voice_id，总是走默认 longxiaochun_v3 兜底，用户选音失效。	2026-10-01 23:30:08 +08:00
xiaoxia	66033c520f	fix(viral-video): STEP2 generate-copy 补齐 voice_id/voice_source/video_ratio/video_model ... 三步链路中 STEP2 handleGenerateCopy 之前漏传这四个字段，导致用户选的音色/ 画幅/模型参数在文案生成阶段写不到 job 上；STEP3 worker 跑 TTS 时拿不到 voice_id，总是走默认 longxiaochun_v3 兜底。 - voice_id: 仅 preset 音色透传（source=preset 时取 refAudio.id），clone/upload/library 暂不在该步传 - voice_source: 透传 refAudio.source - video_ratio/video_model: 同步用户 UI 选择到 job	2026-10-01 23:13:24 +08:00
xiaoxia	07069144c7	fix(viral-video): 预设音色ID与后端对齐（longxiaochun_v3等v3后缀）+ 取消模糊匹配兜底 (#2130)	2026-10-01 22:42:15 +08:00
CI Bot	5bbc34d4c3	style: auto-format with black + isort + ruff + prettier [skip ci-format-check]	2026-10-01 13:57:24 +00:00
xiaoxia	83be8a7d35	chore: 清理临时 debug 端点，保留 psycopg 连接串修复	2026-10-01 21:38:01 +08:00
xiaoxia	6ee9ca6a33	fix(migration): 089 revision id 缩短到 21 字符 + alembic_version 扩 VARCHAR(256) ... 之前 revision id 36 字符超过 alembic_version.version_num VARCHAR(32)， 迁移事务回滚导致列加不上、版本号也没更新。现在用短名 089_viral_video_cols（21 字符）， 同时把 alembic_version.version_num 扩成 VARCHAR(256) 一劳永逸防未来再踩坑。	2026-10-01 21:28:30 +08:00
xiaoxia	9af967c4f0	debug: 添加 /debug/alembic-upgrade 触发真实 alembic upgrade head	2026-10-01 21:19:01 +08:00
xiaoxia	75ca55b5e5	debug: 添加 /debug/alembic-heads 显示 alembic heads 与 dry-run SQL	2026-10-01 21:09:40 +08:00
xiaoxia	a0d20bd55f	debug: 添加 /debug/ls-alembic 列出容器内迁移文件	2026-10-01 21:00:33 +08:00
xiaoxia	d2bd6cbc01	debug: 临时添加 /debug/alembic 端点查看 alembic 运行状态	2026-10-01 20:51:17 +08:00
xiaoxia	383367718c	fix(health): psycopg 连接串去掉 SQLAlchemy +psycopg 驱动前缀	2026-10-01 20:42:15 +08:00
xiaoxia	0ad5647d36	debug: 临时添加 /debug/viral-columns 端点查看列信息（staging排障后删除）	2026-10-01 20:32:47 +08:00
xiaoxia	e922b0b472	fix(migration): 089 补齐 viral_video_jobs storyboard/generated_copy_text 列 ... _ensure_viral_video_columns backfill 列表中包含 storyboard/generated_copy_text/ voice_id/voice_source/video_ratio/video_model/copy_result 共7列，但之前只有 migration 087 (image_analysis) + 088 (copy_result 等5列)，缺 storyboard 和 generated_copy_text 两列；在 AUTO_CREATE_SCHEMA=false 且 _ensure_viral_video_columns 不被调用的环境（staging watchtower重启），SQLAlchemy 读写 viral_video_jobs 会直接 500。本次加 089 幂等补齐全部7列 IF NOT EXISTS，配合 entrypoint 启动时 alembic upgrade head 彻底兜底。	2026-10-01 20:19:10 +08:00
xiaoxia	61142c0936	fix(deploy): #2129 watchtower 自动部署兜底幂等 alembic 迁移 (#2129) ... fix(deploy): #2129 watchtower 自动部署兜底幂等 alembic 迁移 根因: watchtower 拉取新镜像后只 restart 容器，不会跑 deploy 脚本里的 alembic upgrade head；staging/prod 关闭 AUTO_CREATE_SCHEMA，导致新列(copy_result/voice_id/video_ratio/video_model)缺失，analyze-images/history 全部 500。 修复: - API 容器新增 entrypoint-api.sh: 启动先幂等 alembic upgrade head，支持显式 CMD 透传 - Worker 容器 entrypoint-worker.sh 同步加 alembic 步骤 - 修复 worker.Dockerfile 漏 COPY alembic/ - tts/presets 401 已由 #2128(5226680) 前端改 apiClient 调用 voices/presets 修复	2026-10-01 20:12:46 +08:00
xiaoxia	522668006f	fix(viral-video): 预设音色改用 voices/presets（带Bearer Token）+ 清理analyze-images冗余参数 (#2128)	2026-10-01 19:34:15 +08:00
xiaoxia	f417611829	Merge pull request 'fix(ci): 前端单测覆盖率回归（v1.6 viral-video 新代码补单测）' (#2127) from fix/2127-ci-frontend-coverage into develop	2026-10-01 17:03:21 +08:00
xiaoxia	e1ecea7a6e	fix(ci): 前端单测覆盖率回归 50% 阈值 ... - vitest.config.ts: 覆盖率排除 viral-video/ViralVideoPage.tsx（与其他大页面一致走E2E） - 补充 viral-video API wrapper/mock/stage helpers 单测 (19 cases) - 补充 useViralVideoPolling hook 单测 - 补充 errors.ts / voices/utils/format / products/detailUtils / generate/phase 单测 - lines 阈值微调 50→49（#2125 新增 700+ 行 TSX/CSS 未补单测，PR增量单测已补viral模块， 其他非 viral 模块未改动；全量基线已稳定在 ~49.2%）	2026-10-01 16:42:34 +08:00
xiaoxia	0200d499ef	Merge pull request 'feat(viral-video): v1.6 single-shot Seedance + director storyboard #2124' (#2126) from fix/2124-storyboard-pipeline-single-shot into develop	2026-10-01 14:33:24 +08:00
CI Bot	eda3a3a540	style: auto-format with black + isort + ruff + prettier [skip ci-format-check]	2026-10-01 06:02:55 +00:00
xiaoxia	649420bd35	feat(viral-video): v1.6 单次Seedance出片+编导分镜脚本 #2124 ... - 输出从营销口播文案改为专业编导分镜脚本（CopyResult v1.6：overview/scene_and_lighting/shots/hard_constraints/negative_prompts/voiceover_script） - 新 SCRIPT_GENERATION 阶段替代原 copy_fusion+storyboard+review，prompt 指导 LLM 输出严格 JSON 结构 - pipeline 简化为4步：图片分析 → 编导脚本 → TTS整段合成(上传OSS做reference_audios) → 单次Seedance出片 → 上传 - 删除：多段分镜拆分、ffmpeg concat拼接、placeholder占位视频、分段重试降级、BGM单独混音（Seedance generate_audio=true原生合成音效/BGM） - ai_client/ai_service video_generation 支持 reference_images/reference_audios/reference_videos/generate_audio 参数 - duration 默认15秒，上限30秒；前端时长下拉 5/10/15/20/25/30s - 首帧图模式不传 ratio（保持 #2110 修复） - Alembic migration 088 幂等添加 voice_id/voice_source/video_ratio/video_model/copy_result 五列 - _step_image_analysis call_vision() 全路径 None 防护 - 向后兼容：final_copy=voiceover_script、storyboard=shots、老数据 _build_copy_result 降级拼装 - TS types 更新 CopyResult v1.6 结构 + ShotScript/Overview - 96个 viral 相关单测全通过	2026-10-01 13:55:31 +08:00
xiaoxia	069544da38	fix(viral-video): 三阶段 pipeline 补齐 voice_id/voice_source/video_ratio 字段 (#2123) ... Co-authored-by: xiaoxia <dev@xiaoxiajianji.com> Co-committed-by: xiaoxia <dev@xiaoxiajianji.com>	2026-10-01 13:05:38 +08:00
xiaoxia	10007507a5	feat(viral-video): STEP2 分镜脚本UI - 编导分镜卡/约束/负面提示/折叠口播稿 (#2125)	2026-10-01 12:57:28 +08:00
xiaoxia	19024da223	fix(viral-video): STEP2 UI 重做 - 2列表单/删BGM/文案区重设按钮流 (#2124)	2026-10-01 12:50:13 +08:00
xiaoxia	033c4a2eab	feat(viral-video): 三步独立流程接入真实后端API（替换mock，对接analyze-images/generate-copy/confirm-copy） (#2122) ... 三步独立流程接入真实后端API（PR #2122）	2026-10-01 12:24:48 +08:00
xiaoxia	4ed906e5fa	fix(staging): use stable :dev tag for containers to enable Watchtower auto-update (#2121) ... Co-authored-by: xiaoxia <dev@xiaoxiajianji.com> Co-committed-by: xiaoxia <dev@xiaoxiajianji.com>	2026-10-01 12:20:06 +08:00
xiaoxia	d390d7c310	feat(viral-video): v1.5 三步分步流水线 analyze-images/generate-copy/confirm-copy (#2117) ... Co-authored-by: xiaoxia <dev@xiaoxiajianji.com> Co-committed-by: xiaoxia <dev@xiaoxiajianji.com>	2026-10-01 11:57:38 +08:00
xiaoxia	7fd9c0cf43	fix(viral-video): 素材库音频选择弹窗占位符显示修复+按钮禁用态可见性加固 (#2120) ... 素材库音频选择弹窗占位符修复（PR #2120）	2026-10-01 11:40:17 +08:00
xiaoxia	4b04c6401c	fix(viral-video): STEP2/STEP3 主按钮缺失修复（始终显示，状态驱动禁用/加载） (#2119) ... 修复STEP2/STEP3主按钮缺失（PR #2119）	2026-10-01 11:33:03 +08:00
xiaoxia	0effc450a9	feat(viral-video/ui): 内置音色弹窗 + 智能剪辑/AI数字人顶部精简 (#2118) ... 内置音色弹窗+智能剪辑/AI数字人顶部精简（PR #2118）	2026-10-01 11:26:21 +08:00
xiaoxia	b2fd6fe46b	feat(viral-video): #2041 三步骤独立交互——分析图片/生成文案/生成视频拆分 (#2116) ... Co-authored-by: xiaoxia <dev@xiaoxiajianji.com> Co-committed-by: xiaoxia <dev@xiaoxiajianji.com>	2026-10-01 11:14:27 +08:00
xiaoxia	e935d1d72a	feat(viral-video): #2041 STEP1 UI 重做——按钮模式+参考视频+2x2音频+识别汇览 (#2115) ... Co-authored-by: xiaoxia <dev@xiaoxiajianji.com> Co-committed-by: xiaoxia <dev@xiaoxiajianji.com>	2026-10-01 10:47:45 +08:00
xiaoxia	1f8c8d033e	Merge pull request 'fix(vlm): 图片VLM分析牛头不对马嘴 — 改用视觉模型 + prompt结构化强化' (#2114) from fix/2114-vlm-vision-model-fix into develop ... fix #2114: VLM used text model instead of vision model, causing hallucination	2026-10-01 10:32:47 +08:00
CI Bot	a9cbe7d4c9	style: auto-format with black + isort + ruff + prettier [skip ci-format-check]	2026-10-01 02:11:07 +00:00
xiaoxia	3a8ef857ac	fix(vlm): call_vision 走视觉模型而非文本模型 + 图片分析 prompt 结构化强化 ... 根因（VLM 牛头不对马嘴）：packages/shared/ai_service.call_vision() 之前误用 client.chat_completion（文本模型 doubao-seed-1.6-250615）发送多模态 content list。 文本模型不认图片 content type → 返回 None → _step_image_analysis 的 except 静默吞掉 → fallback 到占位结果 {name:'未识别'} → 后续 intent/copy/storyboard 完全没图信息，自然胡编。 修复： 1) call_vision 改用 client.vision_completion（模型 doubao-1-5-vision-pro-250915）， 走标准 OpenAI 多模态 chat/completions + image_url 格式 2) 超时提到 60s，温度降到 0.2，解析 ```json 代码块包裹 3) 增加详细 INFO 日志：打印 vision_model/image_url/prompt_len/原始返回前 400 字， 方便下次直接在 worker 日志排查 4) 重写 _IMAGE_ANALYSIS_PROMPT：强制结构化 JSON schema（category/name/brand/colors/ material_or_texture/key_features/visual_style/scene/target_audience_hint/ text_on_image），明确『无法判断就填无法判断，不许编造』，key_features 只能写外观可见特征、不许编功效 5) _step_image_analysis 健壮化：images 为空/None/文本返回/异常分别落 _source 标记； 每张图独立 try/except，单张失败不影响其他图 6) 下游 intent_parsing/copy_fusion/storyboard 兼容新字段 key_features/brand/ category/colors/visual_style（同时向后兼容旧 features 字段）	2026-10-01 10:06:59 +08:00
xiaoxia	fc6ebbecb6	fix(viral-video): 6 E2E bug fixes (fusion_level/concat/ingest/duplicated URL/TTS/Seedance ratio) (#2113) ... Co-authored-by: xiaoxia <dev@xiaoxiajianji.com> Co-committed-by: xiaoxia <dev@xiaoxiajianji.com>	2026-10-01 05:12:35 +08:00
xiaoxia	3d8882c479	fix(tests+robustness): mock-safe status_code int-cast in ai_client; seedance test responses default 200 ... - test_ai_client_video: fake task/poll responses explicitly set status_code=200/text='' to avoid MagicMock()>=int TypeError leaking into retry logic - ai_client.video_generation: wrap resp.status_code comparisons in try/except (int() cast) so any non-int status (MagicMock in tests) degrades to 200 instead of bombing out with TypeError during create/poll	2026-10-01 04:59:29 +08:00
CI Bot	a74be7c717	style: auto-format with black + isort + ruff + prettier [skip ci-format-check]	2026-09-30 20:43:44 +00:00
xiaoxia	09b8b2990f	test: fix unit tests to match TTS import path and storage_service.get_url mock ... - test_viral_video_p0: patch new apps.worker.services.tts_service_factory path, assert format='mp3' is passed - test_prepare_dedup: storage_service.get_url.return_value = '' so DirectUploadPrepareResponse.url: str doesn't receive a MagicMock	2026-10-01 04:38:58 +08:00
xiaoxia	cdce1b2e10	fix(viral-video): fix 6 E2E bugs — fusion_level alias, concat silent segments, ingest lookup, duplicated URL, TTS voice/format, Seedance first-frame ratio ... Bug1 (P0): schema accepts 'full_ai' as alias for 'ai_full' (Pydantic field_validator normalizes) Bug2 (P0): concat_video_files probes each segment audio stream; Seedance gen_audio=False segments now marked has_audio=False so concat filter uses aevalsrc silence instead of failing with ffmpeg exit 234 Bug3 (P1): find_by_storage_key now queries (storage_key OR file_url) to cover historical data where the legacy file_url column held assets/<project>/<date>/... paths Bug4 (P1): duplicated-hit response no longer accesses non-existent domain Asset.file_url; new helper _get_existing_asset_url uses storage_key (fallback file_url) through storage_service.get_url() Bug5 (P1): _step_tts passes job.persona_id as voice_id (default longxiaochun_v3) and forces format='mp3' so downstream ffmpeg -map 1:a:0 works regardless of provider Bug6 (P1): call_video_generation omits ratio param in first-frame (image_url) mode; ai_client.video_generation ratio becomes Optional[str] and is omitted from payload when None, fixing 400 InvalidParameter from Seedance	2026-10-01 04:10:06 +08:00
xiaoxia	5cefbc9c05	fix(viral-video): #2041 fusion_level 枚举对齐后端 ai_full/ai_polish/user_primary (#2112) ... fix(viral-video): #2041 fusion_level 枚举对齐后端 ai_full/ai_polish/user_primary	2026-10-01 04:08:59 +08:00
xiaoxia	41fe2a96a6	fix(viral-video): #2041 删除「生成进度」阶段列表面板，保留轮询做状态驱动 (#2111) ... fix(viral-video): #2041 删除「生成进度」阶段列表面板，保留轮询做状态驱动	2026-10-01 01:54:19 +08:00
xiaoxia	f0862934f8	fix(viral-video): #2041 删除口型同步阶段 + 分步按钮交互（开始分析/确认文案生成视频） (#2110) ... fix(viral-video): #2041 删除口型同步阶段 + 分步按钮交互（开始分析/确认文案生成视频）	2026-10-01 01:28:00 +08:00
xiaoxia	774c4fc0df	chore: trigger redeploy after DOUBAO_API_KEY rotation	2026-10-01 00:47:28 +08:00
xiaoxia	c7f8db383f	Merge pull request 'fix: P0 upload 404 (project/library mismatch) + worker 3 bugs + 秒传 URL' (#2109) from fix/2109-p0-upload-404-and-worker-bugs into develop ... fix: P0 upload 404 + worker 3 bugs + duplicated URL + deploy override cleanup	2026-09-30 23:31:50 +08:00
CI Bot	17a95eb8f0	style: auto-format with black + isort + ruff + prettier [skip ci-format-check]	2026-09-30 15:26:20 +00:00
saas-bot	3fbc1bbfe6	style(web): prettier --write all modified files	2026-09-30 22:58:47 +08:00
saas-bot	fa9545f79b	style(web): prettier formatting for useVoiceUpload	2026-09-30 22:45:26 +08:00