Commit Graph

72 Commits

Author SHA1 Message Date
menghao e4e3cab559 修复画布参考图上传删除链路 (#164)
Project CI / Repository checks (push) Successful in 3m3s
Project CI / Frontend tests (push) Successful in 3m32s
Project CI / Backend tests (push) Successful in 5m45s
Project CI / Native shell tests (push) Successful in 16m17s
主站规范生成支持多张参考图

修复普通生成参考图删除与提交状态不同步

优化触屏设备删除按钮可达性并补充回归测试

---------

Co-authored-by: kdletters <kdletters@qq.com>
Reviewed-on: http://192.168.35.82/git/GenarrativeAI/Genarrative/pulls/164
Co-authored-by: 孟豪 <mh18530625731@163.com>
Co-committed-by: 孟豪 <mh18530625731@163.com>
2026-08-21 11:55:32 +08:00
kdletters 0a6d550d73 修复双 PR 合并后的持续集成问题
Project CI / Repository checks (push) Failing after 51s
Project CI / Frontend tests (push) Successful in 3m27s
Project CI / Backend tests (push) Successful in 4m8s
Project CI / Native shell tests (push) Successful in 16m0s
补齐生成引用槽位参数并恢复 source 稳定标识\n恢复画布导出元数据格式化函数导入\n同步图标规范 GenerationInputs V2 测试契约\n修正音效提交并发测试的画布落层断言\n统一 Rust 格式以通过仓库门禁
2026-08-08 22:21:57 +08:00
kdletters a27798be77 合并生成参数恢复与改造入口修复
合入 PR 135 的 GenerationInputs V2 稳定字段与严格恢复链路
统一缺失来源时保留改造入口并在点击后复检拒绝
保留主分支 SFX V2 自动时长、Loop 与权威元数据语义
将队列引用槽位与场景配方修复移植到 PR 137 的模块结构
2026-08-08 22:03:50 +08:00
k88936 7ce7c11ca4 合并主分支并解决编辑器冲突
保留图片生成 V2 配方中的像素艺术风格断言
合并 AGC 统一端口段、文本编辑器与移动端欢迎页改动
合并 Jenkins 产物权限门禁并同步项目共享文档
2026-08-08 17:15:07 +08:00
k88936 d1562ab65f 合并主分支并适配图标精灵图编辑
Project CI / Repository checks (pull_request) Successful in 1m35s
Project CI / Frontend tests (pull_request) Successful in 3m18s
Project CI / Backend tests (pull_request) Successful in 4m14s
Project CI / Native shell tests (pull_request) Successful in 16m18s
合入统一 Lexical 文本编辑器与滚动条样式调整
保留图标规范字段优化、字符限制与提交门禁
补齐图标生成及画布集成测试的 Lexical 交互适配
2026-08-08 15:23:29 +08:00
k88936 676bd524ee 滑动条样式调整 (#149)
Project CI / Repository checks (push) Successful in 1m23s
Project CI / Frontend tests (push) Successful in 3m1s
Project CI / Backend tests (push) Successful in 3m51s
Project CI / Native shell tests (push) Successful in 13m32s
示例来自微信:
![shotmd-1786091231.jpg](/attachments/b59bae77-00ca-4669-8a54-a9786ae53c93)

before:
- firefox:
![shotmd-1786091298.jpg](/attachments/80bbd46e-551d-4b44-abe4-d3d0ecaa34dd)

- edge:
![shotmd-1786091062.jpg](/attachments/7918c523-1b5b-4439-926d-591e0029c5b9)

after:
- firefox:
![shotmd-1786110119.jpg](/attachments/31b22537-236b-4782-bc47-28ac5ecb0f9f)
![shotmd-1786110060.jpg](/attachments/b48d5343-fd07-451d-943a-55000600e248)

- edge:

![shotmd-1786110050.jpg](/attachments/cfd0d8e4-7d52-45f2-bab8-1b573b3a43b0)

- 使用 Lexical + OverlayScrollbars 实现统一可控的滑动条样式, 取代原来的textarea

Reviewed-on: http://192.168.35.82/git/GenarrativeAI/Genarrative/pulls/149
Co-authored-by: 王德宇 <kvtodev@outlook.com>
Co-committed-by: 王德宇 <kvtodev@outlook.com>
2026-08-08 15:10:16 +08:00
k88936 f8bcf7cf54 修复图标规范上传类型合同
图标规范上传在项目资源与账号素材分支统一登记为 icon-spec。

补充上传后提交图集生成及两条持久化分支回归测试。

同步编辑器设计文档、外部 API Skill 与 Python helper 的 referenceId 合同。
2026-08-07 11:15:11 +08:00
k88936 969afbaf94 合并主分支并解决图标生成模块冲突
保留图标规范与图集生成独立模块边界

融合主分支角色动画与图片序列字段
2026-08-06 17:13:41 +08:00
k88936 291724a984 合并主分支并解决画布生成冲突
同步主分支的角色动作序列、精选展示与外部编辑器契约更新
保留画布生成输入 V2 契约并合并视频、音效时长元数据
修复精选视频对数值型 V2 时长元数据的兼容
2026-08-06 17:09:39 +08:00
k88936 a03f13d0a5 Fix/动作 前端展示下载+后端数据结构+精选 的问题 (#117)
Project CI / Repository checks (push) Successful in 1m28s
Project CI / Frontend tests (push) Successful in 3m19s
Project CI / Backend tests (push) Successful in 4m3s
Project CI / Native shell tests (push) Successful in 13m58s
原问题

```
历史角色动作是一个兼容例外:
后端允许它没有 editor_project_resource 资源行,只要布局自身保存了完整图片序列帧。
```

- 把生成动作的原视频asstetkind改为 video, 只把图片序列作为action
- db asset新增字段image_sequence_frames_json image_seq_duration_ms等, (原来动作的这些数据存在generation-input中,并不合适)
修改了externaljob,把这部分数据正确填写到数据库中。
- 为避免到处fallback,做了数据库迁移, 脚本: `scripts/spacetime-normalize-editor-character-actions.mjs`   手动进行过画布+素材库中动作序列帧+原始视频 迁移的测试

- 移除preview_video_path字段,  reason:
>preview video
    → separate resource/asset with assetKind="video"
  final transparent action
    → assetKind="character-animation"
    → source_resource_id points to the preview-video resource
    → image_sequence_frames_json contains the actual playable result
>
> So preview_video_path on the final action duplicates the source video resource’s image_src.
- 移除每个frame的index字段 原因: 这个只用于后端内部处理时有一个并发请求, 每个赋一个index方便收集, 后续没有再用到,且与数组本身重复

- 清除副产品preview video的generation_input_json,  因为会影响改造功能, 迁移后预览视频不提供改造(参数), 只有序列帧动作有改造

仍存在的共性问题: #134

- 导出:
下载改为完整序列帧, 封面不再作为fallback, 部分帧读取失败时行为:仍生成 ZIP,并记录失败帧, 不变

画布放置:
before:
![shotmd-1785207959-compressed.webp](/attachments/e56454f0-49a3-4a8a-951f-c21960975f99)
after:
![shotmd-1785231365-compressed.webp](/attachments/886f07df-c25b-4ecc-acc9-160e1795a991)
精选模块: 主页展示, 审核部分UI:
![shotmd-1785500046-compressed.webp](/attachments/9ec3b679-0045-494a-9f20-134f7e73fecf)
![shotmd-1785499780-compressed.webp](/attachments/42396e63-5f26-4853-a367-0f3c15111add)
这两处序列帧们的加载设计为惰式的, 只有hover和单独preview才会全部加载

Reviewed-on: http://192.168.35.82/git/GenarrativeAI/Genarrative/pulls/117
Reviewed-by: 段舒康 <kdletters@qq.com>
Co-authored-by: 王德宇 <kvtodev@outlook.com>
Co-committed-by: 王德宇 <kvtodev@outlook.com>
2026-08-06 16:57:38 +08:00
k88936 3fdbe81bfe 重构图标规范与图集生成链路
新增图标规范独立队列任务并将参数补全纳入统一计费操作
将图标图集生成和手工拆分迁入独立图标模块并复用公共生图能力
将图标规范引用收紧为项目资源或素材 ID
同步 SpacetimeDB 引用解析契约、External OpenAPI、前端与 Agent 提交链路
补充相关回归测试和长期架构约定
2026-08-06 15:43:10 +08:00
k88936 aec9b61121 合并主分支游戏创作更新
同步 AI 游戏创作 Runtime、资源图与工作台主线改动
保留画布编辑器生成配方恢复改动
2026-08-06 11:32:00 +08:00
k88936 c932f5e8fb 合并主线并保留图标规范生成逻辑
同步游戏创作运行时、资源依赖画布与像素编辑链路更新
保留图标规范参考图解析与提示词结构调整
补齐合并后的前后端合同、文档与测试
2026-08-06 11:02:15 +08:00
lhk229 8f19964e3f 加入手动完美像素入口;像素艺术勾选时自动补提示词约束 (#124)
Project CI / Repository checks (push) Successful in 1m0s
Project CI / Backend tests (push) Successful in 3m52s
Project CI / Native shell tests (push) Successful in 13m8s
Project CI / Frontend tests (push) Successful in 2m43s
Co-authored-by: 段舒康 <kdletters@qq.com>
Reviewed-on: http://192.168.35.82/git/GenarrativeAI/Genarrative/pulls/124
Reviewed-by: 段舒康 <kdletters@qq.com>
Co-authored-by: Linghong <ink29535@proton.me>
Co-committed-by: Linghong <ink29535@proton.me>
2026-08-05 21:28:40 +08:00
k88936 0d0d4c0915 合并主分支协作 Prompt 约束更新
同步协作 Prompt 跨语义复用禁止规则
保留画布生成配方分支改动
2026-08-05 17:46:49 +08:00
k88936 1f530bc806 合并主线并解决图标规范冲突
同步主线的智能体运行时、画布编辑器与后端生成链路更新
解决图标规范面板、编辑器客户端与长期决策文档冲突
保留图标规范分支的字段合同与生成入口
2026-08-05 14:40:38 +08:00
k88936 3485bad2d3 重构:统一编辑器工作流中的 "icon-spec" 处理
- 将所有相关工作流从 "ui-spec" 迁移至 "icon-spec"。
- 优化 `playSetting` 和 `artStyle` 的输入处理逻辑。
- 更新编辑器提示词、提交模型及测试用例以适配 "icon-spec"。
- 在 UI 及后端调用中集成 `EDITOR_ICON_SPEC_PROMPT_MAX_LENGTH` 及相关校验。
- 新增 "icon-spec" 生成与微调的专用 API 路径。
- 移除旧版 "ui-spec" 相关定义及提示词构建器。
2026-08-04 20:04:07 +08:00
k88936 99d239c8bf style: 画布编辑器多行输入框样式调整 (#130)
Project CI / Repository checks (push) Successful in 50s
Project CI / Backend tests (push) Successful in 3m30s
Project CI / Frontend tests (push) Successful in 3m42s
Project CI / Native shell tests (push) Successful in 11m34s
复用了画布agent的输入框:

* 高度自适应
* 滑动条样式变为浅色和位置不再出界

before
![shotmd-1785745933-compressed.webp](/attachments/95727f6d-63e1-48f7-b806-8203cd73d1eb)
after
![shotmd-1785745957-compressed.webp](/attachments/b5af077d-d3b9-4572-9135-39e288291e24)

Reviewed-on: http://192.168.35.82/git/GenarrativeAI/Genarrative/pulls/130
Co-authored-by: 王德宇 <kvtodev@outlook.com>
Co-committed-by: 王德宇 <kvtodev@outlook.com>
2026-08-04 17:07:30 +08:00
k88936 5fd26a11ee 统一画布生成产物改造契约
新增 generationInputs V2 稳定 action、field id 与 reference id 契约
恢复图片、视频、音频、角色动作等生成面板并保留 legacy 回退
非生成裁扩、去背景和图集拆分不再继承生成配方
匿名精选响应不再返回 owner generationInputs
同步前后端测试与架构文档
2026-08-03 20:07:06 +08:00
kdletters 53f06055a4 修复画布权威快照合并回归
Project CI / Repository checks (push) Successful in 1m19s
Project CI / Frontend tests (push) Failing after 2m10s
Project CI / Native shell tests (push) Failing after 3m27s
Project CI / Backend tests (push) Successful in 3m53s
统一生成完成态与本地布局的生命周期合并
保留同项目权威刷新中的有效画布选择
同步生成集成夹具与真实后端完成契约
补齐画布持久化回归测试和项目记忆
2026-08-03 15:20:44 +08:00
lhk229 5ec7182d65 修复画布图层查询在重渲染窗口下的测试竞态
Project CI / Repository checks (pull_request) Successful in 1m0s
Project CI / Frontend tests (pull_request) Successful in 2m1s
Project CI / Native shell tests (pull_request) Successful in 2m31s
Project CI / Backend tests (pull_request) Successful in 3m41s
only exposes character animation generation 用例先 await findByAltText 取到
第一个图层,再用同步 getByAltText 取第二个。合并 web master 带入的项目封面
快照改动会在图层落位后触发一次重渲染,同步查询正好落进该窗口就取不到元素,
CI 962 因此失败。

该写法在 master 上逐字相同,属既有竞态,非本分支引入;但它挡住本 PR 的 CI,
故改为异步查询容忍这次重渲染,不改变断言语义与产品代码。

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
2026-07-29 13:23:07 +00:00
lhk229 717b94aad8 合并 web master 角色交付尺寸统一
Project CI / Frontend tests (pull_request) Failing after 5m43s
Project CI / Repository checks (pull_request) Successful in 5m48s
Project CI / Native shell tests (pull_request) Successful in 7m49s
Project CI / Backend tests (pull_request) Successful in 8m50s
接受 master 的角色带背景原图与透明图统一交付尺寸方案,像素模式不再
豁免提前归一:逻辑像素规整改在按业务像素矩阵 resize_to_fill(Lanczos
重采样 + 居中裁切)后的交付尺寸图上进行,snapper 自行还原回输入尺寸,
删除原先后置的 nearest 二次恢复及其 pixel_art_applied 开关。

后处理降级告警合并两侧机制:postprocess 失败路径改用 master 的
_with_dimension 重载并继续经 merge_editor_generation_warnings 汇总;
成功路径补回 master 的 dimension_warning,避免自动合并将其丢弃。

结构断言改为只固定 editor_postprocess_fallback_warning 共同前缀。
decision-log 保留双方新决策条目,并修订像素艺术条目中已过时的
"provider 原尺寸分析网格"表述。

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
2026-07-29 09:33:36 +00:00
k88936 6ee33dc561 fix: 首页「最近项目」缩略图与退出前视角不一致,出现偏移或裁切 (#111)
Project CI / Repository checks (push) Successful in 1m4s
Project CI / Frontend tests (push) Successful in 2m9s
Project CI / Native shell tests (push) Successful in 2m38s
Project CI / Backend tests (push) Successful in 3m23s
- 封面使用真实画布尺寸及最终 viewport.x/y/scale。
- 320×240 封面改为居中 cover 裁切,不再从左上角取景。
- 正确绘制画布背景色。

in canva:
![shotmd-1784874081.jpg](/attachments/4ac128cc-64c8-4a78-9e75-c35902049d3e)
![shotmd-1784876202.jpg](/attachments/6197259b-887e-4fd6-95b5-b722ec670a02)

in home:
![shotmd-1784874090.jpg](/attachments/241c7cd3-1eee-4f7b-96f2-4ca0268d2bee)
![shotmd-1784876211.jpg](/attachments/aa4200b6-bef0-4122-a797-ebc082b27a17)

因为原本的布局持久化可能有一些bug,改动不相关,所以我回退到最开始的一版:
- 不额外监听画布尺寸变化,给最小尺寸兜底,(大了比小了好)
- 封面不要求强一致性,生成/上传时机仍然按照原来随布局持久化

优化了封面生成:
- 改用webp格式
- 尽量利用图片缓存

最终生成封面从700ms加速到150ms以内, 加上必须的上传和确认150ms 和 布局持久化的时间相差不多,不是特别离谱的手速都能保证两者一致

---------

Co-authored-by: kdletters <kdletters@qq.com>
Reviewed-on: http://192.168.35.82/git/GenarrativeAI/Genarrative/pulls/111
Co-authored-by: 王德宇 <kvtodev@outlook.com>
Co-committed-by: 王德宇 <kvtodev@outlook.com>
2026-07-29 16:09:30 +08:00
lhk229 1ed8064d2f 统一图标图集全连通域拆分
移除前端从提示词解析素材数量的逻辑,完整提示词作为单个请求元素提交。

删除后端按描述数量截断切片的路径,自动拆分与手动拆分复用全连通域算法。

统一切片命名和尺寸、像素、数量限制,保留手动拆分按钮与接口。

更新既有测试、OpenAPI、编辑器文档和共享决策记录。
2026-07-29 04:52:29 +00:00
lhk229 732fcf0b45 新增图片生成像素艺术后处理
新增确定性像素网格规整模块并保留第三方 MIT 许可证。

为普通图片、角色和图标图集接入像素艺术选项、并发预算和失败降级。

同步前端交互、请求契约、OpenAPI、项目文档与测试。
2026-07-29 03:46:03 +00:00
menghao 47f3297e84 修复图片改造后二维清晰度降级问题 (#90)
前端按图层真实尺寸恢复图片改造比例与清晰度

后端在角色去背景后恢复生成原图像素尺寸

补充前后端二千像素尺寸回归测试

同步更新图片画布技术文档与项目决策记录

---------

Co-authored-by: MengHao <menghao@MengHaodeMacBook-Air.local>
Reviewed-on: https://git.genarrative.world/git/GenarrativeAI/Genarrative/pulls/90
Reviewed-by: 段舒康 <kdletters@qq.com>
Co-authored-by: 孟豪 <mh18530625731@163.com>
Co-committed-by: 孟豪 <mh18530625731@163.com>
2026-07-18 16:50:17 +08:00
kdletters 5468bb3212 修复画布生成器恢复串台
区分生成结果归属与派生输入关系,避免角色动作和快速编辑面板串台
补全角色动作、视频、音效、背景音乐与宣发素材的生成器恢复类型
统一清理剪切和素材删除后的关联生成状态,避免孤立占位
补充生成器恢复、宣发类型和删除清理回归测试并同步编辑器文档
2026-07-17 21:23:12 +08:00
lhk229 aecabdacfd 新抠图算法 (#85)
Co-authored-by: 段舒康 <kdletters@qq.com>
Reviewed-on: https://git.genarrative.world/git/GenarrativeAI/Genarrative/pulls/85
Co-authored-by: Linghong <ink29535@proton.me>
Co-committed-by: Linghong <ink29535@proton.me>
2026-07-16 18:33:06 +08:00
kdletters 9807295a7d 调整画布生成输入与图集快速编辑
隐藏生成面板资源名称字段并恢复提示词输入边框

恢复图标图集快速编辑并仅对单图标禁用

补充工具栏右键菜单工作流回归测试与编辑器文档
2026-07-14 23:26:35 +08:00
kdletters 0baa8cf8cc 修复退款复核与发布安全边界
补齐退款人工复核、刷新恢复、会话隔离和旧冷备导入兼容
强化真实微信退款对账开关、重复环境变量和生产发布门禁
优化公开作品资产授权查询并保持实时撤销语义
统一后台微信渠道、退款追回流水和画布正式资源测试契约
补齐生成与素材侧栏回归测试
2026-07-14 16:31:22 +08:00
kdletters 4e7bcb8f24 补齐生成中间产物持久化与拆图提示
图标和UI图集拆分失败时保留主图集并在内联、队列及刷新路径提示
角色、图片修改、图标UI和角色动作链路把可恢复中间产物写入项目与素材库
外部生成摘要新增独立告警字段并清理成功任务的历史错误
同步共享契约、SpacetimeDB迁移与绑定、OpenAPI、设计文档和回归测试
2026-07-13 21:41:46 +08:00
kdletters eae00292a1 新增画板图集自动拆分
生成图标图集后自动拆分并持久化独立素材
新增图集工具栏手动拆分入口与权威画布回填
修复连通域切片阅读顺序并补齐前后端测试
同步外部接口与编辑器设计文档
2026-07-13 19:10:30 +08:00
kdletters 754d364227 修复画布生成刷新后完成同步
任务列表识别当前画布项目的未确认完成任务并发出完成通知。

编辑器收到任务完成通知后重新加载项目快照并刷新素材库。

补充刷新后任务已完成和轮询完成的回归测试。
2026-07-07 19:45:16 +08:00
lhk229 308d440816 新抠图算法(轻量版)上线 (#73)
新抠图算法(轻量版)上线,生图背景色同步改为多颜色+自动决策

---------

Co-authored-by: kdletters <kdletters@qq.com>
Reviewed-on: https://git.genarrative.world/git/GenarrativeAI/Genarrative/pulls/73
Reviewed-by: kdletters <kdletters@qq.com>
Co-authored-by: Linghong <ink29535@proton.me>
Co-committed-by: Linghong <ink29535@proton.me>
2026-07-07 14:57:10 +08:00
kdletters d707613a8b 修复画布快速编辑回写与原图元数据
快速编辑回到原图面板流程,避免新建生成器占位

提交快速编辑时传入目标图层并在后端成功后回写画布

快速编辑生成输入记录原图引用,图片信息可展示原图来源

同步前后端测试与图片画布文档记忆
2026-07-02 23:05:48 +08:00
kdletters 53c5639620 完善画布任务与标准绿幕抠除链路
统一编辑器标准绿幕提示词和本地绿幕透明化后处理

生成标准绿幕资产前保存源图到 OSS 并下沉字节级抠绿到 platform-image

将手动去背景接入外部生成队列与画布任务侧栏

修复任务侧栏分页、滚动、计时、焦点与完成信息展示

补充右键菜单关闭和素材复制下载交互

同步编辑器相关文档和项目记忆

(cherry picked from commit a8c19e132d)
2026-06-30 17:12:13 +08:00
kdletters b3501f6fed 修复画布快速编辑还原生成器
直接点击生成图时还原原生成器对话框

非生成图直接点击只选中不打开生成器

快速编辑按钮和右键入口按素材元数据打开对应生成器

补齐资源元数据回填和入口边界测试

(cherry picked from commit 45e4b6e64a)
2026-06-30 17:12:13 +08:00
kdletters bb51c7e0d2 快速编辑直接覆盖原图
将图片快速编辑提交切到 /api/editor/images/edits,并用结果回写源图层

保留重绘作为右侧新结果图层,角色重绘继续走 character 生成契约

更新快速编辑文档和定向回归测试
2026-06-27 20:06:24 +08:00
kdletters a6671e7da7 修正图标素材只生成透明图集
后端图标素材生成只持久化扣绿后的 spritesheet 图集。

响应兼容保留 iconImageSrcs 但图标生成固定返回空数组。

前端测试改为验证只回填图集且不生成独立图标层。

同步 OpenAPI、编辑器文档和项目记忆中的图标素材生成约定。
2026-06-27 18:52:48 +08:00
kdletters e4ac57c97d 快速编辑改造接入统一画布生成器,视频改造走视频流程
快速编辑不再使用独立 quickEditPanel,改为 openPlacedCanvasGenerationDialog 创建 quick-edit 模式的画布生成器占位及作曲面板
改造操作区分媒体类型:视频层改造走 createVideoRedrawGenerationDialogDraft 自动打开视频生成面板
appendGenerationReference 扩展支持 quick-edit 模式,pickQuickEditReferenceFromLayer / updateQuickEditSelection 走统一 generateDialog 状态
去掉 appendQuickEditReference / archiveActiveCanvasGenerationDialog 调用,新增 createVideoRedrawGenerationDialogDraft
集成测试更新对话框 aria-label 为统一生成器标签
2026-06-25 16:55:41 +08:00
kdletters 0c2f743010 要求UI素材图集使用绿幕背景
后端UI设计素材提取固定提示词加入纯绿色绿幕要求

前端UI设计素材提取常量与历史字段断言同步绿幕要求

更新编辑器文档和项目记忆中的素材提取约定
2026-06-25 14:58:09 +08:00
kdletters 6bf8b0051b 完善画板生成器与素材导出体验
图片改造改为独立 quick-edit 生成器占位并补齐提交链路
生成完成应用项目快照后同步刷新素材库
序列帧导出补充 Spine JSON 与独立验证器
序列帧播放预加载帧资源以减少加载闪烁
本地 dev 默认启用 all 进程角色并更新文档
2026-06-25 00:12:21 +08:00
JenkenB 7830d201c4 Server-side billing for editor generation
Move editor generation pricing to backend: remove priceMudPoints from public request schemas and docs, require backend to compute model prices and perform pre-charge via execute_billable_asset_operation_with_cost. Update character animation and video handlers to use authenticated owner IDs and wrap generation requests in billing; remove client-side price validation. Adjust OpenAPI, project-memory docs, pricing management docs, and related tests to reflect that front-end only displays prices and must not submit priceMudPoints. Also document new user registration reward and wallet refund outbox behavior. Code and test changes applied across api-server handlers, shared contracts, and image-editor client/tests.
2026-06-22 23:11:23 +08:00
JenkenB a1677027f3 Merge branch 'codex/editor-asset-library' of https://git.genarrative.world/git/GenarrativeAI/Genarrative into codex/editor-asset-library 2026-06-22 21:27:54 +08:00
JenkenB 2007f1f9ce Update editor generation pricing defaults
Unify and lower default mud-point prices across editor generation features and update related API/schema, docs and tests. Changes set gpt-image-2 to 1K=3 / 2K=5, audio1.0 to 5, and chirp-v5 to 12, and update server default JSON, frontend models/workflows, OpenAPI schema (require aspectRatio/imageSize/priceMudPoints for UI extraction), multiple UX docs (quick-edit viewport, UI extraction panel, pricing guidance) and accompanying tests to match the new defaults and validation rules.
2026-06-22 21:10:43 +08:00
wuxiangwanzi 25421140bf 优化画布泥点入口
让画布顶部泥点余额按钮复用创作页充值和兑换码入口

抽取账户充值支付结果与确认弹窗组件供创作页和画布页共用

补充画布泥点入口回归测试和编辑器技术文档
2026-06-22 21:08:33 +08:00
kdletters 1176c71f72 角色动作素材改为序列帧
将角色动作结果图层主类型改为 image-sequence,并持久化完整帧列表与预览视频来源

在画布中增加序列帧播放器,展示透明 PNG 帧而不是预览视频

将角色动作单素材下载和画布素材导出改为序列帧 ZIP/frames 目录

更新角色动作、导出方案和项目记忆文档,补齐相关前端测试
2026-06-22 20:49:51 +08:00
JenkenB 55469a2198 UI extraction selection + local FFmpeg dev env
Require an explicit red-box selection before extracting UI-design assets: add rectangular/ellipse/brush selection UI, render red-outline marks into a marked Data URL on the client, and update the extraction prompt to "仅提取被红色框框选的素材并整理成spritesheet". Add frontend overlay/raster models and tests to support the selection flow and adjust numerous image-editor tests and views to wire up the selection lifecycle. On the dev ops side, enhance scripts/dev to detect a local %LOCALAPPDATA%/Genarrative/ffmpeg/bin installation on Windows, prepend it to PATH and inject CHARACTER_ANIMATION_FFMPEG_PATH/FFPROBE_PATH (with tests), and export the helper. Server config now prefers GENARRATIVE_*-prefixed FFmpeg env vars and includes a Rust test to validate reading the prefixed variables. Update docs to describe the new UI extraction behavior and local FFmpeg handling.
2026-06-22 19:13:37 +08:00
JenkenB 163b2fa077 Add editor generation model pricing admin UI & API
Introduce centralized model pricing for editor generations: add EditorGenerationPricing types and admin API client methods (GET/POST), backend default config and runtime override support, and server route/handler changes to expose admin editor-generation-pricing. Add an AdminEditorGenerationPricingPage component, styles, route entry, icon, and tests, plus a test for admin routes. Update docs and decision log to describe the pricing contract and management workflow. This enables managing per-model pricing (perGeneration / perSecond, flat or tiered) from the admin panel and exposes runtime pricing to the main site.
2026-06-22 16:57:29 +08:00
JenkenB c836ac604d Treat character-action generations as video & UI tweaks
Handle character-action generation results as real video assets: use backend previewVideoPath as layer src, set mediaType="video" and assetKind="character-animation", and store the first frame only as thumbnail/poster. Update docs to document the new behavior and preserve backend frame extraction/green-screen workflows. Update UI/UX: compact parameter labels (e.g. "16:9·1K"), quick-edit treats original image as an implicit last reference (hidden from the reference bar) and allows editable bottom capsules, and ensure exports preserve true media file types. Change upload toolbar icon to Upload and add tests to cover the icon and video-layer behavior; also add serialization/hydration test for video thumbnails and adjust many integration/unit tests to reflect new labels, sizes and generation semantics. Files touched include multiple docs under docs/, edited components and tests under src/components/image-editor/, and minor stylesheet updates.
2026-06-22 13:45:20 +08:00