统一音视频时长展示元数据
删除 EditorAsset、CanvasLayer 与画布布局中的通用 durationSeconds 链路。 将上传及生成请求时长统一写入 generationInputs.fields，并由服务端去重补齐。 音频播放只使用 loadedmetadata，素材详情、导出和精选成本读取展示字段。 补充前后端回归测试并同步架构文档与共享决策。
This commit is contained in:
@@ -5809,7 +5809,7 @@
|
||||
|
||||
- 决策：`assetKind` 是编辑器资源和素材唯一权威媒体类别；不新增或返回 `mediaType`。`character-animation` 渲染为序列帧，`video` 渲染为视频，`audio/sound-effect/background-music` 渲染为音频，其余类别渲染为图片。前端内部可保留派生的 `CanvasMediaType` 选择渲染器，但不能把它作为后端事实。
|
||||
- 持久化边界：`editor_project_resource` 与 `editor_asset` 表尾只保存 `image_sequence_frames_json` 与 `image_sequence_duration_ms`，默认均为 `None`；`editor_showcase_asset` 作为提交时冻结的审核与公开快照，同样在表尾保存这两个字段并从账号素材逐字段复制，旧行默认均为 `None`。正式帧对象和角色动作生成响应都不保存或返回 `frameIndex`，数组位置是唯一播放顺序；帧数取数组长度，FPS 由帧数和毫秒时长即时推导，不持久化 `frame_count`、`fps` 或通用 `duration_seconds`。后端处理抽帧和逐帧去背时仍保留内部 `frame_index`，仅用于乱序并发收口、OSS 命名、日志和错误定位；仓库内旧 helper 同步升级，不为它保留外部兼容字段。
|
||||
- 时长口径：`image_sequence_duration_ms` 只表示角色图片序列完整播放一次的毫秒时长，与音频 / 视频生成请求中的 `durationSeconds` 完全分离。角色动作与视频生成响应仍可携带各自既有的请求 / 结果级 `frameCount/fps/durationSeconds`，音视频画布布局也可保留既有运行时秒数；不得为音频生成响应新增 `durationSeconds`，也不得把任何音视频秒数写入资源 / 素材或图片序列字段。前端播放器使用 `imageSequenceDurationMs / imageSequenceFrames.length`，Spine 导出时才换算秒数并推导 FPS。
|
||||
- 时长口径：`image_sequence_duration_ms` 只表示角色图片序列完整播放一次的毫秒时长，与音频 / 视频生成请求中的 `durationSeconds` 完全分离。角色动作与视频生成响应仍可携带各自既有的请求 / 结果级 `frameCount/fps/durationSeconds`；通用音视频秒数不进入资源 / 素材正式列、`EditorAsset`、`CanvasLayer` 或画布 layout，只允许把上传探测值或生成请求值格式化为用户可见字符串后写入 `generation_inputs_json.fields[]` 供素材详情展示。素材放置与工程恢复不做媒体探测，也不从 layout / resource 回退；音频播放只使用媒体 `loadedmetadata.duration`。不得为音频生成响应新增 `durationSeconds`，也不得把音视频秒数写入图片序列字段。前端角色动作播放器使用 `imageSequenceDurationMs / imageSequenceFrames.length`，Spine 导出时才换算秒数并推导 FPS。
|
||||
- 写入与复用边界：`assetKind=character-animation` 必须在 SpacetimeDB storage/procedure 边界同时提供至少两帧有效数组与大于 0 的图片序列毫秒时长；其他类别不得携带图片序列字段。同项目同源同媒体资源只允许 `None → Some` 单调回填，非空冲突失败关闭，延迟重试的 `updated_at` 取请求时间与既有时间的较大值。
|
||||
- 生成关联边界：角色动作生成响应同时返回已经持久化的最终 `resource` / `asset`；带项目上下文时前端结果图层必须直接使用 `resource.resourceId` 及其预览视频来源血缘，缺少 resource 直接失败。无项目放置也必须绑定响应中的正式账号素材，不能从响应 `frames` 构造无资产真相的本地动作结果。
|
||||
- repair 幂等边界：legacy 音频 repair 不再派生或回填任何资源级通用时长，因此新列上线前已完成 repair 的重放继续按原字段精确匹配；图片序列字段在音频行上均为 `None`。
|
||||
|
||||
@@ -429,7 +429,7 @@ Responses 的终态载荷既是工具调用的恢复源，也是正文的恢复
|
||||
|
||||
- `asset_kind` 是资源 / 素材唯一权威媒体类别，不新增或返回 `media_type` / `mediaType`。角色动作预览 MP4 使用 `asset_kind = video`，最终透明帧集使用 `asset_kind = character-animation`；前端据此派生具体渲染器。
|
||||
- `editor_project_resource` 与 `editor_asset` 表尾只追加 `image_sequence_frames_json: Option<String>` 和 `image_sequence_duration_ms: Option<u64>`；`editor_showcase_asset` 作为提交时冻结的审核与公开快照，也在表尾追加并从账号素材复制相同两字段。前者保存完整有效帧数组，数组位置是唯一播放顺序，正式帧对象不保存或返回 `frameIndex`；后者只表示该图片序列完整播放一次的毫秒时长。帧数始终取数组长度，FPS 在播放或导出时即时推导，不持久化 `frame_count` 或 `fps`。
|
||||
- 图片序列时长不能复用音频 / 视频生成请求的 `durationSeconds`，资源和素材也不保存通用 `duration_seconds`。角色动作与视频生成响应保留各自既有的请求 / 结果级秒数，音频 / 视频图层的既有运行时秒数可留在画布布局中；音频生成响应不得新增 `durationSeconds`，音频实际时长仍可由 `<audio>` 的 `loadedmetadata.duration` 获取，但这些值都不得映射到资源、素材或图片序列字段。
|
||||
- 图片序列时长不能复用音频 / 视频生成请求的 `durationSeconds`，资源和素材也不保存通用 `duration_seconds`。角色动作与视频生成响应保留各自既有的请求 / 结果级秒数；音频 / 视频的用户可见时长只作为字符串展示项写入 `generation_inputs_json.fields[]`，上传媒体使用本次上传探测值，带时长选项的生成任务使用用户提交值，不再复制到 `EditorAsset`、`CanvasLayer` 或画布 layout，也不在素材放置 / 工程恢复时探测或从 layout、resource 做双来源回退。音频播放控件只信任 `<audio>` 的 `loadedmetadata.duration`，展示字段不得参与播放、裁切或变速。音频生成响应不得新增 `durationSeconds`，这些展示值也不得映射到资源、素材正式列或图片序列字段。
|
||||
- 角色动作 worker 在透明帧全部持久化后创建最终项目资源与账号素材，并写入帧数组与 `image_sequence_duration_ms`；生成响应可继续返回请求 / 结果层面的 `frameCount`、`fps`、`durationSeconds`，但它们不是持久化真相。预览视频仍作为独立 `asset_kind = video` 中间素材保留并承担生成成本，最终派生素材成本为 0。
|
||||
- `create_editor_project_resource`、`create_editor_asset` 和同源资源回填必须在 procedure/storage 边界验证最终状态：`asset_kind = character-animation` 必须同时包含至少两帧的有效数组和大于 0 的 `image_sequence_duration_ms`；其他类别不得携带任一图片序列字段。同源资源只允许从 `None` 单调补齐字段，非空冲突失败关闭，补写后的 `updated_at` 不得早于既有时间。
|
||||
- `generation_inputs_json` 只保存用户生成 / 重放输入，例如 `fields`、`references`、`artSpec`；动作输出和内部背景决策色不得再写入该 JSON。背景决策审计继续走独立审计链路。
|
||||
|
||||
@@ -74,7 +74,7 @@ use crate::{
|
||||
build_editor_canvas_resource_placement_item, complete_editor_canvas_generation_with_items,
|
||||
persist_editor_generated_media_asset,
|
||||
remove_editor_generated_screen_background_with_bgfilter,
|
||||
resolve_editor_reference_object_key_for_owner,
|
||||
resolve_editor_reference_object_key_for_owner, with_editor_media_duration_generation_input,
|
||||
},
|
||||
editor_screen_background_decision::{
|
||||
EditorScreenBackgroundDecisionInput, EditorScreenBackgroundDecisionKind,
|
||||
@@ -1088,6 +1088,8 @@ pub(crate) async fn generate_editor_video_for_owner(
|
||||
})?;
|
||||
let normalized = normalize_editor_video_request_with_pricing(payload, &pricing)
|
||||
.map_err(|error| editor_video_error_response(&request_context, error))?;
|
||||
let generation_inputs =
|
||||
with_editor_media_duration_generation_input(generation_inputs, normalized.duration_seconds);
|
||||
let settings = require_editor_video_settings(&state, normalized.model.as_str())
|
||||
.map_err(|error| editor_video_error_response(&request_context, error))?;
|
||||
let http_client = build_upstream_http_client(settings.ark.request_timeout_ms)
|
||||
@@ -1171,7 +1173,6 @@ pub(crate) async fn generate_editor_video_for_owner(
|
||||
task_id: Some(task_id.clone()),
|
||||
object_key: generated.object_key.clone(),
|
||||
asset_object_id: generated.asset_object_id.clone(),
|
||||
duration_seconds: Some(normalized.duration_seconds),
|
||||
source_resource_id,
|
||||
asset_kind: Some(asset_kind.clone()),
|
||||
generation_inputs,
|
||||
|
||||
@@ -671,7 +671,6 @@ pub(crate) struct EditorCanvasGeneratedLayerInput {
|
||||
pub(crate) task_id: Option<String>,
|
||||
pub(crate) object_key: Option<String>,
|
||||
pub(crate) asset_object_id: Option<String>,
|
||||
pub(crate) duration_seconds: Option<u32>,
|
||||
pub(crate) source_resource_id: Option<String>,
|
||||
pub(crate) asset_kind: Option<String>,
|
||||
pub(crate) generation_inputs: Option<Value>,
|
||||
@@ -8104,7 +8103,6 @@ pub(crate) fn build_editor_canvas_generated_layer_item(
|
||||
"taskId": input.task_id,
|
||||
"objectKey": input.object_key,
|
||||
"assetObjectId": input.asset_object_id,
|
||||
"durationSeconds": input.duration_seconds,
|
||||
"sourceResourceId": input.source_resource_id,
|
||||
"assetKind": input.asset_kind,
|
||||
"generationInputs": input.generation_inputs,
|
||||
@@ -8535,6 +8533,41 @@ pub(crate) fn serialize_editor_generation_inputs(
|
||||
serialize_editor_asset_metadata(metadata)
|
||||
}
|
||||
|
||||
pub(crate) fn with_editor_media_duration_generation_input(
|
||||
metadata: Option<Value>,
|
||||
duration_seconds: u32,
|
||||
) -> Option<Value> {
|
||||
if duration_seconds == 0 {
|
||||
return metadata;
|
||||
}
|
||||
let mut object = match metadata {
|
||||
Some(Value::Object(object)) => object,
|
||||
None => serde_json::Map::new(),
|
||||
Some(value) => return Some(value),
|
||||
};
|
||||
let fields = object
|
||||
.entry("fields".to_string())
|
||||
.or_insert_with(|| Value::Array(Vec::new()));
|
||||
let Some(fields) = fields.as_array_mut() else {
|
||||
return Some(Value::Object(object));
|
||||
};
|
||||
fields.retain(|field| {
|
||||
field
|
||||
.get("title")
|
||||
.and_then(Value::as_str)
|
||||
.map(str::trim)
|
||||
.is_none_or(|title| title != "时长" && title != "duration")
|
||||
});
|
||||
fields.push(json!({
|
||||
"title": "时长",
|
||||
"value": format!("{duration_seconds}秒"),
|
||||
}));
|
||||
object
|
||||
.entry("references".to_string())
|
||||
.or_insert_with(|| Value::Array(Vec::new()));
|
||||
Some(Value::Object(object))
|
||||
}
|
||||
|
||||
pub(crate) fn serialize_editor_image_sequence_frames(
|
||||
frames: Option<Value>,
|
||||
) -> Result<Option<String>, AppError> {
|
||||
@@ -10160,6 +10193,68 @@ mod tests {
|
||||
}
|
||||
}
|
||||
|
||||
#[test]
|
||||
fn generated_media_canvas_item_keeps_duration_only_in_generation_inputs() {
|
||||
let generation_inputs = with_editor_media_duration_generation_input(
|
||||
Some(json!({
|
||||
"fields": [
|
||||
{ "title": "视频描述", "value": "角色挥手" },
|
||||
{ "title": "duration", "value": "4秒" }
|
||||
],
|
||||
"references": []
|
||||
})),
|
||||
5,
|
||||
);
|
||||
let completion = EditorCanvasGenerationCompletionRequest {
|
||||
dialog_id: Some("dialog-video".to_string()),
|
||||
title: "生成视频".to_string(),
|
||||
placeholder: EditorCanvasGenerationPlaceholderPayload {
|
||||
x: 120.0,
|
||||
y: 80.0,
|
||||
width: 560.0,
|
||||
height: 315.0,
|
||||
original_width: 1280.0,
|
||||
original_height: 720.0,
|
||||
},
|
||||
};
|
||||
let item = build_editor_canvas_generated_layer_item(
|
||||
&completion,
|
||||
EditorCanvasGeneratedLayerInput {
|
||||
layer_id: "layer-video".to_string(),
|
||||
resource_id: "resource-video".to_string(),
|
||||
title: "生成视频".to_string(),
|
||||
src: "/generated/video.mp4".to_string(),
|
||||
thumbnail_src: None,
|
||||
original_width: 1280,
|
||||
original_height: 720,
|
||||
display_width: 560.0,
|
||||
display_height: 315.0,
|
||||
source_type: "generated".to_string(),
|
||||
prompt: Some("角色挥手".to_string()),
|
||||
actual_prompt: None,
|
||||
model: Some("seedance2.0-fast".to_string()),
|
||||
provider: Some("VectorEngine".to_string()),
|
||||
task_id: Some("task-video".to_string()),
|
||||
object_key: Some("generated/video.mp4".to_string()),
|
||||
asset_object_id: Some("asset-object-video".to_string()),
|
||||
source_resource_id: None,
|
||||
asset_kind: Some("video".to_string()),
|
||||
generation_inputs,
|
||||
},
|
||||
);
|
||||
|
||||
assert!(item.get("durationSeconds").is_none());
|
||||
assert_eq!(
|
||||
item["generationInputs"]["fields"].as_array().map(Vec::len),
|
||||
Some(2)
|
||||
);
|
||||
assert_eq!(
|
||||
item["generationInputs"]["fields"][1]["title"],
|
||||
json!("时长")
|
||||
);
|
||||
assert_eq!(item["generationInputs"]["fields"][1]["value"], json!("5秒"));
|
||||
}
|
||||
|
||||
#[test]
|
||||
fn editor_project_revision_conflicts_map_to_http_conflict() {
|
||||
let error = map_editor_project_error(SpacetimeClientError::Procedure(
|
||||
|
||||
@@ -23,7 +23,7 @@ use crate::{
|
||||
editor_project::{
|
||||
EditorCanvasGeneratedLayerInput, PersistEditorGeneratedAssetRequest,
|
||||
build_editor_canvas_generated_layer_item, complete_editor_canvas_generation_with_items,
|
||||
persist_editor_generated_media_asset,
|
||||
persist_editor_generated_media_asset, with_editor_media_duration_generation_input,
|
||||
},
|
||||
http_error::AppError,
|
||||
request_context::RequestContext,
|
||||
@@ -220,6 +220,10 @@ pub(crate) async fn generate_editor_sound_effect_for_owner(
|
||||
.map_err(|error| error.into_response_with_context(Some(&request_context)))?;
|
||||
let normalized = normalize_editor_sound_effect_request_with_pricing(payload, &pricing)
|
||||
.map_err(|error| error.into_response_with_context(Some(&request_context)))?;
|
||||
let generation_inputs = with_editor_media_duration_generation_input(
|
||||
generation_inputs,
|
||||
u32::from(normalized.duration),
|
||||
);
|
||||
let settings = require_vector_engine_audio_settings(&state)
|
||||
.map_err(|error| error.into_response_with_context(Some(&request_context)))?;
|
||||
let http_client = platform_audio::build_vector_engine_audio_http_client(&settings)
|
||||
@@ -316,7 +320,6 @@ pub(crate) async fn generate_editor_sound_effect_for_owner(
|
||||
task_id: Some(generated.task_id.clone()),
|
||||
object_key: generated.object_key.clone(),
|
||||
asset_object_id: generated.asset_object_id.clone(),
|
||||
duration_seconds: Some(u32::from(normalized.duration)),
|
||||
source_resource_id: None,
|
||||
asset_kind: Some("sound-effect".to_string()),
|
||||
generation_inputs,
|
||||
@@ -556,7 +559,6 @@ pub(crate) async fn generate_editor_background_music_for_owner(
|
||||
task_id: Some(generated.task_id.clone()),
|
||||
object_key: generated.object_key.clone(),
|
||||
asset_object_id: generated.asset_object_id.clone(),
|
||||
duration_seconds: None,
|
||||
source_resource_id: None,
|
||||
asset_kind: Some("background-music".to_string()),
|
||||
generation_inputs,
|
||||
|
||||
@@ -1,6 +1,7 @@
|
||||
import { describe, expect, it } from 'vitest';
|
||||
|
||||
import type { EditorProjectResourceSnapshot } from '../../services/image-editor/editorProjectClient';
|
||||
import { calculateEditorVideoPrice } from '../image-editor/ImageCanvasGenerationModel';
|
||||
import { buildCreationShowcaseItems } from './creationShowcaseModel';
|
||||
|
||||
function createResource(
|
||||
@@ -544,6 +545,34 @@ describe('creationShowcaseModel', () => {
|
||||
expect(items[0]?.cost).toBe('5泥点');
|
||||
});
|
||||
|
||||
it('infers video cost from the duration field instead of legacy top-level metadata', () => {
|
||||
const items = buildCreationShowcaseItems({
|
||||
activeTab: 'all',
|
||||
projectResources: [
|
||||
createResource({
|
||||
resourceId: 'video-1',
|
||||
label: '角色宣传片',
|
||||
assetKind: 'video',
|
||||
imageSrc: '/generated-editor-videos/video.mp4',
|
||||
model: 'seedance2.0-fast',
|
||||
generationInputs: {
|
||||
fields: [
|
||||
{ title: '清晰度', value: '720p' },
|
||||
{ title: '时长', value: '6秒' },
|
||||
],
|
||||
references: [],
|
||||
durationSeconds: 4,
|
||||
},
|
||||
}),
|
||||
],
|
||||
});
|
||||
|
||||
expect(items).toHaveLength(1);
|
||||
expect(items[0]?.cost).toBe(
|
||||
`${calculateEditorVideoPrice('seedance2.0-fast', '720p', 6)}泥点`,
|
||||
);
|
||||
});
|
||||
|
||||
it('keeps character animation videos out of the marketing tab', () => {
|
||||
const items = buildCreationShowcaseItems({
|
||||
activeTab: 'marketing',
|
||||
|
||||
@@ -819,7 +819,7 @@ function inferAssetCost(asset: EditorAssetSnapshot) {
|
||||
inferVideoResolution(asset, ['480p', '720p', '1080p'], '480p') as
|
||||
'480p' | '720p' | '1080p',
|
||||
normalizeInt(
|
||||
asset.generationInputs?.durationSeconds,
|
||||
inferMediaDurationSeconds(asset),
|
||||
DEFAULT_VIDEO_DURATION_SECONDS,
|
||||
4,
|
||||
15,
|
||||
@@ -832,6 +832,15 @@ function inferAssetCost(asset: EditorAssetSnapshot) {
|
||||
return null;
|
||||
}
|
||||
|
||||
function inferMediaDurationSeconds(asset: EditorAssetSnapshot) {
|
||||
const durationField = asset.generationInputs?.fields.find((field) => {
|
||||
const title = field.title.trim().toLowerCase();
|
||||
return title === '时长' || title === 'duration';
|
||||
});
|
||||
const matchedValue = durationField?.value.match(/\d+(?:\.\d+)?/u)?.[0];
|
||||
return matchedValue ? Number(matchedValue) : undefined;
|
||||
}
|
||||
|
||||
function inferVideoResolution(
|
||||
asset: EditorAssetSnapshot,
|
||||
allowedValues: string[],
|
||||
|
||||
@@ -737,7 +737,7 @@ describe('ImageCanvasEditorModel', () => {
|
||||
});
|
||||
});
|
||||
|
||||
it('hydrates audio media type and sound effect asset kind from saved layout', () => {
|
||||
it('hydrates audio display duration only from resource generation inputs', () => {
|
||||
const hydrated = hydrateLayer(
|
||||
{
|
||||
layerId: 'layer-audio',
|
||||
@@ -753,13 +753,16 @@ describe('ImageCanvasEditorModel', () => {
|
||||
sourceType: 'generated',
|
||||
mediaType: 'audio',
|
||||
assetKind: 'sound-effect',
|
||||
durationSeconds: 3.5,
|
||||
},
|
||||
new Map([
|
||||
[
|
||||
'resource-audio',
|
||||
{
|
||||
imageSrc: '/generated-character-drafts/editor-audios/sfx.mp3',
|
||||
generationInputs: {
|
||||
fields: [{ title: '时长', value: '3.5秒' }],
|
||||
references: [],
|
||||
},
|
||||
},
|
||||
],
|
||||
]),
|
||||
@@ -770,8 +773,12 @@ describe('ImageCanvasEditorModel', () => {
|
||||
src: '/generated-character-drafts/editor-audios/sfx.mp3',
|
||||
mediaType: 'audio',
|
||||
assetKind: 'sound-effect',
|
||||
durationSeconds: 3.5,
|
||||
generationInputs: {
|
||||
fields: [{ title: '时长', value: '3.5秒' }],
|
||||
references: [],
|
||||
},
|
||||
});
|
||||
expect(hydrated).not.toHaveProperty('durationSeconds');
|
||||
});
|
||||
|
||||
it('hydrates character animation sequence fields from the project resource', () => {
|
||||
|
||||
@@ -205,7 +205,6 @@ export function createLayerFromAsset(
|
||||
objectKey: asset.objectKey,
|
||||
assetObjectId: asset.assetObjectId,
|
||||
sourceResourceId: asset.sourceResourceId,
|
||||
durationSeconds: asset.durationSeconds,
|
||||
sourceAssetId: asset.id,
|
||||
resourceAssetKind: asset.assetKind ?? assetKind ?? null,
|
||||
assetKindOverride: null,
|
||||
@@ -335,7 +334,6 @@ export function serializeLayer(layer: CanvasLayer): EditorProjectLayerSnapshot {
|
||||
taskId: layer.taskId,
|
||||
objectKey: layer.objectKey,
|
||||
assetObjectId: layer.assetObjectId,
|
||||
durationSeconds: layer.durationSeconds,
|
||||
sourceResourceId: layer.sourceResourceId,
|
||||
sourceAssetId: layer.sourceAssetId,
|
||||
}),
|
||||
@@ -875,10 +873,6 @@ export function hydrateLayer(
|
||||
? stringOrNull(resource?.assetObjectId)
|
||||
: (stringOrNull(snapshot.assetObjectId) ??
|
||||
stringOrNull(resource?.assetObjectId)),
|
||||
durationSeconds: isFormalCharacterAnimation
|
||||
? undefined
|
||||
: (audioDurationOrNull(snapshot.durationSeconds) ??
|
||||
audioDurationOrNull(resource?.durationSeconds)),
|
||||
imageSequenceDurationMs,
|
||||
sourceResourceId: isFormalCharacterAnimation
|
||||
? stringOrNull(resource?.sourceResourceId)
|
||||
@@ -1207,7 +1201,6 @@ export type CanvasLayerResourceMetadata = {
|
||||
width?: number | null;
|
||||
height?: number | null;
|
||||
sourceType?: string | null;
|
||||
durationSeconds?: number | null;
|
||||
imageSequenceDurationMs?: number | null;
|
||||
sourceResourceId?: string | null;
|
||||
assetKind?: string | null;
|
||||
|
||||
@@ -60,7 +60,6 @@ export type EditorAsset = {
|
||||
showcaseSubmitError?: string | null;
|
||||
assetKind?: CanvasAssetKind | null;
|
||||
generationInputs?: CanvasGenerationInputs | null;
|
||||
durationSeconds?: number;
|
||||
imageSequenceFrames?: EditorImageSequenceFrameResult[];
|
||||
imageSequenceDurationMs?: number;
|
||||
uploadStatus?: 'uploading' | 'failed';
|
||||
@@ -117,7 +116,6 @@ export type CanvasLayer = {
|
||||
assetKindOverride?: CanvasAssetKind | null;
|
||||
assetKind?: CanvasAssetKind | null;
|
||||
generationInputs?: CanvasGenerationInputs | null;
|
||||
durationSeconds?: number;
|
||||
imageSequenceFrames?: EditorImageSequenceFrameResult[];
|
||||
imageSequenceDurationMs?: number;
|
||||
previewVideoPath?: string | null;
|
||||
|
||||
@@ -1451,7 +1451,7 @@ describe('ImageCanvasEditorView', () => {
|
||||
fields: [
|
||||
{ title: '音效提示词', value: '金币跳出音' },
|
||||
{ title: 'model', value: 'audio1.0' },
|
||||
{ title: 'duration', value: '8秒' },
|
||||
{ title: '时长', value: '8秒' },
|
||||
],
|
||||
references: [
|
||||
{
|
||||
|
||||
@@ -194,7 +194,7 @@ describe('ImageCanvasExportModel', () => {
|
||||
},
|
||||
{ title: '用户输入', value: '用户填写的视觉要求' },
|
||||
{ title: '游戏分类', value: '游戏音效' },
|
||||
{ title: 'duration', value: '5秒' },
|
||||
{ title: '时长', value: '5秒' },
|
||||
{ title: '处理模型', value: 'birefnet' },
|
||||
],
|
||||
references: [],
|
||||
@@ -207,7 +207,7 @@ describe('ImageCanvasExportModel', () => {
|
||||
fields: [
|
||||
{ title: '用户输入', value: '用户填写的视觉要求' },
|
||||
{ title: '游戏分类', value: '游戏音效' },
|
||||
{ title: 'duration', value: '5秒' },
|
||||
{ title: '时长', value: '5秒' },
|
||||
],
|
||||
references: [],
|
||||
});
|
||||
|
||||
@@ -12,7 +12,6 @@ import {
|
||||
isEditorUserVisibleGenerationInputField,
|
||||
UI_DESIGN_ASSET_EXTRACTION_PROMPT,
|
||||
} from './ImageCanvasGenerationModel';
|
||||
import { formatCanvasDurationMetric } from './ImageCanvasMediaModel';
|
||||
|
||||
export function sanitizeExportFilePart(value: string, fallback: string) {
|
||||
const safeValue = value
|
||||
@@ -563,20 +562,12 @@ export function buildLayerVisibleExportMetadata(layer: CanvasLayer) {
|
||||
object: layer.objectKey ?? layer.assetObjectId ?? '-',
|
||||
};
|
||||
|
||||
if (layer.mediaType === 'audio') {
|
||||
return {
|
||||
...base,
|
||||
duration: formatCanvasDurationMetric(layer.durationSeconds).replace(
|
||||
/^时长\s*/u,
|
||||
'',
|
||||
),
|
||||
};
|
||||
}
|
||||
|
||||
return {
|
||||
...base,
|
||||
resolution: `${layer.originalWidth} x ${layer.originalHeight} px`,
|
||||
};
|
||||
return layer.mediaType === 'audio'
|
||||
? base
|
||||
: {
|
||||
...base,
|
||||
resolution: `${layer.originalWidth} x ${layer.originalHeight} px`,
|
||||
};
|
||||
}
|
||||
|
||||
export function buildLayerExportMetadata(
|
||||
|
||||
@@ -292,7 +292,7 @@ describe('ImageCanvasGenerationDialogModel', () => {
|
||||
fields: [
|
||||
{ title: 'prompt', value: '金币收集音' },
|
||||
{ title: 'model', value: 'audio1.0' },
|
||||
{ title: 'duration', value: '7秒' },
|
||||
{ title: '时长', value: '7秒' },
|
||||
],
|
||||
references: [],
|
||||
},
|
||||
@@ -361,7 +361,7 @@ describe('ImageCanvasGenerationDialogModel', () => {
|
||||
title: '游戏背景音乐',
|
||||
prompt: '【系统内置】完整背景音乐提示词',
|
||||
generationInputs: {
|
||||
fields: [{ title: 'duration', value: '10秒' }],
|
||||
fields: [{ title: '时长', value: '10秒' }],
|
||||
references: [],
|
||||
},
|
||||
}),
|
||||
|
||||
@@ -529,7 +529,10 @@ function resolveAudioRedrawSoundModel(
|
||||
}
|
||||
|
||||
function resolveAudioRedrawSoundDuration(sourceLayer: CanvasLayer) {
|
||||
const fieldValue = findGenerationInputFieldValue(sourceLayer, ['duration']);
|
||||
const fieldValue = findGenerationInputFieldValue(sourceLayer, [
|
||||
'时长',
|
||||
'duration',
|
||||
]);
|
||||
const matchedValue = fieldValue?.match(/\d+/u)?.[0];
|
||||
const duration = matchedValue ? Number.parseInt(matchedValue, 10) : null;
|
||||
if (!duration || !Number.isFinite(duration)) {
|
||||
|
||||
@@ -501,7 +501,10 @@ describe('ImageCanvasGenerationLayerModel', () => {
|
||||
originalHeight: 120,
|
||||
},
|
||||
generationInputs: {
|
||||
fields: [{ title: 'prompt', value: '金币掉落叮当声' }],
|
||||
fields: [
|
||||
{ title: 'prompt', value: '金币掉落叮当声' },
|
||||
{ title: '时长', value: '75秒' },
|
||||
],
|
||||
references: [],
|
||||
},
|
||||
});
|
||||
@@ -526,11 +529,18 @@ describe('ImageCanvasGenerationLayerModel', () => {
|
||||
objectKey: 'generated-character-drafts/editor-audios/sfx.mp3',
|
||||
assetObjectId: 'assetobj-audio-1',
|
||||
sourceAssetId: 'asset-audio-bff',
|
||||
durationSeconds: 75,
|
||||
generationInputs: {
|
||||
fields: [
|
||||
{ title: 'prompt', value: '金币掉落叮当声' },
|
||||
{ title: '时长', value: '75秒' },
|
||||
],
|
||||
references: [],
|
||||
},
|
||||
generatedAssetSnapshot: expect.objectContaining({
|
||||
assetId: 'asset-audio-bff',
|
||||
}),
|
||||
});
|
||||
expect(layer).not.toHaveProperty('durationSeconds');
|
||||
expect(layer.provider).toBeUndefined();
|
||||
});
|
||||
|
||||
@@ -579,8 +589,8 @@ describe('ImageCanvasGenerationLayerModel', () => {
|
||||
objectKey: 'generated-character-drafts/editor-videos/video.mp4',
|
||||
assetObjectId: 'assetobj-video-1',
|
||||
thumbnailSrc: '/generated-character-drafts/editor-videos/cover.png',
|
||||
durationSeconds: 5,
|
||||
});
|
||||
expect(layer).not.toHaveProperty('durationSeconds');
|
||||
expect(layer.generatedAssetSnapshot?.assetId).toBe('asset-video-bff');
|
||||
expect(layer.provider).toBeUndefined();
|
||||
});
|
||||
|
||||
@@ -457,7 +457,6 @@ export function createVideoResultLayer({
|
||||
sourceAssetId: asset?.assetId,
|
||||
groupId: sourceLayer?.groupId,
|
||||
generationInputs: resource?.generationInputs ?? generationInputs,
|
||||
durationSeconds: generated.durationSeconds,
|
||||
generatedAssetSnapshot: asset ?? generated.asset ?? undefined,
|
||||
};
|
||||
}
|
||||
@@ -513,7 +512,6 @@ export function createAudioResultLayer({
|
||||
objectKey: resource?.objectKey ?? generated.objectKey,
|
||||
assetObjectId: resource?.assetObjectId ?? generated.assetObjectId,
|
||||
sourceAssetId: asset?.assetId,
|
||||
durationSeconds: generated.durationSeconds ?? undefined,
|
||||
generationInputs: resource?.generationInputs ?? generationInputs,
|
||||
generatedAssetSnapshot: asset ?? undefined,
|
||||
};
|
||||
|
||||
@@ -546,7 +546,7 @@ describe('ImageCanvasGenerationModel', () => {
|
||||
fields: [
|
||||
{ title: 'prompt', value: '金币掉落叮当声' },
|
||||
{ title: 'model', value: 'audio1.0' },
|
||||
{ title: 'duration', value: '7秒' },
|
||||
{ title: '时长', value: '7秒' },
|
||||
],
|
||||
references: [],
|
||||
});
|
||||
|
||||
@@ -1031,6 +1031,7 @@ export function buildImageGenerationInputs(
|
||||
|
||||
export function buildVideoGenerationInputs(
|
||||
prompt: string,
|
||||
durationSeconds: number,
|
||||
references?: CharacterReferenceImage[],
|
||||
): CanvasGenerationInputs {
|
||||
const mediaIndexes = {
|
||||
@@ -1039,7 +1040,10 @@ export function buildVideoGenerationInputs(
|
||||
audio: 0,
|
||||
};
|
||||
return {
|
||||
fields: createGenerationInputField('视频描述', prompt),
|
||||
fields: [
|
||||
...createGenerationInputField('视频描述', prompt),
|
||||
...createGenerationInputField('时长', `${durationSeconds}秒`),
|
||||
],
|
||||
references: (references ?? []).flatMap((reference) => {
|
||||
const mediaType =
|
||||
reference.mediaType === 'video' || reference.mediaType === 'audio'
|
||||
@@ -1066,7 +1070,7 @@ export function buildSoundEffectGenerationInputs(
|
||||
fields: [
|
||||
...createGenerationInputField('prompt', prompt),
|
||||
...createGenerationInputField('model', model),
|
||||
...createGenerationInputField('duration', `${duration}秒`),
|
||||
...createGenerationInputField('时长', `${duration}秒`),
|
||||
],
|
||||
references: [],
|
||||
};
|
||||
|
||||
@@ -851,6 +851,10 @@ describe('ImageCanvasGenerationSubmissionModel', () => {
|
||||
},
|
||||
result: {
|
||||
generationInputs: {
|
||||
fields: [
|
||||
{ title: '视频描述', value: '根据参考生成角色宣传片' },
|
||||
{ title: '时长', value: '5秒' },
|
||||
],
|
||||
references: [
|
||||
{
|
||||
title: '参考图 1',
|
||||
@@ -1031,7 +1035,7 @@ describe('ImageCanvasGenerationSubmissionModel', () => {
|
||||
fields: [
|
||||
{ title: 'prompt', value: '金币掉落叮当声' },
|
||||
{ title: 'model', value: 'audio1.0' },
|
||||
{ title: 'duration', value: '7秒' },
|
||||
{ title: '时长', value: '7秒' },
|
||||
],
|
||||
references: [],
|
||||
},
|
||||
@@ -1063,7 +1067,7 @@ describe('ImageCanvasGenerationSubmissionModel', () => {
|
||||
fields: [
|
||||
{ title: 'prompt', value: '按钮确认短促音' },
|
||||
{ title: 'model', value: 'audio1.0' },
|
||||
{ title: 'duration', value: '5秒' },
|
||||
{ title: '时长', value: '5秒' },
|
||||
],
|
||||
},
|
||||
},
|
||||
|
||||
@@ -502,6 +502,7 @@ export function buildImageGenerationSubmissionPlan({
|
||||
),
|
||||
generationInputs: buildVideoGenerationInputs(
|
||||
normalizedPrompt,
|
||||
durationSeconds,
|
||||
dialog.generationReferences,
|
||||
),
|
||||
},
|
||||
|
||||
@@ -1,33 +1,3 @@
|
||||
export function normalizeCanvasDurationSeconds(value: unknown) {
|
||||
return typeof value === 'number' && Number.isFinite(value) && value > 0
|
||||
? value
|
||||
: null;
|
||||
}
|
||||
|
||||
export function formatCanvasDurationLabel(value: unknown) {
|
||||
const durationSeconds = normalizeCanvasDurationSeconds(value);
|
||||
if (durationSeconds === null) {
|
||||
return '--:--';
|
||||
}
|
||||
|
||||
const totalSeconds = Math.max(1, Math.round(durationSeconds));
|
||||
const hours = Math.floor(totalSeconds / 3600);
|
||||
const minutes = Math.floor((totalSeconds % 3600) / 60);
|
||||
const seconds = totalSeconds % 60;
|
||||
const paddedSeconds = String(seconds).padStart(2, '0');
|
||||
|
||||
if (hours > 0) {
|
||||
return `${hours}:${String(minutes).padStart(2, '0')}:${paddedSeconds}`;
|
||||
}
|
||||
|
||||
return `${minutes}:${paddedSeconds}`;
|
||||
}
|
||||
|
||||
export function formatCanvasDurationMetric(value: unknown) {
|
||||
// 中文注释：音频对象没有视觉分辨率语义，画布角标统一显示时长。
|
||||
return `时长 ${formatCanvasDurationLabel(value)}`;
|
||||
}
|
||||
|
||||
type MediaPreviewTone = 'audio' | 'video';
|
||||
|
||||
type MediaPreviewMarker = {
|
||||
|
||||
@@ -224,7 +224,7 @@ describe('ImageCanvasMetadataModalView', () => {
|
||||
expect(onClose).toHaveBeenCalledTimes(1);
|
||||
});
|
||||
|
||||
it('renders generated audio metadata with duration instead of resolution', () => {
|
||||
it('renders audio duration from generation inputs instead of layer metadata', () => {
|
||||
render(
|
||||
<ImageCanvasMetadataModalView
|
||||
layer={createLayer({
|
||||
@@ -236,9 +236,11 @@ describe('ImageCanvasMetadataModalView', () => {
|
||||
originalHeight: 120,
|
||||
model: 'suno',
|
||||
taskId: 'audio-task-75',
|
||||
durationSeconds: 75,
|
||||
generationInputs: {
|
||||
fields: [{ title: '音乐描述', value: '紧张的地下城循环音乐' }],
|
||||
fields: [
|
||||
{ title: '音乐描述', value: '紧张的地下城循环音乐' },
|
||||
{ title: '时长', value: '75秒' },
|
||||
],
|
||||
references: [],
|
||||
},
|
||||
})}
|
||||
@@ -253,7 +255,7 @@ describe('ImageCanvasMetadataModalView', () => {
|
||||
expect(within(dialog).getByText('音乐描述')).toBeTruthy();
|
||||
expect(within(dialog).getByText('紧张的地下城循环音乐')).toBeTruthy();
|
||||
expect(within(dialog).getByText('时长')).toBeTruthy();
|
||||
expect(within(dialog).getByText('1:15')).toBeTruthy();
|
||||
expect(within(dialog).getByText('75秒')).toBeTruthy();
|
||||
expect(within(dialog).queryByText('Resolution')).toBeNull();
|
||||
expect(within(dialog).queryByText('420 x 120 px')).toBeNull();
|
||||
});
|
||||
|
||||
@@ -6,7 +6,6 @@ import {
|
||||
getEditorLayerModelDisplayName,
|
||||
isEditorUserVisibleGenerationInputField,
|
||||
} from './ImageCanvasGenerationModel';
|
||||
import { formatCanvasDurationMetric } from './ImageCanvasMediaModel';
|
||||
|
||||
export type ImageCanvasMetadataModalViewProps = {
|
||||
layer: CanvasLayer | null;
|
||||
@@ -17,13 +16,6 @@ function formatReferenceType(refType: 'project-resource' | 'asset') {
|
||||
return refType === 'project-resource' ? '项目资源' : '素材';
|
||||
}
|
||||
|
||||
function formatLayerDurationForDisplay(layer: CanvasLayer) {
|
||||
return formatCanvasDurationMetric(layer.durationSeconds).replace(
|
||||
/^时长\s*/u,
|
||||
'',
|
||||
);
|
||||
}
|
||||
|
||||
export function ImageCanvasMetadataModalView({
|
||||
layer,
|
||||
onClose,
|
||||
@@ -104,19 +96,14 @@ export function ImageCanvasMetadataModalView({
|
||||
<dd>
|
||||
{getEditorLayerModelDisplayName(layer.model)}
|
||||
</dd>
|
||||
{layer.mediaType === 'audio' ? (
|
||||
<>
|
||||
<dt>时长</dt>
|
||||
<dd>{formatLayerDurationForDisplay(layer)}</dd>
|
||||
</>
|
||||
) : (
|
||||
{layer.mediaType !== 'audio' ? (
|
||||
<>
|
||||
<dt>Resolution</dt>
|
||||
<dd>
|
||||
{layer.originalWidth} x {layer.originalHeight} px
|
||||
</dd>
|
||||
</>
|
||||
)}
|
||||
) : null}
|
||||
<dt>Task</dt>
|
||||
<dd>{formatTaskIdForDisplay(layer.taskId)}</dd>
|
||||
</dl>
|
||||
|
||||
@@ -14,6 +14,7 @@ import {
|
||||
applyUploadAssetReadFailure,
|
||||
applyUploadAssetReadSuccess,
|
||||
bindUploadLayerToPersistedAsset,
|
||||
buildUploadedMediaGenerationInputs,
|
||||
createUploadCanvasLayer,
|
||||
createUploadedGenerationReference,
|
||||
createUploadingAssetPlaceholder,
|
||||
@@ -539,6 +540,7 @@ describe('ImageCanvasUploadModel', () => {
|
||||
fileName: '画布视频.mp4',
|
||||
imageSrc: '/generated/video.mp4',
|
||||
mediaType: 'video',
|
||||
durationSeconds: 4.25,
|
||||
imageDimensions: { width: 1280, height: 720 },
|
||||
canvasPoint: { x: 110, y: 120 },
|
||||
canvasSize: { width: 900, height: 640 },
|
||||
@@ -555,7 +557,19 @@ describe('ImageCanvasUploadModel', () => {
|
||||
height: 720,
|
||||
originalWidth: 1280,
|
||||
originalHeight: 720,
|
||||
generationInputs: {
|
||||
fields: [{ title: '时长', value: '4.25秒' }],
|
||||
references: [],
|
||||
},
|
||||
});
|
||||
expect(layer).not.toHaveProperty('durationSeconds');
|
||||
});
|
||||
|
||||
it('omits invalid or image durations from uploaded media metadata', () => {
|
||||
expect(buildUploadedMediaGenerationInputs('image', 4)).toBeUndefined();
|
||||
expect(
|
||||
buildUploadedMediaGenerationInputs('audio', Number.NaN),
|
||||
).toBeUndefined();
|
||||
});
|
||||
|
||||
it('uses viewport center fallback for invalid upload points', () => {
|
||||
|
||||
Some files were not shown because too many files have changed in this diff Show More
Reference in New Issue
Block a user