auto-save 2026-05-13 21:24 (~6)

api/main.py

@@ -115,6 +115,11 @@ class GeneratedVideo(BaseModel):
     created_at: float = 0.0
 
 
+class VideoSourceRef(BaseModel):
+    kind: Literal["image", "source_video"] = "image"
+    url: str = ""
+
+
 class StoryboardScene(BaseModel):
     """分镜头编排：每个 selected 分镜对应一个 scene 描述
     v2: 4 图槽 + 时长（复制粘贴模式）— 主体 / 场景 / 产品 / 动作 各一张图
@@ -265,6 +270,9 @@ def storyboard_ref_path(job_id: str, ref: dict | None) -> Path | None:
     except Exception:
         return None
     if kind == "keyframe":
+        clean = job_dir(job_id) / "cleaned" / f"{frame_idx:03d}.jpg"
+        if clean.exists():
+            return clean
         p = job_dir(job_id) / "frames" / f"{frame_idx:03d}.jpg"
         return p if p.exists() else None
     if kind == "cutout":
@@ -1643,6 +1651,7 @@ class GenerateStoryboardVideoReq(BaseModel):
     scene_image: dict | None = None
     product_image: dict | None = None
     action_image: dict | None = None
+    source_ref: VideoSourceRef | None = None
     model: str = ""
     size: str = "720x1280"
 
@@ -1749,18 +1758,27 @@ def ark_reference_data_url(ref_img: Path) -> str:
     return f"data:{mime};base64,{base64.b64encode(ref_img.read_bytes()).decode('ascii')}"
 
 
-def submit_video_create(client, url: str, headers: dict, ref_img: Path, payload: dict):
+def submit_video_create(client, url: str, headers: dict, ref_img: Path, payload: dict, source_ref: VideoSourceRef | None = None):
     if video_uses_ark():
+        content = [{"type": "text", "text": payload["prompt"]}]
+        if source_ref and source_ref.kind == "source_video" and source_ref.url:
+            content.append(
+                {
+                    "type": "video_url",
+                    "video_url": {"url": source_ref.url},
+                    "role": "reference_video",
+                }
+            )
+        content.append(
+            {
+                "type": "image_url",
+                "image_url": {"url": ark_reference_data_url(ref_img)},
+                "role": "first_frame",
+            }
+        )
         data = {
             "model": payload["model"],
-            "content": [
-                {"type": "text", "text": payload["prompt"]},
-                {
-                    "type": "image_url",
-                    "image_url": {"url": ark_reference_data_url(ref_img)},
-                    "role": "first_frame",
-                },
-            ],
+            "content": content,
             "ratio": size_to_video_ratio(str(payload.get("size", ""))),
             "duration": int(float(str(payload.get(VIDEO_DURATION_FIELD, 5)))),
             "watermark": False,
@@ -1783,7 +1801,7 @@ def submit_video_create(client, url: str, headers: dict, ref_img: Path, payload:
         )
 
 
-def render_storyboard_video(job_id: str, local_id: str, provider_id: str, ref_path: Path, prompt: str, model: str, seconds: str, size: str) -> None:
+def render_storyboard_video(job_id: str, local_id: str, provider_id: str, ref_path: Path, prompt: str, model: str, seconds: str, size: str, source_ref: VideoSourceRef | None = None) -> None:
     import httpx
 
     out_dir = job_dir(job_id) / "storyboard_videos" / local_id
@@ -1801,7 +1819,10 @@ def render_storyboard_video(job_id: str, local_id: str, provider_id: str, ref_pa
             create = None
             create_errors: list[str] = []
             for create_path in VIDEO_CREATE_PATHS:
-                resp = submit_video_create(client, f"{base}{video_path(create_path)}", headers, ref_img, payload)
+                resp = submit_video_create(client, f"{base}{video_path(create_path)}", headers, ref_img, payload, source_ref)
+                if video_uses_ark() and source_ref and resp.status_code in {400, 422}:
+                    create_errors.append(f"{video_path(create_path)} + reference_video -> HTTP {resp.status_code}: {resp.text[:160]}")
+                    resp = submit_video_create(client, f"{base}{video_path(create_path)}", headers, ref_img, payload, None)
                 if resp.status_code < 400:
                     create = resp
                     break
@@ -1881,7 +1902,10 @@ def generate_storyboard_video(job_id: str, idx: int, req: GenerateStoryboardVide
         created_at=time.time(),
     )
     update(job, generated_videos=[item] + job.generated_videos, message=f"视频生成已提交 · 分镜 {idx + 1}")
-    bg.add_task(render_storyboard_video, job_id, local_id, "", ref_path, prompt, model, seconds, req.size)
+    source_ref = req.source_ref
+    if source_ref and source_ref.kind == "source_video" and not source_ref.url:
+        source_ref = None
+    bg.add_task(render_storyboard_video, job_id, local_id, "", ref_path, prompt, model, seconds, req.size, source_ref)
     return job