Spaces:
Runtime error
Runtime error
File size: 18,594 Bytes
1c97e64 0bca353 1c97e64 0bca353 1c97e64 c9807e8 1c97e64 c9807e8 1c97e64 c9807e8 1c97e64 0bca353 1c97e64 c9807e8 1c97e64 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 | import os
import json
import time
import shutil
import uuid
import threading
import html
from pathlib import Path
from PIL import Image
LEGACY_OUTPUT_DIR = Path("outputs")
OUTPUT_DIR = Path(os.environ.get("WAN_OUTPUT_DIR", "/data/outputs" if Path("/data").is_dir() else "outputs"))
UPLOADS_DIR = OUTPUT_DIR / "uploads"
VIDEOS_DIR = OUTPUT_DIR / "videos"
JOBS_FILE = OUTPUT_DIR / "jobs.json"
def unpack_image_files(input_image):
if not input_image:
return []
raw_items = []
if isinstance(input_image, (list, tuple)):
raw_items = list(input_image)
else:
raw_items = [input_image]
extracted_paths = []
for item in raw_items:
if item is None:
continue
path = None
if isinstance(item, str):
path = item
elif hasattr(item, "path") and item.path:
path = item.path
elif hasattr(item, "name") and item.name:
path = item.name
elif isinstance(item, dict):
path = item.get("path") or item.get("name")
elif isinstance(item, Image.Image) or isinstance(item, np.ndarray):
extracted_paths.append(item)
continue
if path:
if os.path.isdir(path):
for root, _, files in os.walk(path):
for f in sorted(files):
if f.lower().endswith((".png", ".jpg", ".jpeg", ".webp", ".bmp")):
extracted_paths.append(os.path.join(root, f))
else:
extracted_paths.append(path)
return extracted_paths
class QueueManager:
def __init__(self):
# Some UI actions (for example clear_queue) persist while already
# holding the state lock, so this must be re-entrant.
self.lock = threading.RLock()
self.gpu_lock = threading.Lock()
self.is_running = False
self.worker_thread = None
self.inference_fn = None
self._ensure_dirs()
self.jobs = self._load_jobs()
self._recover_interrupted_jobs()
def _ensure_dirs(self):
# Preserve any queued/completed work created before outputs moved into
# the mounted Space bucket. This is a copy, never a destructive move.
if OUTPUT_DIR != LEGACY_OUTPUT_DIR and not JOBS_FILE.exists() and LEGACY_OUTPUT_DIR.exists():
shutil.copytree(LEGACY_OUTPUT_DIR, OUTPUT_DIR, dirs_exist_ok=True)
OUTPUT_DIR.mkdir(parents=True, exist_ok=True)
UPLOADS_DIR.mkdir(parents=True, exist_ok=True)
VIDEOS_DIR.mkdir(parents=True, exist_ok=True)
if not JOBS_FILE.exists():
with open(JOBS_FILE, "w", encoding="utf-8") as f:
json.dump([], f)
def _load_jobs(self):
with self.lock:
if JOBS_FILE.exists():
try:
with open(JOBS_FILE, "r", encoding="utf-8") as f:
return json.load(f)
except Exception as e:
print(f"[QueueManager] Error loading jobs.json: {e}")
return []
return []
def _save_jobs(self):
with self.lock:
with open(JOBS_FILE, "w", encoding="utf-8") as f:
json.dump(self.jobs, f, indent=2, ensure_ascii=False)
def _recover_interrupted_jobs(self):
updated = False
for job in self.jobs:
if job.get("status") == "running":
job["status"] = "queued"
updated = True
if updated:
self._save_jobs()
def add_job(self, input_image, last_image, prompt, steps, negative_prompt,
duration_seconds, sequential_renders, guidance_scale, guidance_scale_2, seed,
randomize_seed, quality, scheduler, flow_shift, frame_multiplier,
lora_groups):
image_paths = unpack_image_files(input_image)
if not image_paths:
return "β Error: Please select at least one valid input image before adding to queue.", self.render_queue_html()
import random
added_ids = []
for idx, img_src in enumerate(image_paths):
try:
if isinstance(img_src, str) and os.path.exists(img_src):
img_obj = Image.open(img_src)
elif isinstance(img_src, Image.Image):
img_obj = img_src
else:
img_obj = Image.fromarray(img_src)
except Exception as e:
print(f"[QueueManager] Error opening image index {idx}: {e}")
continue
timestamp_str = time.strftime("%Y%m%d_%H%M%S")
short_id = uuid.uuid4().hex[:6]
job_id = f"job_{timestamp_str}_{short_id}_{idx+1}"
input_image_path = UPLOADS_DIR / f"{job_id}_input.png"
img_obj.save(input_image_path)
last_image_path = None
if last_image is not None:
last_image_path = str(UPLOADS_DIR / f"{job_id}_last.png")
try:
last_src = last_image.name if hasattr(last_image, "name") else (last_image.path if hasattr(last_image, "path") else last_image)
if isinstance(last_src, Image.Image):
last_src.save(last_image_path)
elif isinstance(last_src, str) and os.path.exists(last_src):
shutil.copy(last_src, last_image_path)
else:
Image.fromarray(last_src).save(last_image_path)
except Exception as e:
print(f"[QueueManager] Warning: failed to save last image: {e}")
last_image_path = None
current_seed = random.randint(0, 2147483647) if randomize_seed else int(seed if seed is not None else 42)
job = {
"id": job_id,
"created_at": time.strftime("%Y-%m-%d %H:%M:%S"),
"status": "queued",
"prompt": prompt or "",
"input_image_path": str(input_image_path).replace("\\", "/"),
"last_image_path": str(last_image_path).replace("\\", "/") if last_image_path else None,
"video_path": None,
"params": {
"steps": int(steps) if steps is not None else 4,
"negative_prompt": negative_prompt or "",
"duration_seconds": float(duration_seconds) if duration_seconds is not None else 10.0,
"sequential_renders": max(1, min(12, int(sequential_renders or 1))),
"guidance_scale": float(guidance_scale) if guidance_scale is not None else 1.0,
"guidance_scale_2": float(guidance_scale_2) if guidance_scale_2 is not None else 1.0,
"seed": current_seed,
"randomize_seed": bool(randomize_seed),
"quality": int(quality) if quality is not None else 6,
"scheduler": str(scheduler) if scheduler else "UniPCMultistep",
"flow_shift": float(flow_shift) if flow_shift is not None else 3.0,
"frame_multiplier": int(frame_multiplier) if frame_multiplier is not None else 16,
"lora_groups": lora_groups if isinstance(lora_groups, list) else []
},
"error": None,
"completed_at": None
}
self.jobs.insert(0, job)
added_ids.append(job_id)
self._save_jobs()
if len(added_ids) == 1:
msg = f"β
1 job ({added_ids[0]}) added to queue."
elif len(added_ids) > 1:
msg = f"β
{len(added_ids)} separate jobs added to queue using prompt."
else:
msg = "β No valid images found to queue."
return msg, self.render_queue_html()
def start_queue(self, inference_fn):
self.inference_fn = inference_fn
self.is_running = True
if self.worker_thread is None or not self.worker_thread.is_alive():
self.worker_thread = threading.Thread(target=self._worker_loop, daemon=True)
self.worker_thread.start()
return "βΆοΈ Queue started.", self.render_queue_html()
def stop_queue(self):
self.is_running = False
return "βΈοΈ Queue stopped.", self.render_queue_html()
def clear_queue(self):
with self.lock:
# Keep currently running job if active, clear all other jobs
self.jobs = [j for j in self.jobs if j.get("status") == "running"]
self._save_jobs()
return "ποΈ Queue cleared.", self.render_queue_html()
def get_error_log(self):
failed_jobs = [j for j in self.jobs if j.get("status") == "failed"]
logs = []
if os.path.exists("app.log"):
try:
with open("app.log", "r", encoding="utf-8", errors="ignore") as f:
lines = f.readlines()
err_lines = [l.strip() for l in lines if any(k in l.lower() for k in ["error", "exception", "traceback", "failed"])]
if err_lines:
logs.append("=== RECENT APP LOG ERRORS ===")
logs.extend(err_lines[-25:])
except Exception as e:
logs.append(f"Error reading app.log: {e}")
if failed_jobs:
logs.append("\n=== FAILED QUEUE JOBS ===")
for j in failed_jobs:
logs.append(f"[{j.get('created_at')}] Job {j.get('id')}: {j.get('error', 'Unknown error')}")
if not logs:
return "β
No errors found in logs or queue jobs! Everything is running smoothly."
return "\n".join(logs)
def check_queue(self):
running_count = sum(1 for j in self.jobs if j["status"] == "running")
queued_count = sum(1 for j in self.jobs if j["status"] == "queued")
completed_count = sum(1 for j in self.jobs if j["status"] == "completed")
msg = f"π Queue checked. Total: {len(self.jobs)} | Queued: {queued_count} | Running: {running_count} | Finished: {completed_count}"
return msg, self.render_queue_html()
def _worker_loop(self):
print("[QueueManager] Background worker thread started.")
while self.is_running:
# Find next queued job (oldest created job that is still queued)
next_job = None
for job in reversed(self.jobs):
if job.get("status") == "queued":
next_job = job
break
if next_job is None:
time.sleep(2)
continue
job_id = next_job["id"]
print(f"[QueueManager] Processing job: {job_id}")
# Enforce single job GPU execution
with self.gpu_lock:
next_job["status"] = "running"
self._save_jobs()
try:
input_img = Image.open(next_job["input_image_path"])
last_img = Image.open(next_job["last_image_path"]) if next_job.get("last_image_path") else None
params = next_job.get("params", {})
if self.inference_fn is None:
raise ValueError("Inference function not bound to QueueManager.")
res = self.inference_fn(
input_image=input_img,
last_image=last_img,
prompt=next_job.get("prompt", ""),
steps=params.get("steps", 4),
negative_prompt=params.get("negative_prompt", ""),
duration_seconds=params.get("duration_seconds", 10.0),
sequential_renders=params.get("sequential_renders", 1),
guidance_scale=params.get("guidance_scale", 1.0),
guidance_scale_2=params.get("guidance_scale_2", 1.0),
seed=params.get("seed", 42),
randomize_seed=params.get("randomize_seed", False),
quality=params.get("quality", 6),
scheduler=params.get("scheduler", "UniPCMultistep"),
flow_shift=params.get("flow_shift", 3.0),
frame_multiplier=params.get("frame_multiplier", 16),
lora_groups=params.get("lora_groups"),
video_component=True,
artifact_dir=str(VIDEOS_DIR / job_id),
)
temp_video = None
if isinstance(res, tuple) and len(res) >= 2:
temp_video = res[1] or res[0]
elif isinstance(res, str):
temp_video = res
dest_video = VIDEOS_DIR / f"{job_id}.mp4"
if temp_video and os.path.exists(temp_video):
shutil.copy(temp_video, dest_video)
next_job["status"] = "completed"
next_job["video_path"] = str(dest_video).replace("\\", "/")
next_job["completed_at"] = time.strftime("%Y-%m-%d %H:%M:%S")
print(f"[QueueManager] Job {job_id} completed successfully.")
else:
next_job["status"] = "failed"
next_job["error"] = "Output video was not generated or file not found."
print(f"[QueueManager] Job {job_id} failed: Video file not created.")
except Exception as e:
print(f"[QueueManager] Exception in job {job_id}: {e}")
next_job["status"] = "failed"
next_job["error"] = str(e)
self._save_jobs()
time.sleep(1)
print("[QueueManager] Background worker thread stopped.")
def render_queue_html(self):
if not self.jobs:
return """
<div class="queue-container">
<div class="queue-empty">No jobs in queue yet.<br>Upload an image and click <b>Add to Queue</b> to begin.</div>
</div>
"""
status_class_map = {
"queued": "badge-queued",
"running": "badge-running",
"completed": "badge-completed",
"failed": "badge-failed"
}
status_label_map = {
"queued": "Queued",
"running": "Running β³",
"completed": "Finished β
",
"failed": "Failed β"
}
queue_status = "ACTIVE βΆοΈ" if self.is_running else "STOPPED βΈοΈ"
status_color = "#00b894" if self.is_running else "#ea868f"
cards = []
for job in self.jobs:
status = job.get("status", "queued")
badge_cls = status_class_map.get(status, "badge-queued")
badge_lbl = status_label_map.get(status, status)
job_id = job.get("id", "")
created_at = job.get("created_at", "")
prompt = job.get("prompt", "")
video_path = job.get("video_path")
params = job.get("params", {})
lora_groups = params.get("lora_groups") or []
lora_text = ", ".join(html.escape(str(name)) for name in lora_groups) if lora_groups else "None"
# Gradio 6 serves allowed local files through /gradio_api/file=.
# The legacy /file= route returns 404 even when the file exists.
file_url = f"/gradio_api/file={video_path}" if video_path else ""
# Class attribute for auto-download triggering via JS
auto_dl_attr = ""
if status == "completed" and file_url:
auto_dl_attr = f'class="pending-auto-download" data-download-url="{file_url}" data-job-id="{job_id}" data-filename="{job_id}.mp4"'
video_html = ""
if status == "completed" and video_path and os.path.exists(video_path):
video_html = f"""
<div class="queue-card-video" {auto_dl_attr}>
<video src="{file_url}" controls width="100%" style="border-radius: 8px; border: 1px solid rgba(255,255,255,0.12); margin-top: 8px; max-height: 240px; background: #000;"></video>
<div style="margin-top: 8px; display: flex; gap: 8px; align-items: center; flex-wrap: wrap;">
<button onclick="loadPreview('{file_url}')" class="btn-preview-link" type="button">ποΈ Preview in Output Box</button>
<a href="{file_url}" download="{job_id}.mp4" class="btn-dl-link">π₯ Download Video</a>
<a href="{file_url}" target="_blank" class="btn-view-link">π Direct Link</a>
</div>
</div>
"""
elif status == "failed":
err_msg = job.get("error", "Unknown error")
video_html = f'<div class="queue-error">β οΈ Error: {err_msg}</div>'
elif status == "running":
video_html = '<div class="queue-processing">βοΈ Generating video on GPU...</div>'
elif status == "queued":
video_html = '<div class="queue-pending">β³ Waiting in queue...</div>'
cards.append(f"""
<div class="queue-card {status}">
<div class="queue-card-header">
<span class="queue-job-id">{job_id}</span>
<span class="queue-badge {badge_cls}">{badge_lbl}</span>
</div>
<div class="queue-card-body">
<div class="queue-prompt"><b>Prompt:</b> {prompt}</div>
<div class="queue-loras"><b>LoRAs:</b> {lora_text}</div>
<div class="queue-meta">Created: {created_at} | Dur: {params.get('duration_seconds', 10.0)}s x {params.get('sequential_renders', 1)} segments | Steps: {params.get('steps', 4)} | FPS: {params.get('frame_multiplier', 16)}</div>
{video_html}
</div>
</div>
""")
cards_html = "\n".join(cards)
return f"""
<div class="queue-container">
<div class="queue-header-bar">
<span>Worker Status: <span style="color: {status_color};">{queue_status}</span></span>
<span>Total Jobs: {len(self.jobs)}</span>
</div>
<div class="queue-cards-list">
{cards_html}
</div>
</div>
"""
|