Skip to content

Apply debug fixes from AGENTS #7

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
LHAMNS wants to merge 1 commit into
base: main
Choose a base branch
from
Open

Apply debug fixes from AGENTS #7

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
6 changes: 6 additions & 0 deletions converter/__init__.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -41,6 +41,12 @@
(255, 128, 0), # 橙
]

# 验证调色板完整性
assert len(COLOR_PALETTE_16) == 16, "Color palette must have exactly 16 colors"
for i, color in enumerate(COLOR_PALETTE_16):
assert len(color) == 3, f"Color {i} must be RGB tuple"
assert all(0 <= c <= 255 for c in color), f"Color {i} values must be in range 0-255"

# 视频参数预设
VIDEO_PRESETS = {
"4K": {"width": 3840, "height": 2160},
Expand Down
27 changes: 12 additions & 15 deletions requirements.txt
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,26 +1,23 @@
# Core dependencies
numpy>=1.21.0
numba>=0.55.1
opencv-python>=4.5.5.64
flask>=2.0.1
flask-socketio>=5.1.1
numpy>=1.21.0,<2.0.0
numba==0.59.1
opencv-python>=4.9.0.0
flask>=2.0.1,<3.0.0
flask-socketio>=5.1.1,<6.0.0
Pillow>=8.3.1
psutil>=5.9.0
watchdog>=2.1.6
tqdm>=4.62.3
python-magic>=0.4.24

# GPU acceleration (optional but recommended)
# Install appropriate version based on CUDA version:
# GPU acceleration - Choose ONE based on CUDA version
# For CUDA 12.x:
cupy-cuda12x>=12.0.0
# For CUDA 11.x (comment above and uncomment below):
# cupy-cuda11x>=11.0.0

cupy-cuda12x==13.2.0
numba==0.59.1
# For CUDA 11.x (comment above, uncomment below):
# cupy-cuda11x>=11.0.0,<12.0.0

# Raptor codes implementation
pyldpc==0.5.2
opencv-python>=4.9.0.0

# Error correction (removed reedsolo, using GPU Raptor codes instead)
# reedsolo>=1.5.4 # Replaced by GPU implementation
# Remove old error correction dependency
# reedsolo>=1.5.4 # REMOVED - replaced by GPU implementation
284 changes: 149 additions & 135 deletions web_ui/server.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -22,9 +22,51 @@
import hashlib
import numpy as np

from converter.gpu_error_correction import get_optimal_error_corrector
from converter.gpu_frame_generator import GPUFrameGenerator
from converter.video_raptor_encoder import VideoRaptorEncoder
# GPU模块条件导入 - 完整fallback机制
try:
from converter.gpu_error_correction import get_optimal_error_corrector
GPU_ERROR_CORRECTION_AVAILABLE = True
logger.info("GPU error correction available")
except ImportError as e:
logger.warning(f"GPU error correction not available: {e}")
GPU_ERROR_CORRECTION_AVAILABLE = False

def get_optimal_error_corrector(redundancy_ratio):
from converter.error_correction import ReedSolomonEncoder
redundancy_bytes = max(1, int(255 * redundancy_ratio))
return ReedSolomonEncoder(redundancy_bytes=redundancy_bytes)

try:
from converter.gpu_frame_generator import GPUFrameGenerator
GPU_FRAME_GENERATION_AVAILABLE = True
logger.info("GPU frame generation available")
except ImportError as e:
logger.warning(f"GPU frame generation not available: {e}")
GPU_FRAME_GENERATION_AVAILABLE = False

try:
from converter.video_raptor_encoder import VideoRaptorEncoder
RAPTOR_ENCODER_AVAILABLE = True
logger.info("Raptor encoder available")
except ImportError as e:
logger.warning(f"Raptor encoder not available: {e}")
RAPTOR_ENCODER_AVAILABLE = False

class VideoRaptorEncoder:
def __init__(self, width, height, fps):
self.width = width
self.height = height
self.fps = fps

def create_metadata_frame(self, file_info):
return np.zeros((self.height, self.width, 3), dtype=np.uint8)

def create_calibration_frame(self):
return np.zeros((self.height, self.width, 3), dtype=np.uint8)

def _add_sync_pattern(self, frame):
frame[::20, :] = 255
frame[:, ::20] = 255

# 添加项目根目录到路径
# Add project root to path
Expand Down Expand Up @@ -314,154 +356,75 @@ def _frame_generated_callback(self, frame_idx, total_frames, frame):
except Exception as e:
logger.error(f"进度更新错误 [{self.task_id}]: {e}", exc_info=True)
def _verify_output_video(self):
"""
验证输出视频文件是否有效

Returns:
tuple: (是否有效, 错误消息)
"""
"""验证输出AVI文件 - 针对未压缩AVI优化"""
if not self.output_path.exists():
return False, "输出文件不存在"

if self.output_path.stat().st_size == 0:
return False, "输出文件大小为0"

try:
# 首先尝试修复AVI文件
logger.info(f"尝试修复AVI文件: {self.output_path}")

# 创建临时文件路径
temp_output = Path(str(self.output_path) + ".fixed.avi")

# 首先尝试将文件复制到新位置以修复可能的问题
fix_cmd = [
"ffmpeg",
"-v", "warning",
"-i", str(self.output_path),
"-c", "copy",
"-movflags", "faststart", # 这个选项会将元数据移到文件开头，解决moov atom问题
str(temp_output)
]

logger.info(f"执行修复命令: {' '.join(fix_cmd)}")
fix_result = subprocess.run(
fix_cmd,
stderr=subprocess.PIPE,
stdout=subprocess.PIPE,
timeout=60
)

if fix_result.returncode == 0 and temp_output.exists() and temp_output.stat().st_size > 0:
# 修复成功，替换原文件
logger.info(f"AVI文件修复成功: {self.output_path}")
shutil.copy2(temp_output, self.output_path)
temp_output.unlink()
else:
# 修复失败，记录错误
error_msg = fix_result.stderr.decode('utf-8', errors='ignore')
logger.warning(f"AVI文件修复失败: {error_msg}")
if temp_output.exists():
temp_output.unlink()

# 使用ffprobe验证视频
info_cmd = [
"ffprobe",
"-v", "error",
"-select_streams", "v:0",
"-show_entries", "stream=width,height,codec_name,duration,nb_frames",
"-of", "json",
str(self.output_path)
]

logger.info(f"执行验证命令: {' '.join(info_cmd)}")
info_result = subprocess.run(
info_cmd,
stderr=subprocess.PIPE,
stdout=subprocess.PIPE,
timeout=30
from converter.avi_validator import validate_avi_file

validation_results = validate_avi_file(
str(self.output_path),
callback=None
)

if info_result.returncode == 0:
# 尝试解析视频信息
try:
video_info = json.loads(info_result.stdout)
logger.info(f"视频信息获取成功: {json.dumps(video_info, indent=2)}")

if "streams" in video_info and len(video_info["streams"]) > 0:
stream_info = video_info["streams"][0]
width = stream_info.get("width")
height = stream_info.get("height")
codec = stream_info.get("codec_name")

# 验证基本参数是否合理
if width and height and codec:
logger.info(f"视频有效: {width}x{height}, 编码: {codec}")
return True, "视频文件有效"
else:
logger.warning(f"视频缺少关键信息: {stream_info}")
return False, "视频信息不完整"
else:
logger.warning("无视频流信息")
return False, "未找到视频流"
except json.JSONDecodeError as e:
logger.error(f"JSON解析错误: {e}")
return False, f"解析视频信息失败: {e}"
else:
error_msg = info_result.stderr.decode('utf-8', errors='ignore')
logger.error(f"视频文件验证失败: {error_msg}")

# 尝试更简单的验证方法 - 仅检查文件是否可以打开
try_cmd = [
"ffmpeg",
"-v", "error",
"-i", str(self.output_path),
"-t", "0.1", # 只读取前0.1秒
"-f", "null",
"-"
]

logger.info(f"尝试简单验证: {' '.join(try_cmd)}")
try_result = subprocess.run(
try_cmd,
stderr=subprocess.PIPE,
stdout=subprocess.PIPE,
timeout=30

if validation_results['overall_success']:
structure = validation_results['structure_validation']
logger.info(
f"AVI验证成功: {structure['resolution']}, "
f"{structure['frame_count']} frames, {structure['fps']} fps"
)

if try_result.returncode == 0:
logger.info("简单验证通过，视频可以读取")
return True, "简单验证通过，视频可以读取"
else:
try_error = try_result.stderr.decode('utf-8', errors='ignore')
logger.error(f"简单验证失败: {try_error}")
return False, f"视频验证失败: {error_msg}\n简单验证失败: {try_error}"

except subprocess.TimeoutExpired:
return False, "视频验证超时"
return True, "AVI文件验证通过"
else:
error_details = []
if 'structure_validation' in validation_results:
error_details.extend(validation_results['structure_validation'].get('error_details', []))
if 'integrity_validation' in validation_results:
error_details.extend(validation_results['integrity_validation'].get('error_details', []))

error_msg = "; ".join(error_details) if error_details else "未知验证错误"
logger.error(f"AVI验证失败: {error_msg}")
return False, f"AVI验证失败: {error_msg}"

except ImportError:
logger.warning("AVI validator not available, using basic validation")
return True, "基础验证通过（验证器不可用）"
except Exception as e:
logger.error(f"视频验证出错: {e}", exc_info=True)
return False, f"视频验证出错: {str(e)}"
logger.error(f"AVI验证过程出错: {e}", exc_info=True)
return False, f"验证过程出错: {str(e)}"

def _conversion_worker(self):
"""GPU-aware conversion worker thread"""
try:
self._update_task_status("initializing")

# ---------- 1. GPU error-correction ----------
if self.error_correction_enabled:
logger.info(f"Initializing Raptor error correction [{self.task_id}]")
from converter.encoder import StreamingDirectAVIEncoder # lazy import

# ---------- 1. GPU error-correction ----------
if self.error_correction_enabled and GPU_ERROR_CORRECTION_AVAILABLE:
logger.info(f"Initializing GPU error correction [{self.task_id}]")
self.error_correction = get_optimal_error_corrector(self.error_correction_ratio)
elif self.error_correction_enabled:
logger.info(f"Using CPU error correction [{self.task_id}]")
self.error_correction = ReedSolomonEncoder(redundancy_bytes=int(255 * self.error_correction_ratio))

# ---------- 2. Frame generator ----------
# ---------- 2. Frame generator ----------
try:
self.frame_generator = GPUFrameGenerator(
resolution=self.resolution,
fps=self.fps,
color_count=self.color_count,
nine_to_one=self.nine_to_one
)
logger.info(f"Using GPU-accelerated frame generator [{self.task_id}]")
except RuntimeError:
if GPU_FRAME_GENERATION_AVAILABLE:
self.frame_generator = GPUFrameGenerator(
resolution=self.resolution,
fps=self.fps,
color_count=self.color_count,
nine_to_one=self.nine_to_one
)
logger.info(f"Using GPU-accelerated frame generator [{self.task_id}]")
else:
raise RuntimeError("GPU not available")
except (RuntimeError, Exception) as e:
logger.info(f"GPU frame generation failed, using CPU: {e}")
generator_class = OptimizedFrameGenerator if self.use_optimized_generator else FrameGenerator
self.frame_generator = generator_class(
resolution=self.resolution,
Expand All @@ -488,7 +451,7 @@ def _conversion_worker(self):
self.video_encoder.start()

# ---------- 4. Metadata / calibration frames ----------
if self.params.get("metadata_frames", True):
if self.params.get("metadata_frames", True) and RAPTOR_ENCODER_AVAILABLE:
logger.info(f"Adding metadata frames [{self.task_id}]")

file_info = {
Expand Down Expand Up @@ -907,6 +870,57 @@ def download_file_by_name(filename):
return jsonify({"error": str(e)}), 500


@app.route('/api/verify-video/<task_id>', methods=['POST'])
def verify_video(task_id):
"""验证生成的AVI视频文件"""
try:
with task_lock:
if task_id not in task_registry:
return jsonify({"error": "找不到指定的任务"}), 404

task_info = task_registry[task_id]

if task_info["status"] != "completed":
return jsonify({"error": "任务尚未完成"}), 400

if not task_info.get("output_path"):
return jsonify({"error": "没有可用的输出文件"}), 404

output_path = Path(task_info["output_path"])

if not output_path.exists():
return jsonify({"error": "输出文件不存在"}), 404

from converter.avi_validator import validate_avi_file

def progress_callback(percentage, frame_num, bytes_processed):
socketio.emit('verification_progress', {
'task_id': task_id,
'percentage': percentage,
'frame': frame_num,
'bytes': bytes_processed
})

validation_results = validate_avi_file(
str(output_path),
callback=progress_callback
)

socketio.emit('verification_complete', {
'task_id': task_id,
'results': validation_results
})

return jsonify({
"success": True,
"validation_results": validation_results
})

except Exception as e:
logger.error(f"视频验证错误: {e}", exc_info=True)
return jsonify({"error": str(e)}), 500


# 添加缺失的清理缓存API
@app.route('/api/clear-cache', methods=['POST'])
def clear_cache():
Expand Down