""" 工具函数模块 提供字幕封面生成过程中需要的通用工具函数 """ import os import sys from pathlib import Path from typing import Tuple, Optional, List, Dict, Any # 尝试导入loguru，如果不可用则使用print作为替代 try: from loguru import logger except ImportError: # Fallback logger that uses print class logger: @staticmethod def info(msg, *args, **kwargs): print(f"[INFO] {msg}", file=sys.stderr) @staticmethod def warning(msg, *args, **kwargs): print(f"[WARN] {msg}", file=sys.stderr) @staticmethod def error(msg, *args, **kwargs): print(f"[ERROR] {msg}", file=sys.stderr) @staticmethod def debug(msg, *args, **kwargs): print(f"[DEBUG] {msg}", file=sys.stderr) # 尝试导入图像处理库 try: import cv2 import numpy as np CV2_AVAILABLE = True except ImportError: CV2_AVAILABLE = False logger.warning("opencv-python不可用，图像处理功能受限") try: from PIL import Image PIL_AVAILABLE = True except ImportError: PIL_AVAILABLE = False logger.warning("PIL不可用，图像处理功能受限") def ensure_dir(path: str) -> Path: """确保目录存在""" path_obj = Path(path) path_obj.mkdir(parents=True, exist_ok=True) return path_obj def get_video_info(video_path: str) -> Dict[str, Any]: """获取视频信息""" if not CV2_AVAILABLE: # 如果opencv不可用，返回基本信息 try: import os file_size = os.path.getsize(video_path) return { 'fps': 30, # 默认值 'frame_count': 0, 'width': 1920, # 默认值 'height': 1080, # 默认值 'duration': 0, # 未知 'path': video_path, 'file_size': file_size, 'note': '视频信息不完整，opencv不可用' } except Exception as e: logger.error(f"获取基本视频信息失败: {e}") raise try: cap = cv2.VideoCapture(video_path) if not cap.isOpened(): raise ValueError(f"无法打开视频文件: {video_path}") # 获取视频属性 fps = cap.get(cv2.CAP_PROP_FPS) frame_count = int(cap.get(cv2.CAP_PROP_FRAME_COUNT)) width = int(cap.get(cv2.CAP_PROP_FRAME_WIDTH)) height = int(cap.get(cv2.CAP_PROP_FRAME_HEIGHT)) duration = frame_count / fps if fps > 0 else 0 cap.release() return { 'fps': fps, 'frame_count': frame_count, 'width': width, 'height': height, 'duration': duration, 'path': video_path } except Exception as e: logger.error(f"获取视频信息失败: {e}") raise def extract_frame_at_time(video_path: str, time_seconds: float): """在指定时间抽取视频帧""" if not CV2_AVAILABLE: logger.warning("opencv不可用，无法抽取视频帧") # 返回一个占位符 return None try: cap = cv2.VideoCapture(video_path) if not cap.isOpened(): raise ValueError(f"无法打开视频文件: {video_path}") # 设置帧位置 fps = cap.get(cv2.CAP_PROP_FPS) frame_number = int(time_seconds * fps) cap.set(cv2.CAP_PROP_POS_FRAMES, frame_number) ret, frame = cap.read() cap.release() if not ret: raise ValueError(f"无法读取帧: {frame_number}") return frame except Exception as e: logger.error(f"抽帧失败: {e}") raise def save_image(image, output_path: str, quality: int = 95) -> None: """保存图像 如果输出路径是PNG格式，会保留alpha通道（透明背景） 如果是JPEG格式，会转换为RGB格式 """ try: ensure_dir(os.path.dirname(output_path)) is_png = output_path.lower().endswith('.png') if CV2_AVAILABLE and isinstance(image, np.ndarray): # 使用OpenCV保存 if is_png and image.shape[2] == 4: # PNG格式且有alpha通道，保存为BGRA（OpenCV使用BGR格式） # 确保图像是BGRA格式（B, G, R, A） # 如果输入是RGBA（R, G, B, A），需要转换为BGRA if image.dtype != np.uint8: image = image.astype(np.uint8) # OpenCV的imwrite会自动处理BGRA格式 success = cv2.imwrite(output_path, image, [cv2.IMWRITE_PNG_COMPRESSION, 3]) elif is_png: # PNG格式但没有alpha通道，转换为RGB if image.shape[2] == 3: success = cv2.imwrite(output_path, image, [cv2.IMWRITE_PNG_COMPRESSION, 3]) else: # 如果是单通道，转换为3通道 if len(image.shape) == 2: image = cv2.cvtColor(image, cv2.COLOR_GRAY2BGR) success = cv2.imwrite(output_path, image, [cv2.IMWRITE_PNG_COMPRESSION, 3]) else: # JPEG格式，确保是3通道BGR if image.shape[2] == 4: # 有alpha通道，先合成到白色背景 bgr = image[:, :, :3] alpha = image[:, :, 3:4] / 255.0 white_bg = np.ones_like(bgr) * 255 image = (bgr * alpha + white_bg * (1 - alpha)).astype(np.uint8) elif len(image.shape) == 2: image = cv2.cvtColor(image, cv2.COLOR_GRAY2BGR) success = cv2.imwrite(output_path, image, [cv2.IMWRITE_JPEG_QUALITY, quality]) if not success: raise ValueError(f"保存图像失败: {output_path}") elif PIL_AVAILABLE and hasattr(image, 'save'): # 使用PIL保存 if is_png: image.save(output_path, 'PNG', compress_level=3) else: # JPEG格式，确保是RGB if image.mode == 'RGBA': # 合成到白色背景 rgb = Image.new('RGB', image.size, (255, 255, 255)) rgb.paste(image, mask=image.split()[3]) image = rgb image.save(output_path, 'JPEG', quality=quality) else: raise ValueError("没有可用的图像保存方法") logger.info(f"图像已保存: {output_path}") except Exception as e: logger.error(f"保存图像失败: {e}") raise def format_time(seconds: float) -> str: """格式化时间为 HH:MM:SS.mmm""" hours = int(seconds // 3600) minutes = int((seconds % 3600) // 60) secs = int(seconds % 60) milliseconds = int((seconds % 1) * 1000) return "02d" def create_temp_dir(prefix: str = "subtitle_cover_") -> str: """创建临时目录""" import tempfile temp_dir = tempfile.mkdtemp(prefix=prefix) logger.info(f"创建临时目录: {temp_dir}") return temp_dir def cleanup_temp_dir(temp_dir: str) -> None: """清理临时目录""" try: import shutil if os.path.exists(temp_dir): shutil.rmtree(temp_dir) logger.info(f"清理临时目录: {temp_dir}") except Exception as e: logger.warning(f"清理临时目录失败: {e}") def validate_file_exists(file_path: str, file_type: str = "文件") -> None: """验证文件是否存在""" if not os.path.exists(file_path): raise FileNotFoundError(f"{file_type}不存在: {file_path}") if not os.path.isfile(file_path): raise ValueError(f"{file_type}不是文件: {file_path}") def get_file_size_mb(file_path: str) -> float: """获取文件大小(MB)""" size_bytes = os.path.getsize(file_path) return size_bytes / (1024 * 1024) def calculate_aspect_ratio(width: int, height: int) -> float: """计算宽高比""" return width / height if height > 0 else 0 def resize_image(image: np.ndarray, target_width: int, target_height: int, keep_aspect_ratio: bool = True) -> np.ndarray: """调整图像大小""" if keep_aspect_ratio: # 保持宽高比 h, w = image.shape[:2] aspect_ratio = w / h if target_width / target_height > aspect_ratio: # 目标更宽，以高度为准 new_width = int(target_height * aspect_ratio) new_height = target_height else: # 目标更高，以宽度为准 new_width = target_width new_height = int(target_width / aspect_ratio) resized = cv2.resize(image, (new_width, new_height)) else: # 不保持宽高比，直接缩放 resized = cv2.resize(image, (target_width, target_height)) return resized def blend_images(background: np.ndarray, foreground: np.ndarray, position: Tuple[int, int] = (0, 0)) -> np.ndarray: """将前景图像合成到背景图像上""" x, y = position h, w = foreground.shape[:2] # 确保位置不超出边界 bg_h, bg_w = background.shape[:2] x = max(0, min(x, bg_w - w)) y = max(0, min(y, bg_h - h)) # 创建ROI roi = background[y:y+h, x:x+w] # 如果前景有alpha通道，进行透明合成 if foreground.shape[2] == 4: # 分离颜色和alpha通道 foreground_rgb = foreground[:, :, :3] alpha = foreground[:, :, 3] / 255.0 # 扩展alpha到3通道 alpha = np.stack([alpha] * 3, axis=2) # 透明合成 blended = foreground_rgb * alpha + roi * (1 - alpha) background[y:y+h, x:x+w] = blended.astype(np.uint8) else: # 直接覆盖 background[y:y+h, x:x+w] = foreground return background def apply_blur(image: np.ndarray, kernel_size: int = 15) -> np.ndarray: """应用模糊效果""" return cv2.GaussianBlur(image, (kernel_size, kernel_size), 0) def apply_gradient_overlay(image: np.ndarray, color: Tuple[int, int, int], opacity: float = 0.5) -> np.ndarray: """应用渐变覆盖""" overlay = np.full_like(image, color, dtype=np.uint8) return cv2.addWeighted(image, 1 - opacity, overlay, opacity, 0)