#!/usr/bin/env python3
"""
检查音素数据集文件的内容和统计信息
"""

import pickle
import os

def inspect_phoneme_dataset(file_path):
    """检查音素数据集文件"""

    if not os.path.exists(file_path):
        print(f"文件不存在: {file_path}")
        return

    print(f"正在加载文件: {file_path}")

    try:
        with open(file_path, 'rb') as f:
            dataset = pickle.load(f)

        print(f"\n=== 数据集统计信息 ===")
        print(f"音素类型数量: {len(dataset)}")

        total_segments = 0
        print(f"\n各音素片段数量:")

        for phoneme, segments in dataset.items():
            segment_count = len(segments)
            total_segments += segment_count
            print(f"  {phoneme}: {segment_count} 个片段")

        print(f"\n总片段数: {total_segments}")

        # 查看第一个音素的第一个片段示例
        if dataset:
            first_phoneme = list(dataset.keys())[2]
            first_segment = dataset[first_phoneme][0]

            print(f"\n=== 数据片段示例 (音素: {first_phoneme}) ===")
            for key, value in first_segment.items():
                if key == 'neural_features':
                    print(f"  {key}: shape {value.shape}, dtype {value.dtype}")
                else:
                    print(f"  {key}: {value}")

    except Exception as e:
        print(f"加载文件时出错: {e}")

if __name__ == "__main__":
    # 检查指定的音素数据集文件
    file_path = "./data_analyse/phoneme_segmented_data/phoneme_dataset_20251008_233045.pkl"

    if os.path.exists(file_path):
        inspect_phoneme_dataset(file_path)
    else:
        print(f"文件不存在: {file_path}")

        # 尝试查找其他可能的位置
        possible_dirs = [
            "./phoneme_segmented_data",
            "../phoneme_segmented_data",
            "../../phoneme_segmented_data"
        ]

        for data_dir in possible_dirs:
            if os.path.exists(data_dir):
                print(f"\n在 {data_dir} 中找到以下文件:")
                files = os.listdir(data_dir)
                for f in files:
                    if f.endswith('.pkl'):
                        print(f"  {f}")
                break