import re
import numpy as np
import matplotlib.pyplot as plt
from matplotlib.font_manager import FontProperties

# 动态加载字体文件
font_path = '/usr/local/sunlogin/res/font/wqy-zenhei.ttc'  # 替换为实际字体路径
my_font = FontProperties(fname=font_path)

# 定义日志文件路径
log_file_path = 'number.txt'
output_file_path = 'backward_point_cloud_stats.txt'

# 正则表达式匹配点云数量和类型
pattern = re.compile(r'\[(\d+) 3 \d+\]')

# 存储后向点云数量的列表
backward_point_cloud_counts = []

# 读取日志文件
with open(log_file_path, 'r') as file:
    for line in file:
        matches = pattern.findall(line)
        if matches:
            backward_point_cloud_counts.extend([int(count) for count in matches])

# 检查数据提取
if backward_point_cloud_counts:
    print(f"成功提取到 {len(backward_point_cloud_counts)} 个后向点云数量。")
else:
    print("没有找到后向点云的相关数据。")

# 统计计算
max_value = np.max(backward_point_cloud_counts)
min_value = np.min(backward_point_cloud_counts)
mean_value = np.mean(backward_point_cloud_counts)
variance_value = np.var(backward_point_cloud_counts)
lower_percentile = np.percentile(backward_point_cloud_counts, 50)
upper_percentile = np.percentile(backward_point_cloud_counts, 95)

# 打印统计结果
print(f"最大值: {max_value}")
print(f"最小值: {min_value}")
print(f"平均值: {mean_value}")
print(f"方差: {variance_value}")
print(f"90%的数据位于区间 [{lower_percentile}, {upper_percentile}] 之间")

# 保存结果
with open(output_file_path, 'w') as output_file:
    output_file.write("后向点云数量统计结果:\n")
    output_file.write(f"最大值: {max_value}\n")
    output_file.write(f"最小值: {min_value}\n")
    output_file.write(f"平均值: {mean_value}\n")
    output_file.write(f"方差: {variance_value}\n")
    output_file.write(f"90%的数据位于区间 [{lower_percentile}, {upper_percentile}] 之间\n")

# 可视化
# 生成以 1000 为间隔的 bins
bin_edges = np.arange(0, max_value + 1000, 1000)
counts, bins, patches = plt.hist(backward_point_cloud_counts, bins=bin_edges, edgecolor='black')

# 过滤掉频次为0的区间
non_zero_bins = bins[:-1][counts > 0]
non_zero_counts = counts[counts > 0]

# 绘制非零频次柱状图
plt.clf()  # 清除之前的图
plt.bar(non_zero_bins + 500, non_zero_counts, width=1000, edgecolor='black')

# 添加频次标签
for count, bin_edge in zip(non_zero_counts, non_zero_bins):
    plt.text(bin_edge + 500, count, f'{int(count)}', ha='center', va='bottom', fontproperties=my_font)

# 设置标题和标签
plt.title('后向点云数量分布', fontproperties=my_font)
plt.xlabel('点云数量', fontproperties=my_font)
plt.ylabel('频率', fontproperties=my_font)
plt.grid(True)

# 设置横轴刻度，只显示非零频次的区间
plt.xticks(non_zero_bins + 500, [f'{int(bin_edge)}' for bin_edge in non_zero_bins], fontproperties=my_font)

plt.show()