154 lines
5.9 KiB
Python
154 lines
5.9 KiB
Python
import os
|
||
import re
|
||
import pandas as pd
|
||
|
||
|
||
def extract_parameters_from_log(log_path):
|
||
"""
|
||
从 log.txt 中提取三个关键参数。
|
||
如果未找到，返回 None。
|
||
"""
|
||
if not os.path.exists(log_path):
|
||
return None, None, None
|
||
|
||
with open(log_path, 'r', encoding='utf-8') as f:
|
||
content = f.read()
|
||
|
||
# 正则表达式定义
|
||
# 1. Percolation diameter (原来的 Step 2)
|
||
# 匹配模式: # Percolation diameter (A): 1.06
|
||
re_percolation = r"Percolation diameter \(A\):\s*([\d\.]+)"
|
||
|
||
# 2. Minimum of d (原来的 Step 3)
|
||
# 匹配模式: the minium of d \n 3.862140561244235
|
||
re_min_d = r"the minium of d\s*\n\s*([\d\.]+)"
|
||
|
||
# 3. Maximum node length (原来的 Step 4)
|
||
# 匹配模式: # Maximum node length detected: 1.332 A
|
||
re_max_node = r"Maximum node length detected:\s*([\d\.]+)\s*A"
|
||
|
||
# 提取数据
|
||
match_perc = re.search(re_percolation, content)
|
||
match_d = re.search(re_min_d, content)
|
||
match_node = re.search(re_max_node, content)
|
||
|
||
# 获取值，如果没匹配到则为空字符串或None
|
||
val_perc = match_perc.group(1) if match_perc else None
|
||
val_d = match_d.group(1) if match_d else None
|
||
val_node = match_node.group(1) if match_node else None
|
||
|
||
return val_perc, val_d, val_node
|
||
|
||
|
||
def process_folder_recursively(base_input_folder, base_output_folder):
|
||
"""
|
||
递归遍历文件夹，提取数据并生成 CSV。
|
||
逻辑：
|
||
1. 遍历 base_input_folder 下的第一层子文件夹（通常是阴离子类别，如 O, S, O+S 等）。
|
||
2. 如果是单阴离子（如 O），直接处理其下的材料文件夹。
|
||
3. 如果是混合阴离子（如 O+S），需要进入下一层（如 O+S/O 和 O+S/S），分别处理。
|
||
4. 结果保存在 base_output_folder 下保持相同的目录结构。
|
||
"""
|
||
|
||
# 获取 after_step1 下的所有顶层目录 (例如 O, S, Cl, S+O ...)
|
||
if not os.path.exists(base_input_folder):
|
||
print(f"输入目录 {base_input_folder} 不存在")
|
||
return
|
||
|
||
top_dirs = [d for d in os.listdir(base_input_folder) if os.path.isdir(os.path.join(base_input_folder, d))]
|
||
|
||
for top_dir in top_dirs:
|
||
top_path = os.path.join(base_input_folder, top_dir)
|
||
|
||
# 判断是否是混合阴离子目录（名字包含 +）
|
||
if "+" in top_dir:
|
||
# 混合阴离子情况：例如 S+O
|
||
# 需要遍历其子目录：S+O/S 和 S+O/O
|
||
sub_anions = [d for d in os.listdir(top_path) if os.path.isdir(os.path.join(top_path, d))]
|
||
for sub_anion in sub_anions:
|
||
# 构建路径：../data/after_step1/S+O/S
|
||
current_process_path = os.path.join(top_path, sub_anion)
|
||
# 构建输出 CSV 路径：../output/S+O/S/S.csv (或者 S+O_S.csv，这里按你要求的 O+S/O/O.csv 格式)
|
||
# 输出目录: ../output/S+O/S
|
||
output_dir = os.path.join(base_output_folder, top_dir, sub_anion)
|
||
csv_filename = f"{sub_anion}.csv"
|
||
|
||
extract_and_save(current_process_path, output_dir, csv_filename)
|
||
else:
|
||
# 单一阴离子情况：例如 O
|
||
# 路径：../data/after_step1/O
|
||
current_process_path = top_path
|
||
# 输出目录: ../output/O
|
||
output_dir = os.path.join(base_output_folder, top_dir)
|
||
csv_filename = f"{top_dir}.csv"
|
||
|
||
extract_and_save(current_process_path, output_dir, csv_filename)
|
||
|
||
|
||
def extract_and_save(input_dir, output_dir, csv_name):
|
||
"""
|
||
实际执行提取和保存的函数。
|
||
input_dir: 包含各个材料文件夹的目录 (例如 .../O/)
|
||
output_dir: CSV 保存目录
|
||
csv_name: CSV 文件名
|
||
"""
|
||
data_list = []
|
||
|
||
# input_dir 下面应该是各个材料的文件夹，例如 141, 142 ...
|
||
if not os.path.exists(input_dir):
|
||
return
|
||
|
||
# 遍历下面的所有材料文件夹
|
||
material_folders = [f for f in os.listdir(input_dir) if os.path.isdir(os.path.join(input_dir, f))]
|
||
|
||
print(f"正在处理目录: {input_dir}, 发现 {len(material_folders)} 个材料文件夹")
|
||
|
||
for material_id in material_folders:
|
||
material_path = os.path.join(input_dir, material_id)
|
||
# 根据新的 step1 逻辑，log 文件名为 log.txt
|
||
log_path = os.path.join(material_path, "log.txt")
|
||
|
||
# 提取数据
|
||
perc, min_d, max_node = extract_parameters_from_log(log_path)
|
||
|
||
# 只要有一个数据存在，就记录（或者你可以改为必须全部存在）
|
||
# 这里设置为只要有记录就加入，方便排查错误
|
||
if perc or min_d or max_node:
|
||
data_list.append({
|
||
"Filename": material_id,
|
||
"Percolation Diameter (A)": perc,
|
||
"Minimum of d": min_d,
|
||
"Maximum Node Length (A)": max_node
|
||
})
|
||
else:
|
||
# 如果 log.txt 不存在或者提取不到数据，可以选择记录空值
|
||
data_list.append({
|
||
"Filename": material_id,
|
||
"Percolation Diameter (A)": None,
|
||
"Minimum of d": None,
|
||
"Maximum Node Length (A)": None
|
||
})
|
||
|
||
# 如果有数据，保存为 CSV
|
||
if data_list:
|
||
if not os.path.exists(output_dir):
|
||
os.makedirs(output_dir)
|
||
|
||
csv_path = os.path.join(output_dir, csv_name)
|
||
df = pd.DataFrame(data_list)
|
||
# 调整列顺序
|
||
df = df[["Filename", "Percolation Diameter (A)", "Minimum of d", "Maximum Node Length (A)"]]
|
||
|
||
df.to_csv(csv_path, index=False)
|
||
print(f"数据已保存至: {csv_path}")
|
||
else:
|
||
print(f"目录 {input_dir} 未提取到有效数据")
|
||
|
||
|
||
if __name__ == "__main__":
|
||
# 输入基础路径 (假设数据在 step1 处理后)
|
||
input_base = "../data/after_step1"
|
||
# 输出基础路径 (你提到的 output 文件夹)
|
||
output_base = "../output"
|
||
|
||
process_folder_recursively(input_base, output_base) |