CSM及TET,CS

This commit is contained in:
2025-12-07 22:30:46 +08:00
parent e885893484
commit cea5ab6d3f
3 changed files with 133 additions and 85 deletions

View File

@@ -1,69 +1,113 @@
from pymatgen.core import Structure
from pymatgen.core.periodic_table import Element, Specie
from pymatgen.io.cif import CifWriter
from crystal_2 import crystal
import crystal_2
import os
import shutil
def get_anion_type(structure):
"""
判断阴离子类型。
仅识别 O, S, Cl, Br 及其组合。
其他非金属元素(如 P, N, F 等)将被忽略。
"""
# 仅保留这四种目标阴离子
valid_anions = {'O', 'S', 'Cl', 'Br'}
# 获取结构中的所有元素符号
elements = set([e.symbol for e in structure.composition.elements])
# 取交集找到当前结构包含的目标阴离子
found_anions = elements.intersection(valid_anions)
if not found_anions:
return "Unknown"
# 如果有多个阴离子,按字母顺序排序并用 '+' 连接
sorted_anions = sorted(list(found_anions))
return "+".join(sorted_anions)
def read_files_check_basic(folder_path):
file_contents = []
"""
读取 CIF 文件,进行基础检查 (check_basic)
通过筛选后按自定义阴离子规则分类并整理到 after_step1 文件夹。
"""
# 输出基础路径
output_base = "../data/after_step1"
if not os.path.exists(folder_path):
print(f"{folder_path} 文件夹不存在")
return file_contents
return
for filename in os.listdir(folder_path):
# 确保输出目录存在
if not os.path.exists(output_base):
os.makedirs(output_base)
cif_files = [f for f in os.listdir(folder_path) if f.endswith(".cif")]
print(f"{folder_path} 发现 {len(cif_files)} 个 CIF 文件,开始筛选与整理...")
count_pass = 0
for filename in cif_files:
file_path = os.path.join(folder_path, filename)
if os.path.isfile(file_path):
try:
temp = crystal(file_path)
file_contents.append(temp)
except Exception as e:
print(e)
continue # 如果出错跳过当前循环避免temp未定义报错
print(f"正在处理{filename}")
# 1. 调用 crystal_2 进行基础筛选
try:
temp = crystal(file_path)
# 进行基础检查 (电荷平衡、化学式检查等)
temp.check_basic()
if temp.check_basic_result:
# 获取不带后缀的文件名,用于创建同名文件夹
file_base_name = os.path.splitext(filename)[0]
if not temp.check_basic_result:
print(f"Skipped: {filename} (未通过 check_basic)")
continue
if not "+" in temp.anion:
# 单一阴离子情况
# 路径变为: ../data/after_step1/Anion/FileBaseName/
base_anion_folder = os.path.join("../data/after_step1", f"{temp.anion}")
target_folder = os.path.join(base_anion_folder, file_base_name)
except Exception as e:
print(f"Error checking {filename}: {e}")
continue
# 2. 筛选通过,进行分类整理
try:
print(f"Processing: {filename} (Passed)")
count_pass += 1
# 为了确保分类逻辑与 Direct 版本一致,重新读取结构判断阴离子
# (忽略 crystal_2 内部可能基于 P/N 等元素的命名)
struct = Structure.from_file(file_path)
anion_type = get_anion_type(struct)
# 获取不带后缀的文件名 (ID)
file_base_name = os.path.splitext(filename)[0]
# --- 构建目标路径逻辑 (Anion/ID/ID.cif) ---
if "+" in anion_type:
# 混合阴离子情况 (如 S+O)
# 分别复制到 S+O/S 和 S+O/O 下
sub_anions = anion_type.split("+")
for sub in sub_anions:
# 路径: ../data/after_step1/S+O/S/123/123.cif
target_folder = os.path.join(output_base, anion_type, sub, file_base_name)
if not os.path.exists(target_folder):
os.makedirs(target_folder)
# 目标文件路径
target_file_path = os.path.join(target_folder, filename)
# 复制文件到目标文件夹
shutil.copy(file_path, target_file_path)
print(f"文件 {filename}通过基本筛选,已复制到 {target_folder}")
else:
# 混合阴离子情况
anions = temp.anion.split("+")
for anion in anions:
# 路径变为: ../data/after_step1/AnionCombination/Anion/FileBaseName/
base_group_folder = os.path.join("../data/after_step1", f"{temp.anion}")
base_anion_folder = os.path.join(base_group_folder, anion)
target_folder = os.path.join(base_anion_folder, file_base_name)
target_file = os.path.join(target_folder, filename)
shutil.copy(file_path, target_file)
else:
# 单一阴离子或 Unknown: ../data/after_step1/S/123/123.cif
target_folder = os.path.join(output_base, anion_type, file_base_name)
if not os.path.exists(target_folder):
os.makedirs(target_folder)
if not os.path.exists(target_folder):
os.makedirs(target_folder)
target_file = os.path.join(target_folder, filename)
shutil.copy(file_path, target_file)
# 目标文件路径
target_file_path = os.path.join(target_folder, filename)
# 复制文件到目标文件夹
shutil.copy(file_path, target_file_path)
print(f"文件 {filename}通过基本筛选,已复制到 {target_folder}")
except Exception as e:
print(f"Error copying {filename}: {e}")
print(f"处理完成。共 {len(cif_files)} 个文件,通过筛选 {count_pass} 个。")
if __name__ == "__main__":
# 根据你的 readmeMP数据在 input_preICSD在 input
# 这里默认读取 input你可以根据实际情况修改
read_files_check_basic("../data/input")