175 lines
6.3 KiB
Python
175 lines
6.3 KiB
Python
#!/usr/bin/env python3
|
||
"""
|
||
源代码清理脚本 - 用于软件著作权申请
|
||
功能：移除C++源代码中的注释和空行，生成干净的源代码文档
|
||
"""
|
||
|
||
import os
|
||
import re
|
||
import shutil
|
||
from pathlib import Path
|
||
|
||
class SourceCodeCleaner:
|
||
def __init__(self, source_dir, output_dir):
|
||
self.source_dir = Path(source_dir)
|
||
self.output_dir = Path(output_dir)
|
||
self.file_extensions = ['.h', '.cpp', '.hpp', '.c', '.cc']
|
||
|
||
def clean_directory(self):
|
||
"""清理整个目录的源代码"""
|
||
if self.output_dir.exists():
|
||
shutil.rmtree(self.output_dir)
|
||
self.output_dir.mkdir(parents=True, exist_ok=True)
|
||
|
||
cleaned_files = []
|
||
total_lines = 0
|
||
|
||
# 遍历源目录
|
||
for root, dirs, files in os.walk(self.source_dir):
|
||
for file in files:
|
||
if any(file.endswith(ext) for ext in self.file_extensions):
|
||
source_file = Path(root) / file
|
||
relative_path = source_file.relative_to(self.source_dir)
|
||
output_file = self.output_dir / relative_path
|
||
|
||
# 创建输出目录
|
||
output_file.parent.mkdir(parents=True, exist_ok=True)
|
||
|
||
# 清理文件
|
||
lines_cleaned = self.clean_file(source_file, output_file)
|
||
total_lines += lines_cleaned
|
||
cleaned_files.append(str(relative_path))
|
||
|
||
return cleaned_files, total_lines
|
||
|
||
def clean_file(self, input_file, output_file):
|
||
"""清理单个文件，移除注释和空行"""
|
||
try:
|
||
with open(input_file, 'r', encoding='utf-8', errors='ignore') as f:
|
||
content = f.read()
|
||
|
||
# 移除多行注释 /* ... */
|
||
content = re.sub(r'/\*.*?\*/', '', content, flags=re.DOTALL)
|
||
|
||
# 移除单行注释 //
|
||
content = re.sub(r'//.*$', '', content, flags=re.MULTILINE)
|
||
|
||
# 移除空行
|
||
lines = content.split('\n')
|
||
cleaned_lines = []
|
||
|
||
for line in lines:
|
||
stripped_line = line.strip()
|
||
if stripped_line: # 非空行
|
||
cleaned_lines.append(line)
|
||
|
||
# 写入清理后的内容
|
||
with open(output_file, 'w', encoding='utf-8') as f:
|
||
f.write('\n'.join(cleaned_lines))
|
||
|
||
return len(cleaned_lines)
|
||
|
||
except Exception as e:
|
||
print(f"清理文件 {input_file} 时出错: {e}")
|
||
return 0
|
||
|
||
def generate_source_document(self, output_file):
|
||
"""生成源代码文档"""
|
||
all_cleaned_files = []
|
||
|
||
for root, dirs, files in os.walk(self.output_dir):
|
||
for file in files:
|
||
if any(file.endswith(ext) for ext in self.file_extensions):
|
||
file_path = Path(root) / file
|
||
all_cleaned_files.append(file_path)
|
||
|
||
# 按文件类型和重要性排序
|
||
all_cleaned_files.sort()
|
||
|
||
with open(output_file, 'w', encoding='utf-8') as f:
|
||
f.write("# 大单检测软件系统 清理后源代码文档\n\n")
|
||
f.write("## 说明\n")
|
||
f.write("本文档包含软件著作权申请所需的清理后源代码，已移除所有注释和空行。\n\n")
|
||
f.write("## 源代码文件列表\n\n")
|
||
|
||
for file_path in all_cleaned_files:
|
||
relative_path = file_path.relative_to(self.output_dir)
|
||
f.write(f"- {relative_path}\n")
|
||
|
||
f.write("\n## 清理后源代码内容\n\n")
|
||
f.write("=" * 80 + "\n\n")
|
||
|
||
for file_path in all_cleaned_files:
|
||
relative_path = file_path.relative_to(self.output_dir)
|
||
f.write(f"文件: {relative_path}\n")
|
||
f.write("=" * 60 + "\n")
|
||
|
||
try:
|
||
with open(file_path, 'r', encoding='utf-8', errors='ignore') as src_file:
|
||
content = src_file.read()
|
||
|
||
# 写入文件内容，每行添加行号
|
||
lines = content.split('\n')
|
||
for i, line in enumerate(lines, 1):
|
||
f.write(f"{i:4d}: {line}\n")
|
||
|
||
f.write("\n" + "=" * 60 + "\n\n")
|
||
|
||
except Exception as e:
|
||
f.write(f"读取文件时出错: {e}\n\n")
|
||
|
||
def main():
|
||
# # 配置路径
|
||
# source_dirs = [
|
||
# "QMainwindow",
|
||
# "Sqbase",
|
||
# "FTAPI",
|
||
# "common_structures",
|
||
# "data_processing",
|
||
# "network_communication",
|
||
# "core"
|
||
# ]
|
||
# 配置路径
|
||
source_dirs = [
|
||
"QMainwindow",
|
||
"Sqbase",
|
||
"common_structures",
|
||
"data_processing",
|
||
"network_communication",
|
||
"core"
|
||
]
|
||
|
||
output_base_dir = "cleaned_source_code"
|
||
source_document_file = "软著申请材料/清理后源代码文档.md"
|
||
|
||
print("开始清理源代码...")
|
||
|
||
all_cleaned_files = []
|
||
total_lines = 0
|
||
|
||
for source_dir in source_dirs:
|
||
if os.path.exists(source_dir):
|
||
print(f"清理目录: {source_dir}")
|
||
output_dir = os.path.join(output_base_dir, source_dir)
|
||
cleaner = SourceCodeCleaner(source_dir, output_dir)
|
||
cleaned_files, lines = cleaner.clean_directory()
|
||
all_cleaned_files.extend(cleaned_files)
|
||
total_lines += lines
|
||
print(f" 已清理 {len(cleaned_files)} 个文件，共 {lines} 行代码")
|
||
else:
|
||
print(f"警告: 目录 {source_dir} 不存在")
|
||
|
||
# 生成源代码文档
|
||
print("生成源代码文档...")
|
||
cleaner = SourceCodeCleaner(output_base_dir, output_base_dir) # 使用自身作为源
|
||
cleaner.generate_source_document(source_document_file)
|
||
|
||
print(f"\n清理完成!")
|
||
print(f"总文件数: {len(all_cleaned_files)}")
|
||
print(f"总代码行数: {total_lines}")
|
||
print(f"源代码文档已生成: {source_document_file}")
|
||
print(f"清理后的源代码保存在: {output_base_dir}")
|
||
|
||
if __name__ == "__main__":
|
||
main()
|