把项目转换为md知识库，提供给gpts使用
python
import os
import chardet
project_path='' # 项目目录路径
project_name='' # 项目名称
# 定义全局字典，映射文件后缀名到Markdown代码块的语言
language_mapping = {
'.java': 'java',
'.py': 'python',
'.js': 'javascript',
'.html': 'html',
'.md': 'markdown',
'.properties': 'properties',
'.xml': 'xml',
'.vue': 'vue',
'.tsx': 'tsx',
'.yml': 'yaml',
'.sql': 'sql',
'.json': 'json',
'.txt': 'text',
'.csv': 'csv',
'.ini': 'ini',
'.c': 'c',
'.cpp': 'cpp',
'.go': 'go',
'.php': 'php',
'.rb': 'ruby',
}
def detect_encoding(file_path):
"""检测文件编码"""
with open(file_path, 'rb') as file:
raw_data = file.read()
result = chardet.detect(raw_data)
return result['encoding']
def write_file_content_to_utf8(file_path, output_file):
"""读取文件并以UTF-8编码写入到txt文件中"""
# 获取文件后缀名
_, ext = os.path.splitext(file_path)
# 检查文件大小（20KB = 20480 字节）
# if os.path.getsize(file_path) > 91136 and ext != '.md':
if os.path.getsize(file_path) > 91136:
print(f"忽略大文件：{file_path}")
return
encoding = detect_encoding(file_path)
try:
with open(file_path, 'r', encoding=encoding) as file:
content = file.read()
relative_path = os.path.relpath(file_path, project_path) # 计算相对路径
output_file.write(f"## 文件：{project_name}\\{relative_path}\n\n") # Markdown标题
if not ext == '.md':
# 根据文件后缀名确定代码块的语言
language = language_mapping.get(ext, 'text') # 默认为'text'
output_file.write(f"```{language}\n") # 语法高亮
output_file.write(content)
if not ext == '.md':
output_file.write("\n```\n\n") # 结束代码块
else:
output_file.write("\n\n") # 结束代码块
except UnicodeDecodeError:
print(f"读取失败（编码不支持）：{file_path}")
def should_ignore(directory,ignored_directories):
"""检查是否应该忽略给定的目录"""
for ignored_dir in ignored_directories:
if ignored_dir in directory:
return True
return False
def process_directory(directory, output_file,ignored_directories):
"""递归处理目录下的所有文件和子目录"""
if should_ignore(directory,ignored_directories):
return # 忽略整个目录
for item in os.listdir(directory):
path = os.path.join(directory, item)
if os.path.isdir(path):
process_directory(path, output_file,ignored_directories)
elif any(path.endswith(suffix) for suffix in file_suffixes):
print('path===============',path)
write_file_content_to_utf8(path, output_file)
def create_knowledge_base(project_directory, output_path):
"""创建知识库文件"""
with open(output_path, 'w', encoding='utf-8') as output_file:
process_directory(project_directory, output_file,ignored_directories)
def create_knowledge_base(paths, output_path):
"""创建知识库文件"""
global project_path,project_name
with open(output_path, 'w', encoding='utf-8', errors='ignore') as output_file:
for project_directory in paths:
project_path=project_directory
project_name = os.path.basename(project_path) # 提取项目名称
output_file.write(f"# 项目：{project_name}\n\n") # Markdown标题
process_directory(project_directory, output_file,ignored_directories)
if __name__ == '__main__':
ignored_directories = ['target','.idea','node_modules','build','dist','out','logs','log','lib','js_sdk','uni_modules','nativeplugins','static','public']
file_suffixes=['.java', '.properties', '.yml', '.md', '.txt','.js','.html','.vue','.xml','.json','.sql','.csv','.py',
'.tsx','.ts','.less','.scss','.gradle'
]
project_dirs = [
'H:\\下载\edge浏览器下载\\spring-security-6.1.6',
]
output_path = 'spring-security-6.1.6.md' # 知识库文件的输出路径
create_knowledge_base(project_dirs, output_path)detect_encoding(file_path): 检测并返回文件的编码类型。
write_file_content_to_utf8(file_path, output_file): 读取文件内容，根据文件扩展名决定是否加上 Markdown 代码块的格式，并写入到指定的输出文件中。
should_ignore(directory, ignored_directories): 判断给定的目录是否应该被忽略。
process_directory(directory, output_file, ignored_directories): 递归处理给定目录下的所有文件和子目录，转换为 Markdown 格式。
create_knowledge_base(paths, output_path): 遍历给定的项目目录列表，为每个项目创建 Markdown 格式的知识库文件。