oscarz
36 lines
970 B
Python
36 lines
970 B
Python
import re
|
||
import os
|
||
import json
|
||
import time
|
||
import csv
|
||
from datetime import datetime
|
||
from urllib.parse import urlparse
|
||
import logging
|
||
import config
|
||
from urllib.parse import urlparse, urlunparse, parse_qs, urlencode
|
||
|
||
def replace_lang_param(url: str) -> str:
|
||
"""
|
||
将URL中的lang参数统一替换为'en'，支持路径中包含lang的情况
|
||
"""
|
||
parsed = urlparse(url)
|
||
|
||
# 处理路径中的lang参数（如 /ja/model/... 或 /en/model/...）
|
||
path_parts = parsed.path.split('/')
|
||
if len(path_parts) >= 2 and path_parts[1] in ['en', 'ja', 'zh']:
|
||
path_parts[1] = 'en' # 替换第二个路径段为'en'
|
||
new_path = '/'.join(path_parts)
|
||
else:
|
||
new_path = parsed.path
|
||
|
||
# 处理查询参数中的lang（如有）
|
||
query = parse_qs(parsed.query)
|
||
|
||
# 构建新URL
|
||
new_parsed = parsed._replace(
|
||
path=new_path,
|
||
query=urlencode(query, doseq=True)
|
||
)
|
||
return urlunparse(new_parsed)
|
||
|