zhuchaowe/maxkb
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

fix: garbled zip import file names (#2747)

Browse Source
This commit is contained in:
shaohuzhang1 2025-03-31 16:22:39 +08:00 committed by GitHub
parent a2b6620b10
commit 9750c6d605
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
1 changed files with 12 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

12
apps/common/handle/impl/zip_split_handle.py
View File

@ -14,6 +14,7 @@ import zipfile
from typing import List from typing import List
from urllib.parse import urljoin from urllib.parse import urljoin
from charset_normalizer import detect
from django.db.models import QuerySet from django.db.models import QuerySet
from common.handle.base_split_handle import BaseSplitHandle from common.handle.base_split_handle import BaseSplitHandle
@ -100,6 +101,15 @@ def get_image_list(result_list: list, zip_files: List[str]):
return image_file_list return image_file_list
def get_file_name(file_name):
try:
file_name_code = file_name.encode('cp437')
charset = detect(file_name_code)['encoding']
return file_name_code.decode(charset)
except Exception as e:
return file_name
def filter_image_file(result_list: list, image_list): def filter_image_file(result_list: list, image_list):
image_source_file_list = [image.get('source_file') for image in image_list] image_source_file_list = [image.get('source_file') for image in image_list]
return [r for r in result_list if not image_source_file_list.__contains__(r.get('name', ''))] return [r for r in result_list if not image_source_file_list.__contains__(r.get('name', ''))]
@ -121,6 +131,8 @@ class ZipSplitHandle(BaseSplitHandle):
with zip_ref.open(file) as f: with zip_ref.open(file) as f:
# 对文件内容进行处理 # 对文件内容进行处理
try: try:
# 处理一下文件名
f.name = get_file_name(f.name)
value = file_to_paragraph(f, pattern_list, with_filter, limit) value = file_to_paragraph(f, pattern_list, with_filter, limit)
if isinstance(value, list): if isinstance(value, list):
result = [*result, *value] result = [*result, *value]