CIRCL
/
PyCIRCLean
mirror of https://github.com/CIRCL/PyCIRCLean
2
0
Fork 0
Code Issues Releases Wiki Activity

fix: Make sure the test suite passes

pull/38/head
Raphaël Vinot 2022-11-11 14:59:41 +01:00
parent 2eff2b7127
commit cc081a921d
4 changed files with 25 additions and 31 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

31
filecheck/filecheck.py
View File

@ -11,7 +11,7 @@ import shutil
import time
import hashlib
from pathlib import Path
from typing import Dict, List, Tuple, Callable, Optional
from typing import Dict, List, Tuple, Callable, Optional, Union
import oletools.oleid # type: ignore
import olefile # type: ignore
@ -34,7 +34,7 @@ class Config:
mimes_rtf: Tuple[str, ...] = ('rtf', 'richtext',)
mimes_pdf: Tuple[str, ...] = ('pdf', 'postscript',)
mimes_xml: Tuple[str, ...] = ('xml',)
mimes_csv: Tuple[str, ...] = ('csv','text/csv')
mimes_csv: Tuple[str, ...] = ('csv', 'text/csv')
mimes_ms: Tuple[str, ...] = ('dosexec',)
mimes_compressed: Tuple[str, ...] = ('zip', 'rar', 'x-rar', 'bzip2', 'lzip', 'lzma', 'lzop',
'xz', 'compress', 'gzip', 'tar',)
@ -49,20 +49,20 @@ class Config:
mimes_metadata: Tuple[str, ...] = ('image/jpeg', 'image/tiff', 'image/png',)
# Mimetype aliases
aliases: Dict[str, str] = {
aliases: Dict[str, Union[str, List[str]]] = {
# Win executables
'application/x-msdos-program': 'application/x-dosexec',
'application/x-dosexec': 'application/x-msdos-program',
# Other apps with confusing mimetypes
'application/rtf': 'text/rtf',
'application/rar': 'application/x-rar',
'application/vnd.rar': 'application/x-rar',
'application/ogg': 'audio/ogg',
'audio/ogg': 'application/ogg'
}
# Mime Type / Extension fix. TODO: Doesn't quite work....????
mimetypes.add_type('text/plain','.csv',False)
mimetypes.add_type('text/csv','.csv',False)
mimetypes.add_type('text/plain', '.csv', False)
mimetypes.add_type('text/csv', '.csv', False)
mimetypes.add_type('application/vnd.apple.numbers', '.numbers', True)
mimetypes.add_type('application/vnd.apple.pages', '.pages', False)
mimetypes.add_type('application/vnd.apple.keynote', '.keynote', False)
@ -126,11 +126,11 @@ class Config:
# In [12]: mimetypes.guess_type('toot.tar.gz', strict=False)
# Out[12]: ('application/x-tar', 'gzip')
# It works as expected if you do mimetypes.guess_type('application/gzip', strict=False)
override_ext: Dict[str, str] = {'.gz': 'application/gzip'
, '.csv': 'text/csv' #,'text/plain' )
, '.numbers': 'application/vnd.apple.numbers' #,'application/zip')
, '.pages': 'application/vnd.apple.pages' #,'application/zip')
, '.keynote': 'application/vnd.apple.keynote' #,'application/zip')
override_ext: Dict[str, str] = {'.gz': 'application/gzip',
'.csv': 'text/csv', # ,'text/plain' )
'.numbers': 'application/vnd.apple.numbers', # ,'application/zip')
'.pages': 'application/vnd.apple.pages', # ,'application/zip')
'.keynote': 'application/vnd.apple.keynote' # ,'application/zip')
}
@ -209,6 +209,9 @@ class File(FileBase):
expected_mimetypes = [expected_mimetype]
if expected_mimetype in Config.aliases:
if isinstance(Config.aliases[expected_mimetype], list):
expected_mimetypes += Config.aliases[expected_mimetype]
else:
expected_mimetypes.append(Config.aliases[expected_mimetype])
if (encoding is None) and (os.path.getsize(self.src_path) == 0):
is_empty_file = True
@ -833,12 +836,12 @@ class KittenGroomerFileCheck(KittenGroomerBase):
Performs a depth-first traversal of the file tree.
"""
skipped_files = ( '.Trashes', '._.Trashes', '.DS_Store', '.fseventsd', '.Spotlight-V100','System Volume Information')
skipped_files = ('.Trashes', '._.Trashes', '.DS_Store', '.fseventsd', '.Spotlight-V100', 'System Volume Information')
queue = []
for path in sorted(os.listdir(root_dir_path), key=lambda x: str.lower(x)):
full_path = root_dir_path / path
filename = full_path.name
if not filename in skipped_files and not filename.startswith('._'):
if filename not in skipped_files and not filename.startswith('._'):
# check for symlinks first to prevent getting trapped in infinite symlink recursion
if full_path.is_symlink():
queue.append(full_path)
@ -849,7 +852,7 @@ class KittenGroomerFileCheck(KittenGroomerBase):
elif full_path.is_file():
queue.append(full_path)
else:
print("SKIPPING: "+filename)
print(f"SKIPPING: {filename}")
return queue
def run(self):

11
kittengroomer/helpers.py
View File

@ -243,17 +243,8 @@ class FileBase(object):
mimetype = 'inode/symlink'
self.set_property('symlink_path', os.readlink(file_path))
else:
try:
mt = magic.from_file(file_path, mime=True)
# libmagic always returns something, even if it's just 'data'
except UnicodeEncodeError as e:
self.add_error(e, '')
mt = None
try:
mimetype = mt.decode("utf-8") # type: ignore
except Exception:
# FIXME: what should the exception be if mimetype isn't utf-8?
mimetype = 'application/octet-stream'
mimetype = magic.from_file(file_path, mime=True)
return mimetype
def _split_mimetype(self, mimetype: str) -> Tuple[Union[str, None], Union[str, None]]:

6
poetry.lock generated
View File

@ -197,9 +197,9 @@ lxml = "*"
[package.source]
type = "git"
url = "https://github.com/grierforensics/officedissector.git"
url = "https://github.com/Rafiot/officedissector.git"
reference = "HEAD"
resolved_reference = "2059a5ba08fa139362e3936578f99c4da9a9b55d"
resolved_reference = "e3d9e8e155cc01180524c9b45b9fbec232206121"
[[package]]
name = "olefile"
@ -444,7 +444,7 @@ python-versions = "*"
[metadata]
lock-version = "1.1"
python-versions = "^3.8"
content-hash = "31838fedc10762d58f75f2822a5a557756f2043f62ed70e3e3bb4ef806958a9b"
content-hash = "dd6a8563c8232219ba31a071d662bee519adaf95317f66ae81b2022269609510"
[metadata.files]
attrs = [

2
pyproject.toml
View File

@ -26,7 +26,7 @@ pillow = "^9.3.0"
olefile = "^0.46"
oletools = "^0.60.1"
python-magic = "^0.4.27"
officedissector = {git = "https://github.com/grierforensics/officedissector.git"}
officedissector = {git = "https://github.com/Rafiot/officedissector.git"}
[tool.poetry.dev-dependencies]
tox = "^3.27.0"