|
1
|
|
|
#!/usr/bin/env python3 |
|
2
|
|
|
# -*- coding: utf-8 -*- |
|
3
|
|
|
|
|
4
|
|
|
""" |
|
5
|
|
|
Module to download and to detach/strip/remove attachments |
|
6
|
|
|
from e-mails on IMAP servers. |
|
7
|
|
|
""" |
|
8
|
|
|
|
|
9
|
|
|
from __future__ import print_function |
|
10
|
|
|
|
|
11
|
|
|
import email |
|
12
|
|
|
import email.mime.text |
|
13
|
|
|
import email.utils |
|
14
|
|
|
from email.parser import HeaderParser |
|
15
|
|
|
import hashlib |
|
16
|
|
|
import logging |
|
17
|
|
|
import os.path |
|
18
|
|
|
import re |
|
19
|
|
|
import shutil |
|
20
|
|
|
import tempfile |
|
21
|
|
|
import unicodedata |
|
22
|
|
|
import emlx |
|
23
|
|
|
|
|
24
|
|
|
# pylint: disable=R0801 |
|
25
|
|
|
__author__ = "Alexander Willner" |
|
26
|
|
|
__copyright__ = "Copyright 2020, Alexander Willner" |
|
27
|
|
|
__credits__ = ["github.com/guido4000", |
|
28
|
|
|
"github.com/halteproblem", "github.com/jamesridgway"] |
|
29
|
|
|
__license__ = "MIT" |
|
30
|
|
|
__version__ = "1.0.0" |
|
31
|
|
|
__maintainer__ = "Alexander Willner" |
|
32
|
|
|
__email__ = "[email protected]" |
|
33
|
|
|
__status__ = "Development" |
|
34
|
|
|
|
|
35
|
|
|
|
|
36
|
|
|
class MailboxCleanerMessage(): |
|
37
|
|
|
""" |
|
38
|
|
|
Class to represent an e-mail. |
|
39
|
|
|
""" |
|
40
|
|
|
|
|
41
|
|
|
_PLACEHOLDER = """ |
|
42
|
|
|
=========================================================== |
|
43
|
|
|
This message contained an attachment that was stripped out. |
|
44
|
|
|
The filename was: "%(filename)s". |
|
45
|
|
|
The size was: %(size)d KB. |
|
46
|
|
|
The type was: %(type)s. |
|
47
|
|
|
Tool: https://github.com/AlexanderWillner/MailboxCleanup |
|
48
|
|
|
=========================================================== |
|
49
|
|
|
""" |
|
50
|
|
|
|
|
51
|
|
|
def __init__(self, args): |
|
52
|
|
|
self.args = args |
|
53
|
|
|
|
|
54
|
|
|
def download_and_detach_attachments(self, msg): |
|
55
|
|
|
"""Download attachments and remove them from the mail.""" |
|
56
|
|
|
|
|
57
|
|
|
modified = False |
|
58
|
|
|
|
|
59
|
|
|
# Iterate over each part of the email |
|
60
|
|
|
for part in msg.walk(): |
|
61
|
|
|
if self.is_non_detachable_part(part): |
|
62
|
|
|
continue |
|
63
|
|
|
success = self.download_attachment(part) |
|
64
|
|
|
if success: |
|
65
|
|
|
self.detach_attachment(part) |
|
66
|
|
|
modified = True |
|
67
|
|
|
|
|
68
|
|
|
return modified |
|
69
|
|
|
|
|
70
|
|
|
def is_non_detachable_part(self, part): |
|
71
|
|
|
"""Only process certain types and sizes of attachments.""" |
|
72
|
|
|
|
|
73
|
|
|
msg_size = len(str(part)) / 1024 |
|
74
|
|
|
logging.debug(' Part\t: %d KB / %d KB (type: %s)', |
|
75
|
|
|
msg_size, self.args.max_size, |
|
76
|
|
|
part.get_content_maintype()) |
|
77
|
|
|
|
|
78
|
|
|
return part.get_content_maintype() == 'multipart' or \ |
|
79
|
|
|
part.get('Content-Disposition') is None or \ |
|
80
|
|
|
msg_size <= self.args.max_size |
|
81
|
|
|
|
|
82
|
|
|
def download_attachment(self, part) -> bool: |
|
83
|
|
|
"""Download the attachment from a part of an email.""" |
|
84
|
|
|
|
|
85
|
|
|
if self.args.skip_download: |
|
86
|
|
|
logging.info(' Downloading\t: skipped (disabled)') |
|
87
|
|
|
return True |
|
88
|
|
|
|
|
89
|
|
|
file_attached = self.convert_filename(part.get_filename()) |
|
90
|
|
|
|
|
91
|
|
|
if file_attached == "unknown": |
|
92
|
|
|
logging.warning('Warning\t: Unknown attachment ' |
|
93
|
|
|
'(skipping this attachment)') |
|
94
|
|
|
return False |
|
95
|
|
|
|
|
96
|
|
|
if not os.path.exists(self.args.target): |
|
97
|
|
|
os.mkdir(self.args.target) |
|
98
|
|
|
with tempfile.NamedTemporaryFile() as file_temp: |
|
99
|
|
|
logging.info(' Downloading\t: "%s" (%s)', |
|
100
|
|
|
file_attached, part.get_content_maintype()) |
|
101
|
|
|
logging.debug(' Downloading\t: To "%s"', file_temp.name) |
|
102
|
|
|
payload = part.get_payload(decode=True) |
|
103
|
|
|
file_temp.write(payload) |
|
104
|
|
|
self._copy_file(file_temp.name, file_attached) |
|
105
|
|
|
|
|
106
|
|
|
return True |
|
107
|
|
|
|
|
108
|
|
|
def _copy_file(self, source, target_name, iterator=0): |
|
109
|
|
|
"""Copy file, check for duplicates via hash value.""" |
|
110
|
|
|
|
|
111
|
|
|
target_base, target_extension = os.path.splitext(target_name) |
|
112
|
|
|
if iterator > 0: |
|
113
|
|
|
target_base = target_base + "-" + str(iterator) |
|
114
|
|
|
target = os.path.join(self.args.target, target_base + target_extension) |
|
115
|
|
|
if iterator == 0: |
|
116
|
|
|
logging.debug(' Moving\t: From "%s" to "%s".', source, target) |
|
117
|
|
|
|
|
118
|
|
|
if not os.path.isfile(target): |
|
119
|
|
|
shutil.copy2(source, target) |
|
120
|
|
|
else: |
|
121
|
|
|
source_hash = MailboxCleanerMessage.get_hash(source) |
|
122
|
|
|
target_hash = MailboxCleanerMessage.get_hash(target) |
|
123
|
|
|
if source_hash != target_hash: |
|
124
|
|
|
if iterator == 0: |
|
125
|
|
|
logging.debug( |
|
126
|
|
|
' Conflict\t: Resolving same file / other hash...') |
|
127
|
|
|
self._copy_file(source, target_name, iterator + 1) |
|
128
|
|
|
else: |
|
129
|
|
|
logging.debug(' Moving\t: Already exists (same hash)') |
|
130
|
|
|
|
|
131
|
|
|
def process_directory(self, handler, folder=None): |
|
132
|
|
|
"""Upload messages from a local directory.""" |
|
133
|
|
|
|
|
134
|
|
|
source = self.args.upload if folder is None else folder |
|
135
|
|
|
if os.path.isfile(source): |
|
136
|
|
|
filenames = [os.path.dirname(source)] |
|
137
|
|
|
else: |
|
138
|
|
|
filenames = os.listdir(source) |
|
139
|
|
|
|
|
140
|
|
|
for i, filename in enumerate(filenames, start=1): |
|
141
|
|
|
filename = os.path.join(source, filename) |
|
142
|
|
|
|
|
143
|
|
|
# Recursive walker |
|
144
|
|
|
if os.path.isdir(filename): |
|
145
|
|
|
self.process_directory(handler, filename) |
|
146
|
|
|
|
|
147
|
|
|
# Only take eml files into account |
|
148
|
|
|
if not filename.lower().endswith(".eml") and\ |
|
149
|
|
|
not filename.lower().endswith(".emlx"): |
|
150
|
|
|
continue |
|
151
|
|
|
|
|
152
|
|
|
logging.warning('Files\t: %d / %d', i, len(filenames)) |
|
153
|
|
|
|
|
154
|
|
|
with open(filename) as filepointer: |
|
155
|
|
|
# Specific handling of emlx files |
|
156
|
|
|
if filename.lower().endswith(".emlx"): |
|
157
|
|
|
msg = emlx.read(filename) |
|
158
|
|
|
else: |
|
159
|
|
|
msg = email.message_from_file(filepointer) |
|
160
|
|
|
|
|
161
|
|
|
# Logging |
|
162
|
|
|
msg_subject = self.get_subject(msg) |
|
163
|
|
|
logging.warning(' File\t: %s (%s)', filename, msg_subject) |
|
164
|
|
|
|
|
165
|
|
|
# Remove attachments |
|
166
|
|
|
self.download_and_detach_attachments(msg) |
|
167
|
|
|
|
|
168
|
|
|
# Post process message (e.g. upload or save it) |
|
169
|
|
|
handler(msg) |
|
170
|
|
|
|
|
171
|
|
|
@staticmethod |
|
172
|
|
|
def detach_attachment(msg): |
|
173
|
|
|
"""Replace large attachment with dummy text.""" |
|
174
|
|
|
|
|
175
|
|
|
# Get message details |
|
176
|
|
|
msg_content = msg.get_content_type() |
|
177
|
|
|
msg_filename = MailboxCleanerMessage.convert_filename( |
|
178
|
|
|
msg.get_filename()) |
|
179
|
|
|
msg_size = len(str(msg)) / 1024 |
|
180
|
|
|
msg_type = msg.get_content_disposition() |
|
181
|
|
|
|
|
182
|
|
|
# Remove some old headers |
|
183
|
|
|
del msg['Content-Transfer-Encoding'] |
|
184
|
|
|
del msg['Content-Disposition'] |
|
185
|
|
|
del msg['Content-Description'] |
|
186
|
|
|
for k, _v in msg.get_params()[1:]: |
|
187
|
|
|
msg.del_param(k) |
|
188
|
|
|
|
|
189
|
|
|
# Make sure different clients visualize the removed content properly |
|
190
|
|
|
msg.set_type('text/plain') |
|
191
|
|
|
msg.set_charset('utf-8') |
|
192
|
|
|
if msg_type == 'attachment': |
|
193
|
|
|
msg.add_header('Content-Disposition', 'inline') |
|
194
|
|
|
else: |
|
195
|
|
|
msg.add_header('Content-Disposition', 'attachment', |
|
196
|
|
|
filename='removed-%s.txt' % msg_filename) |
|
197
|
|
|
msg.add_header('Content-Description', |
|
198
|
|
|
'removed-%s.txt' % msg_filename) |
|
199
|
|
|
|
|
200
|
|
|
# Replace content |
|
201
|
|
|
msg_details = dict(type=msg_content, |
|
202
|
|
|
filename=msg_filename, |
|
203
|
|
|
size=msg_size) |
|
204
|
|
|
msg_placeholder = MailboxCleanerMessage._PLACEHOLDER % msg_details |
|
205
|
|
|
msg_placeholder = email.mime.text.MIMEText(msg_placeholder, |
|
206
|
|
|
'text', 'utf-8') |
|
207
|
|
|
msg.set_payload(msg_placeholder.get_payload()) |
|
208
|
|
|
|
|
209
|
|
|
@staticmethod |
|
210
|
|
|
def get_uid(message) -> str: |
|
211
|
|
|
"""Get UID of message.""" |
|
212
|
|
|
|
|
213
|
|
|
parser = HeaderParser() |
|
214
|
|
|
header = parser.parsestr(message.as_string()) |
|
215
|
|
|
uid = email.utils.parseaddr(header['message-id']) |
|
216
|
|
|
return uid[1] |
|
217
|
|
|
|
|
218
|
|
|
@staticmethod |
|
219
|
|
|
def get_subject(message) -> str: |
|
220
|
|
|
"""Get shortened message subject for visualization.""" |
|
221
|
|
|
|
|
222
|
|
|
if 'subject' in message: |
|
223
|
|
|
subject = message['subject'] |
|
224
|
|
|
else: |
|
225
|
|
|
subject = "unknown" # very rarely messages have no subject |
|
226
|
|
|
subject, encoding = email.header.decode_header(subject)[0] |
|
227
|
|
|
encoding = 'utf-8' if encoding is None else encoding |
|
228
|
|
|
subject = subject.decode(encoding, errors='replace')\ |
|
229
|
|
|
if hasattr(subject, 'decode') else subject |
|
230
|
|
|
subject = subject[:75] + (subject[75:] and '...') |
|
231
|
|
|
subject = subject.replace('\r\n', '') |
|
232
|
|
|
subject = subject.replace('\t', ' ') |
|
233
|
|
|
|
|
234
|
|
|
return subject |
|
235
|
|
|
|
|
236
|
|
|
@staticmethod |
|
237
|
|
|
def get_hash(filename: str) -> str: |
|
238
|
|
|
"""Get hash from filename to detect duplicates.""" |
|
239
|
|
|
|
|
240
|
|
|
hash_value = hashlib.sha256() |
|
241
|
|
|
with open(filename, "rb") as file: |
|
242
|
|
|
for byte_block in iter(lambda: file.read(4096), b""): |
|
243
|
|
|
hash_value.update(byte_block) |
|
244
|
|
|
return hash_value.hexdigest() |
|
245
|
|
|
|
|
246
|
|
|
@staticmethod |
|
247
|
|
|
def slugify_filename(value): |
|
248
|
|
|
"""Make sure attachments contain only valid characters.""" |
|
249
|
|
|
|
|
250
|
|
|
value = str(value) |
|
251
|
|
|
value = unicodedata.normalize('NFKC', value) |
|
252
|
|
|
value = re.sub(r'[^.\w\s-]', '_', value) |
|
253
|
|
|
return value |
|
254
|
|
|
|
|
255
|
|
|
@staticmethod |
|
256
|
|
|
def convert_filename(file_struct) -> str: |
|
257
|
|
|
"""Decode the name of some attachments.""" |
|
258
|
|
|
|
|
259
|
|
|
filename = 'unknown' |
|
260
|
|
|
if file_struct is not None: |
|
261
|
|
|
file_struct = email.header.decode_header(file_struct)[0] |
|
262
|
|
|
encoding = file_struct[1] |
|
263
|
|
|
if encoding is not None: |
|
264
|
|
|
filename = file_struct[0].decode(encoding) |
|
265
|
|
|
else: |
|
266
|
|
|
filename = file_struct[0] |
|
267
|
|
|
|
|
268
|
|
|
return MailboxCleanerMessage.slugify_filename(filename) |
|
269
|
|
|
|
AlexanderWillner /
MailboxCleanup
