1
|
|
|
#! /usr/bin/env python |
2
|
|
|
# -*- coding: utf-8 -*- |
3
|
|
|
# Compatible with python3 and python2 (tested with at least 2.4) |
4
|
|
|
|
5
|
|
|
# flake8: noqa |
6
|
|
|
# pylint: skip-file |
7
|
|
|
# Originally from https://github.com/LRGH/emlx2eml |
8
|
|
|
|
9
|
|
|
import sys |
10
|
|
|
import os |
11
|
|
|
import logging |
12
|
|
|
import struct |
13
|
|
|
import email |
14
|
|
|
import base64 |
15
|
|
|
import mimetypes |
16
|
|
|
|
17
|
|
|
log = logging.getLogger("emlx2eml") |
18
|
|
|
console_handler = logging.StreamHandler() |
19
|
|
|
console_handler.setFormatter(logging.Formatter("%(levelname)-5s: %(message)s")) |
20
|
|
|
log.addHandler(console_handler) |
21
|
|
|
log.setLevel(logging.DEBUG) |
22
|
|
|
# log.setLevel(logging.ERROR) |
23
|
|
|
|
24
|
|
|
|
25
|
|
|
def find_emlx(input): |
26
|
|
|
if os.path.islink(input): |
27
|
|
|
return [] |
28
|
|
|
elif os.path.isdir(input): |
29
|
|
|
files = [] |
30
|
|
|
for x in os.listdir(input): |
31
|
|
|
files += find_emlx(os.path.join(input, x)) |
32
|
|
|
return files |
33
|
|
|
elif input.endswith(".emlx"): |
34
|
|
|
return [input] |
35
|
|
|
else: |
36
|
|
|
return [] |
37
|
|
|
|
38
|
|
|
|
39
|
|
|
# Some definitions, to enforce compatibility with python2 and python3 |
40
|
|
|
newline = struct.pack("B", 10) |
41
|
|
|
if sys.version_info[0] == 2: |
42
|
|
|
message_from_bytes = email.message_from_string |
43
|
|
|
def message_as_bytes(msg): return msg.as_string(unixfrom=True) |
44
|
|
|
else: |
45
|
|
|
message_from_bytes = email.message_from_bytes |
46
|
|
|
def message_as_bytes(msg): return msg.as_bytes(unixfrom=True) |
47
|
|
|
|
48
|
|
|
|
49
|
|
|
def copy_emlx(emlx, out_dir): |
50
|
|
|
# Get the numeric id |
51
|
|
|
id = get_numeric_id(emlx) |
52
|
|
|
|
53
|
|
|
# Create output file |
54
|
|
|
if not os.path.exists(out_dir): |
55
|
|
|
os.mkdir(out_dir) |
56
|
|
|
eml = os.path.join(out_dir, id+".eml") |
57
|
|
|
log.debug("Extract %s to %s", emlx, eml) |
58
|
|
|
if os.path.exists(eml): |
59
|
|
|
log.error("%s already exists", eml) |
60
|
|
|
return False |
61
|
|
|
# Parse the EMLX file |
62
|
|
|
msg = parse_emlx(emlx) |
63
|
|
|
msg.set_unixfrom("From emlx2eml Thu Apr 19 00:00:00 2012") |
64
|
|
|
# TODO: generate relevant values for unixfrom |
65
|
|
|
open(eml, "wb").write(message_as_bytes(msg)) |
66
|
|
|
|
67
|
|
|
|
68
|
|
|
def get_numeric_id(filename): |
69
|
|
|
id = os.path.basename(filename) |
70
|
|
|
assert(id.endswith(".emlx")) |
71
|
|
|
id = id[:-5] |
72
|
|
|
if id.endswith(".partial"): |
73
|
|
|
id = id[:-8] |
74
|
|
|
return id |
75
|
|
|
|
76
|
|
|
|
77
|
|
|
def parse_emlx(filename): |
78
|
|
|
# Read file |
79
|
|
|
content = open(filename, "rb").read() |
80
|
|
|
|
81
|
|
|
# Extract parts |
82
|
|
|
eol = content.find(newline) |
83
|
|
|
length = int(content[:eol]) |
84
|
|
|
body = content[eol+1:eol+1+length] |
85
|
|
|
# TODO: parse the content of 'plist', e.g. using plistlib |
86
|
|
|
# plist = content[eol+1+length:] |
87
|
|
|
msg = message_from_bytes(body) |
88
|
|
|
|
89
|
|
|
# Find where attachments may be |
90
|
|
|
id = get_numeric_id(filename) |
91
|
|
|
attach_dir = os.path.dirname(filename) |
92
|
|
|
if attach_dir == "": |
93
|
|
|
attach_dir = "." |
94
|
|
|
attach_dir += "/../Attachments/" + id |
95
|
|
|
|
96
|
|
|
# Make complete eml |
97
|
|
|
parse_msg(attach_dir, msg, []) |
98
|
|
|
return msg |
99
|
|
|
|
100
|
|
|
|
101
|
|
|
def parse_msg(attach_dir, msg, depth): |
102
|
|
|
log.debug("%sPART %s %r of type %s", " "*len(depth), |
103
|
|
|
".".join([str(_+1) for _ in depth]), msg, msg.get_content_type()) |
104
|
|
|
if msg.is_multipart(): |
105
|
|
|
for idx, part in enumerate(msg.get_payload()): |
106
|
|
|
parse_msg(attach_dir, part, depth+[idx]) |
107
|
|
|
include_attachment(attach_dir, part, depth+[idx]) |
108
|
|
|
|
109
|
|
|
|
110
|
|
|
# When the attachment has no explicit filename, Mail.app generates a name |
111
|
|
|
# which we want to guess. |
112
|
|
|
base_filename = u"Mail Attachment" |
113
|
|
|
mimetypes.add_type('image/pjpeg', '.jpg', strict=True) |
114
|
|
|
mimetypes.add_type('image/jpg', '.jpg', strict=True) |
115
|
|
|
|
116
|
|
|
|
117
|
|
|
def include_attachment(attach_dir, part, depth): |
118
|
|
|
if "X-Apple-Content-Length" not in part: |
119
|
|
|
return |
120
|
|
|
file = part.get_filename() |
121
|
|
|
mime_type = part.get_content_type() |
122
|
|
|
if file is None: |
123
|
|
|
file = base_filename + mimetypes.guess_extension(mime_type) |
124
|
|
|
dirpath = attach_dir + "/" + ".".join([str(_+1) for _ in depth]) |
125
|
|
|
try: |
126
|
|
|
data = open(dirpath+"/"+file, "rb").read() |
127
|
|
|
except FileNotFoundError: |
128
|
|
|
log.error("%s Attachment '%s' not found in %s", |
129
|
|
|
" "*len(depth), file, dirpath) |
130
|
|
|
return |
131
|
|
|
log.debug("%s Attachment '%s' found", " "*len(depth), file) |
132
|
|
|
cte = part["Content-Transfer-Encoding"] |
133
|
|
|
if cte is None: |
134
|
|
|
pass |
135
|
|
|
elif cte == "base64": |
136
|
|
|
data = base64.b64encode(data) |
137
|
|
|
data = newline.join([data[i*76:(i+1)*76] |
138
|
|
|
for i in range(len(data)//76+1)]) |
139
|
|
|
else: |
140
|
|
|
log.error("Attachment dir is %s", attach_dir) |
141
|
|
|
log.error(" CTE %r", cte) |
142
|
|
|
log.error(" CD %r", part["Content-Disposition"]) |
143
|
|
|
part.set_payload(data) |
144
|
|
|
|
145
|
|
|
|
146
|
|
|
if __name__ == "__main__": |
147
|
|
|
try: |
148
|
|
|
input, out_dir = sys.argv[1:] |
149
|
|
|
except ValueError: |
150
|
|
|
print("Syntax: emlx2eml.py <source> <output_dir>") |
151
|
|
|
print(" <source> can be an EMLX file, or a directory that will") |
152
|
|
|
print(" be recursively searched for EMLX files.") |
153
|
|
|
sys.exit(1) |
154
|
|
|
log.debug("Input %s; Output %s", input, out_dir) |
155
|
|
|
for emlx in find_emlx(input): |
156
|
|
|
copy_emlx(emlx, out_dir) |
157
|
|
|
|