X-Git-Url: https://git.phdru.name/?a=blobdiff_plain;f=mimedecode.py;h=48a6aba65ff01450825798e128319e0479d4e917;hb=1a01d40ea2ce5272480a09e6af36ac6f7162b4e0;hp=9af561e90ddc4b87616c2d84f5041eae06be8ef9;hpb=6f1f365730f3a886ce31abe39cc4cc0f992206f0;p=mimedecode.git diff --git a/mimedecode.py b/mimedecode.py index 9af561e..48a6aba 100755 --- a/mimedecode.py +++ b/mimedecode.py @@ -1,10 +1,9 @@ #! /usr/bin/env python """Decode MIME message""" -from mimedecode_version import __version__, __author__, __copyright__, __license__ - import sys, os -import email +from mimedecode_version import __version__, \ + __author__, __copyright__, __license__ me = os.path.basename(sys.argv[0]) @@ -34,12 +33,9 @@ def output_headers(msg): output("\n") # End of headers -def recode(s, charset): - return unicode(s, charset, "replace").encode(g.default_encoding, "replace") - def recode_if_needed(s, charset): if charset and charset.lower() <> g.default_encoding: - s = recode(s, charset) + s = unicode(s, charset, "replace").encode(g.default_encoding, "replace") return s @@ -77,9 +73,6 @@ def decode_header(msg, header): set_header(msg, header, new_value) -def _decode_header_param(s): - return recode_if_needed(s[2], s[0]) - def decode_header_param(msg, header, param): "Decode mail header's parameter (if exists) and put it back, if it was encoded" @@ -87,7 +80,7 @@ def decode_header_param(msg, header, param): value = msg.get_param(param, header=header) if value: if isinstance(value, tuple): - new_value = _decode_header_param(value) + new_value = recode_if_needed(value[2], value[0]) else: new_value = _decode_header(value) if new_value <> value: # do not bother to touch msg if not changed @@ -269,6 +262,18 @@ def totext(msg, instring): return s +mimetypes = None + +def _guess_extension(ctype): + global mimetypes + if mimetypes is None: + import mimetypes + mimetypes.init() + user_mime_type = os.path.expanduser('~/.mime.types') + if os.path.exists(user_mime_type): + mimetypes._db.read(user_mime_type) + return mimetypes.guess_extension(ctype) + def _save_message(msg, outstring, save_headers=False, save_body=False): for header, param in ( ("Content-Disposition", "filename"), @@ -276,16 +281,27 @@ def _save_message(msg, outstring, save_headers=False, save_body=False): ): fname = msg.get_param(param, header=header) if fname: + if isinstance(fname, tuple): + fname = fname[2] # Do not recode if it isn't recoded yet + try: + for forbidden in chr(0), '/', '\\': + if forbidden in fname: + raise ValueError + except ValueError: + continue fname = '-' + fname break else: fname = '' g.save_counter += 1 fname = str(g.save_counter) + fname + if '.' not in fname: + ext = _guess_extension(msg.get_content_type()) + if ext: fname += ext global output save_output = output - outfile = open(os.path.join(g.destination_dir, fname), 'w') + outfile = open_output_file(fname) output = outfile.write if save_headers: output_headers(msg) @@ -305,14 +321,16 @@ def decode_part(msg): ctype = msg.get_content_type() if ctype: masks.append(ctype) - mtype = msg.get_content_maintype() - if mtype: + mtype = ctype.split('/')[0] masks.append(mtype + '/*') masks.append('*/*') left_binary = False for content_type in masks: - if content_type in g.binary_mask: + if content_type in g.totext_mask or \ + content_type in g.decoded_binary_mask: + break + elif content_type in g.binary_mask: left_binary = True break @@ -338,7 +356,7 @@ def decode_part(msg): output("\nMessage body of type %s skipped.\n" % ctype) break elif content_type in g.error_mask: - raise ValueError, "content type %s prohibited" % ctype + break else: # Neither content type nor masks were listed - decode by default outstring = totext(msg, outstring) @@ -351,6 +369,10 @@ def decode_part(msg): elif content_type in g.save_message_mask: _save_message(msg, outstring, save_headers=True, save_body=True) + for content_type in masks: + if content_type in g.error_mask: + raise ValueError, "content type %s prohibited" % ctype + def decode_multipart(msg): "Decode multipart" @@ -394,6 +416,19 @@ def decode_message(msg): output(msg.as_string()) +def open_output_file(filename): + fullpath = os.path.abspath(os.path.join(g.destination_dir, filename)) + full_dir = os.path.dirname(fullpath) + create = not os.path.isdir(full_dir) + if create: + os.makedirs(full_dir) + try: + return open(fullpath, 'w') + except: + if create: + os.removedirs(full_dir) + + class GlobalOptions: from m_lib.defenc import default_encoding recode_charset = 1 # recode charset of message body @@ -520,7 +555,7 @@ if __name__ == "__main__": g.input_filename = '-' infile = sys.stdin if g.output_filename: - outfile = open(os.path.join(g.destination_dir, g.output_filename), 'w') + outfile = open_output_file(g.output_filename) else: g.output_filename = '-' outfile = sys.stdout @@ -533,7 +568,7 @@ if __name__ == "__main__": infile = open(arguments[0], 'r') if la == 1: if g.output_filename: - outfile = open(os.path.join(g.destination_dir, g.output_filename), 'w') + outfile = open_output_file(g.output_filename) else: g.output_filename = '-' outfile = sys.stdout @@ -545,7 +580,7 @@ if __name__ == "__main__": outfile = sys.stdout else: g.output_filename = arguments[1] - outfile = open(os.path.join(g.destination_dir, g.output_filename), 'w') + outfile = open_output_file(g.output_filename) else: usage(1, 'Too many arguments') @@ -561,10 +596,11 @@ if __name__ == "__main__": g.outfile = outfile output = outfile.write + import email msg = email.message_from_file(infile) for header, value in g.set_header_value: - msg[header] = value + set_header(msg, header, value) for header, param, value in g.set_header_param: if header in msg: