X-Git-Url: https://git.phdru.name/?a=blobdiff_plain;f=mimedecode.py;h=8d50b7d0d56453ab0d148cc0d66311c2d6b13081;hb=bc6318c1a631d838dfe6bfe1c94638c59ea626bf;hp=3a8cbbe533bbd29f4e7589b40ee44d8b635fa8bd;hpb=44f274a71a4b30c8d6f54b7ea374c8d7f8e95eed;p=mimedecode.git diff --git a/mimedecode.py b/mimedecode.py index 3a8cbbe..8d50b7d 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]) @@ -18,28 +17,25 @@ Broytman mimedecode.py version %s, %s def usage(code=0, errormsg=''): version(0) sys.stdout.write("""\ - Usage: %s [-h|--help] [-V|--version] [-cCDP] [-H|--host=hostname] [-f charset] [-d header1[,h2,...]|*[,-h1,...]] [-p header1[,h2,h3,...]:param1[,p2,p3,...]] [-r header1[,h2,...]|*[,-h1,...]] [-R header1[,h2,h3,...]:param1[,p2,p3,...]] [--set-header header:value] [--set-param header:param=value] [-Bbeit mask] [-o output_file] [input_file [output_file]] + Usage: %s [-h|--help] [-V|--version] [-cCDP] [-H|--host=hostname] [-f charset] [-d header1[,h2,...]|*[,-h1,...]] [-p header1[,h2,h3,...]:param1[,p2,p3,...]] [-r header1[,h2,...]|*[,-h1,...]] [-R header1[,h2,h3,...]:param1[,p2,p3,...]] [--set-header header:value] [--set-param header:param=value] [-Bbeit mask] [--save-headers|body|message mask] [-O dest_dir] [-o output_file] [input_file [output_file]] """ % me) if errormsg: - sys.stderr.write(errormsg + '\n') + sys.stderr.write(errormsg + os.linesep) sys.exit(code) def output_headers(msg): unix_from = msg.get_unixfrom() if unix_from: - output(unix_from + '\n') + output(unix_from + os.linesep) for key, value in msg.items(): - output("%s: %s\n" % (key, value)) - output("\n") # End of headers - + output("%s: %s%s" % (key, value, os.linesep)) + output(os.linesep) # 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 @@ -229,9 +222,9 @@ def decode_body(msg, s): if not command: return s - file = open(filename, 'w') - file.write(s) - file.close() + outfile = open(filename, 'wb') + outfile.write(s) + outfile.close() pipe = os.popen(command, 'r') s = pipe.read() @@ -266,6 +259,56 @@ def totext(msg, instring): output_headers(msg) output(s) + 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"), + ("Content-Type", "name"), + ): + 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_output_file(fname) + output = outfile.write + if save_headers: + output_headers(msg) + if save_body: + output(outstring) + outfile.close() + output = save_output def decode_part(msg): @@ -278,16 +321,20 @@ 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 + elif content_type in g.fully_ignore_mask: + return encoding = msg["Content-Transfer-Encoding"] if left_binary or encoding in (None, '', '7bit', '8bit', 'binary'): @@ -299,51 +346,82 @@ def decode_part(msg): for content_type in masks: if content_type in g.totext_mask: - totext(msg, outstring) - return + outstring = totext(msg, outstring) + break elif content_type in g.binary_mask or \ content_type in g.decoded_binary_mask: output_headers(msg) output(outstring) - return + break elif content_type in g.ignore_mask: output_headers(msg) - output("\nMessage body of type `%s' skipped.\n" % content_type) - return + output("%sMessage body of type %s skipped.%s" % (os.linesep, ctype, os.linesep)) + break elif content_type in g.error_mask: - raise ValueError, "content type `%s' prohibited" % content_type + break + else: + # Neither content type nor masks were listed - decode by default + outstring = totext(msg, outstring) - # Neither content type nor masks were listed - decode by default - totext(msg, outstring) + for content_type in masks: + if content_type in g.save_headers_mask: + _save_message(msg, outstring, save_headers=True, save_body=False) + if content_type in g.save_body_mask: + _save_message(msg, outstring, save_headers=False, save_body=True) + if 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" decode_headers(msg) + boundary = msg.get_boundary() + + masks = [] + ctype = msg.get_content_type() + if ctype: + masks.append(ctype) + mtype = ctype.split('/')[0] + masks.append(mtype + '/*') + masks.append('*/*') + + for content_type in masks: + if content_type in g.fully_ignore_mask: + return + elif content_type in g.ignore_mask: + output_headers(msg) + output("%sMessage body of type %s skipped.%s" % (os.linesep, ctype, os.linesep)) + if boundary: + output("%s--%s--%s" % (os.linesep, boundary, os.linesep)) + return + elif content_type in g.error_mask: + raise ValueError, "content type %s prohibited" % ctype + output_headers(msg) if msg.preamble: # Preserve the first part, it is probably not a RFC822-message output(msg.preamble) # Usually it is just a few lines of text (MIME warning) if msg.preamble is not None: - output("\n") + output(os.linesep) first_subpart = True - boundary = msg.get_boundary() - for subpart in msg.get_payload(): if boundary: if first_subpart: first_subpart = False else: - output("\n") - output("--%s\n" % boundary) + output(os.linesep) + output("--%s%s" % (boundary, os.linesep)) # Recursively decode all parts of the subpart decode_message(subpart) if boundary: - output("\n--%s--\n" % boundary) + output("%s--%s--%s" % (os.linesep, boundary, os.linesep)) if msg.epilogue: output(msg.epilogue) @@ -360,6 +438,19 @@ 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, 'wb') + except: + if create: + os.removedirs(full_dir) + + class GlobalOptions: from m_lib.defenc import default_encoding recode_charset = 1 # recode charset of message body @@ -389,11 +480,18 @@ class GlobalOptions: totext_mask = [] # A list of content-types to decode binary_mask = [] # A list of content-types to pass through decoded_binary_mask = [] # A list of content-types to pass through (content-transfer-decoded) - ignore_mask = [] # Ignore (skip, do not decode and do not include into output) + ignore_mask = [] # Ignore (do not decode and do not include into output) but output a warning instead of the body + fully_ignore_mask = [] # Completely ignore - no headers, no body, no warning error_mask = [] # Raise error if encounter one of these + save_counter = 0 + save_headers_mask = [] + save_body_mask = [] + save_message_mask = [] + input_filename = None output_filename = None + destination_dir = os.curdir g = GlobalOptions @@ -403,8 +501,10 @@ def get_opts(): try: options, arguments = getopt(sys.argv[1:], - 'hVcCDPH:f:d:p:r:R:b:B:e:i:t:o:', - ['help', 'version', 'host=', 'set-header=', 'set-param=']) + 'hVcCDPH:f:d:p:r:R:b:B:e:I:i:t:O:o:', + ['help', 'version', 'host=', + 'save-headers=', 'save-body=', 'save-message=', + 'set-header=', 'set-param=']) except GetoptError: usage(1) @@ -450,10 +550,20 @@ def get_opts(): g.binary_mask.append(value) elif option == '-b': g.decoded_binary_mask.append(value) + elif option == '-I': + g.fully_ignore_mask.append(value) elif option == '-i': g.ignore_mask.append(value) elif option == '-e': g.error_mask.append(value) + elif option == '--save-headers': + g.save_headers_mask.append(value) + elif option == '--save-body': + g.save_body_mask.append(value) + elif option == '--save-message': + g.save_message_mask.append(value) + elif option == '-O': + g.destination_dir = value elif option == '-o': g.output_filename = value else: @@ -470,7 +580,7 @@ if __name__ == "__main__": g.input_filename = '-' infile = sys.stdin if g.output_filename: - outfile = open(g.output_filename, 'w') + outfile = open_output_file(g.output_filename) else: g.output_filename = '-' outfile = sys.stdout @@ -483,7 +593,7 @@ if __name__ == "__main__": infile = open(arguments[0], 'r') if la == 1: if g.output_filename: - outfile = open(g.output_filename, 'w') + outfile = open_output_file(g.output_filename) else: g.output_filename = '-' outfile = sys.stdout @@ -495,7 +605,7 @@ if __name__ == "__main__": outfile = sys.stdout else: g.output_filename = arguments[1] - outfile = open(arguments[1], 'w') + outfile = open_output_file(g.output_filename) else: usage(1, 'Too many arguments') @@ -511,10 +621,11 @@ 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: