X-Git-Url: https://git.phdru.name/?a=blobdiff_plain;f=mimedecode.py;h=f333a146d3ca689e4c9ad7de035f2cfd010203e9;hb=d8b1f5d6cb8bc4998899dabf35c1339c1d3572af;hp=48f6f7e4293963f756d079b5d3ac479a0e312567;hpb=49058d6afeb5ed7df0cda2815d41edfa2474dd17;p=mimedecode.git diff --git a/mimedecode.py b/mimedecode.py index 48f6f7e..f333a14 100755 --- a/mimedecode.py +++ b/mimedecode.py @@ -17,25 +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] [--save-headers|body|message mask] [-O dest_dir] [-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_if_needed(s, charset): - if charset and charset.lower() <> g.default_encoding: - s = unicode(s, charset, "replace").encode(g.default_encoding, "replace") + if charset and charset.lower() != g.default_encoding: + s = s.decode(charset, "replace").encode(g.default_encoding, "replace") return s @@ -43,9 +43,9 @@ def _decode_header(s): """Return a decoded string according to RFC 2047. NOTE: This is almost the same as email.Utils.decode. """ - import email.Header + import email.header - L = email.Header.decode_header(s) + L = email.header.decode_header(s) if not isinstance(L, list): # s wasn't decoded return s @@ -66,24 +66,24 @@ def _decode_header(s): def decode_header(msg, header): "Decode mail header (if exists) and put it back, if it was encoded" - if msg.has_key(header): + if header in msg: value = msg[header] new_value = _decode_header(value) - if new_value <> value: # do not bother to touch msg if not changed + if new_value != value: # do not bother to touch msg if not changed set_header(msg, header, new_value) def decode_header_param(msg, header, param): "Decode mail header's parameter (if exists) and put it back, if it was encoded" - if msg.has_key(header): + if header in msg: value = msg.get_param(param, header=header) if value: if isinstance(value, tuple): 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 + if new_value != value: # do not bother to touch msg if not changed msg.set_param(param, new_value, header) @@ -181,7 +181,7 @@ def decode_headers(msg): def set_header(msg, header, value): "Replace header" - if msg.has_key(header): + if header in msg: msg.replace_header(header, value) else: msg[header] = value @@ -211,8 +211,8 @@ def decode_body(msg, s): entries = mailcap.lookup(caps, content_type, "view") for entry in entries: - if entry.has_key('copiousoutput'): - if entry.has_key('test'): + if 'copiousoutput' in entry: + if 'test' in entry: test = mailcap.subst(entry['test'], content_type, filename) if test and os.system(test) != 0: continue @@ -222,9 +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() @@ -241,7 +241,7 @@ def recode_charset(msg, s): "Recode charset of the message to the default charset" save_charset = charset = msg.get_content_charset() - if charset and charset.lower() <> g.default_encoding: + if charset and charset.lower() != g.default_encoding: s = recode_if_needed(s, charset) content_type = msg.get_content_type() set_content_type(msg, content_type, g.default_encoding) @@ -321,8 +321,7 @@ 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('*/*') @@ -334,6 +333,8 @@ def decode_part(msg): 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'): @@ -354,7 +355,7 @@ def decode_part(msg): break elif content_type in g.ignore_mask: output_headers(msg) - output("\nMessage body of type %s skipped.\n" % ctype) + output("%sMessage body of type %s skipped.%s" % (os.linesep, ctype, os.linesep)) break elif content_type in g.error_mask: break @@ -365,42 +366,90 @@ def decode_part(msg): for content_type in masks: if content_type in g.save_headers_mask: _save_message(msg, outstring, save_headers=True, save_body=False) - elif content_type in g.save_body_mask: + if content_type in g.save_body_mask: _save_message(msg, outstring, save_headers=False, save_body=True) - elif content_type in g.save_message_mask: + 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 + 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 + + for content_type in masks: + if content_type in g.save_body_mask or \ + content_type in g.save_message_mask: + _out_l = [] + first_subpart = True + for subpart in msg.get_payload(): + if first_subpart: + first_subpart = False + else: + _out_l.append(os.linesep) + _out_l.append("--%s%s" % (boundary, os.linesep)) + _out_l.append(subpart.as_string()) + _out_l.append("%s--%s--%s" % (os.linesep, boundary, os.linesep)) + outstring = ''.join(_out_l) + break + else: + outstring = None + + 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) + 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) @@ -424,7 +473,7 @@ def open_output_file(filename): if create: os.makedirs(full_dir) try: - return open(fullpath, 'w') + return open(fullpath, 'wb') except: if create: os.removedirs(full_dir) @@ -459,7 +508,8 @@ 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 @@ -479,7 +529,7 @@ def get_opts(): try: options, arguments = getopt(sys.argv[1:], - 'hVcCDPH:f:d:p:r:R:b:B:e:i:t:O:o:', + '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=']) @@ -528,6 +578,8 @@ 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':