-#! /usr/local/bin/python -O
-"""Decode MIME message.
-
-Author: Oleg Broytmann <phd@phd.pp.ru>
-Copyright: (C) 2001-2002 PhiloSoft Design
-License: GPL
-"""
-
-__version__ = "1.1.7"
+#! /usr/bin/env python
+"""Decode MIME message"""
+from mimedecode_version import __version__, __author__, __copyright__, __license__
import sys, os
-import mimetools
+import email
try:
- from cStringIO import StringIO
+ from cStringIO import StringIO
except ImportError:
- from StringIO import StringIO
-
-
-import socket
-host_name = socket.gethostname()
+ from StringIO import StringIO
me = os.path.basename(sys.argv[0])
-def usage(code=0):
- sys.stdout.write("""\
-Usage: %s [-h|--help] [-V|--version] [-cCDP] [-f charset] [-d header] [-p header:param] [-beit mask] [filename]
-""" % me)
- sys.exit(code)
-
-
-def version():
- sys.stdout.write("""\
-BroytMann mimedecode.py version %s
-""" % __version__)
- sys.exit(0)
+def version(exit=1):
+ sys.stdout.write("""\
+Broytman mimedecode.py version %s, %s
+""" % (__version__, __copyright__))
+ if exit: sys.exit(0)
+def usage(code=0, errormsg=''):
+ version(0)
+ sys.stdout.write("""\
+Usage: %s [-h|--help] [-V|--version] [-cCDP] [-H|--host=hostname] [-f charset] [-d header] [-p header:param] [-beit mask] [-o output_file] [input_file [output_file]]
+""" % me)
+ if errormsg:
+ sys.stderr.write(errormsg + '\n')
+ sys.exit(code)
-def output(s):
- sys.stdout.write(s)
def output_headers(msg):
- if msg.unixfrom:
- output(msg.unixfrom)
- output("%s\n" % msg)
+ unix_from = msg.get_unixfrom()
+ if unix_from:
+ output(unix_from + '\n')
+ for key, value in msg.items():
+ output("%s: %s\n" % (key, value))
+ output("\n") # End of headers
def recode(s, charset):
- return unicode(s, charset, "replace").encode(GlobalOptions.default_charset, "replace")
-
-
-def recode2(s, charset):
- if charset and charset <> GlobalOptions.default_charset:
- charset = charset.lower()
- s = recode(s, charset)
- return s
-
-
-def getparam(msg, header, param):
- "Get parameter from the header; return the header without the parameter, parameter itself and rfc2231 flag"
-
- if not msg.has_key(header):
- return None, None, 0
-
- header = msg[header]
- parts = [part.strip() for part in header.split(';')]
-
- new_parts = [parts[0]] # The header itself
- del parts[0]
-
- new_value = None
- rfc2231_encoded = 0
-
- import re, rfc822
- rfc2231_continuation = re.compile("^%s\\*[0-9]+\\*?$" % param)
- rfc2231_header = []
-
- for part in parts:
- name, value = part.split('=', 1)
- # The code is incomplete. Continuations in rfc2231-encoded paramters
- # (header*1, header*2, etc) are not yet supported
- if (name == param) or (name == param + '*'):
- new_value = rfc822.unquote(value)
- rfc2231_encoded += (name <> param)
- elif rfc2231_continuation.match(name):
- rfc2231_header.append(rfc822.unquote(value))
- rfc2231_encoded = 1
- else:
- new_parts.append(part)
-
- if rfc2231_header:
- new_value = ''.join(rfc2231_header)
-
- if new_value is not None:
- return "; ".join(new_parts), new_value, rfc2231_encoded
-
- return None, None, 0
-
+ return unicode(s, charset, "replace").encode(gopts.default_encoding, "replace")
+
+def recode_if_needed(s, charset):
+ if charset and charset.lower() <> gopts.default_encoding:
+ s = recode(s, charset)
+ return s
+
+
+def _decode_header(s):
+ """Return a decoded string according to RFC 2047.
+ NOTE: This is almost the same as email.Utils.decode.
+ """
+ from types import ListType
+ import email.Header
+
+ L = email.Header.decode_header(s)
+ if not isinstance(L, ListType):
+ # s wasn't decoded
+ return s
+
+ rtn = []
+ for atom, charset in L:
+ if charset is None:
+ rtn.append(atom)
+ else:
+ rtn.append(recode_if_needed(atom, charset))
+ rtn.append(' ')
+ del rtn[-1] # remove the last space
+
+ # Now that we've decoded everything, we just need to join all the parts
+ # together into the final string.
+ return ''.join(rtn)
def decode_header(msg, header):
- "Decode mail header (if exists) and put it back, if it was encoded"
-
- if msg.has_key(header):
- value = msg[header]
- new_value = decode_rfc2047(value)
- if value <> new_value: # do not bother to touch msg if not changed
- 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"
+ "Decode mail header (if exists) and put it back, if it was encoded"
- if msg.has_key(header):
- new_value, pstr, rfc2231_encoded = getparam(msg, header, param)
- if pstr is not None:
- if rfc2231_encoded:
- new_str = decode_rfc2231(pstr)
- else:
- new_str = decode_rfc2047(pstr)
- if pstr <> new_str: # do not bother to touch msg if not changed
- msg[header] = "%s; %s=\"%s\"" % (new_value, param, new_str)
+ if msg.has_key(header):
+ value = msg[header]
+ new_value = _decode_header(value)
+ if new_value <> value: # do not bother to touch msg if not changed
+ set_header(msg, header, new_value)
-def decode_rfc2047(s):
- "Decode string according to rfc2047"
+def _decode_header_param(s):
+ return recode_if_needed(s[2], s[0])
- parts = s.split() # by whitespaces
- new_parts = []
- got_encoded = 0
-
- for s in parts:
- l = s.split('?')
-
- if l[0] <> '=' or l[4] <> '=': # assert correct format
- new_parts.append(' ')
- new_parts.append(s) # if not encoded - just put it into output
- got_encoded = 0
- continue
-
- if not got_encoded:
- new_parts.append(' ') # no space between encoded parts, one space otherwise
- got_encoded = 1
-
- charset = l[1].lower()
- encoding = l[2].lower()
- s = l[3]
-
- if '*' in charset:
- charset, language = charset.split('*', 1) # language ignored
-
- infile = StringIO(s)
- outfile = StringIO()
-
- if encoding == "b":
- from base64 import decode
- elif encoding == "q":
- from quopri import decode
- else:
- raise ValueError, "wrong encoding `%s' (expected 'b' or 'q')" % encoding
-
- decode(infile, outfile)
- s = outfile.getvalue()
-
- if charset == GlobalOptions.default_charset:
- new_parts.append(s) # do not recode
- continue
-
- s = recode(s, charset)
- new_parts.append(s)
-
- if new_parts and new_parts[0] == ' ':
- del new_parts[0]
- return ''.join(new_parts)
-
-
-def decode_rfc2231(s):
- "Decode string according to rfc2231"
+def decode_header_param(msg, header, param):
+ "Decode mail header's parameter (if exists) and put it back, if it was encoded"
- charset, language, s = s.split("'", 2) # language ignored
+ if msg.has_key(header):
+ value = msg.get_param(param, header=header)
+ if value:
+ from types import TupleType
+ if isinstance(value, TupleType):
+ new_value = _decode_header_param(value)
+ else:
+ new_value = _decode_header(value)
+ if new_value <> value: # do not bother to touch msg if not changed
+ msg.set_param(param, new_value, header)
- i = 0
- result = []
- while i < len(s):
- c = s[i]
- if c == '%': # hex
- i += 1
- c = chr(int(s[i:i+2], 16))
- i += 1
- result.append(c)
- i += 1
+def decode_headers(msg):
+ "Decode message headers according to global options"
- s = ''.join(result)
- s = recode2(s, charset)
- return s
+ for header in gopts.decode_headers:
+ decode_header(msg, header)
+ for header, param in gopts.decode_header_params:
+ decode_header_param(msg, header, param)
-def decode_headers(msg):
- "Decode message headers according to global options"
- for header in GlobalOptions.decode_headers:
- decode_header(msg, header)
+def set_header(msg, header, value):
+ "Replace header"
- for header, param in GlobalOptions.decode_header_params:
- decode_header_param(msg, header, param)
- if header.lower() == "content-type" and msg.has_key(header):
- # reparse type
- msg.typeheader = msg[header]
- msg.parsetype() # required for plist...
- msg.parseplist() #... and reparse decoded plist
+ if msg.has_key(header):
+ msg.replace_header(header, value)
+ else:
+ msg[header] = value
def set_content_type(msg, newtype, charset=None):
- plist = msg.getplist()
- if plist:
- if charset:
- newplist = []
- for p in plist:
- if p.split('=')[0] == "charset":
- p = "charset=%s" % charset
- newplist.append(p)
- plist = newplist
-
- elif charset:
- plist = ["charset=%s" % charset]
+ msg.set_type(newtype)
- else:
- plist = []
-
- if plist and plist[0]: plist.insert(0, '')
- msg["Content-Type"] = "%s%s" % (newtype, ";\n ".join(plist))
+ if charset:
+ msg.set_param("charset", charset, "Content-Type")
caps = None # Globally stored mailcap database; initialized only if needed
def decode_body(msg, s):
- "Decode body to plain text using first copiousoutput filter from mailcap"
-
- import mailcap, tempfile
-
- global caps
- if caps is None:
- caps = mailcap.getcaps()
-
- content_type = msg.gettype()
- filename = tempfile.mktemp()
- command = None
-
- entries = mailcap.lookup(caps, content_type, "view")
- for entry in entries:
- if entry.has_key('copiousoutput'):
- if entry.has_key('test'):
- test = mailcap.subst(entry['test'], content_type, filename)
- if test and os.system(test) != 0:
- continue
- command = mailcap.subst(entry["view"], content_type, filename)
- break
-
- if not command:
- return s
+ "Decode body to plain text using first copiousoutput filter from mailcap"
+
+ import mailcap, tempfile
+
+ global caps
+ if caps is None:
+ caps = mailcap.getcaps()
+
+ content_type = msg.get_content_type()
+ filename = tempfile.mktemp()
+ command = None
+
+ entries = mailcap.lookup(caps, content_type, "view")
+ for entry in entries:
+ if entry.has_key('copiousoutput'):
+ if entry.has_key('test'):
+ test = mailcap.subst(entry['test'], content_type, filename)
+ if test and os.system(test) != 0:
+ continue
+ command = mailcap.subst(entry["view"], content_type, filename)
+ break
- file = open(filename, 'w')
- file.write(s)
- file.close()
+ if not command:
+ return s
- pipe = os.popen(command, 'r')
- s = pipe.read()
- pipe.close()
- os.remove(filename)
+ file = open(filename, 'w')
+ file.write(s)
+ file.close()
- set_content_type(msg, "text/plain")
- msg["X-MIME-Body-Autoconverted"] = "from %s to text/plain by %s id %s" % (content_type, host_name, command.split()[0])
+ pipe = os.popen(command, 'r')
+ s = pipe.read()
+ pipe.close()
+ os.remove(filename)
- msg.maintype = "text"
- msg.subtype = "plain"
- msg.type = "text/plain"
+ set_content_type(msg, "text/plain")
+ msg["X-MIME-Autoconverted"] = "from %s to text/plain by %s id %s" % (content_type, gopts.host_name, command.split()[0])
- return s
+ return s
def recode_charset(msg, s):
- "Recode charset of the message to the default charset"
-
- save_charset = charset = msg.getparam("charset")
- if charset and charset <> GlobalOptions.default_charset:
- s = recode2(s, charset)
- content_type = msg.gettype()
- set_content_type(msg, content_type, GlobalOptions.default_charset)
- msg["X-MIME-Charset-Autoconverted"] = "from %s to %s by %s id %s" % (save_charset, GlobalOptions.default_charset, host_name, me)
- return s
+ "Recode charset of the message to the default charset"
+ save_charset = charset = msg.get_content_charset()
+ if charset and charset.lower() <> gopts.default_encoding:
+ s = recode_if_needed(s, charset)
+ content_type = msg.get_content_type()
+ set_content_type(msg, content_type, gopts.default_encoding)
+ msg["X-MIME-Autoconverted"] = "from %s to %s by %s id %s" % (save_charset, gopts.default_encoding, gopts.host_name, me)
+ return s
-def totext(msg, infile):
- "Convert infile (StringIO) content to text"
- if msg.getmaintype() == "multipart": # Recursively decode all parts of the multipart message
- newfile = StringIO("%s\n%s" % (msg, infile.getvalue()))
- decode_file(newfile)
- return
+def totext(msg, instring):
+ "Convert instring content to text"
- # Decode body and recode charset
- s = decode_body(msg, infile.getvalue())
- if GlobalOptions.recode_charset:
- s = recode_charset(msg, s)
+ # Decode body and recode charset
+ s = decode_body(msg, instring)
+ if gopts.recode_charset:
+ s = recode_charset(msg, s)
- output_headers(msg)
- output(s)
+ output_headers(msg)
+ output(s)
-def decode_part(msg, infile):
- "Decode one part of the message"
+def decode_part(msg):
+ "Decode one part of the message"
- encoding = msg.getencoding()
- outfile = StringIO()
+ decode_headers(msg)
+ encoding = msg["Content-Transfer-Encoding"]
- if encoding in ('', '7bit', '8bit', 'binary'):
- mimetools.copyliteral(infile, outfile)
- else: # Decode from transfer ecoding to text or binary form
- mimetools.decode(infile, outfile, encoding)
- msg["Content-Transfer-Encoding"] = "8bit"
- msg["X-MIME-Autoconverted"] = "from %s to 8bit by %s id %s" % (encoding, host_name, me)
+ if encoding in (None, '', '7bit', '8bit', 'binary'):
+ outstring = str(msg.get_payload())
+ else: # Decode from transfer ecoding to text or binary form
+ outstring = str(msg.get_payload(decode=1))
+ set_header(msg, "Content-Transfer-Encoding", "8bit")
+ msg["X-MIME-Autoconverted"] = "from %s to 8bit by %s id %s" % (encoding, gopts.host_name, me)
- decode_headers(msg)
+ # Test all mask lists and find what to do with this content type
+ masks = []
+ ctype = msg.get_content_type()
+ if ctype:
+ masks.append(ctype)
+ mtype = msg.get_content_maintype()
+ if mtype:
+ masks.append(mtype + '/*')
+ masks.append('*/*')
- # Test all mask lists and find what to do with this content type
+ for content_type in masks:
+ if content_type in gopts.totext_mask:
+ totext(msg, outstring)
+ return
+ elif content_type in gopts.binary_mask:
+ output_headers(msg)
+ output(outstring)
+ return
+ elif content_type in gopts.ignore_mask:
+ output_headers(msg)
+ output("\nMessage body of type `%s' skipped.\n" % content_type)
+ return
+ elif content_type in gopts.error_mask:
+ raise ValueError, "content type `%s' prohibited" % content_type
- for content_type in msg.gettype(), msg.getmaintype()+"/*", "*/*":
- if content_type in GlobalOptions.totext_mask:
- totext(msg, outfile)
- return
- elif content_type in GlobalOptions.binary_mask:
- output_headers(msg)
- output(outfile.getvalue())
- return
- elif content_type in GlobalOptions.ignore_mask:
- output_headers(msg)
- output("\nMessage body of type `%s' skipped.\n" % content_type)
- return
- elif content_type in GlobalOptions.error_mask:
- raise ValueError, "content type `%s' prohibited" % content_type
+ # Neither content type nor masks were listed - decode by default
+ totext(msg, outstring)
- # Neither content type nor masks were listed - decode by default
- totext(msg, outfile)
+def decode_file(infile):
+ "Decode the entire message"
-def decode_file(infile, seekable=1):
- "Decode the entire message"
+ msg = email.message_from_file(infile)
- m = mimetools.Message(infile)
- boundary = m.getparam("boundary")
+ if msg.is_multipart():
+ decode_headers(msg)
+ output_headers(msg)
- if not boundary:
- if not m.getheader("Content-Type"): # Not a message, just text - copy it literally
- output(infile.read())
+ 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)
- else: # Simple one-part message - decode it
- decode_part(m, infile)
+ boundary = msg.get_boundary()
- else: # MIME message - decode all parts; may be recursive
- decode_headers(m)
- output_headers(m)
+ for subpart in msg.get_payload():
+ output("\n--%s\n" % boundary)
- import multifile
- mf = multifile.MultiFile(infile, seekable)
- mf.push(boundary)
+ if subpart.is_multipart(): # Recursively decode all parts of the subpart
+ newfile = StringIO(subpart.as_string())
+ newfile.seek(0)
+ decode_file(newfile)
+ else:
+ decode_part(subpart)
- if not seekable: # Preserve the first part, it is probably not a RFC822-message
- output(mf.read()) # Usually it is just a few lines of text (MIME warning)
+ output("\n--%s--\n" % boundary)
- while 1:
- m = mimetools.Message(mf)
- decode_part(m, mf)
+ if msg.epilogue:
+ output(msg.epilogue)
- if not mf.next():
- break
- output("\n--%s\n" % boundary)
+ else:
+ if msg.has_key("Content-Type"): # Simple one-part message - decode it
+ decode_part(msg)
- mf.pop()
- output("\n--%s--\n" % boundary)
+ else: # Not a message, just text - copy it literally
+ output(msg.as_string())
class GlobalOptions:
- default_charset = sys.getdefaultencoding()
- recode_charset = 1 # recode charset of message body
-
- decode_headers = ["Subject", "From"] # A list of headers to decode
- decode_header_params = [("Content-Type", "name"),
- ("Content-Disposition", "filename")
- ] # A list of headers' parameters to decode
-
- totext_mask = [] # A list of content-types to decode
- binary_mask = [] # A list to pass through
- ignore_mask = [] # Ignore (skip, do not decode and do not include into output)
- error_mask = [] # Raise error if encounter one of these
-
-
-def init():
- from getopt import getopt, GetoptError
-
- try:
- options, arguments = getopt(sys.argv[1:], 'hVcCDPf:d:p:b:e:i:t:',
- ['help', 'version'])
- except GetoptError:
- usage(1)
-
- for option, value in options:
- if option == '-h':
- usage()
- elif option == '--help':
- usage()
- elif option == '-V':
- version()
- elif option == '--version':
- version()
- elif option == '-c':
- GlobalOptions.recode_charset = 1
- elif option == '-C':
- GlobalOptions.recode_charset = 0
- elif option == '-f':
- GlobalOptions.default_charset = value
- elif option == '-d':
- GlobalOptions.decode_headers.append(value)
- elif option == '-D':
- GlobalOptions.decode_headers = []
- elif option == '-p':
- GlobalOptions.decode_header_params.append(value.split(':', 1))
- elif option == '-P':
- GlobalOptions.decode_header_params = []
- elif option == '-t':
- GlobalOptions.totext_mask.append(value)
- elif option == '-b':
- GlobalOptions.binary_mask.append(value)
- elif option == '-i':
- GlobalOptions.ignore_mask.append(value)
- elif option == '-e':
- GlobalOptions.error_mask.append(value)
- else:
- usage(1)
-
- return arguments
+ from m_lib.defenc import default_encoding
+ recode_charset = 1 # recode charset of message body
+
+ host_name = None
+
+ decode_headers = ["From", "Subject"] # A list of headers to decode
+ decode_header_params = [
+ ("Content-Type", "name"),
+ ("Content-Disposition", "filename"),
+ ] # A list of headers' parameters to decode
+
+ totext_mask = [] # A list of content-types to decode
+ binary_mask = [] # A list to pass through
+ ignore_mask = [] # Ignore (skip, do not decode and do not include into output)
+ error_mask = [] # Raise error if encounter one of these
+
+ input_filename = None
+ output_filename = None
+
+gopts = GlobalOptions
+
+
+def get_opt():
+ from getopt import getopt, GetoptError
+
+ try:
+ options, arguments = getopt(sys.argv[1:], 'hVcCDPH:f:d:p:b:e:i:t:o:',
+ ['help', 'version', 'host'])
+ except GetoptError:
+ usage(1)
+
+ for option, value in options:
+ if option in ('-h', '--help'):
+ usage()
+ elif option in ('-V', '--version'):
+ version()
+ elif option == '-c':
+ gopts.recode_charset = 1
+ elif option == '-C':
+ gopts.recode_charset = 0
+ elif option in ('-H', '--host'):
+ gopts.host_name = value
+ elif option == '-f':
+ gopts.default_encoding = value
+ elif option == '-d':
+ gopts.decode_headers.append(value)
+ elif option == '-D':
+ gopts.decode_headers = []
+ elif option == '-p':
+ gopts.decode_header_params.append(value.split(':', 1))
+ elif option == '-P':
+ gopts.decode_header_params = []
+ elif option == '-t':
+ gopts.totext_mask.append(value)
+ elif option == '-b':
+ gopts.binary_mask.append(value)
+ elif option == '-i':
+ gopts.ignore_mask.append(value)
+ elif option == '-e':
+ gopts.error_mask.append(value)
+ elif option == '-o':
+ gopts.output_filename = value
+ else:
+ usage(1)
+
+ return arguments
if __name__ == "__main__":
- arguments = init()
-
- seekable = 0
- if len(arguments) == 0:
- infile = sys.stdin
- elif len(arguments) <> 1:
- usage(1)
- elif arguments[0] == '-':
- infile = sys.stdin
- else:
- infile = open(arguments[0], 'r')
- seekable = 1
-
- decode_file(infile, seekable)
+ arguments = get_opt()
+
+ la = len(arguments)
+ if la == 0:
+ gopts.input_filename = '-'
+ infile = sys.stdin
+ if gopts.output_filename:
+ outfile = open(gopts.output_filename, 'w')
+ else:
+ gopts.output_filename = '-'
+ outfile = sys.stdout
+ elif la in (1, 2):
+ if (arguments[0] == '-'):
+ gopts.input_filename = '-'
+ infile = sys.stdin
+ else:
+ gopts.input_filename = arguments[0]
+ infile = open(arguments[0], 'r')
+ if la == 1:
+ if gopts.output_filename:
+ outfile = open(gopts.output_filename, 'w')
+ else:
+ gopts.output_filename = '-'
+ outfile = sys.stdout
+ elif la == 2:
+ if gopts.output_filename:
+ usage(1, 'Too many output filenames')
+ if (arguments[1] == '-'):
+ gopts.output_filename = '-'
+ outfile = sys.stdout
+ else:
+ gopts.output_filename = arguments[1]
+ outfile = open(arguments[1], 'w')
+ else:
+ usage(1, 'Too many arguments')
+
+ if (infile is sys.stdin) and sys.stdin.isatty():
+ if (outfile is sys.stdout) and sys.stdout.isatty():
+ usage()
+ usage(1, 'Filtering from console is forbidden')
+
+ if not gopts.host_name:
+ import socket
+ gopts.host_name = socket.gethostname()
+
+ gopts.outfile = outfile
+ output = outfile.write
+
+ try:
+ decode_file(infile)
+ finally:
+ infile.close()
+ outfile.close()