]>
git.proxmox.com Git - mirror_edk2.git/blob - AppPkg/Applications/Python/Python-2.7.2/Lib/email/message.py
1 # Copyright (C) 2001-2006 Python Software Foundation
3 # Contact: email-sig@python.org
5 """Basic message object for the email package object model."""
13 from cStringIO
import StringIO
15 # Intrapackage imports
17 from email
import utils
18 from email
import errors
22 # Regular expression that matches `special' characters in parameters, the
23 # existence of which force quoting of the parameter value.
24 tspecials
= re
.compile(r
'[ \(\)<>@,;:\\"/\[\]\?=]')
28 def _splitparam(param
):
29 # Split header parameters. BAW: this may be too simple. It isn't
30 # strictly RFC 2045 (section 5.1) compliant, but it catches most headers
31 # found in the wild. We may eventually need a full fledged parser
33 a
, sep
, b
= param
.partition(';')
35 return a
.strip(), None
36 return a
.strip(), b
.strip()
38 def _formatparam(param
, value
=None, quote
=True):
39 """Convenience function to format and return a key=value pair.
41 This will quote the value if needed or if quote is true. If value is a
42 three tuple (charset, language, value), it will be encoded according
45 if value
is not None and len(value
) > 0:
46 # A tuple is used for RFC 2231 encoded parameter values where items
47 # are (charset, language, value). charset is a string, not a Charset
49 if isinstance(value
, tuple):
50 # Encode as per RFC 2231
52 value
= utils
.encode_rfc2231(value
[2], value
[0], value
[1])
53 # BAW: Please check this. I think that if quote is set it should
54 # force quoting even if not necessary.
55 if quote
or tspecials
.search(value
):
56 return '%s="%s"' % (param
, utils
.quote(value
))
58 return '%s=%s' % (param
, value
)
67 while end
> 0 and (s
.count('"', 0, end
) - s
.count('\\"', 0, end
)) % 2:
68 end
= s
.find(';', end
+ 1)
74 f
= f
[:i
].strip().lower() + '=' + f
[i
+1:].strip()
75 plist
.append(f
.strip())
80 def _unquotevalue(value
):
81 # This is different than utils.collapse_rfc2231_value() because it doesn't
82 # try to convert the value to a unicode. Message.get_param() and
83 # Message.get_params() are both currently defined to return the tuple in
84 # the face of RFC 2231 parameters.
85 if isinstance(value
, tuple):
86 return value
[0], value
[1], utils
.unquote(value
[2])
88 return utils
.unquote(value
)
93 """Basic message object.
95 A message object is defined as something that has a bunch of RFC 2822
96 headers and a payload. It may optionally have an envelope header
97 (a.k.a. Unix-From or From_ header). If the message is a container (i.e. a
98 multipart or a message/rfc822), then the payload is a list of Message
99 objects, otherwise it is a string.
101 Message objects implement part of the `mapping' interface, which assumes
102 there is exactly one occurrence of the header per message. Some headers
103 do in fact appear multiple times (e.g. Received) and for those headers,
104 you must use the explicit API to set or get all the headers. Not all of
105 the mapping methods are implemented.
109 self
._unixfrom
= None
112 # Defaults for multipart messages
113 self
.preamble
= self
.epilogue
= None
115 # Default content type
116 self
._default
_type
= 'text/plain'
119 """Return the entire formatted message as a string.
120 This includes the headers, body, and envelope header.
122 return self
.as_string(unixfrom
=True)
124 def as_string(self
, unixfrom
=False):
125 """Return the entire formatted message as a string.
126 Optional `unixfrom' when True, means include the Unix From_ envelope
129 This is a convenience method and may not generate the message exactly
130 as you intend because by default it mangles lines that begin with
131 "From ". For more flexibility, use the flatten() method of a
134 from email
.generator
import Generator
137 g
.flatten(self
, unixfrom
=unixfrom
)
140 def is_multipart(self
):
141 """Return True if the message consists of multiple parts."""
142 return isinstance(self
._payload
, list)
147 def set_unixfrom(self
, unixfrom
):
148 self
._unixfrom
= unixfrom
150 def get_unixfrom(self
):
151 return self
._unixfrom
154 # Payload manipulation.
156 def attach(self
, payload
):
157 """Add the given payload to the current payload.
159 The current payload will always be a list of objects after this method
160 is called. If you want to set the payload to a scalar object, use
161 set_payload() instead.
163 if self
._payload
is None:
164 self
._payload
= [payload
]
166 self
._payload
.append(payload
)
168 def get_payload(self
, i
=None, decode
=False):
169 """Return a reference to the payload.
171 The payload will either be a list object or a string. If you mutate
172 the list object, you modify the message's payload in place. Optional
173 i returns that index into the payload.
175 Optional decode is a flag indicating whether the payload should be
176 decoded or not, according to the Content-Transfer-Encoding header
179 When True and the message is not a multipart, the payload will be
180 decoded if this header's value is `quoted-printable' or `base64'. If
181 some other encoding is used, or the header is missing, or if the
182 payload has bogus data (i.e. bogus base64 or uuencoded data), the
183 payload is returned as-is.
185 If the message is a multipart and the decode flag is True, then None
189 payload
= self
._payload
190 elif not isinstance(self
._payload
, list):
191 raise TypeError('Expected list, got %s' % type(self
._payload
))
193 payload
= self
._payload
[i
]
195 if self
.is_multipart():
197 cte
= self
.get('content-transfer-encoding', '').lower()
198 if cte
== 'quoted-printable':
199 return utils
._qdecode
(payload
)
200 elif cte
== 'base64':
202 return utils
._bdecode
(payload
)
203 except binascii
.Error
:
206 elif cte
in ('x-uuencode', 'uuencode', 'uue', 'x-uue'):
209 uu
.decode(StringIO(payload
+'\n'), sfp
, quiet
=True)
210 payload
= sfp
.getvalue()
212 # Some decoding problem
214 # Everything else, including encodings with 8bit or 7bit are returned
218 def set_payload(self
, payload
, charset
=None):
219 """Set the payload to the given value.
221 Optional charset sets the message's default character set. See
222 set_charset() for details.
224 self
._payload
= payload
225 if charset
is not None:
226 self
.set_charset(charset
)
228 def set_charset(self
, charset
):
229 """Set the charset of the payload to a given character set.
231 charset can be a Charset instance, a string naming a character set, or
232 None. If it is a string it will be converted to a Charset instance.
233 If charset is None, the charset parameter will be removed from the
234 Content-Type field. Anything else will generate a TypeError.
236 The message will be assumed to be of type text/* encoded with
237 charset.input_charset. It will be converted to charset.output_charset
238 and encoded properly, if needed, when generating the plain text
239 representation of the message. MIME headers (MIME-Version,
240 Content-Type, Content-Transfer-Encoding) will be added as needed.
244 self
.del_param('charset')
247 if isinstance(charset
, basestring
):
248 charset
= email
.charset
.Charset(charset
)
249 if not isinstance(charset
, email
.charset
.Charset
):
250 raise TypeError(charset
)
251 # BAW: should we accept strings that can serve as arguments to the
252 # Charset constructor?
253 self
._charset
= charset
254 if 'MIME-Version' not in self
:
255 self
.add_header('MIME-Version', '1.0')
256 if 'Content-Type' not in self
:
257 self
.add_header('Content-Type', 'text/plain',
258 charset
=charset
.get_output_charset())
260 self
.set_param('charset', charset
.get_output_charset())
261 if isinstance(self
._payload
, unicode):
262 self
._payload
= self
._payload
.encode(charset
.output_charset
)
263 if str(charset
) != charset
.get_output_charset():
264 self
._payload
= charset
.body_encode(self
._payload
)
265 if 'Content-Transfer-Encoding' not in self
:
266 cte
= charset
.get_body_encoding()
270 self
._payload
= charset
.body_encode(self
._payload
)
271 self
.add_header('Content-Transfer-Encoding', cte
)
273 def get_charset(self
):
274 """Return the Charset instance associated with the message's payload.
279 # MAPPING INTERFACE (partial)
282 """Return the total number of headers, including duplicates."""
283 return len(self
._headers
)
285 def __getitem__(self
, name
):
286 """Get a header value.
288 Return None if the header is missing instead of raising an exception.
290 Note that if the header appeared multiple times, exactly which
291 occurrence gets returned is undefined. Use get_all() to get all
292 the values matching a header field name.
294 return self
.get(name
)
296 def __setitem__(self
, name
, val
):
297 """Set the value of a header.
299 Note: this does not overwrite an existing header with the same field
300 name. Use __delitem__() first to delete any existing headers.
302 self
._headers
.append((name
, val
))
304 def __delitem__(self
, name
):
305 """Delete all occurrences of a header, if present.
307 Does not raise an exception if the header is missing.
311 for k
, v
in self
._headers
:
312 if k
.lower() != name
:
313 newheaders
.append((k
, v
))
314 self
._headers
= newheaders
316 def __contains__(self
, name
):
317 return name
.lower() in [k
.lower() for k
, v
in self
._headers
]
319 def has_key(self
, name
):
320 """Return true if the message contains the header."""
322 return self
.get(name
, missing
) is not missing
325 """Return a list of all the message's header field names.
327 These will be sorted in the order they appeared in the original
328 message, or were added to the message, and may contain duplicates.
329 Any fields deleted and re-inserted are always appended to the header
332 return [k
for k
, v
in self
._headers
]
335 """Return a list of all the message's header values.
337 These will be sorted in the order they appeared in the original
338 message, or were added to the message, and may contain duplicates.
339 Any fields deleted and re-inserted are always appended to the header
342 return [v
for k
, v
in self
._headers
]
345 """Get all the message's header fields and values.
347 These will be sorted in the order they appeared in the original
348 message, or were added to the message, and may contain duplicates.
349 Any fields deleted and re-inserted are always appended to the header
352 return self
._headers
[:]
354 def get(self
, name
, failobj
=None):
355 """Get a header value.
357 Like __getitem__() but return failobj instead of None when the field
361 for k
, v
in self
._headers
:
362 if k
.lower() == name
:
367 # Additional useful stuff
370 def get_all(self
, name
, failobj
=None):
371 """Return a list of all the values for the named field.
373 These will be sorted in the order they appeared in the original
374 message, and may contain duplicates. Any fields deleted and
375 re-inserted are always appended to the header list.
377 If no such fields exist, failobj is returned (defaults to None).
381 for k
, v
in self
._headers
:
382 if k
.lower() == name
:
388 def add_header(self
, _name
, _value
, **_params
):
389 """Extended header setting.
391 name is the header field to add. keyword arguments can be used to set
392 additional parameters for the header field, with underscores converted
393 to dashes. Normally the parameter will be added as key="value" unless
394 value is None, in which case only the key will be added. If a
395 parameter value contains non-ASCII characters it must be specified as a
396 three-tuple of (charset, language, value), in which case it will be
397 encoded according to RFC2231 rules.
401 msg.add_header('content-disposition', 'attachment', filename='bud.gif')
404 for k
, v
in _params
.items():
406 parts
.append(k
.replace('_', '-'))
408 parts
.append(_formatparam(k
.replace('_', '-'), v
))
409 if _value
is not None:
410 parts
.insert(0, _value
)
411 self
._headers
.append((_name
, SEMISPACE
.join(parts
)))
413 def replace_header(self
, _name
, _value
):
416 Replace the first matching header found in the message, retaining
417 header order and case. If no matching header was found, a KeyError is
420 _name
= _name
.lower()
421 for i
, (k
, v
) in zip(range(len(self
._headers
)), self
._headers
):
422 if k
.lower() == _name
:
423 self
._headers
[i
] = (k
, _value
)
426 raise KeyError(_name
)
429 # Use these three methods instead of the three above.
432 def get_content_type(self
):
433 """Return the message's content type.
435 The returned string is coerced to lower case of the form
436 `maintype/subtype'. If there was no Content-Type header in the
437 message, the default type as given by get_default_type() will be
438 returned. Since according to RFC 2045, messages always have a default
439 type this will always return a value.
441 RFC 2045 defines a message's default type to be text/plain unless it
442 appears inside a multipart/digest container, in which case it would be
446 value
= self
.get('content-type', missing
)
448 # This should have no parameters
449 return self
.get_default_type()
450 ctype
= _splitparam(value
)[0].lower()
451 # RFC 2045, section 5.2 says if its invalid, use text/plain
452 if ctype
.count('/') != 1:
456 def get_content_maintype(self
):
457 """Return the message's main content type.
459 This is the `maintype' part of the string returned by
462 ctype
= self
.get_content_type()
463 return ctype
.split('/')[0]
465 def get_content_subtype(self
):
466 """Returns the message's sub-content type.
468 This is the `subtype' part of the string returned by
471 ctype
= self
.get_content_type()
472 return ctype
.split('/')[1]
474 def get_default_type(self
):
475 """Return the `default' content type.
477 Most messages have a default content type of text/plain, except for
478 messages that are subparts of multipart/digest containers. Such
479 subparts have a default content type of message/rfc822.
481 return self
._default
_type
483 def set_default_type(self
, ctype
):
484 """Set the `default' content type.
486 ctype should be either "text/plain" or "message/rfc822", although this
487 is not enforced. The default content type is not stored in the
490 self
._default
_type
= ctype
492 def _get_params_preserve(self
, failobj
, header
):
493 # Like get_params() but preserves the quoting of values. BAW:
494 # should this be part of the public interface?
496 value
= self
.get(header
, missing
)
500 for p
in _parseparam(';' + value
):
502 name
, val
= p
.split('=', 1)
506 # Must have been a bare attribute
509 params
.append((name
, val
))
510 params
= utils
.decode_params(params
)
513 def get_params(self
, failobj
=None, header
='content-type', unquote
=True):
514 """Return the message's Content-Type parameters, as a list.
516 The elements of the returned list are 2-tuples of key/value pairs, as
517 split on the `=' sign. The left hand side of the `=' is the key,
518 while the right hand side is the value. If there is no `=' sign in
519 the parameter the value is the empty string. The value is as
520 described in the get_param() method.
522 Optional failobj is the object to return if there is no Content-Type
523 header. Optional header is the header to search instead of
524 Content-Type. If unquote is True, the value is unquoted.
527 params
= self
._get
_params
_preserve
(missing
, header
)
528 if params
is missing
:
531 return [(k
, _unquotevalue(v
)) for k
, v
in params
]
535 def get_param(self
, param
, failobj
=None, header
='content-type',
537 """Return the parameter value if found in the Content-Type header.
539 Optional failobj is the object to return if there is no Content-Type
540 header, or the Content-Type header has no such parameter. Optional
541 header is the header to search instead of Content-Type.
543 Parameter keys are always compared case insensitively. The return
544 value can either be a string, or a 3-tuple if the parameter was RFC
545 2231 encoded. When it's a 3-tuple, the elements of the value are of
546 the form (CHARSET, LANGUAGE, VALUE). Note that both CHARSET and
547 LANGUAGE can be None, in which case you should consider VALUE to be
548 encoded in the us-ascii charset. You can usually ignore LANGUAGE.
550 Your application should be prepared to deal with 3-tuple return
551 values, and can convert the parameter to a Unicode string like so:
553 param = msg.get_param('foo')
554 if isinstance(param, tuple):
555 param = unicode(param[2], param[0] or 'us-ascii')
557 In any case, the parameter value (either the returned string, or the
558 VALUE item in the 3-tuple) is always unquoted, unless unquote is set
561 if header
not in self
:
563 for k
, v
in self
._get
_params
_preserve
(failobj
, header
):
564 if k
.lower() == param
.lower():
566 return _unquotevalue(v
)
571 def set_param(self
, param
, value
, header
='Content-Type', requote
=True,
572 charset
=None, language
=''):
573 """Set a parameter in the Content-Type header.
575 If the parameter already exists in the header, its value will be
576 replaced with the new value.
578 If header is Content-Type and has not yet been defined for this
579 message, it will be set to "text/plain" and the new parameter and
580 value will be appended as per RFC 2045.
582 An alternate header can specified in the header argument, and all
583 parameters will be quoted as necessary unless requote is False.
585 If charset is specified, the parameter will be encoded according to RFC
586 2231. Optional language specifies the RFC 2231 language, defaulting
587 to the empty string. Both charset and language should be strings.
589 if not isinstance(value
, tuple) and charset
:
590 value
= (charset
, language
, value
)
592 if header
not in self
and header
.lower() == 'content-type':
595 ctype
= self
.get(header
)
596 if not self
.get_param(param
, header
=header
):
598 ctype
= _formatparam(param
, value
, requote
)
600 ctype
= SEMISPACE
.join(
601 [ctype
, _formatparam(param
, value
, requote
)])
604 for old_param
, old_value
in self
.get_params(header
=header
,
607 if old_param
.lower() == param
.lower():
608 append_param
= _formatparam(param
, value
, requote
)
610 append_param
= _formatparam(old_param
, old_value
, requote
)
614 ctype
= SEMISPACE
.join([ctype
, append_param
])
615 if ctype
!= self
.get(header
):
619 def del_param(self
, param
, header
='content-type', requote
=True):
620 """Remove the given parameter completely from the Content-Type header.
622 The header will be re-written in place without the parameter or its
623 value. All values will be quoted as necessary unless requote is
624 False. Optional header specifies an alternative to the Content-Type
627 if header
not in self
:
630 for p
, v
in self
.get_params(header
=header
, unquote
=requote
):
631 if p
.lower() != param
.lower():
633 new_ctype
= _formatparam(p
, v
, requote
)
635 new_ctype
= SEMISPACE
.join([new_ctype
,
636 _formatparam(p
, v
, requote
)])
637 if new_ctype
!= self
.get(header
):
639 self
[header
] = new_ctype
641 def set_type(self
, type, header
='Content-Type', requote
=True):
642 """Set the main type and subtype for the Content-Type header.
644 type must be a string in the form "maintype/subtype", otherwise a
645 ValueError is raised.
647 This method replaces the Content-Type header, keeping all the
648 parameters in place. If requote is False, this leaves the existing
649 header's quoting as is. Otherwise, the parameters will be quoted (the
652 An alternative header can be specified in the header argument. When
653 the Content-Type header is set, we'll always also add a MIME-Version
656 # BAW: should we be strict?
657 if not type.count('/') == 1:
659 # Set the Content-Type, you get a MIME-Version
660 if header
.lower() == 'content-type':
661 del self
['mime-version']
662 self
['MIME-Version'] = '1.0'
663 if header
not in self
:
666 params
= self
.get_params(header
=header
, unquote
=requote
)
669 # Skip the first param; it's the old type.
670 for p
, v
in params
[1:]:
671 self
.set_param(p
, v
, header
, requote
)
673 def get_filename(self
, failobj
=None):
674 """Return the filename associated with the payload if present.
676 The filename is extracted from the Content-Disposition header's
677 `filename' parameter, and it is unquoted. If that header is missing
678 the `filename' parameter, this method falls back to looking for the
682 filename
= self
.get_param('filename', missing
, 'content-disposition')
683 if filename
is missing
:
684 filename
= self
.get_param('name', missing
, 'content-type')
685 if filename
is missing
:
687 return utils
.collapse_rfc2231_value(filename
).strip()
689 def get_boundary(self
, failobj
=None):
690 """Return the boundary associated with the payload if present.
692 The boundary is extracted from the Content-Type header's `boundary'
693 parameter, and it is unquoted.
696 boundary
= self
.get_param('boundary', missing
)
697 if boundary
is missing
:
699 # RFC 2046 says that boundaries may begin but not end in w/s
700 return utils
.collapse_rfc2231_value(boundary
).rstrip()
702 def set_boundary(self
, boundary
):
703 """Set the boundary parameter in Content-Type to 'boundary'.
705 This is subtly different than deleting the Content-Type header and
706 adding a new one with a new boundary parameter via add_header(). The
707 main difference is that using the set_boundary() method preserves the
708 order of the Content-Type header in the original message.
710 HeaderParseError is raised if the message has no Content-Type header.
713 params
= self
._get
_params
_preserve
(missing
, 'content-type')
714 if params
is missing
:
715 # There was no Content-Type header, and we don't know what type
716 # to set it to, so raise an exception.
717 raise errors
.HeaderParseError('No Content-Type header found')
720 for pk
, pv
in params
:
721 if pk
.lower() == 'boundary':
722 newparams
.append(('boundary', '"%s"' % boundary
))
725 newparams
.append((pk
, pv
))
727 # The original Content-Type header had no boundary attribute.
728 # Tack one on the end. BAW: should we raise an exception
730 newparams
.append(('boundary', '"%s"' % boundary
))
731 # Replace the existing Content-Type header with the new value
733 for h
, v
in self
._headers
:
734 if h
.lower() == 'content-type':
736 for k
, v
in newparams
:
740 parts
.append('%s=%s' % (k
, v
))
741 newheaders
.append((h
, SEMISPACE
.join(parts
)))
744 newheaders
.append((h
, v
))
745 self
._headers
= newheaders
747 def get_content_charset(self
, failobj
=None):
748 """Return the charset parameter of the Content-Type header.
750 The returned string is always coerced to lower case. If there is no
751 Content-Type header, or if that header has no charset parameter,
755 charset
= self
.get_param('charset', missing
)
756 if charset
is missing
:
758 if isinstance(charset
, tuple):
759 # RFC 2231 encoded, so decode it, and it better end up as ascii.
760 pcharset
= charset
[0] or 'us-ascii'
762 # LookupError will be raised if the charset isn't known to
763 # Python. UnicodeError will be raised if the encoded text
764 # contains a character not in the charset.
765 charset
= unicode(charset
[2], pcharset
).encode('us-ascii')
766 except (LookupError, UnicodeError):
768 # charset character must be in us-ascii range
770 if isinstance(charset
, str):
771 charset
= unicode(charset
, 'us-ascii')
772 charset
= charset
.encode('us-ascii')
775 # RFC 2046, $4.1.2 says charsets are not case sensitive
776 return charset
.lower()
778 def get_charsets(self
, failobj
=None):
779 """Return a list containing the charset(s) used in this message.
781 The returned list of items describes the Content-Type headers'
782 charset parameter for this message and all the subparts in its
785 Each item will either be a string (the value of the charset parameter
786 in the Content-Type header of that part) or the value of the
787 'failobj' parameter (defaults to None), if the part does not have a
788 main MIME type of "text", or the charset is not defined.
790 The list will contain one string for each part of the message, plus
791 one for the container message (i.e. self), so that a non-multipart
792 message will still return a list of length 1.
794 return [part
.get_content_charset(failobj
) for part
in self
.walk()]
796 # I.e. def walk(self): ...
797 from email
.iterators
import walk