email bug?

Re: email bug?

Stuart D. Gathman:[color=blue]
> Content-Type: image/pjpeg; name="Jim&& amp;Jill"[/color]
[color=blue]
> What IE apparently gets is:
>
> [('image/pjpeg', ''), ('name', '"Jim&& Jill"')]
>
> Is this a bug (in the email package, I mean - obviously IE is buggy)?
>
> Do I have to write my own custom param parsing routines to handle this?[/color]

BTW, I verified this in 2.3.

Looks like the Content-Type syntax is defined in

5.1. Syntax of the Content-Type Header Field

content := "Content-Type" ":" type "/" subtype
*(";" parameter)

parameter := attribute "=" value

value := token / quoted-string

token := 1*<any (US-ASCII) CHAR except SPACE, CTLs,
or tspecials>

tspecials := "(" / ")" / "<" / ">" / "@" /
"," / ";" / ":" / "\" / <">
"/" / "[" / "]" / "?" / "="
; Must be in quoted-string,
; to use within parameter values

So the ";" must be in a quoted string. That's defined in
RFC 822, http://www.faqs.org/rfcs/rfc822.html
(now obsolete)

quoted-string = <"> *(qtext/quoted-pair) <">

qtext = <any CHAR excepting <">, ; => may be folded
"\" & CR, and including
linear-white-space>

CHAR = <any ASCII character>

The ';' is in CHAR and is not "\" nor CR so it's in qtext,
so it's part of quoted-string, so it's allowed in a value
without extra interpretation.

I looks like 2822 (the updated version of 822) a
http://www.faqs.org/rfcs/rfc2822.html agrees.

So I think it's a bug in the email module's parser.

The actual bug is in email/Parser.py with

# Regular expression used to split header parameters. BAW: this may be too
# simple. It isn't strictly RFC 2045 (section 5.1) compliant, but it
catches
# most headers found in the wild. We may eventually need a full fledged
# parser eventually.
paramre = re.compile(r'\s *;\s*')

A quick scan of the code suggests that it isn't a quick fix (eg,
not just a matter of tweaking that regexp.

Could you file a bug report against it?

Andrew
dalke@dalkescie ntific.com

Re: email bug?

On Sun, 24 Aug 2003 00:14:45 -0400, Andrew Dalke wrote:

[color=blue]
> A quick scan of the code suggests that it isn't a quick fix (eg, not
> just a matter of tweaking that regexp.[/color]

Here is my quick (and probably incorrect) fix:

from email.Message import Message
from email.Utils import unquote

# helper to split params while ignoring ';' inside quotes
def _parseparam(str ):
plist = []
while str[:1] == ';':
str = str[1:]
end = str.find(';')
while end > 0 and (str.count('"', 0,end) & 1):
end = str.find(';',en d + 1)
if end < 0: end = len(str)
f = str[:end]
if '=' in f:
i = f.index('=')
f = f[:i].strip().lower( ) + \
'=' + f[i+1:].strip()
plist.append(f. strip())
str = str[end:]
return plist

class MimeMessage(Mes sage):

def getparam(self,n ame,header='con tent-type'):
for key,val in self.getparams( header):
if key == name: return unquote(val)
return None

# like get_params but obey quotes
def getparams(self, header='content-type'):
"Return all parameter names and values. Use parser that handles quotes."
val = self.get(header )
result = []
if val:
plist = _parseparam(';' + val)
for p in plist:
i = p.find('=')
if i >= 0: result.append(( p[:i].lower(),unquot e(p[i+1:])))
return result

--
Stuart D. Gathman <[email protected] m>
Business Management Systems Inc. Phone: 703 591-0911 Fax: 703 591-6154
"Confutatis maledictis, flamis acribus addictis" - background song for
a Microsoft sponsored "Where do you want to go from here?" commercial.

Re: email bug?

Stuart D. Gathman:[color=blue]
> Here is my quick (and probably incorrect) fix:[/color]

There's a test suite in email.tests. It includes tests for
getparams, and I see some commented out code which
lists a test known to fail.

You could use that to check the validity of your code.

Andrew
dalke@dalkescie ntific.com