Cc'ing Chris on this because he's usually the one to say
"hmm, I think that's an illegal encoding", and Fejj, because
he's usually the one to say "Charles and Chris, you bozos,
you're doing it wrong. You were *supposed* to do xyz..."

Honestly, the only thing I'm good for is pestering Chris and
Fejj with "have you fixed this charset encoding bug yet?"
and then saying "oh, I get it now" once I read the diff. :)

from rfc2047:

   IMPORTANT: 'encoded-word's are designed to be recognized as 'atom's
   by an RFC 822 parser.

from rfc822, Appendix D:

     atom        =  1*<any CHAR except specials, SPACE and CTLs>

     specials    =  "(" / ")" / "<" / ">" / "@"  ; Must be in quoted-
                 /  "," / ";" / ":" / "\" / <">  ;  string, to use
                 /  "." / "[" / "]"              ;  within a word.

     word        =  atom / quoted-string

In plain English: 'This is an illegal encoding' :)   Specifically, the
colon shouldn't be there (not as a literal anyway).

We've been down this road before and the consensus was that we can't
address every non-compliant agent out there.

ok, so I looked over some header fields, which pan doesn't decode as 
author of specific news client meant. this one is afaik rfc compliant:

Subject: Re: !!! =?windows-1250?Q?Nab=EDz=EDm_scanov=E1n=ED_negativ=F9?=
 =?windows-1250?Q?=2C_p=F8edloh_do_A4=2C_=E8/b_lasertov=FD_ti?=
 =?windows-1250?Q?sk_a_=E8/b_inkoutov=FD_tisk_do_A2!!!?=

pan decodes only first line here (to the end of first =?widows-1250xyz?=)

yea, it looks like it should be decoding. I'll take a look tonight
when I get home :-)

looks like my gmime-utils.c:quoted_decode() function was a bit too
aggresive with erroring out. If it encountered a SPACE or ESPECIAL it
would abort. The check for ESPECIAL was wrong, since ESPECIALs are
actually allowed to be in the encoded-text section of an encoded-word
(rfc2047). The SPACE check was just plain redundant since our caller
does this, so I removed that whole else-if section.

this is now fixed in gmime-2, gmime-1 and pan HEAD

I also updated gmime-table-private.h since I found some problems with
it as well.

thanks for the bug report, otherwise I probably would never have
noticed ;-)