Mailing-List: contact mime4j-dev-help@james.apache.org; run by ezmlm
Precedence: bulk
Reply-To: mime4j-dev@james.apache.org
Received-SPF: pass (athena.apache.org: domain of markus.wiederkehr@gmail.com
 designates 209.85.218.171 as permitted sender)
DomainKey-Signature: a=rsa-sha1; c=nofws;
        d=gmail.com; s=gamma;
        h=mime-version:in-reply-to:references:date:message-id:subject:from:to
         :content-type:content-transfer-encoding;
        b=RKD1bx/gEFFDNVPBeU2zxmgTZ4o4oBZRPMS+IbIBQF1vCaZg6w2yU2+KTPBEAKX+TQ
         O+i3pNZuIWb95CbdB/IuLp/x/jevWgpR9lJsD5u70n5eXABgC18qt9uZxltpctAbOrqI
         AUhCW5dj1sPYpWJ56InXFSDltsW6b+3/EVyTs=
MIME-Version: 1.0
In-Reply-To: <49CBF784.8010200@ufal.mff.cuni.cz>
References: <49CBF784.8010200@ufal.mff.cuni.cz>
Date: Thu, 26 Mar 2009 23:51:52 +0100
Message-ID: <a92573960903261551t5ae11679x71b64b962540c600@mail.gmail.com>
Subject: Re: accented characters in e-mail addresses
From: Markus Wiederkehr <markus.wiederkehr@gmail.com>
To: mime4j-dev@james.apache.org
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: quoted-printable

E-mail header fields may contain us-ascii characters only. To overcome
this restriction the "name" part of an e-mail address is usually
encoded by a mechanism called an "encoded word". Your mail client then
knows how to interpret these encoded words and is able to display the
original name.

Just look into the source code of your e-mails to see what I mean.
You'll occasionally see e-mail addresses such as
"=3D?ISO-8859-1?Q?Hans_M=3DFCller?=3D <hans.mueller@acme.org>" which is
equivalent to "Hans M=C3=BCller <hans.mueller@acme.org>".

Mime4j should be capable of decoding them too..

Markus

On Thu, Mar 26, 2009 at 10:45 PM, Ondrej Bojar <bojar@ufal.mff.cuni.cz> wro=
te:
> Dear Mime4J developers,
>
> I use android and both the builtin Email client and the K-9 replacement
> delete e-mail addresses containing accented characters. (If say "P=C3=A9t=
=C3=A9r
> <peter@peter.com>" sends me an e-mail and I hit 'Reply', the 'To' field
> becomes blank.)
>
> I can barely read Java, but I understood from K-9 source they use your
> Mime4J for e-mail address parsing (and thus validation).
>
> I was not able to compile the code downloaded from your site (I know noth=
ing
> about Maven, I installed it but running 'mvn test' tried to download
> something and failed.)
>
> I compiled K-9 (the source of which includes a version of mime4j) and I
> guess this exception is exactly the reason why they remove addresses with
> accented characters:
>
> 22:39 vaio classes$java org.apache.james.mime4j.field.address.AddressList
>> P=C3=A9t=C3=A9r <peter@peter.com>
> P=C3=A9t=C3=A9r <peter@peter.com>
> org.apache.james.mime4j.field.address.parser.ParseException: Lexical erro=
r
> at line 1, column 2. =C2=A0Encountered: "\u00e9" (233), after : ""
> =C2=A0 =C2=A0 =C2=A0 =C2=A0at
> org.apache.james.mime4j.field.address.parser.AddressListParser.parse(Addr=
essListParser.java:42)
> =C2=A0 =C2=A0 =C2=A0 =C2=A0at
> org.apache.james.mime4j.field.address.AddressList.parse(AddressList.java:=
116)
> =C2=A0 =C2=A0 =C2=A0 =C2=A0at
> org.apache.james.mime4j.field.address.AddressList.main(AddressList.java:1=
32)
>
>
> I've read your remark somewhere that you're deliberately not handling Bas=
e64
> or Quoted-Printable, but this is plain UTF-8 so that shouldn't pose a
> problem.
>
> My question is simple: who should I blame ;-)
>
> With apologies for a question from a non-Javist,
> =C2=A0Ondrej Bojar.
>
> --
> Ondrej Bojar (mailto:obo@cuni.cz / bojar@ufal.mff.cuni.cz)
> http://www.cuni.cz/~obo