I'm trying to make a "Hello World" zimlet like the one described in the wiki, but instead of just "Hello World" I want it to find "こんにちは" (Japanese for "Hello"). I'm having no luck and am looking for some ideas where I'm going wrong.

I modified the wiki's example from this:
Code:
 <zimlet name="com_zimbra_hello" version="1.0" description="Hello World">
     <contentObject>
         <matchOn>
             <regex attrs="ig">Hello World</regex>
         </matchOn>
         <toolTip>Hello World Zimlet</toolTip>
     </contentObject>
 </zimlet>
to this:
Code:
<zimlet name="com_zimbra_hello" version="1.0" description="こんにちは">
        <contentObject>
                <matchOn>
                        <regex attrs="ig">こんにちは</regex>
                        <regex attrs="ig">\u3053\u3093\u306B\u3061\u306F</regex>
                        <regex attrs="ig">Hello World</regex>
                        <regex attrs="ig">=E3=81=93=E3=82=93=E3=81=AB=E3=81=A1=E3=81=AF</regex>
                        <regex attrs="ig">44GT44KT44Gr44Gh44Gv</regex>
                </matchOn>
                <toolTip>\u3053\u3093\u306B\u3061\u306F Hello World Zimlet</toolTip>
        </contentObject>
</zimlet>
The "Hello World" works finding the text and popping up the tool tip as expected.
But for the "こんにちは" in an email, I get nothing.

As you can see, I've used the literal string:
Code:
<regex attrs="ig">こんにちは</regex>
I've used the Unicode codes:
<regex attrs="ig">\u3053\u3093\u306B\u3061\u306F</regex>
I've tried this search string:
Code:
<regex attrs="ig">=E3=81=93=E3=82=93=E3=81=AB=E3=81=A1=E3=81=AF</regex>
Why? Because I've send myself an email with Japanese text wrapped in English (English+Japanese+English) in both the subject and body, so the encoding came out as "quoted-printable":
Code:
Subject: =?utf-8?Q?Hello_World_=E3=81=93=E3=82=93?=
 =?utf-8?Q?=E3=81=AB=E3=81=A1=E3=81=AF_Hello_World?=
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: quoted-printable
X-Originating-IP: [192.168.0.2]
X-Mailer: Zimbra 5.0.13_GA_2791.RHEL5 (ZimbraWebClient - FF3.0 (Mac)/5.0.13_GA_2791.RHEL5)

Hello World =E3=81=93=E3=82=93=E3=81=AB=E3=81=A1=E3=81=AF Hello World

Lastly, I tried this search string:
Code:
<regex attrs="ig">44GT44KT44Gr44Gh44Gv</regex>
because, I sent myself an email with just Japanese text and it was automatically encoded as base64:
Code:
Subject: =?utf-8?B?44GT44KT44Gr44Gh44Gv?=
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: base64
X-Originating-IP: [192.168.0.2]
X-Mailer: Zimbra 5.0.13_GA_2791.RHEL5 (ZimbraWebClient - FF3.0 (Mac)/5.0.13_GA_2791.RHEL5)

44GT44KT44Gr44Gh44Gv

P.S. side note: The description (description="こんにちは">) has the literal UTF-8 string shows up fine in the admin UI.

Does anyone know how to get the tooltip to popup on Unicode foreign language characters?