Open Discussion (#3) - Composite Unicode (encoded utf-8) (#58) - Message List

Composite Unicode (encoded utf-8)

My language contains of some character á, à, ớ, ế, ề, ờ, ở, ắ, ấ, ầ,... The latest version of Logic Mail can not display it correctly.

I checked this mail and noted that it's been written by Composite Unicode and I recheck again and learned that there's two kind of Unicode for my language, "Precompound Unicode" and "Composite Unicode".

Logic Mail handle "Precompound Unicode" well, but for "Composite Unicode" It is not displayed correctly.

This is the tested mail. (Noted that the subject is encoded and displayed incorrectly)

Received: by with SMTP id u2cs486664fgb;
        Tue, 28 Oct 2008 21:45:09 -0700 (PDT)
Received: by with SMTP id 20mr5210873tiz.29.1225255507365;
        Tue, 28 Oct 2008 21:45:07 -0700 (PDT)
Received: by with HTTP; Tue, 28 Oct 2008 21:45:07 -0700 (PDT)
Message-ID: <>
Date: Wed, 29 Oct 2008 11:45:07 +0700
From: "Gmail Invite" <>
To: "Gmail Invite" <>, "xxxx Bui" <>
Subject: =?UTF-8?Q?Th=C3=A1ng_8/2008_em_th=E1=BA=A5y?= =?UTF-8?Q?_c=C3=B3_l=E1=BB=8Bch_thi_l=E1=BA=A1i_m=C3=B4n?= =?UTF-8?Q?_Lu=E1=BA=ADt_H=C3=ACnh_s=E1=BB=B1_d=C3=A0nh_c?= =?UTF-8?Q?ho_l=E1=BB=9Bp_h=E1=BB=8Dc_th=E1=BB=A9_3-5-7.?=
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: base64
Content-Disposition: inline
  • Message #148

    I did some investigation last night, along with a friend of mine who can read Vietnamese and thus interpret many of the websites I found in my searches.

    Here is what I've found so far:

    • Most text out there on the web is formatted like your sample, and the BlackBerry web browser won't even render it properly
    • The Java 6 API (J2SE) has a Normalizer that can convert this text into a format the BlackBerry can display
    • Fixing this issue would require porting a Unicode normalizer to the embedded Java (J2ME/BlackBerry) version LogicMail runs on.
    • It is definitely do-able, but may require a fair amount of work to implement.

    I'll do more investigation on what is actually involved in doing Unicode normalization inside of LogicMail, and let you know how practical of a task this turns out to be.

  • Message #214

    Just wanted to let you know, as of today, I added an optional "Unicode normalization" feature to LogicMail 1.1 (the "maintenance branch"). If you turn this on, it should fix your problem :-)