TCYElement dissapeared.

Question

Hi, All.
This problem is on the release build of Adobe Flash Builder 4.
Introduction
Now I'm trying to make e-Book Reader which can handle with Japanese vertical text layout.
Of course, TCY(Tate-Chu-Yoko) Layout is needed.
Please see the snapshot below.
Generally say, the TextFlow object which applyed "rl" on BlockProgression shows Japanese vertical Layout properly.
Multi byte Japanese characters are "standing up" same as horizontal layout.
And Single Byte characters (I wonder if this name is proper. ASCII? Like "a", "0", "é" ,etc...) are rotated +90 degree, in other words, "layed down".
So, in ordinary Japanese vertical text layout, independent single character or  independent 2 characters will be TCYed as shown above.
Rest (I mean consecutive 3 or more Single Byte Characters) are "layed down".
Now, the introduction ends.
Problem: TCYElement which contains text are dissapeared on "NORMALIZATION".
On the release build of Flash Builder 4, some Multi Byte characters are not "standing up properly".
These are, for example, "◇", "□", "☆", "①".
For the precaution, "☆" is the character which has code "A1F9" on UTF-8.
To solve this problem, I implemented the function "Auto TCY", which scratches xml to be import and wrap these Not-StandingUp-Characters with <tcy> tag.
It works well as I'd thought, and as a result, I got proper XML.
But all of them(TCY tags) were removed after being passed to TextConverter.ImportToFlow method.
Checked the API Reference, I've found It told that.
TCY blocks which contain no text will be removed from the text flow  during the normalization process.
http://help.adobe.com/en_US/FlashPlatform/reference/actionscript/3/flashx/textLayout/eleme nts/TCYElement.html
Now, the problem is TCYElements are removed despite they have a character.
Anyone has idea?

Answer 1

Hi, robin.
Thanks for you reply.
But I found this problem is not because of TCY, but because of nesting SubParagraphGroupElement inside of FlowLeafElement.
First, the part of markup After "Auto TCY"ed was like this.
<flow:p fontSize="14">
    <flow:span styleName="font3366cc">
      <flow:tcy>★</flow:tcy>
      不快な態度をチェック
    </flow:span>
  </flow:p>
  <flow:p styleName="color3366cc">
    <flow:tcy>□</flow:tcy>
    　話すときに口を手で覆う
</flow:p>
If you haven't installed Japanese font, it wouldn't be displayed properly.
Anyway, The Entire XML is valid.
After applyed to "TextConverter.importToFlow()" method, <flow:tcy>★</flow:tcy> dissapeared, but <flow:tcy>□</flow:tcy> didn't.
After stripping span tags from this markup, "★" was displayed.
New markup is like below.
<flow:p fontSize="14" styleName="font3366cc">
   <flow:tcy>★</flow:tcy>
   不快な態度をチェック
</flow:p>
<flow:p styleName="color3366cc">
   <flow:tcy>□</flow:tcy>
   話すときに口を手で覆う
</flow:p>
So I think this isn't BUG.
Markup is the matter.
Now, I realized the importance of markup structure.
Thanks a lot.