Re: Some questions about Unicode's CJK Unified Ideograph

From: Andrew West <andrewcwest_at_gmail.com>
Date: Sat, 30 May 2015 10:19:03 +0100

On 30 May 2015 at 02:50, Ken Whistler <kenwhistler_at_att.net> wrote:
>
> 1. I have seen a chinese character ⿰言亜 from a Vietnamese dictionary NHAT
> DUNG THUONG DAM DICTIONARY
>
> Extension F is harder to track down, because it has not yet been
> approved by the UTC, and comes in two pieces, with different
> progression so far in the ISO committee. Perhaps somebody on this list
> who has better access to the relevant documents can let you
> know whether ⿰言亜 can be found in those sets.

It's not in my lists of F1 and F2 characters.

> 2. Is combined characters like U+20DD intended to work with all different
> type of characters, or is it some problem related to implementation ? as I
> when i write ゆ⃝ (Japanese Hiragana Letter Yu + Combining Enclosing Circle)
> appear to be separate on most font I use, but if I change the Hiragana Yu
> into a conventional = sign or some latin character, most fonts are at
least
> somehow able to put them together. Or, is there any better/alternative
> representation in unicode that can show japanese hiragana yu in a circle?
>
> Combining enclosing marks in principle could work with most characters,
> but in practice most arbitrary combinations do not work very well,
> because they would require very complicated font support.

It's not that complicated, but I think most fonts don't support arbitrary
combinations with combining enclosing circle because there is little or no
demand for them. BabelStone Han displays Japanese Hiragana Letter Yu +
Combining Enclosing Circle quite well, but on the other hand it does not
work so well with CJK ideographs, and fails with Latin letters and
punctuation.


> 4.In CJK Symbols and Punctuation, Proper name mark and Book name mark are
> not included. While there are charactera like U+2584, U+FE33, U+FE4F, and
> U+FE34 in unicode that is more or less a representation for the two
symbol,
> they do not appear below or on the left of typed characters when text flow
> is horizontal/vertical, and instead, they occupy their own space which
make
> them having little use in daily life, and while the proper name mark and
> book name mark can represented by text editing softwares and css but those
> representation are not ideal and they do match "Criteria for Encoding
> Symbols". Is it possible to make a new unicode symbol, or change some
> current symbol into one that could appear in suitable place of other
> characters when typed? And a property of the symbol is that when used in
> case like 美國紐約 which 美國 and 紐約 are two different proper name (place name),
> so an underline should go below them without any separation between the
> character 美and國 or 紐and約 (when text are written horizontally), but at the
> same time the underline should not be linked between 國 and 紐 as 國 is the
end
> of first place name while 紐 is the start of the other.
>
>
> What you are talking about is, indeed, best handled by text styling
> attributes,rather than by individual character encoding.

I agree. However, if you really do want to represent underlining of proper
names at the character encoding level, then you would have to do something
like put U+0332 Combining Low Line after each character to be underlined,
and select a font that supports Combining Low Line with CJK ideographs.
BabelStone Han supports this low-level method of underlining CJK
ideographs, but if you want a space in the underlining between 美國 and 紐約
you would have to insert a very thin space (U+200A Hair Space in this
example) between the characters.


Andrew

MeiguoNiuyue.png circled_yu.png
Received on Sat May 30 2015 - 04:20:45 CDT

This archive was generated by hypermail 2.2.0 : Sat May 30 2015 - 04:20:46 CDT