Re: stacking diacritcs and fullwidth characters

Date: Wed Jan 30 2008 - 11:43:44 CST

  • Next message: Peter Constable: "RE: Unicode Transliteration Guidelines released"


    > John,
    > We also have people interested in Cipher music notation.


    Actually it was Paul Neeley, who orginally asked me several months
    ago, how I was printing cipher music, to which I replied that I am
    using software I wrote myself, however really it should be possible to
    do almost everything (slurs excepted) using a font.

    > I was given to understand that:
    > Basically, if we set aside slurs and a few special characters <snip>, the
    > core of what we need are just
    > some diacritics to attach to the numbers 1 through 9:

    Namely to U+FF10 -U+FF19

    > - overline
    > - double overline

    In China underline and double underline are used as oppsed to
    overlines (U+0332 Combining Low Line, and U+0333 Cobining Double Line)
    , in theory there could be three lines below. I was more inclined just
    to stack U+332's.

    > - underdot
    > - overdot


    > - forward-slash strike-through
    > - backslash strike-through
    Theese are not, off-hand, required for CKJV characters

    To these one could add


    > Charis SIL/Doulos SIL have all of those with the exception of the
    > backslash strike-through. I believe that will be in the next version of
    > our fonts (but U+20E5 is a combining mark for symbols, rather than in the
    > regular combining mark block, is that a problem?). However, there is the
    > issue of "strike-through characters". We haven't given the strike-through
    > characters any smart fonts so no smart attachment takes place. That would
    > be a problem.

    With Doulos SIL it is possible to display simple cipher music with
    half width numbers. However the full width numbers U+FF10 - U+FF19 ,
    do not appear to have the correct anchors to put the diacritcs at the
    centre of either the top or base. In theory this should be no more
    difficult for U+FF10 -U+FF19 than for U+0030 - U+0039.

    A second prolem is the length of the lines underneath ideally double
    for fullwidth characters. This may not have a simple solutiion.

    > Of course, maybe none of these are issues for Cipher notation in China.
    > These were requested for another country that uses it.

    There are other features sometimes seen, but as you say the above
    covers the basics. For slurs, I was thinking in terms of U+1D178
    MusicalSymbol Begin Slur and U+1D179 Musical Symbol End Slur, which in
    plain text could display like a diacritc at the top of the character.


    > Lorna

    This message sent through Virus Free Email

    This archive was generated by hypermail 2.1.5 : Wed Jan 30 2008 - 11:46:38 CST