Re: Unicode Quranic Glyphs Proposal

[Mohammed Yousif <mhdyousif at gmx dot net>]

On Thursday 08 April 2004 21:59, Nadim Shaikli wrote:
> >
> > Just tried it on Qt, Gtk and MS-Windows giving varying results but
> > all of them are not what we expect. (for example they are fixed at
> > a Y-axis position that is usually not consistent with harakat so
> > it's not suitable for Arabic. Also, the implementations are even
> > making it a spacing character) I think what you mentioned, Nadim,
> > is the appropriate thing to do.
>
> Mohammed, didn't we say we'd be happy with a Y-axis position specific
> glyph ?  Meaning if the OVERLINE glyph was specified so that in the
> font file it was marked to be slightly higher than whatever the
> upper-bound of LAM + haraka.1 + haraka.2 is,

  No, it doesn't do that. It's not expected for it to be used with Arabic
  diacritics at all.
  The current behavior with tashkil is really bad, it's sometimes lower and
  sometimes higher than tashkil depending on the height of the arabic letter
  (Well, not really the line but the tashkil is the thing that is changing
   its Y-Axis position all the time)

> wouldn't that work ?

  What you explained is very acceptable but unfortunately...

> In other words assume LAM takes upto y-axis 80 and haraka.1 (say shadda)
> makes it now 90 and haraka.2 (say fatha) makes it 100, can't we specify
> this OVERLINE to always appear at say y-axis location 150 or similar ?
>
> If not, I'm not really sure what to ask for then - since I can only
> envision a y-axis specified solution (it will also need to be a combining
> character producing a ligature character akin to fatha).  Do please
> tell me what to say in the proposal then to get it done ASAP.
>
  
  We would want something like what you explained, a fixed Y-Axis char that is
  higher than anything else, that is.

  Another thing is the lam alef with floating hamza thing, it's a ligature
  like lam alef and we need someway to be able to put harakat on the lam
  or on the alef and they need to be drawn appropriately.
  I noticed the existence of 0675 (Arabic letter high hamza alef - kazakh),
  there are two shapes of it already existing in unicode tables (isolated
  initial and final) the isolated is the same as the initial one but for some
  reason they are there. This character is the regular Alef Hamza but with
  the hamza to the right of the alef but still on top of its Y-Axis position,
  that is very similar to what we are looking for (lam alef with floating
  hamza have the hamza to the right of the alef).
  So I started looking for a ligature that 0675 is used to produce but didn't
  find any, that is good news.
  I'm thinking of telling Unicode that we want lam with 0675 to produce our
  proposed ligature (lam alef with floating hamza), that way we can overcome
  the tashkil problem with the ligature (What tashkil belongs to lam/and what
  belongs to the alef) so it can be drawn properly.
  So, instead of encoding the Qur'an using the ligature directly, we instead
  use 0644 + 0675 and put the harakat correctly.