Hi Folk,
Just would like to hear from arabic people.
Thanks,
Chookij V.
------------- Begin Forwarded Message -------------
Date: Mon, 12 Jul 1999 17:16:54 -0700
From: Rick McGowan <rmcgowan@apple.com>
Subject: Re: Presentation of unknowned composited sequences for arabic script
To: chookij.vanatham@eng.sun.com
Ah, thanks for an example.
BEH + BEH + FATHATAN + FATHATAN + FATHATAN + FATHATAN + FATHATAN + BEH
In this example, if it was in "proper" Unicode encoding, all of the "beh"
would be nominal forms in backing store of course. The final one would be
displayed as a final beh, and the initial one as an initial beh; they would
connect as usual, etc.
As for the fathatans... I would still recommend stacking them up so that
they "try to be visible" to signal the user that they are ill formed... It
doesn't make any sense, as you say. Maybe we'll see what Arabic people on
the Unicode list say.
Rick
------------- End Forwarded Message -------------
Date: Mon, 12 Jul 1999 16:16:37 -0700 (PDT)
From: Chookij Vanatham <chookij.vanatham@eng.sun.com>
Subject: Re: Presentation of unknowned composited sequences for arabic script
To: chookij.vanatham@eng.sun.com, rmcgowan@apple.com
MIME-version: 1.0
Hi Rick,
I wouldn't use "show hidden" to confuse with "invisible mark" as you said.
So, if within the sequences have 5 or more vowel "fathatan", like below.
BEH + BEH + FATHATAN + FATHATAN + FATHATAN + FATHATAN + FATHATAN + BEH
Moving the diacritic up, wouldn't be enough for lot of vowels. ZWJ and ZWNJ
wouldn't be involved for this situation. I guess that those vowels (FATHATAN)
would be displayed with middle form (FATHATAN middle form U+FE77) if we still
want these vowels displayed connected each other (for cursive script).
My point is if actually, these vowels are not going to have any meaning
anyway, then what if the BEH before the first vowel (FATHATAN) will be "final"
and the last BEH would be isolated form. Would this be acceptable enough for
arabic script or will it look weird or not conform to unicode standard ?
Chookij V.
] Date: Mon, 12 Jul 1999 14:23:26 -0700
] From: Rick McGowan <rmcgowan@apple.com>
] Subject: Re: Presentation of unknowned composited sequences for arabic script
] To: chookij.vanatham@eng.sun.com
]
] > I'm not too sure is what the unknown composited sequences mean.
] > Would that mean the "composited sequences" where we don't expect to be,
] > like incorrect order of combined character sequence ?
]
] Yes, it would mean things you're not expecting. However, to do Arabic
] typography, the non-spacing marks can come anywhere, and your implementation
] will have to deal with them. There are two ways to display non-spacing
] marks: one way is to do simple overlap -- put the mark at the same location.
] Sometimes taht won't be visible, as you know. The other way is to figure
] out some better placement for the mark with relation to the base glyph. It's
] not too hard to move the diacritic up, for example. Or you could have
] metrics for such placements, aside from the regular font information.
]
] All of these display issues have NOTHING to do with "show hidden"
] attributes. That has to do with display of INVISIBLE marks, such as
] zero-width joiners and such. That's why I thought you were confused.
]
] Rick
]
This archive was generated by hypermail 2.1.2 : Tue Jul 10 2001 - 17:20:48 EDT