FontLab Forum
2012-02-09, 05:47:08 *
Welcome, Guest. Please login or register.

Login with username, password and session length
News: Welcome to the FontLab forum, read how to use it! Update: Archives from old MSN forums are now available on our forum.
 
   Home   Help Search Calendar Downloads Tags Login Register  
Del.icio.us Digg FURL FaceBook Stumble Upon Reddit SlashDot

Pages: [1]
  Print  
Author Topic: help about creating fraction and case-sencitive in arabic by msvolt  (Read 6305 times)
ikhlasulhaq
Sr. Member
****

Karma: +0/-0
Pakistan Pakistan

Posts: 7


Email
« on: 2009-05-25, 01:19:18 »


Hello Friends,

How can I define Fractions in MSVOLT? Please provide if some example is available.
Can i define Arabic Case-Sensitive from Latin specified glyphs? Please furnish with exampl if available.

Ikhlas
Logged
tiro_hudson
Beta: FontLab Studio Win
Hero Member
***

Karma: +8/-0
Canada Canada

Posts: 85


WWW Email
« Reply #1 on: 2009-05-26, 08:56:30 »

There are two ways to handle fractions in OpenType. One is by using precomposed fraction glyphs and ligature substitution. The other, more flexible approach, uses independent numerator and denominator variants of numerals. One then performs three substitutions in individual lookups under the 'frac' feature:

1. Change all numeral characters to numerator glyphs, e.g. one -> one.numr

2. Change the generic slash character to the fraction slash: slash -> fraction

3. Contextually change any numerator glyph to a denominator glyph if it follows either the fraction slash or another denominator:

one.numr -> one.dnom
...
| {slash, <denominators>
Logged
tiro_hudson
Beta: FontLab Studio Win
Hero Member
***

Karma: +8/-0
Canada Canada

Posts: 85


WWW Email
« Reply #2 on: 2009-05-26, 08:57:58 »

What do you mean by 'Arabic Case-Sensitive'?
Logged
ikhlasulhaq
Sr. Member
****

Karma: +0/-0
Pakistan Pakistan

Posts: 7


Email
« Reply #3 on: 2009-05-27, 05:40:24 »

I want to replace Some Basic Latin glyphs with Arabic glyphs (this means when I am writing in English the Brackets, Asterk sgould be typed normal but when i change Typing mode to Arabic Brackets & Asterk should replace with Arabic ones as they are different in shapes). Please see below my desired case sensitive substitution lookup.







As you replied earlier about fraction. So my lookup is not as described by you (it is ligature lookup). See below:



As you advised I will make 3 individual lookups. But my question is "I will have to make glyphs of numerator and denominator separately?" & "I will have to set position of numerator and denominator (set according to baseline) or program will automatically do it?"

with thanks

by ikhlasulhaq
Logged
arifkarim
Sr. Member
****

Karma: +0/-0
Posts: 5


Email
« Reply #4 on: 2009-06-07, 21:52:11 »

Aoa,
Ikhlas, this board is normally not very responsiv. We need people like u on our urdu typography forum:
http://www.urduweb.org/mehfil/

plz join, and we will anser ur questions there...

arif
Logged
tiro_hudson
Beta: FontLab Studio Win
Hero Member
***

Karma: +8/-0
Canada Canada

Posts: 85


WWW Email
« Reply #5 on: 2009-06-08, 19:35:46 »

You have a few separate issues here, which I will try to address systematically.

1. Your terminology is confusing. These are not 'case' distinctions, but localised forms, which may be implemented using the <locl> layout feature for the <arab> script tag language systems. However, you will probably find that only some of the symbol characters that you want to change are affected by the <locl> feature. This is not a problem in the feature or, indeed, a font issue at all: it is a problem in how layout engines process text and how they determine where to break glyph runs. In order for glyphs to be subject to layout features under the <arab> script tag, they need to be recognised by the layout engine as either Arabic characters or as characters that should be folded into glyph runs with Arabic characters. The latter applies to things like common punctuation characters that might be used with multiple scripts: these should be rolled into glyph runs with adjacent Arabic characters. But there are a lot of symbol characters that are not treated by layout engines in this way, and which will never be treated as Arabic and, hence, never subject to layout features under the <arab> script tag.

2. Transformation between different styles of numerals, e.g. from European to Arab/Indic numerals, is a character-level transformation. You can't perform this reliably at the glyph level (for the reasons explained above).

3. If you have precomposed fraction glyphs, as in your illustrations, then the kind of ligature substitutions you show should work correctly. However, you probably want to treat these as LTR lookups, not RTL, since the numeral sequences have left-to-right directionality.

If you also want to be able to compose arbitrary fractions, not covered by precomposed glyphs, then you need to provide separate numerator and denominator glyphs and to follow the contextual substitution rules that I explained previously.
Logged
ikhlasulhaq
Sr. Member
****

Karma: +0/-0
Pakistan Pakistan

Posts: 7


Email
« Reply #6 on: 2009-06-11, 09:33:05 »

Mr. Hudson,
With reference to your answer regarding “Case-Sensitive” & “Fractions”, I was unfortunately unable to understand it.


 All that I get is that “in Arabic Language System all Opentype registered tags do not work”. In the following there are registered tags on which I worked in the light of your advice.








In the following I am again trying to explain my project in highlights. And also let me tell you some problems I am facing, so that you may help me once again.

For a personal project I need a standard Unicode Opentype Font for windows ms word 2003 & adobe indesign middle eastern edition which contains full support for English & Arabic (Arabic, Urdu, Persian, Pushtu & Sindhi). I also need in these language systems “Custom Fractions” and “Custom Superscript Footnote’s Numbers”. I designed this font according to the “Arial Regular” font in Font Creator 5.6 (Unicode TTF base). But in my Font, settings for Arabic Unicode glyphs are different from “Arial Regular”.

When I opened my font in MSVOLT to add Arabic Opentype features: “everything was right except Arabic glyphs whose Unicode values were changed with one another”. So I have to manually set Unicode values in MSVOLT.

In my font in Arabic there were some glyphs without any Unicode values. I assigned them Private Use Area but then I have come to know that MS Windows does not support private use area.



Then I used Supplementary Private Use Area-A, but the Unicode values of these glyphs do not appear in Font Creator but it is present in MSVOLT.






My font comprises on following Unicode forms and page codes:



PROBLEMS IN MY FONT
1-   Arabic punctuation



mark except standard Arabic Unicode chart does not appear in Arabic language system. But there are many font developers who worked on Arabic punctuation marks & it works (plz see in previous post the screenshots). How I can activate these punctuation marks in my font?

2   Unicode standard punctuation mark chart already used in my English font.



        On which Unicode value I must define Arabic punctuation marks OR I have to make Custom Keyboard in Arabic to activate these Arabic punctuation marks?
3   How can I define custom footnote superscript signs in my font? Please look at following:




4   How can I define custom Arabic fractions in my font?
        as your advise

i make three individual lookup under frac tag as below:



but when i type in arabic 1/2 so it is in latin 1/2 ligature not in arabic 1/2?

please tell every action with related feature & lookup. if with  example then the best. i am very thank full this help.

regard
ikhlas


Logged
tiro_hudson
Beta: FontLab Studio Win
Hero Member
***

Karma: +8/-0
Canada Canada

Posts: 85


WWW Email
« Reply #7 on: 2009-06-12, 17:29:14 »

Let's look at the fractions:

but when i type in arabic 1/2 so it is in latin 1/2 ligature not in arabic 1/2?

Of course, because '1/2' is the character sequence 0031 002F 0032, all of which are Basic Latin characters and the numerals are European numerals. If you want fractions with Indo-Arabic numerals, then you need to use the character codes for those numerals from the Unicode Arabic block: 0660-0669 (or 06F0-06F9 for Indo-Persian numerals). You can't input Latin characters and expect to get Arabic characters.

Note a couple of other things about the fraction lookups:

They need to be in the correct order to be processed properly. The first lookup changes numerals to numerators, the second changes the slash character to the fraction (or in your case Arabic fraction) glyph, and the third contextually changes the numerators to denominators after the fraction glyph or a preceding denominator. If you put the denominator lookup before the slash -> fraction substitution, it will never be triggered.

The direction of the fraction lookups should be LTR.
Logged
arifkarim
Sr. Member
****

Karma: +0/-0
Posts: 5


Email
« Reply #8 on: 2009-06-13, 10:10:00 »

hmm... ok ikhlas, i have understood fully ur problems. I have been through them before and know every solution. Unfortunately I cant write solutions here, takes too much time. i have added u on msn and also sent mail. Feel free to contact me... Smiley

Arif
Logged
Tags:
Pages: [1]
  Print  
 
Jump to:  

Powered by MySQL Powered by PHP Powered by SMF | SMF © 2011, Simple Machines Valid XHTML 1.0! Valid CSS!