Arabic text

[Sebastian Weber <se...@hmedia.de>]

Hi Kia and All,
There is a file for Unicode normalization tests http://www.unicode.org/Public/UNIDATA/NormalizationTest.txt
To make FOP print Arabic text I took the contents of the normalization test to find the correct Arabic representation form A or B for a glyph that is represented by some characters between 0x0600 and 0x06FF.

For a ligature of length 3 NormalizationTest.txt contains e.g.

FCF2;FCF2;FCF2;0640 064E 0651;0640 064E 0651; # (ﳲ; ﳲ; ﳲ; ـ◌َ◌ّ; ـ◌َ◌ّ; ) ARABIC LIGATURE SHADDA WITH FATHA MEDIAL FORM

That means Unicode characters 0640 064E 0651 should be replaced by FCF2.
But it occurred to me that in real text ligature SHADDA WITH FATHA MEDIAL FORM always appears as 0640 0651 064E which means that the second and third characters are switched. This happens with all ligatures of length 3. Can anyone tell me why? And is it ok to alter the test table and change all ligatures of length 3 by switching characters 2 and 3. The ligature above would look like this:

FCF2;FCF2;FCF2;0640 0651 064E;0640 0651 064E; # (ﳲ; ﳲ; ﳲ; ـ◌َ◌ّ; ـ◌َ◌ّ; ) ARABIC LIGATURE SHADDA WITH FATHA MEDIAL FORM

Or is there another reason why my text is not conform with the Unicode tests???


Regards

--
Sebastian Weber
Software Engineer
Hmedia

phone:  +49 351 479 77 40  
mobile: +49 177 754 94 57

Ludwig-Richter-Allee 6
01445 Radebeul
Germany

www:    http://www.hmedia.de
e-mail: sebastian.weber@hmedia.de