The problem is really the font that should have an advance 0 for the
diacriticals as the Unicode documentation requires. If you use the font
arialuni.ttf the result will be correct.
The fix is to set in the font the char advance of those chars to 0. This
requires the version at itextpdf.sf.net.
For all the chars in the range: \u064b-\u0658,\u0670, \u06d6-\u06dc,
\u06df-\u06e4, \u06e7-\u06e8, \u06ea-\u06ed do:
BaseFont.setCharAdvance(char, 0);
The result will be correct
Best Regards,
Paulo Soares
----- Original Message -----
From: "Ayman Zarroug" <[EMAIL PROTECTED]>
To: "Paulo Soares" <[EMAIL PROTECTED]>;
<[EMAIL PROTECTED]>
Sent: Sunday, December 21, 2003 9:14
Subject: Re: [iText-questions] Problems in Arabic Diactrics (tashkeel)
RE: [iText-questions] Problems in Arabic Diactrics (tashkeel)Thanks for all
your feedback with regards to this issue.
Please find attached to this email two .pdf files:
1. correct_display.pdf (this pdf is generated by converting a word
document. diacritics are correctly displayed in this file.)
2. itext_display.pdf (this pdf is generated by using iText Other
fonts display similar results.)
Please note that using Times, Courier, Arial and Ms Sans Serif fonts with
iText will all generate similar results.
Thanks for your time,
Ayman
----- Original Message -----
From: Paulo Soares
To: Ayman Zarroug ; [EMAIL PROTECTED]
Sent: Thursday, December 18, 2003 5:50 PM
Subject: RE: [iText-questions] Problems in Arabic Diactrics (tashkeel)
The diacritics are being taken care of. What I found out was that some
fonts, notably times and arial from win xp, treat them as spacing chars when
they sould be non spacing. Quite curiosely, arial unicode works as expected.
It might be better for you to give me an unicode arabic sequence and the
expected output result.
Best Regards,
Paulo Soares
-----Original Message-----
From: Ayman Zarroug [SMTP:[EMAIL PROTECTED]
Sent: Wednesday, December 17, 2003 12:35
To: Paulo Soares; [EMAIL PROTECTED]
Subject: Re: [iText-questions] Problems in Arabic Diactrics
(tashkeel)
The method "ColumnText.setArabicOptions(ColumnText.AR_COMPOSEDTASHKEEL)"
is being called, but it doesn't resolve the problem.
Just to clarify the problem, words in arabic may have diacritics
(tashkeel), which are symbols that appear either above or below a letter.
The diacrics should not affect the way in which 2 letters are joined.
Currently, iText treats diactrics as a normal character that affects the way
letters are joined.
The method "ArabicLigaturizer.ligature(nextletter, curchar)" checks the
current character with the next character in order to determine the join
type. The method correctly determines when a word with diacritics is being
processed and returns a value of 1 to distinguish this case.
In the method "ArabicLigaturizer.shape(char text[], StringBuffer string,
int level)" there is a sequence of if-then-else blocks that shape the
characters based on the join type (determined by ligature method). However,
the case for join type 1 (diactric join) is not implemented.
I think the correct behaviour for the diactric join would be:
1. Determine the join type based on the current character and the
character following the diacritic.
2. Draw the diactric either above or below the character preceding the
diactric (depending on which diactric).
Ayman
----- Original Message -----
From: Paulo Soares <mailto:[EMAIL PROTECTED]>
To: Ayman Zarroug <mailto:[EMAIL PROTECTED]> ;
[EMAIL PROTECTED]
<mailto:[EMAIL PROTECTED]>
Sent: Tuesday, December 16, 2003 8:50 PM
Subject: RE: [iText-questions] Problems in Arabic Diactrics (tashkeel)
You may try:
ColumnText.setArabicOptions(ColumnText.AR_COMPOSEDTASHKEEL);
If that's not satisfactory as an arabic speaker you'll be able to give
pointers to correct it. I can produce code but I don't know a word of
arabic.
Best Regards,
Paulo Soares
-----Original Message-----
From: [EMAIL PROTECTED]
<mailto:[EMAIL PROTECTED]>
[SMTP:[EMAIL PROTECTED] On Behalf Of Ayman
Zarroug
Sent: Tuesday, December 16, 2003 14:54
To: [EMAIL PROTECTED]
Subject: [iText-questions] Problems in Arabic Diactrics
(tashkeel)
So glad that arabic is now supported in iText, as this finally allows
us to generate PDF reports using iReport and JasperReports (though it was
necessary to make some modifications in both iReport and JasperReports to
get this running).
The good news is that it generally works. The bad news is that arabic
text with diacritics are not properly displayed in PDFs. This can be seen
by running the sample "hebrew-arabic" code provided by Paulo Soares at
arabic_hebrew.java in itextpdf.sf.net.
Tried the new release itext-paulo-127, but that doesn't seem to fix
the problem.
Looking through iText code leads me to believe that functionality to
handle this case is not implemented.
Are there any plans to solve this problem in the near future???
ayman
-------------------------------------------------------
This SF.net email is sponsored by: IBM Linux Tutorials.
Become an expert in LINUX or just sharpen your skills. Sign up for IBM's
Free Linux Tutorials. Learn everything from the bash shell to sys admin.
Click now! http://ads.osdn.com/?ad_id=1278&alloc_id=3371&op=click
_______________________________________________
iText-questions mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/itext-questions
