| View previous topic :: View next topic |
| Author |
Message |
Walker
Newbie
Joined: 04 Nov 2009
Posts: 2
|
 Posted: Wed Nov 04, 2009 2:17 am Post subject: Hindi character combination |
 |
|
Dear Hindi experts,
I'm currently building a 3D font engine to display fonts in my application.
I want to try to include several languages and I also want to include 'Hindi' and realize the correct display of hindi character combinations.
Therefor I am looking for a specific algorithm to implement the rules which are necessary to create those new ligatures e.g. a certain combination of characters leads to a new combined glyph. I am sorry for missing terminology but that's what I know about what's happening in Hindi due to my internet research.
I wonder if you guys can give me a hint on how Open Office (or other free text-editors) handles these issues or is it something that is coded in the operations systems kernel?
Any help highly appreciated!
Thx in advance...
W A L K E R  |
|
| Back to top |
|
 |
Robert Tucker
Moderator
Joined: 16 Aug 2004
Posts: 3367
Location: Manchester UK
|
| Posted: Wed Nov 04, 2009 6:39 am Post subject: |
|
|
I think it's a matter of the font and the way the system converts keystrokes into character codes.
| Quote: | | The X Window System used on most Unix-like systems today uses X Keyboard Extension (XKB) for translating keystrokes into character codes. |
http://www.linux.com/archive/feature/113715
With Devanagari there is, I believe, the problem of at least one of the vowels going in front of the consonant(s) it is typed after. In this respect you may need to look into BIdi (bi-directional) rendering. See, for example:
http://java.sun.com/j2se/1.4.2/docs/api/java/text/Bidi.html
This thread may also be of interest:
http://www.oooforum.org/forum/viewtopic.phtml?t=43377
_________________
LibreOffice 3.6.6 on Fedora 18, LibreOffice 4.0.2 on Ubuntu 13.04 (Double Boot) |
|
| Back to top |
|
|
Walker
Newbie
Joined: 04 Nov 2009
Posts: 2
|
| Posted: Fri Nov 06, 2009 4:51 am Post subject: |
|
|
Hi Robert,
first of all thanks a lot for your precious help.
As from what I understood the links you posted refer to the implementation methods of the several character combination when they're well known.
My problem is that I don't know nothing about Hindi at all nor do I know anyone who can tell me about the ligatures.
I am using a tool called FONTRAST http://hexforge.com/fontrast.shtml to create the fonst. I need a proper algorithm which includes the mechanics how to adjust/rearrange the Hindi characters to a new glyph if a valid combination of characters exists.
I thought it could be helpful if I ask someone from the OpenOffice Development team how they've managed to get that mechanism working.
The ongoing research tells me that this is rather a modul of the operating system than the job of any text processing application. Currently I'm trying to find something useful in the sourcecode of this one here: SCIM http://www.scim-im.org/
Anyway, if someone has an idea I'd be happy to hear... 
Best regards
W A L K E R |
|
| Back to top |
|
|
|
