Unicode support issue in some of OLD devices

[saqibatiq]

Hello, I have a j2me application with unicode conversion for my language URDU, on latest and even somewhat little old devices it works fine, shows the language as it should. But I have checked it over some more old devices like N73, N70 and Samsung E250, where it displays nothing or box symbols as it indicates it doesn't support unicodes and some devices just miss a few alphabets of my language rather than whole. So what I want to know is that there are some examples of creating your own script for language but I didn't get that, is there someone to provide detailed procedure for that or an alternate to support my language over these devices? thanks

[petrib]

The older devices probably support Unicode just fine, but lack a font that has glyphs for the Urdu language.

With Java there's not much you can do, except including your own fonts (and even then, with J2ME, it means bitmap files which has all your required symbola, and from which you clip the symbol you wish, and show that piece of the bitmap).

http://www.developer.nokia.com/Commu...nts_in_Java_ME

Might also be worth you while to read and digest this article: http://www.joelonsoftware.com/articles/Unicode.html

[saqibatiq]

The older devices probably support Unicode just fine, but lack a font that has glyphs for the Urdu language.

With Java there's not much you can do, except including your own fonts (and even then, with J2ME, it means bitmap files which has all your required symbola, and from which you clip the symbol you wish, and show that piece of the bitmap).

http://www.developer.nokia.com/Commu...nts_in_Java_ME

Might also be worth you while to read and digest this article: http://www.joelonsoftware.com/articles/Unicode.html

Thanks a lot, want to ask that do I have to add each and every possible combination that could be there for the language to make a single word with different alphabets as in my language URDU there are different glyph (shapes) for one alphabet that can make different words by combining multiple glyph or shapes unlike English language.

[grahamhughes]

Yes, you would have to provide an image for each character. As you say, this is particularly difficult in Urdu (and Arabic), because the shape of each character changes depending on the characters on either side. And, of course, Urdu is written right-to-left.

One thing to consider: if the device does not support Urdu characters, then it's user interface is not in Urdu, and presumably the user uses the phone in some other language. So, do you need to support Urdu on such a device? Consider supporting multiple languages, and selecting the appropriate text for the device (see the wiki). Your app could fall back onto English for devices that don't support Urdu.

[saqibatiq]

Yes, you would have to provide an image for each character. As you say, this is particularly difficult in Urdu (and Arabic), because the shape of each character changes depending on the characters on either side. And, of course, Urdu is written right-to-left.

One thing to consider: if the device does not support Urdu characters, then it's user interface is not in Urdu, and presumably the user uses the phone in some other language. So, do you need to support Urdu on such a device? Consider supporting multiple languages, and selecting the appropriate text for the device (see the wiki). Your app could fall back onto English for devices that don't support Urdu.

Thanks for your reply. I am using two languages English and Urdu in my app, falling back to english is no problem but I have to stick with Urdu as an option though because of the understanding language of my area and I simply cannot ignore it. I can try the solution provided by @petrib about using bitmap images but I am a little confuse because I am using Forms in my application and cannot move to canvas or so for any reason.

[grahamhughes]

The "custom font" mechanism petrib suggested requires the use of Canvas, so you cannot use this on a Form. You could consider switching to another U library such as LWUIT, which I think will allow you to use a custom font.

Think also about input... a non-Urdu device will not provide support for Urdu input either.

My point was: if a user has a device that does not support Urdu, presumably he is comfortable using the device in English. You can provide English only for these devices, and Urdu for devices that are capable of supporting it. Some devices simply will not support Urdu without you doing a lot of work.

Use:

/*
    Will give you "en" for English, "ur" for Urdu, etc.
*/
String language = System.getProperty("microedition.locale").substring(0, 2);

to find out the language and/or region for which the device is configured, and you can select the appropriate language text automatically.