Author Topic: How to disallow all but text characters for all languages? (Read 4985 times)

dydya-stepa · « **on:** March 05, 2012, 06:01:31 pm »

I want to allow only text characters for a textbox. How can I filter out all the other characters from evt.Text.Unicode. I'm talking about commas, semicolons and so on.

texus · « **Reply #1 on:** March 05, 2012, 06:13:56 pm »

Every character has a number: http://www.asciitable.com/

For only the letters you will need to do this:

Code: [Select]

if (((evt.Text.Unicode > 64) && (evt.Text.Unicode < 91)) // check for capital letters
 || ((evt.Text.Unicode > 96) && (evt.Text.Unicode < 123))) // and check for lowercase
{
    // add the letters to the textbox
}

dydya-stepa · « **Reply #2 on:** March 05, 2012, 06:58:19 pm »

isn't that for english only? i want ALL languages as stated in the name of the topic

http://en.wikibooks.org/wiki/Unicode/Character_reference/0000-0FFF

3rd an 4th section contain also crappy characters.

texus · « **Reply #3 on:** March 05, 2012, 07:23:56 pm »

Sorry, I didn't read the topic name.

You might want to check out functions like iswalpha.

dydya-stepa · « **Reply #4 on:** March 05, 2012, 07:56:03 pm »

good idea. does sfml support this ? can I use sfml's evt.Text.Unicode with c/c++ build in functions?

Laurent · « **Reply #5 on:** March 05, 2012, 08:11:43 pm »

iswalpha expects a wide character, but it's not clear what encoding it works with.

If you need to classify Unicode characters, you should use a true Unicode library. You won't do anything good with only standard functions; they are not made for Unicode.

Nexus · « **Reply #6 on:** March 13, 2012, 06:40:49 pm »

texus, when working with ASCII, I would rather use the meaningful char literals instead of magic numbers, e.g. 'A'.

texus · « **Reply #7 on:** March 13, 2012, 06:45:39 pm »

You are absolutely right, that would be clearer.