Stereo Tool

I suppose this thing writes UTF-8? So if I just convert UTF-8 in the input to the RDS charset it should be fine?

No, as far as I can see it writes in Latin-1. I have not seen any multi-byte characters in the file.
Most RDS receivers I have seen that don't support lower case convert to upper case, but some old ones substitutes with a space.
Almost all receivers I have seen that don't support funny characters convert them to a space ("NÜRNBERG" -> "N RNBERG"), but I have seen one or two that just leaves them out ("NÜRNBERG" -> "NRNBERG").

That was only regarding PS. I'll echo AlexL in that if RadioText is supported at all, then most likely the entire RDS charset is supported.
Yep, I just took 7.02BETA4 for a spin. Looking good!

--
Ragnar

Heh, you can say that
Especially the last three characters in the Swedish alphabet; åäö. That's why the word "Räksmörgås" is popular when doing character set tests here.

BTW, this is what a Räksmörgås looks like:


--
Ragnar

What's the stuff below the shrimps?

Edit: See edits below

But anyway, I'm getting a big confused.

First, the ASCII characters that to my knowledge are default in Windows are these:
http://www.theasciicode.com.ar/ascii-codes.txt

This is however VERY different from Latin-1 etc. Also unrelated to UTF-8.

Could you check which character you get with ASCII code 129? Normally you get this by typing ALT+129 on the numeric keypad, that's how I make a ü if I need to type something in German.

This is by the way also the character set that's built into Stereo Tool, I don't know why characters above 127 aren't displayed, will check that now.


EDIT:
- Ah, problem in Stereo Tool display found - now all characters ARE displayed
- Ow, what you enter by typing ALT+129 is indeed ASCII code 129 if you use the DOS editor. But not if you do the same thing in Notepad. Then you get 252. Which fits Latin-1!

That looks like CP437 or CP850 - a DOS encoding (I recognize it from the line drawing characters). Thats not Win1252 (the western windows encoding). So those are not the windows "characters".
Thats because Windows imported this feature from DOS as far as I know. If you want to use the real encoding windows uses for applications, you need to prefix your number by a 0. I.e. ALT + 0252 should give you an ü.
So ST is secretly using DOS code internally? Are you sure that a printf of the entered ü gives you a 129 and not a 252?

See my edits in the code above... Indeed I use a DOS font because I made this font a loooong time ago (I actually made it to be able to keep using my MSX printer when I got my first PC ). And yes you are probably right about the printf - and also vice versa, if I type this code in Notepad and Notepad correctly shows a ü, then run the program in a DOS box I get a small 'n' at the top. The same thing happens in Stereo Tool The new font is ok though, it's the monosize font used for entering and displaying texts (mainly RDS) that uses the wrong mapping.

Mayonnaise. Yum!

--
Ragnar

Part 1 should be ok now (conversion of PC to DOS characters, so everything is displayed correctly). Which leaves sending it to the RDS encoder and reading it back from it.

I somehow found a wrong ASCII-list. Most characters were correct but quite a few were different. Should be ok now, in next beta (702-003).

What I needed was this one:
http://www.kostis.net/charsets/cp437.htm