Esperanto mojibake & CS+ font editing

I'm considering making an Esperanto translation of Cave Story, and I have been testing some stuff to see if I will be able to make it work.

The main problem right now is Esperanto letters. Courier New supports them, but when I try to use them in-game, I get mojibake like this:

those first two lines should say:

eĥoŝanĝo ĉiuĵaŭde
EĤOŜANĜO ĈIUĴAŬDE​

(it's a short phrase that includes all 6 letters unique to Esperanto)

Does anyone know a way I can get these letters to display correctly? ASCII substitutes exist (e.g. ehxosxangxo cxiujxauxde), but I'd rather not use them if at all possible. They look a bit ugly.


Also, does anyone know a way to edit the font files of Cave Story+? If there's a way I could add characters, that would be awesome. I'm guessing csfont.fnt tells the game which character is found where on the csfont_00.bmp bitmap, but I can't find a way to read and edit it.

 

cave story only recognizes multi-byte (unicode etc) characters that have the high bit in the high byte. Specifically it does this because that will capture japanese characters and the rest of the text is 7-bit ascii
haven't got good solution for you at this moment as it's quite late for me

 

Aha, I see.

I was actually a little surprised to get this problem, because I've browsed through other threads about translation, and there wan't any mention of this problem there. I get this for French, because its characters are supported in ANSI, but Polish should be having the same problem:

That bottom line should read "ć ń ó ś ź ż ą ę ł", the Polish letters

I can only get the ó to work, by converting to ANSI encoding. I can't figure out how they got the other letters to work.

 

have you tried using one of those international versions as a base instead of english?

 

I tried taking a look at the Russian one, but Russian texts look like this when I run it:

So it looks like languages rely on Windows' setting for non-unicode characters. Which I should have known, because Japanese does this too. I forgot about that.
This is a bit of a problem, because Esperanto does not have a local encoding option like this.

I guess I could make a custom font that replaces unused ASCII symbols with Esperanto letters and hack it into the game. But I'd have to figure out how to do all that, and I'd prefer not having to require players to install a weird font.

*sigh* I thought this would be easier.

 

You clearly have a non-unicode codepage or you wouldn't get anything.

I suspect that your default non-unicode codepage is Windows-1250. Just save as ANSI.

https://en.wikipedia.org/wiki/Windows-1250

Actually I think I might try a bit of a proof of concept later.

 

Just save as ANSI.

That is the problem. ANSI doesn't support the characters I need. ĉ ĝ ĥ ĵ ŝ ŭ are not used in any language marketable enough for Windows to make a code page for them.

I'm pretty sure my computer uses Windows-1252, btw.

 

q=ĉ
w=ĝ
x=ĥ
y=ĵ
{=ŝ
[=ŭ

Q=Ĉ
W=Ĝ
X=Ĥ
Y=Ĵ
}=Ŝ
]=Ŭ

Would this work as a font mapping?

I can build a custom font. So all we need is some way to link it to CS without someone needing to install it first.

 

As long as <> and {} are not used in the game and they don't mess with the code if I use them, that would work, yes.

But I'd make it:
q=ĉ
w=ŭ
x=ŝ
y=ĝ
<=ĵ
{=ĥ

ŭ is basically a w sound, and ĉ and ŝ correspond to q and x in Chinese pinyin. For the other three, ĝ is common, ĵ is uncommon, and ĥ is very rare, so ĝ can have the normal letter (which also happens to be a little visually similar to a handwritten g), and ĥ can have the weird one that is hardest to reach on a keyboard.

That way it's a little easier for me if I want to edit the text after converting my translations to the ASCII substitutes.

 

Actually I probably shouldn't use <> or [] for that exact reason, but {} should be fine.

What about \=ĵ and |=Ĵ (backslash for lowercase and pipe for uppercase)? Would that work?

 

iirc the only "special character" in vanilla TSC is < and #, and sometimes =
anything else is fair game

 

What about \=ĵ and |=Ĵ (backslash for lowercase and pipe for uppercase)? Would that work?

Sure

 

iirc the only "special character" in vanilla TSC is < and #, and sometimes =
anything else is fair game

We need to account for the credits as well. So square brackets are out.