Multi-byte characters support!

QuadMan · Tue 19/06/2007 10:43:33

Dear developers!!!!

Your engine is very great and powerfull. I'm going to make japanese-language learning game, but without multibytecharacters support it wouldn't be able.

Maybe in future release, you will add support for unicode characters? I think it will be great improvement for your engine.

Thank you.
Semenov Alexander.

Rui 'Trovatore' Pires · Tue 19/06/2007 11:03:59

Does it help that you can use TTF fonts? You can, you know.

Gilbert · Tue 19/06/2007 11:34:09

No, it won't.

Rui 'Trovatore' Pires · Tue 19/06/2007 11:35:35

Oh.

...

...how come? Why not?

QuadMan · Tue 19/06/2007 11:37:34

Yes, may be, it will help me for first time, but as far as I know, there are restrinction in 100 characters for one font. So, it's not very comfortable... besause in Japanese there are Hiragana and Katakana (all about 100 characters) and some primitive kanji (about 100-200)... so, it's very uncomfortable. I will try to think about it, but, if you have ability to add support for unicode characters it will be the 100% solution in all cases.

Thanks.

Rui 'Trovatore' Pires · Tue 19/06/2007 11:40:24

Ah, right.

...hmm, you could try SSH's "sprite fonts" module. Not very comfortable either, I know, but it's a possible workaround.

Unless I'm missing something again...

Gilbert · Tue 19/06/2007 12:49:46

If you don't need Kanji for Japanese, you may try my crappy programme though, make sure you read the documents, it's not very friendly anyway.

QuadMan · Tue 19/06/2007 15:22:39

Unfortunatelly, main purpose in learning program is to learn kanji.

Pumaman · Tue 19/06/2007 19:18:46

How was Japanese represented before Unicode was invented? Presumably there must be some sort of 256-character font that can display something reasonable?

Gilbert · Wed 20/06/2007 01:13:53

If you need Kanjis in Japanese doublebyte (multibyte) is needed.
CJK (Chinese-Japanese-Korean) characters cannot be displayed within the 256 limit. Double byte fonts existed LONG before unicode was invented.
It's just because there weren't a unified and consistent format to display all these characters so later we have unicode.

Traveler · Wed 20/06/2007 06:29:22

Working with doublebyte characters is a PITA. Unicode would be a lot better, since then it can represent all known characters.

Just out of curiosity: why would it be a big trouble adding Unicode support? (I assume it would, so correct me if I'm wrong.

)

Monsieur OUXX · Thu 21/06/2007 13:27:32

Quote from: Traveler on Wed 20/06/2007 06:29:22
Just out of curiosity: why would it be a big trouble adding Unicode support? (I assume it would, so correct me if I'm wrong. )

I suppose Chris chose to handle strings in a very simple way, to have a simplier code. So their representation in memory must be very simple. Extending the size of one character in memory would force him to change everycode used to handle strings : allocating strings, iterating through strings, comparing strings, etc. Not to mention the way fonts are mapped to strings!

Pumaman · Thu 21/06/2007 19:38:47

Well, AGS 2 was originally written in 1999, and in those days unicode wasn't so prevelant. Plus, back then I didn't really appreciate the international considerations that would be involved

But the new 2.8 editor will make it easy to support unicode in the editor, however it would then take a fair bit of work to upgrade the engine accordingly.