Sorry your browser is not supported!

Have done a bit of coding in the past - in BASIC originally when I was young, more recently in dedicated MUD codes. so text-based coding is not a problem, although drag-and-drop would be preferred.

I'm an Esperantist and thought it would be a good learning tool to have a game in Esperanto - there are a few already, including an excellent port of Manic Mansion for the nostalgic, but I had an idea of having a language course for newcomers masked as a single-player RPG.

So a couple of questions, bearing in mind I come from a MUD background.

Firstly, I imagine I can do a single player graphical RPG with text (I'm thinking along the lines of Monkey Island, that kind of thing). Perhaps a point and click. The idea is that it would give language lessons in a virtual world kind of format. So, first question is, DarkBASIC, 3D Gamemaker or Realmcrafter? I have no artistic skill at all so something where there are a lot of available textures would be a big plus. Can you import the FPS creation models into any of these?

I appreciate that question has probably already been answered somewhere else, but the second question is very specific - does the in-game text generation support Unicode (Latin Extended-A). I need to be able to say things like this - Ĉu vi komprenas la vorton “naǔ”?

Many thanks - multajn dankojn!

Strig

OK, I asked Support and they said that DBPro supports Unicode with the set text font command. I've tried every character set and none of them seem to work. I even set up a loop and tried every character set number between 1 and about 500 and still no joy. Any ideas?

Just out of curiosity, could I see the code for that?

Binned it now. I'll see if I can recreate it and post it.

Not impressed with support. They told me this thing can do Unicode and it clearly doesn't. Serves me right for asking the devs instead of the users.

As long as the EXE embeds the font I should be able to hack about in the font and put the accented characters over something I don't need and use that character set. It's a lot of faffing about for something that just about every other program can do these days.

Interestingly enough, charset 2 comes up with a lot of wingdings but the codes that are supposed to display the Russian etc. don't seem to help. Copied these from a website and pasted them into a Unicode file in Notepad. Shows up in the edit window but not in EXE. You just get a lot of question marks for everything. Presumably you have to "launder" the text through something else, but not being able to copy and paste Unicode text in is going to be monumental pain in the rear end.

I do wonder if I have to use UTF-8 or launder it into ISO-8859-3 with something like BabelPad.

In this day and age you should be able to cut and paste something out of Notepad into a program and have it work properly, though. If games can't be localised easily then it's going to put a lot of people off.

I saw all the stuff about language problems on the forums so I asked Support if there was Unicode support and they told me there was, for DBPro. Was assuming it had been implemented, therefore. If they have implemented it, it's not very easy to actually make it work...

Sorry if I've already mentioned this - something I said earlier hasn't been approved yet so I don't know what I said...

I just had a thought - perhaps it would be easiest to create a bitmap font with XFont and substitute some of the unused characters? It'd mean I'd have to remember what letter I'd allocated to what character. It would make alphabetical sorting a bit off, but as I had a point and click/multiple choice game in mind, it won't matter too much.

This is something we've had to struggle with for years. Always used to get round it in the pre-Unicode days by putting X's after the accented letters so you get phrases like Lauxsxajne estas rano en mia bideo, which you get used to after a while, but in this 21st Century Unicode world where any old website can actually handle this sort of thing, people expect to see the accents now.

yeah I mean I would love to know what actually is and how to get to it

Looks like maths to me. I don't speak maths. . Where is it from?

It's probably Shift JIS encoded Japanese after bunging it in Google and doing a bit of guesswork. I might write a little program to render it on screen and ask someone what it says.

All this mojibake is what Unicode was invented for. The Esperanto of character encoding.

its from Dark Source... which has a ready made sample of unicode embedding... and this was the Japanese [yup your research paid off as you were correct - well done!] and thus I am confused how am I supposed to knwo how to get this script... as there was no data on it...

Annoyingly, whoever I am emailing at Support seems to think that multiple character sets a la ye olde Windows is Unicode. It's not. Windows Code Pages were ad hoc additions to ANSI and were used in the 80's and 90's before Unicode, and have been superseded. This is NOT Unicode. This is Windows Code Pages which are 20 years old! The set that has Esperanto in is standard ISO/IEC 8859-3 (we share a set with Turkish and Maltese!).

Quoting Microsoft, "Western European editions of Windows 3.x supported three character sets per installation: a single Windows character set (ANSI), an OEM character set, and the symbol character set. Because different language editions of Windows 3.1 supported different default Windows and OEM character sets, sharing documents among different systems was not always feasible. " Here is their list of the decimal Charsets, with the Charset Value, which is what DBPro uses:

http://msdn.microsoft.com/en-us/library/cc194829.aspx

It's basically a dimensioned array of characters. The old school stuff like DBPro uses has a variety of different arrays for text, and you reference the variable for the array that has the characters you need. It means you have to keep changing arrays for different characters. Unicode is a much, much larger array that has just about every language you could ever need, including Klingon, IN ONE ARRAY. Which is why so many of us are clamouring for it.

http://www.joelonsoftware.com/articles/Unicode.html

The upshot of all of this is if you want to see which characters are in which sets, open up Character Map in Windows, go to Character Set and look for all of the ones that start with Windows. For things like Japanese, you also need a suitable font as well. Basically, you need to have an old Japanese computer. f you look then at the Unicode set, there are many, many more characters in it. That's the one I want. I want to be able to use the Cyrillic Capital Letter Ot if I want, straight from Notepad, because it looks like a bottom. Ѿ

Sorry for sounding a bit frustrated but I am. Got this software on the understanding that it does Unicode, and the support are still claiming it is Unicode but it isn't. Not fit for purpose. Stultuloj.

You can search for characters you want in Character Map too. Latin Capital Letter C with Circumflex and Latin Capital Letter U with Breve are two Esperanto ones. Type in Circumflex under "Search For" in there and see what you get. In Unicode, three and a half lines of characters, including most of the ones I need, from Ĉ through to ŝ. Then do breve and you find Ŭ and ŭ which are the other two. Why Zamenhof decided that he needed breves all of a sudden instead of circumflexes I have no idea. But yes, all I need is those 12, which is why I'm considering hacking a font to get round this. it's also possible that the DBPro character codes for the Turkish set are based on ISO/IEC 8859-3 in which case the Esperanto accents are in there somewhere and might be reached by use of chr$, but I'd have to type some very strange stuff in to get the right characters to come out. However, if it's based on the Windows Code Page and not ISO/IEC 8859-3 I'm stuffed, and will have to hack a font.

At least it's a project.

This page also has some interesting things:

http://www.cs.tut.fi/~jkorpela/chars.html

The point is, with Unicode you can just cut and paste and IT WORKS.

Oh, and this is what your Japanese turned out like. I could send it to a Japanese Esperantist I know, but for all I know it might be rude. Hell, why not. Might give her a laugh.

Actually, Klingon isn't in Standard Unicode. It's in a Private Use Area. Got that bit wrong. I only know two words of Klingon anyway.

I am lost again... UNICODE!!!!!!! HEY its a bug so lets stick it in the top three bugs thread?

I don't think it's a bug. I think it's a "feature".

You can actually translate this forum into Esperanto via the pull-down at the top. It reckons it's The Game Kreintoj which is a heroic effort but La Ludkreintoj I would have put.

Basically, DB was written in the heady old days when PC's didn't really need to talk to PCs in other countries. So it still uses the text encoding of ye olde computers. With the theory being that one language per PC is enough.

Things have changed, but the encoding of the text in DB hasn't changed with it. I presume the problem is that to please the minority that want to use Unicode fonts in DB, they'd have to quite seriously hack around at it and they don't want to.

I have had some success, however, using some old fonts that I had kicking around from the pre-Unicode days to solve just these problems. Basically, the characters you want are moved to hide in the slots of ones you don't want. Attached code and exe. I'd appreciate it if someone who doesn't have the SudEuro fonts (I would guess that's pretty much all of you) could tell me if it works. The characters are in the code as "Ææ Øø ¦¶ ¬¼ Þþ Ýý" but should come out as Ĉĉ Ĝĝ Ĥĥ Ĵĵ Ŝŝ Ŭŭ.

Oh I'm loving Google Translate. It's translated me talking about a port of Manic Mansion as "bonega haveno de Manic Mansion"... a great harbour of Manic Mansion.

Anyway, if my little exe works then I'm a feliĉa kuniklo and can actually get on with writing some stuff now.

Oh, and the Japanese apparently says "Dark Basic"