Another issue: when I switch to Russian in LFS, I can't call any hotkey, LFS doesn't recognize them, I need to switch back to English, then back to Russian.
Is it possible to make LFS understand other keyboard layouts? (I hope so, since you've asked us to press a key )
p.s.: got some other corrections, will send them privately.
Hmm, i don't really know what you mean, about switching to Russian and switching to English. LFS is supposed to select the Cyrillic code page as soon as you run it, then you should be able to type Russian characters or English characters, in the normal way that Windows allow you to do it.
Well that is how i understand it - i really don't know anything about Windows interface for switching code pages - i've never used a keyboard that can type non-english characters.
As far as i know, no-one should need to press a key unless they want to use some other code page. Are you talking about a Windows thing or a LFS thing? I think you willl have to explain the bug step by step because i'm quite confused!
Few little problems with "Balkan" letters: Capital ŠČĆ and Ž are not big enough, simply theres no room for the little "thingies" over them, so thats staying as it is I suppose. Small ć looks pretty similar to small č. That "thingie" on ć should be maybe looking more angled to the right somehow.
In Serbia we use both Latin or Cyrillic letters, and switching between them is quite difficult in LFS right now. We use different keyboards for those and in Windows its enough to do just that, switch keyboard (language) and other codepage is selected (I think). In LFS first I have to type ^C then switch to cyrillic keyboard then cyrillic letters are there. And now the same problem as detail said: to type next message I have to switch the keyboard back to latin to type the T as in cyrillic keyboard that key enters a different character (also a T but cyrillic T). We can always use latin so this is not a big problem for us though.
And now the language files. If I enter some of our letters in the strings they don't show up in the game, they show wrongly. They show correctly only if I enter ^E (for Central European) in every string there in language file. What kind of solution is there (will be) for this?
OK enough criticism, its only the small things, otherwise its absolutely great that we have this.
Hmm, maybe i need to make LFS check the currently active codepage more often (once each time you open a new message box - or do you need to change codepage within the same message?). I'll think about this and maybe we can try something tomorrow. I guess this is what detail is talking about. It's quite bad that you have to use the ^L and ^C keys - the whole idea is that LFS should be selecting the correct code page. It was supposed to work intuitively and no need to mess around with those keys.
That's the easiest thing of all You can see the first line in the translation (tx_codepage)... well, that must be set to the correct code page. The available code page numbers are shown in the preview thread.
By the way, which translation are you doing, have you written to Victor about it? We have an online translation system and various translations are being developed at the moment.
I'll try to explain. Well, to type in cyrillic one must switch his keyboar layout to his language, Russian for instance.
But with this layout, when i press "t" on the keyboard, LFS receives cyrillic "е" (similar to latin, but completely different code) and doesn't open the chat dialogue. The same for showing forces (cyrillic "п" instead of "f"): when Russian layout is selected, LFS doesn't change anything.
In Windows and MS Office you can press Ctrl + Z in any layout (in Russian the system will receive Ctrl + Я) the effect will be the same. This is because they've made a special correspondence table for QWERTY and standard russian keyboard.
As for some letters, here is a file with PSD image with proper sizes of two letters.
First, the joints of sticks of "И" aren't wider than the ends of the letter. Second, the letter "Д" is less wide, you'll see in the image a rough example of how it's measured and an illustration of how both letters ought look.
Before this patch my Windows was set to 1252 codepage as language.exe said and I tried the patch first like that. Our characters didn't show when I typed with Serbian (Latin) or Cyrillic keyboard. When I changed to 1250 in Control Panel/Regional and Language Options/Language for non-Unicode programs (and language.exe confirmed that), central european latin characters then worked in LFS with correct keyboard, but switching to cyrillic keyboard did nothing but show weird characters. Not without typing in ^C first. So I guess switching the keyboard in Windows XP doesn't tell LFS anything?
Here's what language.exe says:
Language: English
OEM Codepage: 852
ANSI Codepage: 1250
Language ID: 0409
Locale: 00000409
That part stays the same and the lower part changes when I change the keyboard:
Serbian (Latin):
Keyboard Layout: 081a081a
Keyboard Layout Name: 0000081A
Serbian (Cyrillic):
Keyboard Layout: 0c1a0c1a
Keyboard Layout Name: 00000C1A
English (United States):
Keyboard Layout: 04090409
Keyboard Layout Name: 00000409
We don't need to mix serbian latin and cyrillic in one message, in fact that would be rather hm strange. What is useful, that I can think of, is typing in an english word here and there and that already works by just selecting a keyboard (this does not change the codepage but first half of every codepage is the same - english). Who knows, maybe it'd be useful to someone to mix codepages in one message sometimes?
Sweeet, just added tx_codepage 1250 at the beginning and all is well, thanks.
Didn't know about this translation system when we started but couple of us have already practicaly finished Serbian translation and help weeks ago. Just need to write it in proper characters now, and find a better solution for couple of terms. I'd better contact Victor anyway.
Another minor addition I just thought of would be to have the language name as a string inside the file instead of the filename itself. Because renaming the file for example into the greek word for "greek" doesn't seem to work (LFS doesn't know it's supposed to be handling a different codepage for the given filename and so just uses the latin-1 character set).
OK... so detail and Mikso, you were in fact talking about quite different things.
For detail, everything's ok, you can type in Russian and English - no problem with text entry in text boxes. The only problem is the various unassignable game keys... ok i'll think about that.
For Mikso, the problem is that you want to use two completely different code pages. You are right, switching the keyboard layout doesn't do anything - the only thing that affects LFS is the "ANSI Codepage" reported by language.exe. I think i could make LFS check the ANSI Codepage more often, but i'm not sure that would solve the problem for you. I guess it could solve the problem, if it's easy to change that value in Windows - but maybe that's not as easy to change on the fly, as the keyboard layout? An alternative would be to provide a drop down box, so you can manually select the code page. So LFS just does the auto select once at startup (as it does now) and you can change it any time after that - no need to type ^C every time when you have switched to Cyrillc mode.
axus : I made minor updates to some of the characters. The e looked a bit ugly to me so i gave it that curved edge. But i'm not sure that was a good idea, as unlike in handwriting, a typed e seems to have a flat line in the middle. Do you think it looked better before? It's easy to change back.
Everyone else : Thanks for the comments. I'll go through the thread and see how many of the updates and improvements can be done.
The curved e I like when it's small (i.e. like your names at the bottom of the main menu), but when it's larger, as in the words 'Multiplayer' it looks wrong.
Also has the x changed, or have I only just noticed how odd it is? It's more like an algebraic x (which I have no idea how to type) than a normal x.
What about "CAPITAL LETTER GHE WITH UPTURN"... should the upturn in fact be higher than the top of all the other capital leters? In LFS at the moment, it is not, the flat bar has been lowered - but i think this is wrong...
[ EDIT : it's ok, i've figured it out, the top bar should be the same height as all the other capitals, and the upturn should go up from there, in the "accent area"... so in a sense, the upturn is sort of like an accent but connected to the main character ]
I don't wont to be an ungrateful [beep], but all of the acutes (central european) characters are wrong. I see that Misko already mentioned ć as an example of that. The issue is that the that acute is wrong.
You mean "Г" with upturn? (it's probably Ukrainian, btw ), I think yes, it should be above tops of capitals, as well as the arc of "Й", but I've already noticed that it's impossible right now.
[p.s. just noticed your another message. Good news, thanks! ]
Another problem: I use 3 keyboard layouts, and when I switch to Italian, instead of letters with diacritics now I see cyrillic: ì ù è ò à => м щ и т а.
Okay, enough for today. Wish me success tomorrow, I'm going to pass TOEFL exam, Test of English as a Foreign Language. Abuse me as much as you can since 03:30 till 07:30 CET (according to our superstitions this gives luck ).
)
]
).