multi-language programming

[OogyWaWa]

i'm interested in understanding how programs are written for languages that don't use the latin alphabet. specifically, japanese. i'm assuming that means i have to learn about Unicode and other byte-formats but I'm kind of clueless as to where to start. has anyone done some programming in chineese/korean/thai/japanese/arabic or some other similar language and can point me in the right direction? thanks

 

[tatteredpotato]

I always thought they just used the same stuff we did... I remember a thread about if keywords were translated into other languages, and I thought it was determined the English stuff was pretty widely used.

 

[OogyWaWa]

well, i know for a fact you can't just open up VS2008, make a new console app, and write some console.out lines and instead of english throw some japanese in there. the compiler doesn't know wtf is going on. also, i tried some basica html output and it came out as gibberish. but, i think i forgot to tell the browser what formatting it was in.

the way i understand it, there is one method of multi-lang programming where you set all your texts to property file links. for example, you determine what lang the user is using and based off that info you pull strings from a en-US.properties or ja-JA.properties file that looks something like:

submit=submit
cancel=cancel

submit=?????
cancel=?????

etc.

 

[Fox5]

As far as code goes, I believe English is fairly widely used.
But for translating an actual app, that could be handled in a variety of different ways.

 

[OogyWaWa]

Originally posted by: Fox5
As far as code goes, I believe English is fairly widely used.
But for translating an actual app, that could be handled in a variety of different ways.

thats what i'm talking about. i'm trying to figure out those ways the app (interface, outputs) are outputted to the screen

 

[squatchman]

Yea, you're going to want to know UNICODE.

 

[OogyWaWa]

Originally posted by: squatchman
Yea, you're going to want to know UNICODE.

not sure if this is sarcasm, but unicode text formats encompass almost all written languages in current use, smart guy

if not, my bad

 

[Markbnj]

It is kind of platform dependent. In the .NET world you localize resources by isolating assets into assemblies and setting the culture attribute in the assembly information. You then use the approriate culture-aware resource access methods to load them. Using these techniques it isn't rocket science to provide alternate libraries of UI strings in different cultures, even though there is a fair bit of plumbing involved.

 

[invidia]

My friend immigrated from China. He uses English in his programming but English is also a requirement in most Chinese schools.

 

[squatchman]

Yea I should've been more clear. Multilingual support will vary based on whatever language/framework you're using at the time. Many simply require an external resource file to match against.

However, being unfamiliar with UNICODE leads to situations like what you have run into. You probably won't be able to simply copy-pasta whatever garbage you want into there. In the end, computers really don't give a damn about these fancy shapes the meat bags use to pass messages.