Random translation-related questions
Posted: Tue Sep 05, 2006 3:15 am
Sorry if these are dumb questions, I'm not very familiar with some technical details of the game. =) Basically, I'm translating U7BG to Finnish.
1) So... I take it "wuc" and "wud" are the obsolete tools and "ucc" and "ucxt" are the new, shiny tool for compiling Usecode assembly. The thing is, I used the ideas from SI-French to start up my translation database. Their method uses wud to decompile the Usecode, and now my code in database refers to stuff by the labels wud generates. Does ucxt generate the exact same label numbers when decompiling the exact same Usecode file? (Based on *extremely* cursory glance, it seems like it does.) What's the exact syntax for compiling ucxt-generated assembly? (Does ucc even do this?)
And exactly how does one use the translation table files - do they need to be compiled somehow? (As in, can I just dump my translation database into an XML file, point Exult there, and it works, or what?)
2) Appears that simply using expack to extract the contents of mainshp.flx, replacing files, rebuilding the file and sticking it in patch directory will not replace intro text, credits, etc. Which is weird. What should be done to fix this?
3) Are there any elegant ways to handle multi-line signs that doesn't involve re-programming the whole shebang? For example, in Trinsic there's a sign that spans four lines: "The Avenue of Fellowship" (each word on its own string). In Finnish, this would be in two words: "Veljeskunnan Puistokatu". Fits nicely on two lines. If I leave the unneeded strings empty (''), the thing does compile, but the sign is always four lines high and the text is always on top end of the sign, which looks silly. Putting a single space on lines above and below (' ') seems to work, but leaves the sign a little bit more spacious than needed. Is there a way to make the signs shrink without decompiling and reprogramming the entire huge sign function?
(A hack in the engine side for the latter would involve simply ignoring lines that are empty when drawing signs...)
1) So... I take it "wuc" and "wud" are the obsolete tools and "ucc" and "ucxt" are the new, shiny tool for compiling Usecode assembly. The thing is, I used the ideas from SI-French to start up my translation database. Their method uses wud to decompile the Usecode, and now my code in database refers to stuff by the labels wud generates. Does ucxt generate the exact same label numbers when decompiling the exact same Usecode file? (Based on *extremely* cursory glance, it seems like it does.) What's the exact syntax for compiling ucxt-generated assembly? (Does ucc even do this?)
And exactly how does one use the translation table files - do they need to be compiled somehow? (As in, can I just dump my translation database into an XML file, point Exult there, and it works, or what?)
2) Appears that simply using expack to extract the contents of mainshp.flx, replacing files, rebuilding the file and sticking it in patch directory will not replace intro text, credits, etc. Which is weird. What should be done to fix this?
3) Are there any elegant ways to handle multi-line signs that doesn't involve re-programming the whole shebang? For example, in Trinsic there's a sign that spans four lines: "The Avenue of Fellowship" (each word on its own string). In Finnish, this would be in two words: "Veljeskunnan Puistokatu". Fits nicely on two lines. If I leave the unneeded strings empty (''), the thing does compile, but the sign is always four lines high and the text is always on top end of the sign, which looks silly. Putting a single space on lines above and below (' ') seems to work, but leaves the sign a little bit more spacious than needed. Is there a way to make the signs shrink without decompiling and reprogramming the entire huge sign function?
(A hack in the engine side for the latter would involve simply ignoring lines that are empty when drawing signs...)