ばか！バカ！　馬鹿ー月！

Okay, so this is going to be a little different from the usual petition guidelines, as visual novel requests have never really come here before. Having said that, there are at least three translators who are willing to work on it:

velocity7 (myself)
YoakeNoHikari
Cosmic Eagle

I'm also willing to do administration for the project.

We're looking for a programmer who's willing to parse through the bytecode of the game, and come up with a method to, at the very least, parse the bytecode and be able to generate bytecode. Also if possible, generating bytecode should allow an all-ages and adult version of the game. The script format is something like this:

First, there's the script.pak, from which you can extract the files. Afterwards, you have *.txt and *.bnr files, where *.txt represents SJIS text that can be edited directly, and *.bnr represents bytecode which determines what is loaded, what events fire, etc. Also, if you hide the script.pak file and leave a script directory, it's possible to actually have the English text in the game without rebuilding the PAK. There are other files in the script.pak as well, but I'm not aware of what they do.

However, I don't want to quite post such files on the wiki as it's kind of a hassle to actually manage... if I can have a tool that can parse the bytecode events and SJIS text into some UTF-8 XML that can be edited and parsed back, that would probably be the best way to handle it.

For anyone who is interested, there is already side material translated for the game itself:
http://www.baka-tsuki.org/project/index ... um_2_Omake

I've heard a lot of praise of this visual novel on Twitter so it would be nice to see it translated. Though there doesn't seem to be an all-ages version of it when I checked it out on VNDB.

No, there isn't. At least not yet, anyway.

This is a damned good game. If you want your tragic romance drama this delivers nicely.

Depending how free I am I may even contribute to the translation here and there....provided someone has the text format of course....not going to do it as I play....

velocity7 wrote: We're looking for a programmer who's willing to parse through the bytecode of the game, and come up with a method to, at the very least, parse the bytecode and be able to generate bytecode. Also if possible, generating bytecode should allow an all-ages and adult version of the game. The script format is something like this:

I don't think you'll be able to find a programmer on these forums.

And I think the link you provided is blocked for all non-registered users of its affiliated wiki.

Whoops, fixed the link.

I might decide to spend a few days next week looking into the bytecode for it. Hopefully it's not in some random weird encoding or at least others have hacked it before already so all it reduces to is finding a documentation on the format and writing a parser.

Not this week though, midterms tomorrow and the day after..

That's fine, there's no real rush into getting something started soon. This is pretty much going to be on the same scale as CLANNAD was for a translation, so it's better to have everything all clean and ready to go, rather than rush it all at once. Also, good luck with your exams!

Yay!

If you've never hacked a VN before, this might help.

Good Luck!

Doesn't seem to have any info on .pak files, so it's pretty useless. I have my own favorite variations on the other tools he chooses to use (well, most of them come with cygwin already, no reason to install more random crap just for a GUI). Thanks though.

But it's by Leaf which is quite famous, so I think it's very likely that there have been extractors/packers created for the older but still quite recent Leaf games. I'm hoping that they didn't decide to change the encryption or format too drastically from that. But if it's something like they're using a brand new game engine, then I think I probably won't have much luck with this.

Edit: ohh the tutorial he has on IDA Pro might come in useful. I could never understand how to use that program... it has too many features the GUI just gives me a headache. Luckily I've taken a course on x86 assembly so it shouldn't take too long to refamiliarize myself with that. But yeah, next week.

Look what I just found: http://asmodean.reverse.net/pages/exkizpak.html

Ugh, I have a midterm at 8am tomorrow, what the hell am I doing...

Edit: lol, seems velocity7 already found out more than I did and posted it in the Visual Novels thread. Can you PM me where you found that information about the "scripts" directory?

Edit2: Nvm, figured it out
so it seems all I need to do is write a parser for the .txt to get it into a more manageable form (probably something that allows it to go onto the Wiki page automatically or something) and the translation can start while I try to figure out the bytecode. Ideally I also want to auto-insert hyphens somehow..

Edit3: It seems most of the information is in the .txt file. I have no idea what the .bnr file is for, but it seems to be just a bunch of 4-byte values with 2 special values, FFFFFFFE and FFFFFC18. However since the thing controlling the graphics isn't in the .txt it's probably in the .bnr. I'm currently thinking that the number corresponds to the number of the graphics file or sound file or something, but I really should be going to sleep.

The text file contains information about when movies are played, and who the speaker is (if a line begins with a 「 then the speaker is displayed above. If the previous line was also a sentence or if it was a special line then the speaker's name defaults to 春希, otherwise the speaker's name is pulled from the previous sentence.) The scene to switch to is the last thing in the file. Choices are in the form １． text, ２．text, ３．text, what happens if 1 is chosen, etc, etc. Haven't quite figured out what CATCH, grp, or the <w###>s mean, but with playing the game and seeing what happens they should become obvious. I can definitely write a quick parser next week to translate between this format and wiki format.

It might be possible that I won't even need to figure out what the .bnr files are, and the H scenes are a scene file by themselves, in which case we can just edit the .txt files to skip it. For editing the graphics menus and stuff, if we leave the filenames the same we won't need to worry about the .bnr files either. Don't know how hard it would be to put in a different font though, but hopefully it shouldn't be too hard, since well the current font looks ugly for English.

How about line breaks?

line breaks are done with \n.
pages are separated with ， and commas are 、
It's fine even if you use , for english, since , =/= ，

Well, I'm more concerned about the bnr files because they allow event editing, etc. Control over those would be especially useful, if there are cases where text is hard-coded rather than referencing the txt, or if one wants to attempt making an all-ages variant.

also, the text file is ust a comma-separated array. I tried editing 1001.txt at some point as well, and discovered there was hard-coded text.

What do you mean by hard-coded text? Do you mean on the graphics?

I do agree event editing would be useful, but I think it may be possible to make an all-ages variant without doing so because the branching and progression through the story is controlled by the .txt files. Thus I'm thinking right now that until there is something specific that we want to be able to accomplish that turns out is only possible by editing the .bnr, only then will I seriously put time into looking at it.