ROM Hack Dragon Ball Fusions Hacking Thread

[YugamiSekai]

This is my first time doing one of these kind of threads.

Dragon Ball Heroes RomFS is made up of 1 file (DOF.cpk). The file format is used in Xenoverse and can be easily extracted/repacked with QuickBMS. When extracted, the file consists of files like jtex, jarc, jmdl, usm, etc. I haven't found the files for the text (unlike Ultimate Mission 2, which is in .Mes files). @nobody231 and I are working on hacking the game in my Ultimate Mission 2 translation thread but I decided to move it.

 

[Helmax]

I'd be down to take a look at this with you guys.

 

[focusonme]

nice work guys!

 

[nobody231]

So, I'm an official part of this hacking thing. Agh, I guess I have to stop being a lazy bum now.

EDIT: So the text are inside the .bl files inside UI, they're archives which hold files. I'm not sure how to extract them, however.

 

[samcambolt270]

allright! another translation to look forward to

 

[nobody231]

allright! another translation to look forward to

It might not happen for a little bit, because we're still trying to reverse engineer the game. At least most of the game isn't stuck in giant bin files that we don't understand, unlike Heroes.

 

[samcambolt270]

still looking forward to it. Sadly, the, like 12 projects im following, none have finished. Dqm terrys wonderland is super close though.

 

[Suwazo]

Are the .bl files not some kind of list of other files found in the .jarcs? They seem to be nothing other than file names and then 00s and FFs.
I thought I found unicode text inside pac/system.jarc/message/msg.jmsg but even if it is I don't know where it'd be for. Have no idea what I'm doing really.

 

[Psi-hate]

I'll be glad to help! I'm very interested in translating Dragon Ball Fusions so if you need help, I'll gladly join!

 

[Rioluwott]

Good luck the game is fun but i don´t understand anything
hope you finish this looking forward playing it

 

[nobody231]

I'll be glad to help! I'm very interested in translating Dragon Ball Fusions so if you need help, I'll gladly join!

We'll let you know when we manage to get into the files.

 

[Psi-hate]

Currently tinkering with the bl files, trying to see what I can do.

 

[Helmax]

Would love to take a look at a BL file, but I cannot decrypt this myself for a few days.
Anyways I could get one/two to examine and compare?

 

[Helmax]

The text definately does not appear to be in the BL files.
It looks to be in /pac/*.jarc files.

System.jarc : 0x5c1b8 is where the text seems to start, with a ton of pointers before it (stored in Unicode).
The rest of the file has UTF-8 commands and files inside of it.

I don't know how to extract a JARC yet, but it's all in there.
One of the files in there is called message/msg.jmsg

In unicode, the JMSG in this file starts at 0x17100, seems like there are mulitple pointer tables, and then the pointers start, and theres a TON of them.

 

[Suwazo]

I don't know how to extract a JARC yet, but it's all in there.
One of the files in there is called message/msg.jmsg

In unicode, the JMSG in this file starts at 0x17100, seems like there are mulitple pointer tables, and then the pointers start, and theres a TON of them.

There is a QuickBMS script for JARC. I tried searching and replacing every instance of Kamehame in the msg.jmsg and loading it up and the move name did not change, though I guess it could have changed elsewhere.

 

[Helmax]

-im sure there is a jmsg file n every single jarc file.- ok so i was wrong!

--------------------- MERGED ---------------------------

so you changed every instance, repacked it, and replaced it and used something like NTR to test i assume?

 

[Psi-hate]

I'm able to extract .JARC files. I'll see about getting something done.

 

[Helmax]

I'm able to extract .JARC files. I'll see about getting something done.

I am also able to extract them now, using this method listed above with QuickBMS (thanks to @Suwazo), and i believe there is a repack.bat that comes with QuickBMS you can use to repack the file.

jMsg is still loaded with pointers, and I bet if they arent recalculated, you can't just expand/shrink segments of text.

--------------------- MERGED ---------------------------

Here's what i figured out so far...

MAGIC HEADER - 0x004 - jMSG (6A4D5347)
FILE LENGTH - 0x004 - file length (13D4A8)

I'm not sure what the next 5 0x004 entries are, but the one right after that is A0500400 or 0450A0.
Now this is interesting, because 0450A0 + 18 (the 5 4-byte section and the 4-byte length header, we can discard the magic header for pointer calculations...), is 0450B8
0450B8 is the first text entry and there is furigana included in it.

でも、オレの<6=方,ほう>がちょっと<6=早,はや>いな、
ハッハッハ！

<6= > looks to be our "ruby" tags like in HTML, controlling the furigana.

でも、オレの<6=方,ほう>がちょっと<6=早,はや>いな、
ハッハッハ！

so if we strip those...

でも、オレの方がちょっと早いな、
ハッハッハ！

Hahaha! Sorry, I'm a little bit early! (quick rough no context translation)

--------------------- MERGED ---------------------------

Sure enough then next text entry starts at 045104, subtract 18 for the pointer, and it's 0450EC.
That one also appears in the header.

I don't fully understand the header, otherwise we could dump and recompile this file.

 

[Psi-hate]

Amazing! When I open "msg.jmsg", I can't seem to find out how to find and translate the tect from it. Do you mind explaining?

 

[Suwazo]

That's impressive. Can barely fathom how you would figure all that out. Perhaps the rest of the header is something that doesn't need recalculated.
All I've accomplished is removing the bulk of the file to get this: