Support for Butterfly's Poison; Blood Chains (Steam version)

[Linsey2244]

Hello, would it be possible to add support for Butterfly's Poison; Blood Chains (https://vndb.org/r108105)? It's a visual novel released on Steam and I would like to request compatibility support for the game.

Here are some sample files (the game .exe and also the SCRIPT.PAK file) that might be useful: https://files.catbox.moe/xs93yo.zip
If more files or any additional information are needed feel free to let me know and I’ll be happy to provide them.

I understand if it's not planned or can't be done, thanks!

[thedanill]

Didn't have time to test the plugin properly, but it should work. Let me know when you check it.
The archive needs to be unzipped into the data folder.

ButterflysPoisonBC_plugin.zip

[Linsey2244]

Thank you very much! I tested the files and the scripts are extracted correctly from the game. However, after modifying even just a single line of dialogue, importing it, and running the game, the dialogues show perfectly but later an issue appears: at the first in-game choice, no matter which option I select, the game doesn't progress as it should. Instead, a message pops up saying Clear data (system data) was saved. Sometimes (depending on which or how many lines I change in the script), it jumps to a completely unrelated scene that doesn’t correspond to the chosen option.

Is there a way to fix this so the game progresses normally? I’m not sure if it has to do with the game or executable checking for a specific string length or offset somewhere in the code.

Screenshots showing this issue after reimporting the modified SCRIPT.PAK:

Also, there are two other issues unrelated to SCRIPT.PAK that I’ve encountered but maybe it's because I'm doing something wrong:

1. Several .pak files can’t be extracted. Maybe it is because they seem to be split into parts like "SYSCG.PAK", "SYSCG2.PAK", "SYSCG3.PAK", etc. (although I’m not sure) and when I try to extract them I get an error like this:

The filename, directory name, or volume label syntax is incorrect.
goroutine 1 [running]:
github.com/golang/glog.stacks(0xc00018d200, 0xc00007e5a0, 0xbe, 0x10e)
        /home/runner/go/pkg/mod/github.com/golang/glog@v1.0.0/glog.go:769 +0xbf
github.com/golang/glog.(*loggingT).output(0xe20be0, 0xc000000003, 0xc00016e000, 0xbd0663, 0xd, 0x21, 0x0)
        /home/runner/go/pkg/mod/github.com/golang/glog@v1.0.0/glog.go:720 +0x39a
github.com/golang/glog.(*loggingT).println(0xe20be0, 0x3, 0xc000073d30, 0x1, 0x1)
        /home/runner/go/pkg/mod/github.com/golang/glog@v1.0.0/glog.go:633 +0xee
github.com/golang/glog.Fatalln(...)
        /home/runner/go/pkg/mod/github.com/golang/glog@v1.0.0/glog.go:1141
lucksystem/cmd.glob..func9(0xe01a00, 0xc00005d080, 0x0, 0x6)
        /home/runner/work/LuckSystem/LuckSystem/cmd/pakExtract.go:33 +0x61b
github.com/spf13/cobra.(*Command).execute(0xe01a00, 0xc00005d020, 0x6, 0x6, 0xe01a00, 0xc00005d020)
        /home/runner/go/pkg/mod/github.com/spf13/cobra@v1.5.0/command.go:876 +0x2ec
github.com/spf13/cobra.(*Command).ExecuteC(0xe00100, 0xc000120010, 0x8, 0xf)
        /home/runner/go/pkg/mod/github.com/spf13/cobra@v1.5.0/command.go:990 +0x375
github.com/spf13/cobra.(*Command).Execute(...)
        /home/runner/go/pkg/mod/github.com/spf13/cobra@v1.5.0/command.go:918
lucksystem/cmd.Execute()
        /home/runner/work/LuckSystem/LuckSystem/cmd/root.go:39 +0x45
main.main()
        /home/runner/work/LuckSystem/LuckSystem/main.go:12 +0x27

2. When I try to run the lucksystem pak replace option after modifying (or even without modifying) files from a previously extracted PAK, I get this error:

invalid syntax
goroutine 1 [running]:
github.com/golang/glog.stacks(0xc00018d100, 0xc00017e460, 0x77, 0x9c)
        /home/runner/go/pkg/mod/github.com/golang/glog@v1.0.0/glog.go:769 +0xbf
github.com/golang/glog.(*loggingT).output(0xe20be0, 0xc000000003, 0xc000235180, 0xbd069d, 0xd, 0x26, 0x0)
        /home/runner/go/pkg/mod/github.com/golang/glog@v1.0.0/glog.go:720 +0x39a
github.com/golang/glog.(*loggingT).println(0xe20be0, 0x3, 0xc00011dcb0, 0x1, 0x1)
        /home/runner/go/pkg/mod/github.com/golang/glog@v1.0.0/glog.go:633 +0xee
github.com/golang/glog.Fatalln(...)
        /home/runner/go/pkg/mod/github.com/golang/glog@v1.0.0/glog.go:1141
lucksystem/cmd.glob..func10(0xe01c80, 0xc00005d080, 0x0, 0x6)
        /home/runner/work/LuckSystem/LuckSystem/cmd/pakReplace.go:38 +0x485
github.com/spf13/cobra.(*Command).execute(0xe01c80, 0xc00005d020, 0x6, 0x6, 0xe01c80, 0xc00005d020)
        /home/runner/go/pkg/mod/github.com/spf13/cobra@v1.5.0/command.go:876 +0x2ec
github.com/spf13/cobra.(*Command).ExecuteC(0xe00100, 0xc00011e010, 0x8, 0xf)
        /home/runner/go/pkg/mod/github.com/spf13/cobra@v1.5.0/command.go:990 +0x375
github.com/spf13/cobra.(*Command).Execute(...)
        /home/runner/go/pkg/mod/github.com/spf13/cobra@v1.5.0/command.go:918
lucksystem/cmd.Execute()
        /home/runner/work/LuckSystem/LuckSystem/cmd/root.go:39 +0x45
main.main()
        /home/runner/work/LuckSystem/LuckSystem/main.go:12 +0x27

I’ve attached several .txt files with examples of the errors mentioned, along with the original game files in case they’re helpful for debugging.

Thanks again and sorry to bother you!

Errors.zip
Output when importing modified SCRIPT.PAK.txt
Game files: https://sbupload.com/155c5fea603ea165

[thedanill]

Finally got some free time. Yay.

1. This issue is related to the ONGOTO opcode (seems like this instruction sets "hooks" on other instructions: when the interpreter jumps to this instruction, the expression from ONGOTO is automatically executed).

This instruction is almost never found in games from KEY/Visual Arts (i can only recall Little Busters), but it's present in almost all third-party developers using this engine.

LuckSystem assumes that one address can only be linked to one jump. I don't really want to rewrite all the jump logic, so for now I've written a quick rough fix. Did a quick check — seems like everything works.
Check this build.

LuckSystem_windows_x86_64_ONGOTO.zip

2. The engine doesn't have a system for part-archives, all of them are full archives. If there's a number in the name, it's just because the developer decided to logically split the data into several parts.

Regarding the extracting issue: I might be wrong, but it seems the problem is that the destination folder doesn't exist. You need to create it manually.
Regarding replacing: quite a strange error, I'll try to check it later.

If you need a solution for working with PAK archives right now, you can use the tools from the another repository https://github.com/G2-Games/lbee-utils.

[Linsey2244]

The build you shared seems to work perfectly for the scripts, now when choosing an option the game proceeds as it should. Thank you so much!

Regarding the errors when extracting some of the PAK files: I tried creating the output folder manually before running the command, but I still get the same error. I’m not sure if this helps, but I noticed that in all the cases where the error occurs, there’s one or more strange characters appearing next to the supposed output path the program detects.

For example:

In this case it shows a right arrow (►):

lucksystem pak extract -i FONT2.PAK -o FONT2.txt --all data/font_out  
F0512 18:42:50.758593   21560 pakExtract.go:33] open C:\Users\monic\Downloads\butterfly poison\data\font_out/►: The filename, directory name, or volume label syntax is incorrect.

Here, it shows an up arrow (↑):

lucksystem pak extract -i OTHCG2.PAK -o OTHCG2.txt --all OTHCG2  
F0512 18:51:58.891360   10884 pakExtract.go:33] open C:\Users\monic\Downloads\butterfly poison\OTHCG2/↑: The filename, directory name, or volume label syntax is incorrect.

And in this one, it shows two symbols (e☺):

lucksystem pak extract -i PARTS2.PAK -o PARTS2.txt --all PARTS2  
F0512 18:54:53.039656    6400 pakExtract.go:33] open C:\Users\monic\Downloads\butterfly poison\PARTS2/e☺: The filename, directory name, or volume label syntax is incorrect.

Also, I tried using lbee-utils on those same files and it fails to open them too (the program crashes and closes almost instantly). Interestingly, the files I can extract with LuckSystem also work fine with lbee-utils except for SYSCG3.PAK, which crashes lbee-utils, while LuckSystem can extract it (although it seems like not entirely).

[thedanill]

Awesome to hear everything's working! You can totally use this now, but these changes will hit the repo a bit later once I clean things up.

Took a close look at the PAK archives, and something weird is going on with them. Like, OTHCG2 is missing part of the header that lists the files. It really does seem like there's a system of part archives, which I've never seen in other games before.

Been super busy lately, but I'll try to get back with updates later.

[Linsey2244]

Took a close look at the PAK archives, and something weird is going on with them. Like, OTHCG2 is missing part of the header that lists the files. It really does seem like there's a system of part archives, which I've never seen in other games before.

Been super busy lately, but I'll try to get back with updates later.

Hello again @thedanill, just checking in to see if you've had any chance to look into the weird PAK files (like OTHCG2, PARTS2, etc.). Totally understand if not, just wondering if there's been any update/progress. Thanks!

[thedanill]

Took a close look at the PAK archives, and something weird is going on with them. Like, OTHCG2 is missing part of the header that lists the files. It really does seem like there's a system of part archives, which I've never seen in other games before.
Been super busy lately, but I'll try to get back with updates later.

Hello again @thedanill, just checking in to see if you've had any chance to look into the weird PAK files (like OTHCG2, PARTS2, etc.). Totally understand if not, just wondering if there's been any update/progress. Thanks!

Hello! Don't have much to say yet, but it's on my list. I'll get to it soon.

[Lilislv]

Finally got some free time. Yay.

1. This issue is related to the ONGOTO opcode (seems like this instruction sets "hooks" on other instructions: when the interpreter jumps to this instruction, the expression from ONGOTO is automatically executed).

This instruction is almost never found in games from KEY/Visual Arts (i can only recall Little Busters), but it's present in almost all third-party developers using this engine.

LuckSystem assumes that one address can only be linked to one jump. I don't really want to rewrite all the jump logic, so for now I've written a quick rough fix. Did a quick check — seems like everything works. Check this build.

LuckSystem_windows_x86_64_ONGOTO.zip

2. The engine doesn't have a system for part-archives, all of them are full archives. If there's a number in the name, it's just because the developer decided to logically split the data into several parts.

Regarding the extracting issue: I might be wrong, but it seems the problem is that the destination folder doesn't exist. You need to create it manually. Regarding replacing: quite a strange error, I'll try to check it later.

If you need a solution for working with PAK archives right now, you can use the tools from the another repository https://github.com/G2-Games/lbee-utils.

doesn't work on cartagra, game craches on the opening

[Linsey2244]

Hello! Don't have much to say yet, but it's on my list. I'll get to it soon.

Hello @thedanill, any update on this?