Yeah, it's working fine now. I think I might have only applied the dohnadohna.ain file the first time through. It's kind of funny though, my gimped patch job had certain parts of the UI text translated that the whole patch didn't. Hopefully this is just because I tried re-patching in the middle of a run, but whatever. Certain team members no longer have their moves and stuff translated to english. Meh. Are the badge descriptions translated with any of the patches? I don't really mind, the game is mindless enough I've been playing without them for 16 episodes, just curious.
Yeah, I have the ain on my side. It's alright if you can't, no rush. Just thought it would be easier for people to translate if the sheet had the character names too instead of just the textbox lines.
Sorry for the double post but I believe I should put this out here before I forget.
If anyone wants to restart the project and needs an editor/proofreader, hit me up. I would prefer not to work with MTL though, because that shit breaks minds. Alternatively, we can wait for Eroguysensei to do his video thing, then use that to make a patch. Whatever works. As long as the NTR bad ends aren't in poetry I'm fine with anything.
The thing is i literally auto selected all the lines with ";m" and copy pasted them to the sheet in like 2 minutes, i spent more time writing the side text and giving good "looks" to the sheet XD. If i find a way to make it fast like that with the ;s lines that have the names i will update it.
If anyone knows a way to automatically select or filter like in the photo it would be amazing.
U should try it, Eraea did an amazing job with the patch and like u can see in the photo it has like 70% of the UI already translated, its a shame that he had to stop...
We need the ";s" lines with a name and the ";m" lines under that name to have them in order. I will look into regular expression but idk if u can select or search for diferent "expressions" at the same time.
Basically we want to know what character says the ;m lines. We have the ;m in order already but its easier if u know what characters are talking.
With regular expression, you can basically look for any mix of words/letters/symbol with wildcards,optional symbols and repetitions. It is really powerful and depending on what scripting language you use, they have different functions.
For Python for example:
(;s.*[A-Z]{1,}"\n;s.*\n(;m.*\n){1,})
This would return the specific 3+ lines of text starting with ;s and the name in Upper case letters, the next ;s line, followed by 1 or more ;m lines. (see
;s denotes scripts / engine commands. ;m denotes messages displayed in the textbox.
So the situation now is that our brother Sun there has already had the ;m lines extracted. Problem is, he's got the text but not the speakers (which is stored in the ;s lines). Whoever's gonna be translating this will have to translate without knowing who's speaking. Recipe for disaster? Yes.
Now to fix that, we gotta get the speakers too. Observe.
Crossed-out lines are the other scripts, it does stuff like play music, display images, give you your nude Porno sprite, etc. Basically we don't want to touch them. Now the blue lines are the ;m lines, already done with. Our interest lies with the red lines, which stores speaker name. Thing is, we're not sure how to separate them from other ;s lines. They're all in the format of >>> ;s[<number>] = "<some japanese>". If we can figure this out, we can solve the above problem.
Alright, explanation done. Now then, SUNCRIS1010, I noticed that the lines we're looking for are always positioned like:
So you can get them if you do something like:
if ;m then >>> if previous line ;s[] = " + integer + " then >>> get previous line
Basically always get the lines two lines above ;m lines with a "「". Unless it's a ＊, in which case get the line above that. This explanation is probably too shitty to be understood lmao. Uh well, good luck.
Oh, and disclaimer for the +1 people: no one is working on this project atm.
;s denotes scripts / engine commands. ;m denotes messages displayed in the textbox.
So the situation now is that our brother Sun there has already had the ;m lines extracted. Problem is, he's got the text but not the speakers (which is stored in the ;s lines). Whoever's gonna be translating this will have to translate without knowing who's speaking. Recipe for disaster? Yes.
Now to fix that, we gotta get the speakers too. Observe.
Crossed-out lines are the other scripts, it does stuff like play music, display images, give you your nude Porno sprite, etc. Basically we don't want to touch them. Now the blue lines are the ;m lines, already done with. Our interest lies with the red lines, which stores speaker name. Thing is, we're not sure how to separate them from other ;s lines. They're all in the format of >>> ;s[<number>] = "<some japanese>". If we can figure this out, we can solve the above problem.
Alright, explanation done. Now then, SUNCRIS1010, I noticed that the lines we're looking for are always positioned like:
So you can get them if you do something like:
if ;m then >>> if previous line ;s[] = " + integer + " then >>> get previous line
Basically always get the lines two lines above ;m lines with a "「". Unless it's a ＊, in which case get the line above that. This explanation is probably too shitty to be understood lmao. Uh well, good luck.
Oh, and disclaimer for the +1 people: no one is working on this project atm.
I have filtered the text file in sections. Each section needs to fulfill the following conditions:
starts with ;s, followed by any other caracters any number of times (ending with a "newline" expression)
OPTIONAL: next line is the same as above, but ends with *" (btw, it is not the same asterisk as on an english keyboard)
for next line(s), repeat step 1 and 2. (<- this line essentially contains the number)
next line starts with ;m, followed by any character any number of times (ending with a "newline" expression)
Step 4 can be repeated any number of times.
Overall, this results in 57533 lines that start with ;m.
One problem that i can see is that this won't be able to extract sections that only contains 1 ;s (e.g. cases where the line directly above and below it start with ;m)
