JSK工房 Hgames Translation Thread

[Delexicus]

@My_name_is_Deack
Just out of curiosity, what are you using to do the translation work? Are you just doing the translation in FFDEC? Or are you using something more streamlined like Tengux's text parser? Here's an example of his parser and it's output.

And here's one of the CSV files output by Tengux's parser for Cheeky Devil uploaded to

You must be registered to see the links

.

 

[My_name_is_Deack]

I use JPEXS, just readed manual at top post in this website, a lil bit of testing, and it worked. I have no idea what programms can be used for this type of work, so jpexs is suitable(cant say its very comfortable, but you know, i dont have any option right now). And Notepad++, of course

Here is few things i wanted to ask. Is there any way to browse all ingame content to see how my translation look? Some of text are closed to me, or even far away to look at it, so i need to play game straight to end, just to see how it looks

 

[Delexicus]

I would highly recommend you give Tengux's Text Parser a try. It helps speed up translations significantly. It also has a duplicate line detector/ duplicator feature built in so you don't have to keep re-translating the same lines over and over again. Here's a link to

You must be registered to see the links

, as well as a

You must be registered to see the links

I just wrote up.

There isn't an easy way to browse in-game content. The best approach would be to utilize a 100% save file and skip to where the translation is listed.

EDIT 1) I would recommend looking over some of the other resources available from the

You must be registered to see the links

.

 

[My_name_is_Deack]

Okey, gonna test it now. Maybe it is the better way, thanks a lot, dude.

 

[jesman_8]

Running into an error recompiling using the JSK Parser GUI and I was hoping I could get some assistance. Working on Cornered by a Cheeky Devil.

The issue is when I go to recompile. I'm using the current google doc and run it through the recompile. It goes through the motions and makes the temp folder, but never ends up editing the game. It doesn't seem to write any final product to the disk. It seems as though the program crashes right after reaching the point where it tries to modify the main .swf file. I've looked for an error log of some sort but can't find anything.

Any suggestions as to what to try, or where to find a log file so I can sort it out myself?

Edit: Figured out the issue. Default folder for output is the original game folder making me think it was supposed to overwrite the original files. Pointing to an empty folder resolved the issue!

 

[Delexicus]

Just a heads up. The JSK Community Guide is going to have a different link address. I had to make a copy of the guide Gdoc to get rid of the bloated version history and resolved comments of the original. It was getting increasingly difficult for our editors to effectively stay on top of comments, suggestions, and questions. The old guide link will still work, but it will only be a doc with a quick explanation and a redirect link to the new guide in it. Here's the

You must be registered to see the links

.

 

[jesman_8]

So I've been playing around with the parser GUI and I have to say, it's a great piece of software. I've run into one limitation at this point though and I can't find the actual cause. I tried checking the code in the software to see if I could debug and offer proper advice but it seems I'm much too rusty with java. I'm opting for a bug report instead. Hope the developer sees this!

This is for Cornered by a Cheeky Devil. I've replaced all Japanese text not already translated in the partial with machine translation, but it still missed quite a bit of text. See the attached image for an example. In the .csv generated from both the original game and the partially translated version I'm unable to find the string of characters. The text can be seen in game by lifting the shirt immediately after starting. I'm not sure if this will help with development at all but I hope it does!

 

[Delexicus]

So I've been playing around with the parser GUI and I have to say, it's a great piece of software. I've run into one limitation at this point though and I can't find the actual cause. I tried checking the code in the software to see if I could debug and offer proper advice but it seems I'm much too rusty with java. I'm opting for a bug report instead. Hope the developer sees this!

This is for Cornered by a Cheeky Devil. I've replaced all Japanese text not already translated in the partial with machine translation, but it still missed quite a bit of text. See the attached image for an example. In the .csv generated from both the original game and the partially translated version I'm unable to find the string of characters. The text can be seen in game by lifting the shirt immediately after starting. I'm not sure if this will help with development at all but I hope it does!

I'll make sure to report this to Tengux. He's currently traveling for work so it may be awhile before he has time to fix it though.

 

[jesman_8]

I'll make sure to report this to Tengux. He's currently traveling for work so it may be awhile before he has time to fix it though.

No rush, and I really appreciate you passing it along!

 

[Tengux]

So I've been playing around with the parser GUI and I have to say, it's a great piece of software. I've run into one limitation at this point though and I can't find the actual cause. I tried checking the code in the software to see if I could debug and offer proper advice but it seems I'm much too rusty with java. I'm opting for a bug report instead. Hope the developer sees this!

This is for Cornered by a Cheeky Devil. I've replaced all Japanese text not already translated in the partial with machine translation, but it still missed quite a bit of text. See the attached image for an example. In the .csv generated from both the original game and the partially translated version I'm unable to find the string of characters. The text can be seen in game by lifting the shirt immediately after starting. I'm not sure if this will help with development at all but I hope it does!

Thank you for the compliment, I'm very pleased that you find the parser to be useful.

Sadly I am far away from my usual development workspace for the next few months so I can't provide much assistance. I'm unable to replicate your issue again because I'm away and writing from a small laptop, but based on your report of this bug I have made a small fix. I'm unable to test if this fixed anything or even if this compiled application will run correctly (although I'm pretty sure it will) so I guess I'll leave that up to you.
I put a new version of the parser (GUI_JSK_parser_v1.141_untested) into the Translation and Modding Resources mega directory with the fix for you to try. See if this one finds the lines you were missing and please do tell here if it worked or not

 

[Delexicus]

I tested out v1.41 of Tengux's parser and it looks like it fixed the missing dialogue issue. v1.4 had 41,351 rows of text, compared to v1.41 having 68,005 rows of text for Cheeky Devil. For an out of town, away from your workstation fix, you sure did nail it Tengux! I uploaded a copy of Cheeky Devil with the v1.41 GUI parser extract to the Collaborative translation sheet. It's a separate tab along the bottom as a new sub sheet for comparison.

 

[Tengux]

We must thank jesman_8 for a good bug report that allowed me to easily fix it using just notepad on a laptop that can't even run the application Notepad is all you need.

 

[jesman_8]

We must thank jesman_8 for a good bug report that allowed me to easily fix it using just notepad on a laptop that can't even run the application Notepad is all you need.

You're a legend Tengux! I can't believe how quick your turnaround was on this. Thank you for everything you do, and I'm happy that my bug report proved useful

I must have missed out on the 1.141 version because it's no longer in the Mega directory. Not a problem however thanks to Delexicus putting the new output in the translation doc.

 

[Delexicus]

You're a legend Tengux! I can't believe how quick your turnaround was on this. Thank you for everything you do, and I'm happy that my bug report proved useful

I must have missed out on the 1.141 version because it's no longer in the Mega directory. Not a problem however thanks to Delexicus putting the new output in the translation doc.

I see the problem. The new version is actually v1.141, not 1.41. I must have misread it in my rush to test it out. I checked and v1.141 is in the MEGA repository as GUI_JSK_parser_v1.141_untested.zip

EDIT 1)

The missing file problem was a bit more complicated than I thought. It turns out that I had to reload the the page from the MEGA pull down menu on the right (three horizontal lines). Thanks should go to Tengux for solving this mystery.

 

[TheUnsaid]

Is it possible to put the google translated readme files with the translated game files? They're really helpful if you just want general hints to get the ending instead of a straight-up walkthrough.

 

[Delexicus]

Is it possible to put the google translated readme files with the translated game files? They're really helpful if you just want general hints to get the ending instead of a straight-up walkthrough.

Not a bad idea. I'll put it on the to do list.

 

[Delexicus]

I'm uploading a composite translation Jesman_8 put together for Nano. It has a base of Icevail's v1.2 translation, and has the rest machine translated. I've noticed a few bits are still in Japanese. I'll look into whether they were simply not translated in the Google Sheet, or if the Parser needs a little more tweaking to pick up more lines.

 

[Delexicus]

I found a large problem with the Icevail + machine translation. The scripts sub folders inside the SWFs are renamed between Icevail's version and the stock version. I don't even know how this would have happened. These script folder names are hard linked in the action scripts. This is a problem since Tengux found some errors in the action scripts of Icevail's release. the easy fix would be to overwrite Icevail's translation over top of stock Japanese actions scripts. That is not easily possible with these different script folder names and hard link differences.

Icevail's translation scripts sub folder is named 小悪魔_fla
Japanese stock scripts sub folder is named 小悪魔からは逃げられない_fla

Maybe there was a silent update on JSK's DLsite that would account for this.

EDIT 1) After some thinking, this may be easier to fix than I thought. If the Translation Google Sheet has the 小悪魔_fla in column B changed to 小悪魔からは逃げられない_fla then it should import properly with the parser.

EDIT 2) The above method would only fix text errors, not action script formatting problems. Some more consideration is in order.

 

[TheUnsaid]

I found a large problem with the Icevail + machine translation. The scripts sub folders inside the SWFs are renamed between Icevail's version and the stock version. I don't even know how this would have happened. These script folder names are hard linked in the action scripts. This is a problem since Tengux found some errors in the action scripts of Icevail's release. the easy fix would be to overwrite Icevail's translation over top of stock Japanese actions scripts. That is not easily possible with these different script folder names and hard link differences.

Icevail's translation scripts sub folder is named 小悪魔_fla
Japanese stock scripts sub folder is named 小悪魔からは逃げられない_fla

Maybe there was a silent update on JSK's DLsite that would account for this.

EDIT 1) After some thinking, this may be easier to fix than I thought. If the Translation Google Sheet has the 小悪魔_fla in column B changed to 小悪魔からは逃げられない_fla then it should import properly with the parser.

EDIT 2) The above method would only fix text errors, not action script formatting problems. Some more consideration is in order.

Good luck man. Thanks for all the effort. I'm glad your migrating the best translations into the most recent versions of the flashes.

 

[Delexicus]

So after some rest and a brain reset, I was able to figure out a decent way to find what content was altered in the action script files that was NOT picked up by the parser.

1.) Do a Deconstruct operation on the stock Japanese Nano game. (Tengux Parser)
2.) Do a bulk search and replace on the Stock Japanese temp AS files to fix the 小悪魔からは逃げられない_fla to 小悪魔_fla issue. (PowerGrep)
3.) Do a Deconstruct operation on the Icevail plus Machine Translation Nano game. (Tengux Parser)
4.) Upload The Stock Japanese CSV file a Google Sheet (Google Sheets)
5.) Do a search and replace on column B of the Stock Japanese Google Sheet to fix the 小悪魔からは逃げられない_fla to 小悪魔_fla issue in the Google Sheet. (Google Sheets)
6.) Download the now Icevail compatible Stock Japanese CSV (Google Sheets)
7.) Do a Recompile operation of the Icevail + Machine Translation of Nano using the Icevail compatible Stock Japanese CSV (Tengux Parser)
8.) Do a Deconstruct operation of the just created Icevail + Machine Translation overwritten by stock Japanese. (Tengux Parser)
9.) Do a compare operation Between the temporary_jskparser folder from Icevail compatible Stock Japanese, and the temparary_jskparser folder from Icevail + Machine Translation overwritten by stock Japanese. (Beyond Compare 4 or equivalent)
10.) Browse results to find both formatting errors as well as translation entries that Icevail made manually to the AS files that were outside of the Parser's scope.

Here are the temp AS folders if you'd like to skip steps 1-8.

You must be registered to see the links

234.93 KB

So far I have found some entries that suggest that there WAS a silent update on DLSsite. Icevail's translation is using the newer DLsite version.

The highlighted text translates to Click Lick Nipple. It is missing from the Stock Japanese version.


Edit 1) Curiosity got the better of me and I downloaded a fresh copy of Nano. I compared the fresh Japanese version vs the Japanese copy stored on MEGA. They are the same. Even though Icevail's copy has more entries than the Japanese one…