- Thread starter LordGamer00
- Start date
- Nov 9, 2019
- 6,991
- 47,825
- 883
A big improvement in translation speed. I tested it on a game and it worked, although some things weren't translated. I realized that you shouldn't translate two games in a row. The ideal thing to do is to open a session, translate one game, close the session, and open it again to translate another game. You're doing a great job and gaining more and more knowledge with each challenge.
Thank you
Test game:
https://f95zone.to/threads/natasha-naughty-wife-v0-54-archimedescompany.198161/post-18810227
Thank you
Test game:
https://f95zone.to/threads/natasha-naughty-wife-v0-54-archimedescompany.198161/post-18810227
Thank you very much — you’ve been one of the people who has followed my program and given feedback from the very beginning, and I really appreciate that. You are right, the program is still not in its final form, with every issue and challenge it keeps teaching me new things and continues to improve.A big improvement in translation speed. I tested it on a game and it worked, although some things weren't translated. I realized that you shouldn't translate two games in a row. The ideal thing to do is to open a session, translate one game, close the session, and open it again to translate another game. You're doing a great job and gaining more and more knowledge with each challenge.
Thank you
Test game:
https://f95zone.to/threads/natasha-naughty-wife-v0-54-archimedescompany.198161/post-18810227
The single-use/session problem has been around for a few versions. My plan was to fix other critical issues first and then return to it, but I got so focused on those that I actually forgot to fix it while solving the interface language problem.
The biggest challenge for me has been the translation system. That’s why I reworked it first, and as a result I achieved a 3–4x speed improvement — honestly, even I didn’t expect such a big difference. After that, I reworked the text extraction system: instead of translating directly from the .rpy files, the program now extracts the text first and then translates it. In theory, this should capture and translate all text from the .rpy files, but for some reason there are still parts left untranslated. I’m not entirely sure why.
To be honest, developing with AI assistance doesn’t replace being an expert developer, but I hope I can eventually solve the issue of translating all the text. Thanks again for your comment!
I released a minor update to prevent issues that occur when performing the second translation or canceling the translation, and it also fixes some static text.
You must be registered to see the links
You must be registered to see the links
Hello,
Thank you for sharing your concern. Let me clarify how the translation system in RenLocalizer works.
RenLocalizer’s translation system is designed to work with multiple endpoints to ensure reliability. The primary source is Google Translate, but if Google fails, the program automatically switches to backup endpoints. Specifically, there are three different Lingva Translate endpoints defined in the code.
What is Lingva Translate?
Lingva is an open‑source, ad‑free alternative to Google Translate. It uses Google’s translation backend but provides its own interface and proxy system. The project is community‑driven, transparent, and its source code is publicly available on GitHub.
Some antivirus tools may flag one of these domains (like `translate.plausibility.cloud`) as a potential risk simply because it uses an uncommon address or proxy‑based service. In reality, this is usually a false positive. The code is fully open source and can be reviewed here:
You must be registered to see the links
There is no harmful activity — the connections are only used for translation purposes. To avoid further concerns, I will replace this endpoint with another one in the next update.
Transparency is very important to me, and I truly appreciate your feedback. If you have any other questions, please feel free to reach out — I’ll be glad to help.
Hello,
Thank you for your reply.
I understand better now why my antivirus triggered that alert.
Good luck and thanks again for the tool.
Hello,
Thank you so much for your comment and feedback—I really appreciate it. My hope is to make this tool even better in the future, or if I get stuck, for others to take it further. That's why I decided to make it open source: so that if one day I can't continue developing it, someone else can pick up where I left off.
Right now, the tool is very close to what I envisioned. The only major goal left is to make it capable of translating all text, but I'm still figuring out how to achieve that.
My plan for the next update is to fix the broken translation engines. I was so focused on extracting text from .rpy files and speeding up Google Translate that I didn't notice the other engines barely worked or didn't work at all. I suspected this, but I was too busy with other issues to check.
Therefore, in the next version, I will fix the DeepL API engine and remove Yandex, Bing, and Libre Translator from the program. I may try adding different engines again later. I will also research how to handle the parts that remain untranslated at this time.
Thank you again for your support—it really motivates me to keep developing this project.
Hello, I'm releasing a new update here. I've added a few extra features to the text extraction and translation system, and I've also made some other changes. Deep Scan and RPYC Reader both come enabled by default, but I should mention that despite this, some text still remains untranslated. I haven't been able to figure out how to solve this issue.
RPYC Reader (Experimental):
Scans for hidden strings in init Python blocks, dictionaries, and variable assignments that normal patterns miss.
Smart Pattern Recognition:
Reads compiled .rpyc files directly via AST extraction—no need to decompile to find hidden strings.
RPYC Reader (Experimental):
Scans for hidden strings in init Python blocks, dictionaries, and variable assignments that normal patterns miss.
Smart Pattern Recognition:
Reads compiled .rpyc files directly via AST extraction—no need to decompile to find hidden strings.
You must be registered to see the links
You must be registered to see the links
konstant61
Member
- May 3, 2017
- 206
- 81
- 240
The translation has gotten better...but there are still gaps, although there are much fewer of them now...Is it possible to increase the scanning depth a little more?
Actually, that's what I tried, but I still couldn't cover everything. I need to do more research and understand what I didn't include, meaning what else I need to add to the filter. I checked various wikis, looked at some GitHub projects, and first included Deep Scanning and then the RPYC reader in the project. I plan to make these more detailed, yes, but I'm not entirely sure how to do it. More accurately, I need to understand what I missed. I'll try to develop it as much as possible before fulfilling my military service in my country. Also, if it weren't for the few people who reviewed my project and provided feedback, I probably wouldn't have gotten this far. In short, I want to thank everyone who provided feedback.
konstant61
Member
- May 3, 2017
- 206
- 81
- 240
I've already noticed one peculiarity: if there are a lot of scripts in a folder, the program doesn't find them all...even a deep scan
Thank you for your comment and feedback. I am currently working on a new version. I am trying to improve Deep Scan and the RPYC reader. I will fix this issue in the new version. I hope you will test it when I release the new version. I hope I can solve it.
konstant61
Member
- May 3, 2017
- 206
- 81
- 240
Of course I'll test the new version...the original translator is very good, even at this stage...
️ Update Speech – RenLocalizer 2.2.6
I’ve been working on this update for almost three days. During this time, I added support for several new formats such as `.rpymc` and `.txt`. Despite all these improvements, I still couldn’t manage to translate hidden strings. My tests on Summertime Saga showed that those hidden texts remain inaccessible, simply because I don’t know where they are located.
That said, RenLocalizer has now reached its highest translation coverage compared to all previous versions. To be clear, this statement is only based on RenLocalizer’s own release history — I’m not comparing it to other programs. If a game developer deliberately hides strings, the tool may not be able to find them. But in most cases, where there’s no such intention, translations should work reliably.
Highlights of this release
Known issues
Closing words
In short: the program now translates more than ever (at least based on my tests ). Of course, it still can’t translate absolutely everything, especially those mysterious hidden strings I mentioned. But I wanted to share this working version with you, and I’ll keep improving it.
As always, your feedback is invaluable.
I’ve been working on this update for almost three days. During this time, I added support for several new formats such as `.rpymc` and `.txt`. Despite all these improvements, I still couldn’t manage to translate hidden strings. My tests on Summertime Saga showed that those hidden texts remain inaccessible, simply because I don’t know where they are located.
That said, RenLocalizer has now reached its highest translation coverage compared to all previous versions. To be clear, this statement is only based on RenLocalizer’s own release history — I’m not comparing it to other programs. If a game developer deliberately hides strings, the tool may not be able to find them. But in most cases, where there’s no such intention, translations should work reliably.
Highlights of this release
- Expanded format support (`.rpymc`, `.txt`, and more).
- RPYC Reader integration: direct `.rpyc` file reading via AST extraction.
- Deep Scan algorithm: detects strings in Python blocks, dictionaries, and variable assignments.
- Backup mechanism to preserve original files before modifications.
- Multi-line list and tuple assignment detection.
- Optional scanning of Ren’Py engine files, with safety checks to protect engine integrity.
- Improved handling of non-standard Ren’Py project structures.
- Better error diagnostics and UTF-8 console compatibility on Windows.
Known issues
- The dictionary system may have a bug — I need to investigate further.
- Hidden “spawn-like” strings still remain unresolved. Even after unpacking the game with UnRen and searching all files, I couldn’t find any trace of them. It feels as if they come from outside the project itself, which makes filtering them extremely difficult.
Closing words
In short: the program now translates more than ever (at least based on my tests ). Of course, it still can’t translate absolutely everything, especially those mysterious hidden strings I mentioned. But I wanted to share this working version with you, and I’ll keep improving it.
As always, your feedback is invaluable.
You must be registered to see the links
You must be registered to see the links
konstant61
Member
- May 3, 2017
- 206
- 81
- 240
Hi, The program translates much more, but also makes more translation errors. I tried correcting the errors manually, but it immediately shows a UTF-8 error...
Would you please report the errors? I will need to review them and research how to fix them.
konstant61
Member
- May 3, 2017
- 206
- 81
- 240
Right now, I'll send you the errors from the last translation of the game and then I'll send you the error with the UTF code
konstant61
Member
- May 3, 2017
- 206
- 81
- 240
konstant61
Member
- May 3, 2017
- 206
- 81
- 240
I changed the translation from "California" to "CA" and immediately got this error...
P.S. and the game no longer starts, even when the data is returned in the script
P.S. and the game no longer starts, even when the data is returned in the script