Before I put a large amount of work into this. Anyone know if there's any ongoing translation efforts for the recent versions? According to the game, I'm running violence 2.44 and molestation 2.24, which I assume is the latest version.
My current plan is to take the translations done by Magus2914 for version 2.31 in the post I'm quoting below, and copy that over to the recent game version. Once that's done, I'll work on updating the rest of the lines that are missing translations.
Based on the line counts between the trans_scenario files, the version I'm working with has an additional 11k lines compared to the files provided by Magus2914, but there's still 29k lines that he did translate, so it would save me time if I can use a translation file that has already been updated. No big deal if not, just want to know what's available.
And if anyone has any info they think might be helpful to this translation effort, I'm all ears. I'm kind of just jumping in after not paying attention to this game for a couple of years, and I haven't caught myself up on the translation status as a whole.
I'm not currently working on translating the recent versions of this game, but I can at least explain how I did it in the past and what pitfalls to watch out for. I did most of my work by opening the .csv files in Notepad++ since the files are formatted in an unusual way and I didn't want to mess with that. If you try working in Excel, it adds a bunch of quotation marks and I didn't think that would play well with the game engine.
First Step: Update the old .csv files
My first step when a new game version was released was to use KDiff3 to compare my translated versions of the .csv files to the latest versions included in the patch (2.44 does appear to be the latest version as-of August 11, 2023). I'd use this to slowly walk through the file to identify any lines that were removed or added and manually copy or delete the lines from my translated file. At the end of this step, I'd have a set of .csv files with all my old translations and all the new untranslated lines ready to go.
Second Step: Figure out how much work needs to be done
While I did all my translations in Notepad++, I would still open the .csv files in Excel (I did this using a side copy since accidentally saving would irreversibly ruin all my work). I'd filter the rows to only show the ones that had blanks in the second column (each record's format is "Japanese Text"<Tab Character>"English Text if translated or nothing if not"<Tab Character>"Developer Notes or nothing") and then sort the first column from A to Z. By ignoring the rows for lines that you shouldn't touch (like ones starting with a # or that just say "ver1.00") you can figure out exactly how many lines you'll need to translate to cover everything. You can also use Excel as you make progress to check that you didn't mess up the tabs somewhere and accidentally put text in the third column. I've never not made any mistakes.
Third Step: Get the easy stuff out of the way
As you scroll through Excel after sorting the untranslated lines by the first column, you'll notice that there's a lot of duplicated text like "･･････" or "！". Do yourself a favor and do a Find/Replace in Notepad++ to translate all of these at once. You'll also see a lot of sound effects like "くちゅくちゅくちゅっ♥" which already have translated versions that you can copy and Find/Replace to get them out of the way. Any normal-looking lines that are repeated a few times may also have translations already, so do a quick text search to see of you can take care of these. If you notice any new characters showing up, you'll need to add a row to the trans_name.csv file so their name gets translated. At the end of this step, the only blanks left in the second column will be the lines you'll actually need to translate. I kept track of the number of rows remaining to help motivate me as I went.
Last Step: Actually translate stuff
Again, I'd start with the lines that had duplicates since it feels good to translate a single line and be able to knock your list of remaining lines down by 5 or whatever. I did my translating using Google Translate and manually copying lines into my browser. Once I was left with a set of mostly-unique untranslated lines, I'd instead use Excel to filter out the translated ones but not sort them by the original text so they'd highlight which lines of the file I still needed to translate. I'd take the lines a scene at a time to give myself achievable goals. For style points, once you're done translating and verifying that you didn't mess-up the tabs, you can copy the entire file into Word to check your spelling. You'll get a bunch of false positives, but this is another way to see if you've made any mistakes. I've never failed to make typos, but this also let me find several mistakes in the original translation.
Why I stopped translating
As you can see, translating this game isn't a small amount of work. There's probably a smarter way to do it, but this is what worked for me. The current game updates revolve around walking around the city which is a much more translation-intensive game style than the original stuff on the train. Since the city scenes are still being made, I've had to scrap entire translated scenes when the developer cut them or made significant changes which is demotivating but understandable. The changes to old lines and the significant amount of admin needed to update the translated file every time are what made me stop. I could imagine coming back to this someday once the content is settled, but due to the slow updates and the amount of work spent just updating the old files to be able to translate the new lines, I don't see that happening anytime soon.
Good luck with your own translation efforts! I agree it's a shame that the new content isn't translated.
