App languages

[LenhartStephan]

Hey, I love where this project is going! I wanted to ask if you already accept app translations or if it's still too early for that.

[niksedk]

What OS are you using?

You can retrieve the current translation in SE via a shortcut, via Options - Shortcuts:

You can e.g. translate with a tool like: https://github.com/niksedk/json-content-translator

The translated language file should be in a folder called "Languages" (I think)

Not all string are translated, but many should be.
You will be the very first to test this :)

[LenhartStephan]

I'm using both macOS and Windows.

So I've played a bit with the translations: The translation files must be in the zip archive in src/UI/Assets/Languages.zip. I don't think the files in ./Languages are used at all, but I'm not entirely sure. The whole files in zip situation is a bit odd, especially since archives are not the best for version control. Also after changing the Languages archive (adding/changing a language) cleaning/rebuilding the project was not enough to see the changes. I had to manually delete the generated folder in ./src/UI/bin.

Since SE4 already has a lot of translated strings, I've started creating a mapping for converting the SE4 xml files to the new json-based translations: https://github.com/LenhartStephan/xml-to-json-converter/tree/subtitle-edit. While it's not possible to create a complete translation file that way, since some strings don't exist in SE4, you can at least create a good foundation.

Your json-content-translator works really well by the way! The mapping intentionally creates empty strings for missing translations, so that it's possible to use the "find next empty" function of your tool.

I'm still working on expanding the mapping, but I'm confident this is a great foundation for creating translations for SE5.

Here is an example screenshot from the german translation:

[mtsubs]

Hi, @LenhartStephan

Nice work, I’m also thinking about contributing translations ( #161 ).

You mentioned a map. Is it possible to use a script to bulk extract all previous translations so we can build on the languages we already have?

Can you share more details about the existing translations?

[LenhartStephan]

Hi, I've written a little python script and mapping in https://github.com/LenhartStephan/xml-to-json-converter/tree/subtitle-edit (subtitle-edit branch) that takes the existing translation files from SE4 in XML format and outputs them in the json format for SE5. All missing strings will be empty so you can use eg. https://github.com/niksedk/json-content-translator to fill in the gaps.

I would still strongly recommend to double check all translations, as there may be nuances in the respective language where a translation doesn't quite fit in the respective context.

Once you have the finished translation file, checkout and run the current build with it. If the base-translation was updated in the meantime (very likely), use the shortcut binding as described by niksedk above to get the current merged translation (dont be confused the file is always outputted as english.json; if the shortcut doesn't work, try focusing the main textbox in the SE UI). This file contains your translation as well as all missing translations (in English), which you can then translate again.

[niksedk]

Nice converter :)

I have also changed the language structure (no longer a zip file), so it should be easier see language stuff and make PRs: https://github.com/niksedk/subtitleedit-avalonia/tree/main/src/UI/Assets/Languages

Anyone can also just post language files here or email (if you do not like the hassle with PRs).

[mtsubs]

I’m wondering whether @niksedk could maintain an official mapping file. That would make translations much easier: whenever the UI is updated translators need to know which keys map to which texts, and which entries are new or removed. Right now determining those differences is time‑consuming. Maintaining an official mapping could clearly indicate what needs translation and save a lot of effort.

[niksedk]

SE 4 import is now included in the JsonContentTranslator: https://github.com/niksedk/json-content-translator/releases

Step 1 - Open the json base language file (e.g. English.jsn)
Step 2 - Press "Cancel" when prompted for "Select translation JSON file" (or pick a file if you got one).
Step 3 - Click the button "Import SE 4 xml..." and choose your old language file (e.g. da-DK.xml)

You will get a conversion report + hopefully many language string imported:

[mtsubs]

SE 4 import is now included in the JsonContentTranslator: https://github.com/niksedk/json-content-translator/releases

Great! I’ll look into it later (I’m translating another film project right now).

[niksedk]

And json-content-translator 1.3 with improved SE 4 import:
https://github.com/niksedk/json-content-translator/releases

[ArsenTech]

That's great! Now I can import all of my previous translations into v5 once the SubtitleEdit is released as a stable version