shows
English (UnitedStates-en-us)
Catalan (Spain-ca)
Danish (Denmark-da).
However, in [Scriv 3 dir]\hunspell\dict, the directories are named English-en-us, Catalan-ca, Danish-da. I think this is true for all of the entries. The directories are named by language and abbreviation, and the entries in the dropdowns are “named” by language first, and then location and abbreviation.
Yes, at present the labels are just taken directly from the dictionary files. Is this just curiosity or is there something particular you’re looking for? We may be able to adjust these in the future, so the labels for instance just use the language and region as in 1.x and look a bit tidier, but particularly as we’re in the midst of trying to update and expand the available dictionaries, leaving this as-is, and showing all the information that it does, is probably best for the moment.
Inconsistencies sometimes bug me, and this is one of them. I think having language name and region/dialect would be best. It’s not a big issue, but as I was poking around, I noticed the inconsistency.
I went looking for dictionaries, myself, and discovered several sources (some of which I’m sure you’ve seen, but others perhaps not):
freeoffice.com/en/support/a … ctionaries
What you download is a .sox file. Rename the file to a .zip file (example: if it’s en_us.sox, rename to to en_us.zip (or en_us.sox.zip; it won’t matter)), and unzip it. Twenty flavors of Spanish, plus more languages.
cgit.freedesktop.org/libreoffic … ries/tree/
Scroll down to the folder links; they’re labeled with their language code. For example, af_ZA links to the Afrikaans dictionary directory.
github.com/elastic/hunspell/tree/master/dicts
4 years old. I checked the en_US.doc, and it has 62k entries, which is more than the normal one Scrivener loads. These are made for the elastic search project, so may not be suitable.
github.com/LibreOffice/dictionaries
Updates vary. Some dictionaries have not been updated in a while, and others have been updated within the last few months.
extensions.services.openoffice.o … ctionaries
Once you find the language (if you do), find the “Download Extension” button and click it. It will have an .oxt extension. Rename the file with a .zip extension. In other words, if it was Deutsch.oxt, rename it to Deutsch.zip. Unzip it. The .dic, .aff, and README files are in that folder.
app.aspell.net/create
Lets you create a dictionary for yourself. You get to decide what level of words you wish to include. It’s English only. FYI, I run the INSANE level (350k words).
Thanks for the source suggestions! We’ve already got some we’re working on, but if we’re able to implement the ability to add your own custom dictionaries, this will be useful too for other users.
The interface language is separate from the dictionary language. The interface language is set at:
Options (F12) → General → Language (about halfway down the left side of the page). Default is the System language, whatever that is.
The language you use for spell-check, however (which is what I assume you mean by project language), is set at
Options (F12) → Corrections.
In the Spelling section of that page, about a third of the way down, is the dictionary chosen. In my system, that’s English. Click “Download”, and the first one in the list is Afrikaans.
Click that. Click “Download” at the bottom of that dialog box.
Then click Select. A dialog will show; choose the language you want to use from it, and click OK.
Bottom of the dialog, click “Apply.”
It should immediately start using that language for spell-checking. If it does not, restart Scrivener, and it should then be using that language for spell-checking.
If you’re not happy with the Afrikaans dictionary, you can replace it (not a procedure approved by L&L, but it does work. A post elsewhere shows how to do it). If you find a better hunspell dictionary for Afrikaans, please inform L&L about it.