k, so here are some specific languages that would be useful for a pilot someone hopes to run in minneapolis: cc @ballPointPenguin
- Somali,
so - Amharic,
am(Ethiopia) - Hmong/Miao
hmn[only three-letter ISO 639-2 Code, no two-letter 639-1 code]
Also, a first stab at common indegineou languages might be (roughly top ones spoken in north america):
- Creek/Muskogee
mus - Objibwe
oj - Cree
cre - Dakoto
dak - Cherokee
chr - Inuktitut
iu - Yupik
ypk - Navajo
nv
4 replies
re: native/indigenous language translations. tweeted: https://twitter.com/patconnolly/status/1274060230620516354
1 reply
also, came across this post by Native-Land.ca website, which works with @NativesinTech, so it's likely going about it's request for translator support thoughtfully: https://native-land.ca/hiring-translators/
but that might be a neat thing to figure out (i have no doubt it would be quicker for you to do it yourself :) )
e.g., I'm wondering if
nv-021 / nv-003 (using UN-M49 standards like es-419 does) or nv, might be preferable over nv-US
i'd be interested to seek advice from native speakers, and perhaps support in properly registering the chosen format, if that's required to do this right (and maybe documenting that process, if it's complex)
i haven't pushed it but I wrote a script that generates all the missing languages compared to the list here: https://github.com/unicode-cldr/cldr-localenames-full/blob/master/main/en-001/languages.json#L238
found some bug in the Mojito UI where it doesn't work for language with 3 letter code - I need to dig because we're relying on lib that seem to have a limitation so not a super quick fix
For plural forms i also have a script that gets the data from ICU (so CLDR) but since it doesn't support all locales that's an issue
i'm not familiar with the process for editing/submitting CLDR: http://cldr.unicode.org/ but that's most likely where it should happen
and re: nv.
bc it seems that a region code is always placed after.
but using
US is kinda a bit of a kick for native folks
true...
i can imagine there might be a preference to be explicit with an exception though. depends on what ppl say when i ask. i'd defer
but if someone wished to register with M49 as the region code (which i understand to be an exception for unique circumstances), i'd be stoked to help with the process :)
if its a convo that is going to happen, it can happen after
i might use the PR as an excuse to have a discussion with others, but pls don't assume any convo in there means I'll expect it merged. consider it throwaway :)
thanks @aurambaj -- i should get some sleep now...!
Hi all, I need help with mojito.
How can I translate the same souce one time?
How can I translate the same souce one time?
For example the word "Yes" is repeated x-time. Can I select all the occurrences and set the releated translation?
Can I do this "massive insert" using mysql or command line?
@dusty-82 you can use the
leveraging-copy-tm command. Depending on how many strings you have and precision you want you can use match based on text unit ids some like mojito leveraging-copy-tm -m TUIDS --tuids-mapping "1001:2001;1002:2002" or a little bit looser you can provide just source/target repository (and asset) and match by content something like mojito leveraging-copy-tm -m MD5 -s repo -t repo potentially limiting to some text unit names with --name-regex
if you want to be very precise on what translation you copy: source text unit to target text unit you can provide a mapping of ids with
-m TUIDS and providing the exact mapping --tuids-mapping "1001:2001;1002:2002"