Variable names in different language

General help and assistance with jamovi. Bug reports can be made at our issues page: https://github.com/jamovi/jamovi/issues . (If you're unsure feel free to discuss it here)

by davidville » Wed Feb 06, 2019 2:16 am

Hello,
All is great with JAMOVI and I use it a lot.
I can change the names of variables using a different language (Hebrew) and One-way ANOVA has no problems.
But ANOVA, MANCOVA and linear regression (at least) give me an error:
<text>:1:10: unexpected symbol 1:cbind( .....etc.....
I was wondering if there might be a fix for this?
Thank you so much
davidville
 
Posts: 4
Joined: Thu Oct 04, 2018 12:23 pm

by Ravi » Thu Feb 07, 2019 9:18 pm

Hi david,

Can you attach the .omv file where this happens?\

Cheers,
Ravi
User avatar
Ravi
 
Posts: 131
Joined: Sat Jan 28, 2017 11:18 am

by davidville » Tue Feb 12, 2019 11:57 pm

Here is an example but it happens in other files as well
Attachments
Example.omv
(3.88 KiB) Downloaded 3 times
davidville
 
Posts: 4
Joined: Thu Oct 04, 2018 12:23 pm

by jonathon » Wed Feb 13, 2019 1:34 am

so the issue here is that a lot of R packages don't handle unicode very well, so we re-encode the variable names into what's called base 64 (an encoding which only uses 0-9a-zA-Z_.). then we prefix them with a . because names in R can't begin with an _. the issue here is that hebrew characters have the misfortune of being encoded to something like:

.15DFx

and R looks at the '.15' and says "ok, so this is a number", and spits the dummy when it gets to the DFx.

so it's likely an issue which only affects the hebrew language, because, coincidentally, hebrew characters get encoded to begin with numbers.

anyhow, all i need to do is change the prefix . to something else, say an x

x15DFx

and the problem is solved.

this will be fixed in the next release.

cheers

jonathon
User avatar
jonathon
 
Posts: 674
Joined: Fri Jan 27, 2017 10:04 am


Return to Help

cron