search

etiennebacher / tidypolars

Get the power of polars with the syntax of the tidyverse
https://tidypolars.etiennebacher.com
Other
172 stars 3 forks source link

Is r-multiverse preferred? #112

Closed grantmcdermott closed 5 months ago

grantmcdermott commented 5 months ago

Hey @etiennebacher,

I noticed that both tidypolars and r-polars are available on the new r-multiverse repo. Is this now the canonical source for installing both packages together? Or, should we just see this as potentially convenient duplication beyond your own r-universe builds?

Thanks

eitsupi commented 5 months ago

I believe r-multiverse is still in the testing phase. See also pola-rs/r-polars#888.

etiennebacher commented 5 months ago

Hi @grantmcdermott, as @eitsupi said, it seems they're still building and testing the infrastructure so it's not the preferred way for now.

Sidenote: I saw your repo on polars + duckdb and I'm wondering if you'll include some tidypolars code since you also present duckplyr. I think most dplyr / tidyr functions that apply on a dataframe are now covered (some arguments might be missing in some of them), and the next steps are to improve coverage for more vector functions. I use a relatively limited set of functions in my work so having info on what people use the most would be helpful :)

eitsupi commented 5 months ago

In fact I would like to see r-multiverse become widely available and not bother with CRAN as I find it quite a pain to use Rust on CRAN, but I don't think it's the right time yet. (We have the double burden of stagnation of rustc on Debian and Fedora on CRAN)

grantmcdermott commented 5 months ago

Sidenote: I saw your repo on polars + duckdb and I'm wondering if you'll include some tidypolars code since you also present duckplyr.

Yup, just adding it now! I meant to do so earlier, but have been absolutely snowed at work. Thank goodness, I got the bulk of the workshop materials done over the weekend.

Will be giving a shout-out to both of you ;-)

grantmcdermott commented 5 months ago

Thanks for the clarification on r-multiverse.