Not converting pandas dataframe to R dataframe as of new Pandas 3.0.0 update.

[lq27]

I have a python library which has a function, py_foo() which returns a pandas dataframe. I have an R function r_foo() which returns reticulate::py_foo(). With pandas 2.3.3, this is not an issue, and reticulate automatically converts the pandas dataframe to an R data.frame. But when pandas updated to 3.0.0, reticulate was unable to do the conversion, such that r_foo() returns a python pandas object in my R code.

Using reticulate::py_to_r() did not work; i.e. r_foo() { return reticulate::py_to_r(reticulate::py_foo())doesn't work.

(When I force installed pandas 2.3.3, my original functionality was restored. )

Unfortunately, I don't have a minimal reproducible example, since my libraries and function calls are embedded in two complex library structures.

[t-kalinowski]

Thanks for the report! This has been fixed in the development version of reticulate already. Please install the dev version with:

remotes::install_github("rstudio/reticulate")

I'll be submitting a new verstion to CRAN later this week.

[t-kalinowski]

Fixed in the CRAN release version now.