topic Re: How to work with R tibble objects in Discussions

How to work with R tibble objects

JHX — Fri, 09 Jun 2023 18:12:50 GMT

I really like integrating R code in JSL, but how can I deal with the correct handling of a tibble (dplyr) object. It seems like the object is damaged, when displayed in JMP. The output on my PC is visible in the appended image.

R Init();
R Submit("library(dplyr)");
x = R Execute({},{data_table},"\[ 
    data_table <- tibble(x = c("A", NA, "B", "C"), y = c("D", "E", "F", "G"))
]\"
);
data_table << New Data View();
Show(data_table);
R Term();

Re: How to work with R tibble objects

peng_liu — Sat, 12 Mar 2022 03:18:34 GMT

I think the problem is caused by NA in the first column. JMP does not know how to deal with it. I don't think that JMP differentiates an empty string from a missing value in a Character column. You may consider changing NA to an empty string or some special string, if that works for you.

Re: How to work with R tibble objects

JHX — Fri, 18 Mar 2022 08:07:25 GMT

thank you peng_liu, at least it works. I was not aware that the big dot indication missing values in JMP is only for the numbers and character do not have NA values. I still hope for an better NA handling in the JSL/R interface in comming versions.

R Init();
R Submit("
	library(dplyr)
	library(tidyr)
");

x  = R Execute({},{data_table},"\[
	data_table <- tibble(x = c(1, NA, 2, 3), y = c("D", "E", "F", "G"), z = c("A", NA, "B", "C"), a = c("H", "I", "J", "K")) %>%
		mutate(across(where(is.character), ~replace_na(.x, "")))
	data_table
	]\"
);

data_table << New Data View( );
//limits << New Data View();

Show(data_table);
R Term(); //Exit

Re: How to work with R tibble objects

peng_liu — Fri, 18 Mar 2022 17:32:45 GMT

I am not sure what third party interfacing with R can do. If I remember correctly, NA in R is a specially encoded double precision number, so it does not align well with other elements in the same vector in memory in this case. Please contact support@jmp.com in case there is a better solution.