read_sas does not handle special missing numeric variables

[RaymondBalise]

read_sas silently sets all of the different kinds of missing SAS values to the generic R NA. So I can't use the *_tagged_na function in haven without manually recoding them.

Is it possible to add an option to convert .A through .Z to other values or mark them appropriately with the haven *_tagged_na functions?

This would be wildly helpful for people who are dealing with data that is not missing at random (missing because someone failed to show up for drug screening is very different than the lab tech dropped the sample).

Unfortunately you need SAS for this reprex:

data "c:\blah\blah.sas7bdat";
input a b;
datalines;
1 .a
2 .b
27 .
28 1
;
run;

---

read_sas does not import missing numeric values from SAS.

library(haven)
read_sas("c://blah//blah.sas7bdat")

[RaymondBalise]

Never-mind I see missing values with print_tagged_na(blah$b). It would be very useful if the print method for a tibble imported with missing data showed the NA type.

[lock]

This old issue has been automatically locked. If you believe you have found a related problem, please file a new issue (with reprex) and link to this issue. https://reprex.tidyverse.org/