[R] how to combine logic test on vectors in R?

Luigi Marongiu m@rong|u@|u|g| @end|ng |rom gm@||@com
Thu Sep 30 15:06:37 CEST 2021


Hello,
I have two data frames, each with three rows:
```
df_a <- data.frame(a = letters[1:3], b = LETTERS[1:3], q = c("", "", ""),
stringsAsFactors = FALSE)
df_b <- data.frame(a = letters[4:6], b = LETTERS[4:6], q = c("", "", "1.5"),
stringsAsFactors = FALSE)
```
I need to test whether the dataframe has been selected and if there is
a value in the q column. I combined in the following test:
```
if (nrow(df_a) == 0 || unique(df_a$q) == "") {
print("empty")
}
if (nrow(df_b) == 0 || unique(df_b$q) == "") {
print("empty")
}
```
The test for df_a worked as expected:
```
> nrow(df_a) == 0
[1] FALSE
> unique(df_a$q) == ""
[1] TRUE
> (nrow(df_a) == 0 || unique(df_a$q) == "")
[1] TRUE
> if (nrow(df_a) == 0 || unique(df_a$q) == "") {
+ print("empty")
+ }
[1] "empty"
```
but the one for df_b did not:
```
> nrow(df_b) == 0
[1] FALSE
> unique(df_b$q) == ""
[1]  TRUE FALSE
> (nrow(df_b) == 0 || unique(df_b$q) == "")
[1] TRUE
> unique(df_b$q)
[1] ""    "1.5"
```
I say that it did not work because unique(df_b$q) IS NOT "", hence
`(nrow(df_b) == 0 || unique(df_b$q) == "")` should be FALSE, instead R
evaluated the first element of unique(df_b$q) == "", which is TRUE.
How can I properly implement a logic test on vectors?
 Thank you



More information about the R-help mailing list