Some R gotchas

Some potentially surprising results from R functions which are important to be aware of.

Here is a stackoverflow discussion with some more.

For matrix:

m <- matrix(1:6, ncol=3)
m

##      [,1] [,2] [,3]
## [1,]    1    3    5
## [2,]    2    4    6

m[1,] # vector

## [1] 1 3 5

m[,1] # vector

## [1] 1 2

m[1,,drop=FALSE] # matrix

##      [,1] [,2] [,3]
## [1,]    1    3    5

For data.frame:

df <- data.frame(a=1:3,b=4:6)
df

##   a b
## 1 1 4
## 2 2 5
## 3 3 6

df[1,] # data.frame

##   a b
## 1 1 4

df[,1] # vector

## [1] 1 2 3

df[,1,drop=FALSE] # data.frame

##   a
## 1 1
## 2 2
## 3 3

df <- data.frame(a=c("10","11","12"))
as.numeric(df$a) + 1

## [1] 2 3 4

df <- data.frame(a=c("10","11","12"), stringsAsFactors=FALSE)
as.numeric(df$a) + 1

## [1] 11 12 13

boxplot: names & horizontal barplot: names.arg & horiz stripchart: group.names & vertical

these are not 'gotchas' but good to remember:

c(TRUE,TRUE,FALSE) & c(FALSE,TRUE,TRUE) # element-wise

## [1] FALSE  TRUE FALSE

c(TRUE,TRUE,FALSE) && c(FALSE,TRUE,TRUE) # just the first

## [1] FALSE

x <- "hi"
is.numeric(x) && x + 1 # evaluates left to right

## [1] FALSE

is.numeric(x) & x + 1 # produces error

## Error: non-numeric argument to binary operator

m <- matrix(1:6, ncol=3)
m^2

##      [,1] [,2] [,3]
## [1,]    1    9   25
## [2,]    4   16   36

apply(m, 2, `^`, 2) # column-wise, ok

##      [,1] [,2] [,3]
## [1,]    1    9   25
## [2,]    4   16   36

apply(m, 1, `^`, 2) # gives back row-wise operation as columns

##      [,1] [,2]
## [1,]    1    4
## [2,]    9   16
## [3,]   25   36

df <- data.frame("test-it-#1"=1:2)
df

##   test.it..1
## 1          1
## 2          2

make.names("test-it-#1") # this function is used

## [1] "test.it..1"

df <- data.frame(a=1:2,b=3:4,c=5:6,d=7:8)
df[,-(2:3)] # numeric index ok

##   a d
## 1 1 7
## 2 2 8

df[,-c("b","c")] # not character index

## Error: invalid argument to unary operator

subset(df, select=-c(b,c)) # by name works here

##   a d
## 1 1 7
## 2 2 8

x <- numeric(0)
1:length(x)

## [1] 1 0

seq_len(length(x))

## integer(0)

seq_along(x)

## integer(0)

x <- c(1,2,NA)
which(x == NA)

## integer(0)

which(is.na(x))

## [1] 3

y <- NULL
y == NULL

## logical(0)

is.null(y)

## [1] TRUE

m <- matrix(1:4,ncol=2)
write.csv(m,file="matrix.csv")
read.csv("matrix.csv")

##   X V1 V2
## 1 1  1  3
## 2 2  2  4

read.csv("matrix.csv",row.names=1)

##   V1 V2
## 1  1  3
## 2  2  4

f <- function() {
  y <- 1:10
  form <- ~ 1
  form
}
get("y", environment(f()))

##  [1]  1  2  3  4  5  6  7  8  9 10

mikelove / r-gotchas Goto Github PK