Converting a number to a string is easy. In many cases, you don’t even need to call the converting functions explicitly. However, when it comes to sophisticatedly control the output, you may find as.character()
is not enough and sprintf()
is the cure. I’ll give some real world use cases to demostrate.
Case 1: Always display 2 digits for a number
For example, if the investment return in a report displays 1%
, it’s natural to ask whether the real number is 1.00%
or 1.xx%
. Here sprintf("%0.2f", x)
can help.
as.character(c(0.1, 1.0))
# [1] "0.1" "1"
sprintf("%0.2f", c(0.1, 1.0))
# [1] "0.10" "1.00"
sprintf("%0.2f%%", c(0.001, 0.01) * 100) # even nicer use double `%` to escape
# [1] "0.10%" "1.00%"
Case 2: Display an integer as 000000
For example, you need to convert some integers to the stock codes of China A shares. Use sprintf("%06d", x)
.
as.character(0:3)
#> [1] "0" "1" "2" "3"
sprintf("%06d", 0:3)
#> [1] "000000" "000001" "000002" "000003"
Case 3: You don’t want the scientific formating for a large number.
Use sprintf("%d")
or sprintf("%.0f")
.
x <- c(1e+05, 1e+06, 1e+07)
as.character(x)
#> [1] "1e+05" "1e+06" "1e+07"
sprintf("%d", x)
#> [1] "100000" "1000000" "10000000"
sprintf("%.0f", x)
#> [1] "100000" "1000000" "10000000"
sprintf("%s", x) # see the explaination below
#> [1] "1e+05" "1e+06" "1e+07"
The letter s
is used to denote the string type. The reason of sprintf("%s", 1e5)
works is that R will convert the number 1e5
to the string "1e5"
silently before pass it to the sprintf()
C rountine. So sprintf("%s", x)
is kind of identical to as.character(x)
.
sprintf()
is powerful although it’s a bit of confusing to learn. See help("sprintf")
for more info.
UPDATE If you need to print numbers like 1,900.00
, you may find prettyNum()
or format()
helpful.