4

I am on Mac OS 10.10 with R 3.1.1

Suppose I have the following data frames a and b with the same attributes:

a<- structure(list(X1 = 1:5, X2 = 6:10), .Names = c("X1", "X2"), row.names = c(NA, 
-5L), class = "data.frame")
b<- structure(list(X1 = 11:15, X2 = 16:20), .Names = c("X1", "X2"
), row.names = c(NA, -5L), class = "data.frame")

and suppose I need to do some cleaning/apply some functions (like dropping column and multiply the first column by 2) on both a and b at the same time such that the original data frames reflect the changes with the following desired output:

> a
  X1 
1  2 
2  4 
3  6 
4  8
5 10

> b
  X1
1 22
2 24
3 26
4 28
5 30

I have just learnt the for loops and *apply functions but get confused when applying them on the data frames I have (which are not a and b but much bigger).

mallet
  • 2,454
  • 3
  • 37
  • 64

1 Answers1

5

It is generally encouraged (by the R Core team memebers) that if you have several data sets that you want to simultaneously operate on, you should keep them all in a list. In order to achieve your goal, you can simply use mget and ls to retreive them from the global environment and then simply replace with first column multiplied by 2, e.g.,

lapply(mget(ls(pattern = "[a-z]")), function(x) x <- x[1] * 2)
# $a
# X1
# 1  2
# 2  4
# 3  6
# 4  8
# 5 10
# 
# $b
# X1
# 1 22
# 2 24
# 3 26
# 4 28
# 5 30
David Arenburg
  • 91,361
  • 17
  • 137
  • 196
  • 1
    Another option is `Map('*', lapply(mget(ls(pattern = "[a-z]")),`[`,1), 2)` – akrun Jan 19 '15 at 11:57
  • Hi! thanks @David Arenburg for your reply, it's been also useful for me. The issue is, after following your advice I get in my console output the result I want, but this is not saved in the environment:: lapply(mget(ls(pattern = "items_symptoms")), function(x) names(x) <- ("exces","uncon","leng6","leng1","restl","fatig", "irrit","conce"")) If I type `names(items_symptoms1)`I won't see any changes, they only show in my console. How to actually edit the objects contained into `mget(ls))`? – Gina Zetkin Mar 31 '16 at 06:18
  • 1
    @GinaZetkin This will create a list of data frames that you will need to save somewhere, e.g. `res <- lapply(mget...` and then you will work with res instead with the data sets in the global environment. The idea here is that you won't pollute your environment with many data sets, rather keep them all in one list. – David Arenburg Mar 31 '16 at 07:53