R lapply to remove column from list of dataframes

Question

I am trying to remove a single column from a list of dataframes using lapply with the below code (want to remove column name "ID" from both df1 and df2). I receive the following error message when running through command prompt:

Error in eval(substitute(select), nl, parent.frame()) : object 'ID' not found Calls: lapply -> FUN -> subset.data.frame -> eval -> eval

Any thoughts as to why the code produces an error?

df1 <- data.frame(
 "Qty" = c(15,29,12,53),
 "Type"   = c("A","B","B","E"),
 "ID"   = c("x123","y121","x556","y119"))

df2 <- data.frame(
 "Qty" = c(5,92,25,31),
 "Type"   = c("I","L","Z","K"),
 "ID"   = c("p433","q232","y344","l598"))

df_list <- mget(ls(pattern= "^df"))

df_list <- lapply(df_list, function(x) subset.data.frame(x, select=-c(ID)))

list2env(df_list, .GlobalEnv)

[Here are ways to subset](https://stackoverflow.com/questions/5234117/how-to-drop-columns-by-name-in-a-data-frame). Just apply them in `lapply` — Sotos, Sep 21 '20 at 14:03

score 5 · Answer 1 · answered Sep 21 '20 at 13:47

5

Try this dplyr approach:

library(dplyr)         
#Data
df1 <- data.frame(
  "Qty" = c(15,29,12,53),
  "Type"   = c("A","B","B","E"),
  "ID"   = c("x123","y121","x556","y119"))

df2 <- data.frame(
  "Qty" = c(5,92,25,31),
  "Type"   = c("I","L","Z","K"),
  "ID"   = c("p433","q232","y344","l598"))

df_list <- mget(ls(pattern= "^df"))

df_list <- lapply(df_list, function(x) {x <- x %>% select(-ID)})

list2env(df_list, .GlobalEnv)

The output (which will be released to environment):

df_list
$df1
  Qty Type
1  15    A
2  29    B
3  12    B
4  53    E

$df2
  Qty Type
1   5    I
2  92    L
3  25    Z
4  31    K

answered Sep 21 '20 at 13:47

Duck

39,058
13
42
84

I receive the following error. Error: Can't subset columns that don't exist. x Column `ID` doesn't exist. Maybe it's an issue with my configuration? – Matt Gossett Sep 21 '20 at 13:51
@MattGossett Check that all dataframes in your list have the column `ID`, use `lapply(yourlist,names)` and check that! – Duck Sep 21 '20 at 13:52
thanks! The code works when run through R Studio but not when run through command prompt. Not sure of the issue. This worked for me though lapply(df_list, function(i){i$ID <- NULL; i}) – Matt Gossett Sep 21 '20 at 14:01
@MattGossett It is a valid option too! Fantastic!!! – Duck Sep 21 '20 at 14:05

score 1 · Answer 2 · answered Sep 21 '20 at 14:13

1

Base R solutions:

# Safer, directly specifying the name: 
Map(function(x){x[,names(x) != "ID"]}, df_list)

# If the data.frames have the same column order: 
lapply(df_list, "[", 1:2)

answered Sep 21 '20 at 14:13

hello_friend

5,682
1
11
15

R lapply to remove column from list of dataframes

2 Answers2