How to add a row to data frame if there are different number of columns?

Question

I have several dataframes with a single row per dataframe and I have an empty dataframe (let's name it 'total'). All dataframes have different count of columns, but some of them intersect. Total dataframe has all possible columns, so if I add any row from those dataframes it should match existed columns and fill values in accordance (if a column doesn't exist in adding row it should be filled as 0).

Example of dataframes with data:

A B C      B E      C E K J
1 2 5      4 2      3 2 5 7

Example of total dataframe:

A B C E K J
1 2 5 0 0 0
0 4 0 2 0 0
0 0 3 2 5 7

So, how to do that? I've tried various binds and inserts but they don't work - in some cases added row changes total dataframe column amount, in some cases added row just duplicates previous row etc.

It would be better if you specify in which format you have this mini-dataframes (in a list maybe?). — Wencheng Lau-Medrano, Jul 20 '22 at 12:56

PaulS · Answer 1 · 2022-07-20T13:03:11.630

1

A possible solution:

library(tidyverse)

bind_rows(df1, df2, df3) %>% 
  mutate(across(everything(), ~ replace_na(.x, 0)))

#>   A B C E K J
#> 1 1 2 5 0 0 0
#> 2 0 4 0 2 0 0
#> 3 0 0 3 2 5 7

edited Jul 20 '22 at 13:03

answered Jul 20 '22 at 12:56

PaulS

21,159
2
9
26

somehow replaces become `false` instead of 0 – anatol Jul 20 '22 at 13:37
It is not `replace` but `replace_na`, @anatol. It works perfectly here. – PaulS Jul 20 '22 at 13:42
I see, I didn't say 'I used replace'. I used your code absolutely as is – anatol Jul 20 '22 at 13:47
also, I'm getting an error `Can't combine '..1$A' and '..2$A' `. does that mean my dataframes has different datatypes for same column? – anatol Jul 20 '22 at 13:53
Try this instead: `bind_rows(df1, df2, df3) %>% mutate(across(everything(), ~ ifelse(is.na(.x), 0, .x)))`. – PaulS Jul 20 '22 at 13:54
You may have your `tidyverse` not updated, @anatol. – PaulS Jul 20 '22 at 13:56
1

`ifelse` works fine – anatol Jul 20 '22 at 13:57
Great, @anatol! – PaulS Jul 20 '22 at 13:58

How to add a row to data frame if there are different number of columns?

1 Answers1