R create duplicate columns (bulk)-CodePudding

How to create multiple duplicate columns based on a colname pattern? My actual dataframe is really large, so I'll need a loop or some other efficient code.

df <- data.frame(x001 = c(1,2,3), x002 = c(3,4,5), x003 = c(6,7,8), soccer = c(700,600,300), volley = c(30,22,29))

df
#current df
#   x001 x002 x003 soccer volley
#   1    3    6    700     30
#   2    4    7    600     22
#   3    5    8    300     29

#desired output: all x00 columns need to be duplicated and get a "no2" addition to their name. 
#   x001 x002 x003 soccer volley  x001no2 x002no2 x003no2
#   1    3    6    700     30       1        3    6  
#   2    4    7    600     22       2        4    7 
#   3    5    8    300     29       3        5    8

CodePudding user response：

Simple

tmp=grep("x00",colnames(df))

cbind(
  df,
  setNames(df[,tmp],paste0(colnames(df)[tmp],"no2"))
)

  x001 x002 x003 soccer volley x001no2 x002no2 x003no2
1    1    3    6    700     30       1       3       6
2    2    4    7    600     22       2       4       7
3    3    5    8    300     29       3       5       8

CodePudding user response：

Using dplyr and stringr:

library(dplyr)
library(stringr)

df %>% bind_cols(df %>% select(starts_with('x00')) %>% rename_all( ~ str_c(., 'no2')) )
  x001 x002 x003 soccer volley x001no2 x002no2 x003no2
1    1    3    6    700     30       1       3       6
2    2    4    7    600     22       2       4       7
3    3    5    8    300     29       3       5       8

CodePudding user response：

Data.table solution:

# Import and initialise the data.table package: 
library(data.table)

# Resolve the column names desired to be replicated: 
# df_names => character vectors
df_names <- grep(
  "^x\\d ",
  colnames(df),
  value = TRUE
)

# Replicate the desired vectors: res => data.table
res <- setDT(df)[, paste0(
  df_names,
  "no2"
  ) := df[
    , .SD,
    .SDcols = df_names
  ],
]

Tidyverse solution:

# Import and initialise the tidyverse package: 
library(tidyverse)

# Replicate the desired column vectors: res => data.frame
res <- df %>% 
  bind_cols(
    .,
    df %>% 
      select_if(
        str_detect(
          colnames(.),
          "^x\\d "
        )
      ) %>% 
      set_names(
        .,
        str_c(
          colnames(.),
          "no2"
      )
    )
  )