Skip to contents

Duplicate rows by a count column

Source: R/duplicate.R
duplicate.Rd

Expands a data frame by repeating each row X number of times, specified by a count column. Useful for reconstructing individual-level data from aggregated or frequency-weighted data frames.

Usage

duplicate(data, n_col, drop_na = FALSE)

Arguments

data

A data frame.

n_col

Column name of the column containing duplication counts, supplied either unquoted (n) or quoted ("n"). Rows with NA counts are counted once unless drop_na = TRUE.

drop_na

Logical. If TRUE, rows where n_col is NA are dropped before expansion. Default is FALSE.

Value

A data frame with more rows than data, where each row i appears n_col[i] times (or once if n_col[i] is NA and drop_na = FALSE). Row names are not reset. The n_col column is retained in the output.

Details

Expansion is performed via rep(seq_len(nrow(data)), times = n_col), so the original row order is preserved within each group of duplicates. NA counts are replaced with 1 prior to expansion when drop_na = FALSE.

A console message reports the final row count of the expanded data frame.

See also

deduplicate for the inverse operation,

rep for the underlying row repetition mechanism.

Examples

df <- data.frame(
  group = c("A", "B", "C"),
  value = c(10, 20, 30),
  n     = c(3, 1, 2)
)

# Expand so each row repeats n times
duplicate(df, n_col = n)
#> [duplicate] dataset expanded to 6 rows based on 'n'
#>     group value n
#> 1       A    10 3
#> 1.1     A    10 3
#> 1.2     A    10 3
#> 2       B    20 1
#> 3       C    30 2
#> 3.1     C    30 2

# NA counts default to 1 repetition
df_na <- data.frame(
  group = c("A", "B", "C"),
  n     = c(2, NA, 3)
)
duplicate(df_na, n_col = n)
#> [duplicate] dataset expanded to 6 rows based on 'n'
#>     group n
#> 1       A 2
#> 1.1     A 2
#> 2       B 1
#> 3       C 3
#> 3.1     C 3
#> 3.2     C 3

# Drop rows with NA counts instead
duplicate(df_na, n_col = n, drop_na = TRUE)
#> [duplicate] 1 NA row(s) removed from numbering column
#> [duplicate] dataset expanded to 5 rows based on 'n'
#>     group n
#> 1       A 2
#> 1.1     A 2
#> 3       C 3
#> 3.1     C 3
#> 3.2     C 3