tidyft (version 0.9.20)

nest: Nest and unnest

Description

Analogous function for nest and unnest in tidyr. unnest will automatically remove other list-columns except for the target list-columns (which would be unnested later). Also, squeeze is designed to merge multiple columns into list column.

Usage

nest(.data, ..., mcols = NULL, .name = "ndt")
unnest(.data, ...)
squeeze(.data, ..., .name = "ndt")
chop(.data, ...)
unchop(.data, ...)

Value

data.table, nested or unnested

Arguments

.data

data.table, nested or unnested

...

The variables for nest group(for nest), columns to be nested(for squeeze and chop), or column(s) to be unnested(for unnest). Could recieve anything that select_dt could receive.

mcols

Name-variable pairs in the list, form like

.name

Character. The nested column name. Defaults to "ndt". list(petal="^Pe",sepal="^Se"), see example.

Details

In the nest, the data would be nested to a column named `ndt`, which is short for nested data.table.

The squeeze would not remove the originial columns.

The unchop is the reverse operation of chop.

These functions are experiencing the experimental stage, especially the unnest. If they don't work on some circumtances, try tidyr package.

References

https://www.r-bloggers.com/much-faster-unnesting-with-data-table/

https://stackoverflow.com/questions/25430986/create-nested-data-tables-by-collapsing-rows-into-new-data-tables

See Also

nest, chop

Examples

mtcars = as.data.table(mtcars)
iris = as.data.table(iris)

# examples for nest

# nest by which columns?
 mtcars %>% nest(cyl)
 mtcars %>% nest("cyl")
 mtcars %>% nest(cyl,vs)
 mtcars %>% nest(vs:am)
 mtcars %>% nest("cyl|vs")
 mtcars %>% nest(c("cyl","vs"))

# nest two columns directly
iris %>% nest(mcols = list(petal="^Pe",sepal="^Se"))

# nest more flexibly
iris %>% nest(mcols = list(ndt1 = 1:3,
  ndt2 = "Pe",
  ndt3 = Sepal.Length:Sepal.Width))

# examples for unnest
# unnest which column?
 mtcars %>% nest("cyl|vs") %>%
   unnest(ndt)
 mtcars %>% nest("cyl|vs") %>%
   unnest("ndt")

df <- data.table(
  a = list(c("a", "b"), "c"),
  b = list(c(TRUE,TRUE),FALSE),
  c = list(3,c(1,2)),
  d = c(11, 22)
)

df
df %>% unnest(a)
df %>% unnest(2)
df %>% unnest("c")
df %>% unnest(cols = names(df)[3])

# You can unnest multiple columns simultaneously
df %>% unnest(1:3)
df %>% unnest(a,b,c)
df %>% unnest("a|b|c")

# examples for squeeze
# nest which columns?
iris %>% squeeze(1:2)
iris %>% squeeze("Se")
iris %>% squeeze(Sepal.Length:Petal.Width)

# examples for chop
df <- data.table(x = c(1, 1, 1, 2, 2, 3), y = 1:6, z = 6:1)
df %>% chop(y,z)
df %>% chop(y,z) %>% unchop(y,z)