collateral (version 0.5.2)

collateral_mappers: Map over a list while capturing side effects.

Description

map_safely(), map_quietly() and map_peacefully() are variants of purrr::map() that wrap the supplied function .f using purrr::safely() and/or purrr::quietly() in order to capture various side effects. Lists mapped in this way have an associated class added to them, allowing them to succinctly summarise captured side effects when displayed in a tibble.

Usage

map_safely(.x, .f, otherwise = NULL, quiet = TRUE, ...)
map_quietly(.x, .f, ...)
map_peacefully(.x, .f, ...)
map2_safely(.x, .y, .f, otherwise = NULL, quiet = TRUE, ...)
map2_quietly(.x, .y, .f, ...)
map2_peacefully(.x, .y, .f, ...)
pmap_safely(.l, .f, otherwise = NULL, quiet = TRUE, ...)
pmap_quietly(.l, .f, ...)
pmap_peacefully(.l, .f, ...)
future_map_safely(.x, .f, otherwise = NULL, quiet = TRUE, ...)
future_map_quietly(.x, .f, ...)
future_map_peacefully(.x, .f, ...)
future_map2_safely(.x, .y, .f, otherwise = NULL, quiet = TRUE, ...)
future_map2_quietly(.x, .y, .f, ...)
future_map2_peacefully(.x, .y, .f, ...)
future_pmap_safely(.l, .f, otherwise = NULL, quiet = TRUE, ...)
future_pmap_quietly(.l, .f, ...)
future_pmap_peacefully(.l, .f, ...)

Arguments

.x

A list or atomic vector.

.f

A function, formula or atomic vector, as specified by purrr::as_mapper().

otherwise

Default value to use when an error occurs.

quiet

Hide errors (TRUE, the default), or display them as they occur?

...

Other arguments supplied to purrr::map() or its variants, or to furrr::future_map() or its variants..

.y

A list or atomic vector, of the same length as .x.

.l

A list of lists. The length of .l determines the number of arguments that .f will be called with. List names will be used if present.

Value

A list of the same length as .x. Each element of the returned list is itself a named list, structured according to the captured side effects. The Details section elaborates on these side effects.

Details

map_safely() will summarise the returned list with a fixed-width string of two (spaced) columns:

1. If a result component is present, R appears, and

2. If an error component is present, E appears.

If either component is missing, an underscore (_) appears in its place.

Similarly, map_quietly() will summarise the returned list with a fixed-width string of four (spaced) columns:

1. If a result component is present, R appears,

2. If an output component is present, O appears,

3. If a messages component is present, M appears, and

4. If a warnings component is present, W appears.

If any is missing, an underscore (_) appears in its place.

Variants for iterating over two or more inputs simultaneously are also provided and function identically to their purrr counterparts:

1. map2_safely()

2. map2_quietly()

3. pmap_safely()

4. pmap_quietly()

Further variants, prefixed by future_, allow safe or quiet mapping to happen in parallel if you have the furrr package installed:

1. future_map_safely()

2. future_map_quietly()

3. future_map2_safely()

4. future_map2_quietly()

5. future_pmap_safely()

6. future_pmap_quietly()

Examples

# NOT RUN {
library(tibble)
library(dplyr)
library(tidyr)
library(collateral)

# like map(), these can be used to iterate over vectors or lists
list("a", 10, 100) %>% map_safely(log)
list(5, -12, 103) %>% map_quietly(log)

# if you're using tibbles, you can also iterate over list-columns,
# such as nested data frames
mtcars %>%
  rownames_to_column(var = "car") %>%
  as_tibble() %>%
  select(car, cyl, disp, wt) %>%
  # spike some rows in cyl == 4 to make them fail
  mutate(wt = dplyr::case_when(
    wt < 2 ~ -wt,
    TRUE ~ wt)) %>%
  # nest and do some operations quietly()
  nest(data = -cyl) %>%
  mutate(qlog = map_quietly(data, ~ log(.$wt)))

# }