I have a list of named vectors containing dates in R and I'd like to convert it into a data frame where columns will be named after vector names. Please note that due to differing vector length there will be NAs in the final fata frame.
d5 <- list(structure(c(event13 = 1797, event17 = 7006, event3 = -6796
), class = "Date"), structure(c(event3 = 5984, event14 = 7175,
event11 = -4031, event20 = 8612, event2 = 5158, event5 = 1002,
event8 = -382, event15 = 4367, event11 = 2960), class = "Date"),
structure(c(event7 = 4394, event13 = -1389, event9 = -4407
), class = "Date"), structure(c(event5 = 7729), class = "Date"),
structure(c(event4 = -3384, event10 = 1288, event10 = 7502,
event5 = -5100, event9 = -3177, event8 = -4027, event11 = -3554,
event16 = 9484, event3 = 1386), class = "Date"), structure(c(event10 = -6906,
event3 = 6966, event8 = -975, event14 = -3286, event12 = 744,
event11 = 7111, event15 = 9576, event12 = 2223, event9 = 10771
), class = "Date"), structure(c(event16 = 8764), class = "Date"),
structure(c(event15 = 9795, event14 = 7681, event2 = -1728,
event14 = 10876), class = "Date"), structure(c(event13 = -1341,
event11 = 1202), class = "Date"), structure(c(event3 = -666,
event1 = -4192, event9 = 8808, event4 = -1765), class = "Date"))
> d5
[[1]]
event13 event17 event3
"1974-12-03" "1989-03-08" "1951-05-25"
[[2]]
event3 event14 event11 event20 event2 event5 event8 event15
"1986-05-21" "1989-08-24" "1958-12-19" "1993-07-31" "1984-02-15" "1972-09-29" "1968-12-15" "1981-12-16"
event11
"1978-02-08"
[[3]]
event7 event13 event9
"1982-01-12" "1966-03-14" "1957-12-08"
[[4]]
event5
"1991-03-01"
...
The output would look like this:
| ID | event1 | event2 | event3 | ... |
|---|---|---|---|---|
| 1 | date1 | NA | date3 | ... |
Using do.call with rbind returns a following error"
as.data.frame(do.call(rbind, d5))
event3 event14 event11 event20 event2 event5 event8 event15 event11
1 1797 7006 -6796 1797 7006 -6796 1797 7006 -6796
2 5984 7175 -4031 8612 5158 1002 -382 4367 2960
3 4394 -1389 -4407 4394 -1389 -4407 4394 -1389 -4407
4 7729 7729 7729 7729 7729 7729 7729 7729 7729
5 -3384 1288 7502 -5100 -3177 -4027 -3554 9484 1386
6 -6906 6966 -975 -3286 744 7111 9576 2223 10771
7 8764 8764 8764 8764 8764 8764 8764 8764 8764
8 9795 7681 -1728 10876 9795 7681 -1728 10876 9795
9 -1341 1202 -1341 1202 -1341 1202 -1341 1202 -1341
10 -666 -4192 8808 -1765 -666 -4192 8808 -1765 -666
Warning message:
In (function (..., deparse.level = 1) :
number of columns of result is not a multiple of vector length (arg 8)
How to convert it efficiently (without equalising length of each vector by adding NAs and resorting them to match the columns)?
CodePudding user response：
library(tidyr)
library(dplyr)
# Make each element of the list a dataframe with two columns: x and event_name
dataframed <- lapply(d5, function(x) {
out <- as.data.frame(x)
out$event_name <- names(x)
out
})
# Make a unique dataframe from this list of dataframes
unlisted <- do.call(rbind, dataframed)
row.names(unlisted) <- NULL
# pivot this dataframe to have event_name as column names
unlisted |>
group_by(event_name) |>
mutate(id = seq_len(n())) |>
ungroup() |>
pivot_wider(names_from = event_name, values_from = x) |>
select(-id)
#> # A tibble: 5 × 17
#> event13 event17 event3 event14 event11 event20 event2
#> <date> <date> <date> <date> <date> <date> <date>
#> 1 1974-12-03 1989-03-08 1951-05-25 1989-08-24 1958-12-19 1993-07-31 1984-02-15
#> 2 1966-03-14 NA 1986-05-21 1961-01-02 1960-04-09 NA 1965-04-09
#> 3 1966-05-01 NA 1973-10-18 1991-01-12 1989-06-21 NA NA
#> 4 NA NA 1989-01-27 1999-10-12 1973-04-17 NA NA
#> 5 NA NA 1968-03-06 NA NA NA NA
#> # … with 10 more variables: event5 <date>, event8 <date>, event15 <date>,
#> # event12 <date>, event7 <date>, event9 <date>, event4 <date>,
#> # event10 <date>, event16 <date>, event1 <date>
Created on 2022-10-12 with reprex v2.0.2
CodePudding user response：
Does this satisfy your needs? Note that due to your second element, there are two columns event11.
res <- data.table::rbindlist(lapply(d5, function(x) data.table::as.data.table(as.list(x))),
fill = TRUE,
use.names = TRUE)
res[1:5]
#> event13 event17 event3 event14 event11 event20 event2
#> 1: 1974-12-03 1989-03-08 1951-05-25 <NA> <NA> <NA> <NA>
#> 2: <NA> <NA> 1986-05-21 1989-08-24 1958-12-19 1993-07-31 1984-02-15
#> 3: 1966-03-14 <NA> <NA> <NA> <NA> <NA> <NA>
#> 4: <NA> <NA> <NA> <NA> <NA> <NA> <NA>
#> 5: <NA> <NA> 1973-10-18 <NA> 1960-04-09 <NA> <NA>
#> event5 event8 event15 event11 event7 event9 event4
#> 1: <NA> <NA> <NA> <NA> <NA> <NA> <NA>
#> 2: 1972-09-29 1968-12-15 1981-12-16 1978-02-08 <NA> <NA> <NA>
#> 3: <NA> <NA> <NA> <NA> 1982-01-12 1957-12-08 <NA>
#> 4: 1991-03-01 <NA> <NA> <NA> <NA> <NA> <NA>
#> 5: 1956-01-15 1958-12-23 <NA> <NA> <NA> 1961-04-21 1960-09-26
#> event10 event10 event16 event12 event12 event14 event1
#> 1: <NA> <NA> <NA> <NA> <NA> <NA> <NA>
#> 2: <NA> <NA> <NA> <NA> <NA> <NA> <NA>
#> 3: <NA> <NA> <NA> <NA> <NA> <NA> <NA>
#> 4: <NA> <NA> <NA> <NA> <NA> <NA> <NA>
#> 5: 1973-07-12 1990-07-17 1995-12-20 <NA> <NA> <NA> <NA>
CodePudding user response：
Assuming no duplicated events in each element of the list:
>dplyr::bind_rows(d5)
# A tibble: 10 × 17
event13 event17 event3 event14 event11 event20 event2 event5 event8
<date> <date> <date> <date> <date> <date> <date> <date> <date>
1 1974-12-03 1989-03-08 1951-05-25 NA NA NA NA NA NA
2 NA NA 1986-05-21 1989-08-24 1958-12-19 1993-07-31 1984-02-15 1972-09-29 1968-12-15
3 1966-03-14 NA NA NA NA NA NA NA NA
4 NA NA NA NA NA NA NA 1991-03-01 NA
5 NA NA 1973-10-18 NA 1960-04-09 NA NA 1956-01-15 1958-12-23
6 NA NA 1989-01-27 1961-01-02 1989-06-21 NA NA NA 1967-05-02
7 NA NA NA NA NA NA NA NA NA
8 NA NA NA 1991-01-12 NA NA 1965-04-09 NA NA
9 1966-05-01 NA NA NA 1973-04-17 NA NA NA NA
10 NA NA 1968-03-06 NA NA NA NA NA NA
# … with 8 more variables: event15 <date>, event7 <date>, event9 <date>, event4 <date>, event10 <date>,
# event16 <date>, event12 <date>, event1 <date>
Here, d5 is manually edited to remove duplicated entries:
d5 <- list(structure(c(event13 = 1797, event17 = 7006, event3 = -6796
), class = "Date"), structure(c(event3 = 5984, event14 = 7175,
event11 = -4031, event20 = 8612, event2 = 5158, event5 = 1002,
event8 = -382, event15 = 4367), class = "Date"), structure(c(event7 = 4394,
event13 = -1389, event9 = -4407), class = "Date"), structure(c(event5 = 7729), class = "Date"),
structure(c(event4 = -3384, event10 = 1288, event5 = -5100,
event9 = -3177, event8 = -4027, event11 = -3554, event16 = 9484,
event3 = 1386), class = "Date"), structure(c(event10 = -6906,
event3 = 6966, event8 = -975, event14 = -3286, event11 = 7111,
event15 = 9576, event12 = 2223, event9 = 10771), class = "Date"),
structure(c(event16 = 8764), class = "Date"), structure(c(event15 = 9795,
event14 = 7681, event2 = -1728), class = "Date"), structure(c(event13 = -1341,
event11 = 1202), class = "Date"), structure(c(event3 = -666,
event1 = -4192, event9 = 8808, event4 = -1765), class = "Date"))
CodePudding user response：
With base R this is an option
Get all event names
nn <- unique(unlist(sapply(d5, function(x) names(x))))
Iterate through all event names and name the columns
res <- data.frame(t(sapply(d5, function(x) strftime(x[nn]))))
colnames(res) <- nn
res
event13 event17 event3 event14 event11 event20 event2
1 1974-12-03 1989-03-08 1951-05-25 <NA> <NA> <NA> <NA>
2 <NA> <NA> 1986-05-21 1989-08-24 1958-12-19 1993-07-31 1984-02-15
3 1966-03-14 <NA> <NA> <NA> <NA> <NA> <NA>
4 <NA> <NA> <NA> <NA> <NA> <NA> <NA>
5 <NA> <NA> 1973-10-18 <NA> 1960-04-09 <NA> <NA>
6 <NA> <NA> 1989-01-27 1961-01-02 1989-06-21 <NA> <NA>
7 <NA> <NA> <NA> <NA> <NA> <NA> <NA>
8 <NA> <NA> <NA> 1991-01-12 <NA> <NA> 1965-04-09
9 1966-05-01 <NA> <NA> <NA> 1973-04-17 <NA> <NA>
10 <NA> <NA> 1968-03-06 <NA> <NA> <NA> <NA>
event5 event8 event15 event7 event9 event4 event10
1 <NA> <NA> <NA> <NA> <NA> <NA> <NA>
2 1972-09-29 1968-12-15 1981-12-16 <NA> <NA> <NA> <NA>
3 <NA> <NA> <NA> 1982-01-12 1957-12-08 <NA> <NA>
4 1991-03-01 <NA> <NA> <NA> <NA> <NA> <NA>
5 1956-01-15 1958-12-23 <NA> <NA> 1961-04-21 1960-09-26 1973-07-12
6 <NA> 1967-05-02 1996-03-21 <NA> 1999-06-29 <NA> 1951-02-04
7 <NA> <NA> <NA> <NA> <NA> <NA> <NA>
8 <NA> <NA> 1996-10-26 <NA> <NA> <NA> <NA>
9 <NA> <NA> <NA> <NA> <NA> <NA> <NA>
10 <NA> <NA> <NA> <NA> 1994-02-12 1965-03-03 <NA>
event16 event12 event1
1 <NA> <NA> <NA>
2 <NA> <NA> <NA>
3 <NA> <NA> <NA>
4 <NA> <NA> <NA>
5 1995-12-20 <NA> <NA>
6 <NA> 1972-01-15 <NA>
7 1993-12-30 <NA> <NA>
8 <NA> <NA> <NA>
9 <NA> <NA> <NA>
10 <NA> <NA> 1958-07-11