Extract a dplyr tbl column as a vector

Question

Is there a more succinct way to get one column of a dplyr tbl as a vector, from a tbl with database back-end (i.e. the data frame/table can't be subset directly)?

require(dplyr)
db <- src_sqlite(tempfile(), create = TRUE)
iris2 <- copy_to(db, iris)
iris2$Species
# NULL

That would have been too easy, so

collect(select(iris2, Species))[, 1]
# [1] "setosa"     "setosa"     "setosa"     "setosa"  etc.

But it seems a bit clumsy.

Answer 1

With dplyr >= 0.7.0, you can use pull to get a vector from a tbl.

library("dplyr")
#> 
#> Attaching package: 'dplyr'
#> The following objects are masked from 'package:stats':
#> 
#>     filter, lag
#> The following objects are masked from 'package:base':
#> 
#>     intersect, setdiff, setequal, union
db <- src_sqlite(tempfile(), create = TRUE)
iris2 <- copy_to(db, iris)
vec <- pull(iris2, Species)
head(vec)
#> [1] "setosa" "setosa" "setosa" "setosa" "setosa" "setosa"

Answer 2

As per the comment from @nacnudus, it looks like a pull function was implemented in dplyr 0.6:

iris2 %>% pull(Species)

---

For older versions of dplyr, here's a neat function to make pulling out a column a bit nicer (easier to type, and easier to read):

pull <- function(x,y) {x[,if(is.name(substitute(y))) deparse(substitute(y)) else y, drop = FALSE][[1]]}

This lets you do either of these:

iris2 %>% pull('Species')
iris2 %>% pull(Species)
iris2 %>% pull(5)

Resulting in...

 [1] 21.0 21.0 22.8 21.4 18.7 18.1 14.3 24.4 22.8 19.2 17.8 16.4 17.3 15.2 10.4 10.4 14.7 32.4 30.4 33.9 21.5 15.5 15.2 13.3 19.2 27.3 26.0 30.4 15.8 19.7 15.0 21.4

And it also works fine with data frames:

> mtcars %>% pull(5)
 [1] 3.90 3.90 3.85 3.08 3.15 2.76 3.21 3.69 3.92 3.92 3.92 3.07 3.07 3.07 2.93 3.00 3.23 4.08 4.93 4.22 3.70 2.76 3.15 3.73 3.08 4.08 4.43
[28] 3.77 4.22 3.62 3.54 4.11

---

A nice way to do this in v0.2 of dplyr:

iris2 %>% select(Species) %>% collect %>% .[[5]]

Or if you prefer:

iris2 %>% select(Species) %>% collect %>% .[["Species"]]

Or if your table isn't too big, simply...

iris2 %>% collect %>% .[["Species"]]

Answer 3

You can also use unlist which I find easier to read because you do not need to repeat the name of the column or specify the index.

iris2 %>% select(Species) %>% unlist(use.names = FALSE)

Answer 4

I would use the extract2 convenience function from magrittr:

library(magrittr)
library(dplyr)

iris2 %>%
  select(Species) %>%
  extract2(1)  

Answer 5

I'd probably write:

collect(select(iris2, Species))[[1]]

Since dplyr is designed for working with tbls of data, there's no better way to get a single column of data.

Answer 6

@Luke1018 proposed this solution in one of the comments:

You can also use the magrittr exposition operator (%$%) to pull a vector from a data frame.

For example:

iris2 %>% select(Species) %>% collect() %$% Species

I thought it deserved its own answer.

Answer 7

If you are used to using square brackets for indexing, another option is to just to wrap the usual indexing approach in a call to deframe(), e.g.:

library(tidyverse)

iris2 <- as_tibble(iris)

# using column name
deframe(iris2[, 'Sepal.Length'])

# [1] 5.1 4.9 4.7 4.6 5.0 5.4

# using column number
deframe(iris2[, 1])

# [1] 5.1 4.9 4.7 4.6 5.0 5.4

That and pull() are both pretty good ways of getting a tibble column.