dplyr error: object not found using rle() in mutate()

Question

I'm trying to use dplyr to insert run lengths of column value into my data for each group.

tdf <- tbl_df(structure(list(group = structure(c(1L, 1L, 1L, 1L, 1L, 1L, 1L, 
1L, 1L, 1L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L), .Label = c("A", 
"B"), class = "factor"), value = c(TRUE, TRUE, TRUE, TRUE, TRUE, 
TRUE, FALSE, FALSE, TRUE, FALSE, FALSE, FALSE, FALSE, FALSE, 
FALSE, TRUE, TRUE, FALSE, FALSE, FALSE)), class = c("tbl_df", 
"tbl", "data.frame"), .Names = c("group", "value"), row.names = c(NA, 
-20L)))

My data looks like this...

 > tdf
Source: local data frame [20 x 2]

       group value
    1      A  TRUE
    2      A  TRUE
    3      A  TRUE
    4      A  TRUE
    5      A  TRUE
    6      A  TRUE
    7      A FALSE
    8      A FALSE
    9      A  TRUE
    10     A FALSE
    11     B FALSE
    12     B FALSE
    13     B FALSE
    14     B FALSE
    15     B FALSE
    16     B  TRUE
    17     B  TRUE
    18     B FALSE
    19     B FALSE
    20     B FALSE

And my desired output looks like this... (notice that the run lengths restart between groups)

   group value run_length
1      A  TRUE          6
2      A  TRUE          6
3      A  TRUE          6
4      A  TRUE          6
5      A  TRUE          6
6      A  TRUE          6
7      A FALSE          2
8      A FALSE          2
9      A  TRUE          1
10     A FALSE          1
11     B FALSE          5
12     B FALSE          5
13     B FALSE          5
14     B FALSE          5
15     B FALSE          5
16     B  TRUE          2
17     B  TRUE          2
18     B FALSE          3
19     B FALSE          3
20     B FALSE          3

I figured it would just be a matter of doing this in dplyr...

group_by(tdf, group) %.% 
  mutate(run_len = rep(rle(value)$lengths,rle(value)$lengths))

But I get the error:

Error in rle(value) : object 'value' not found

I have a solution outside of dplyr using split and lapply, but would like to know how this would work in dplyr.

Your code works for me with dplyr. I suspect you might have plyr loaded as well which is messing up mutate. — Ramnath

npjc npjc · Accepted Answer · 2014-05-23T03:45:35

Solution:

tbl_dt() seems to be what you need.

from your tdf this works as expected:

tbl_dt(tdf) %>% 
group_by(group) %>%
mutate( run_len = rep( rle(value)$lengths,rle(value)$lengths ) )

bug?

the equivalent tbl_df() does not work...I do not know why...

tbl_df(tdf) %>% 
group_by(group) %>%
mutate( run_len = rep( rle(value)$lengths,rle(value)$lengths ) )

gives familiar error: Error in rle(value) : object 'value' not found

dplyr error: object not found using rle() in mutate()

1 Answers

Solution:

bug?