Unexpected behavior of flatten_dfr

[expersso]

Consider the following:

> l <- list(c(1, 2), c(3, 4))
> l
[[1]]
[1] 1 2

[[2]]
[1] 3 4

> do.call(rbind, l)
     [,1] [,2]
[1,]    1    2
[2,]    3    4
> flatten_dfr(l)
Error in bind_rows_(x, .id) : 'getCharCE' must be called on a CHARSXP

I would expect flatten_dfr(l) == do.call(rbind, l).

> sessionInfo()
R version 3.4.1 (2017-06-30)
Platform: i386-w64-mingw32/i386 (32-bit)
Running under: Windows 7 x64 (build 7601) Service Pack 1

Matrix products: default

locale:
[1] LC_COLLATE=English_United Kingdom.1252  LC_CTYPE=English_United Kingdom.1252    LC_MONETARY=English_United Kingdom.1252
[4] LC_NUMERIC=C                            LC_TIME=English_United Kingdom.1252    

attached base packages:
[1] stats     graphics  grDevices utils     datasets  methods   base     

other attached packages:
[1] dplyr_0.7.2        purrr_0.2.3.9000   readr_1.1.1        tidyr_0.6.3        tibble_1.3.3       ggplot2_2.2.1.9000 tidyverse_1.1.1  

[elong0527]

And the function will report error if the input is a list of matrix.

library(tidyverse)

tmp <- data.frame(xx = rnorm(10), yy = rnorm(10), grp = rbinom(10, size = 1, prob = 0.5))
foo <- ~  coef(summary(lm(xx~yy, data = .)))

x <- tmp %>% 
     split(.$grp) %>%
     map(foo) 

x %>% flatten_dfr()

Error in bind_rows_(x, .id) : Argument 1 must have names

[yutannihilation]

data.frame and matrix are different things in terms of data structure, so I feel it is natural that flatten_dfr() throws error because it accepts data.frame, not matrix.

is(do.call(rbind, list(c(1, 2), c(3, 4))))
#> [1] "matrix"    "array"     "structure" "vector"

@elong0527
In your case, you can convert it to data.frame before using flatten_dfr() theoretically. But because of #374, you cannot do this for now.

x %>% map(as.data.frame) %>% flatten_dfr(.id = "id")
#> # A tibble: 2 x 5
#>      id    Estimate `Std. Error`  `t value` `Pr(>|t|)`
#>   <chr>       <dbl>        <dbl>      <dbl>      <dbl>
#> 1     1  0.09920347    0.5399731  0.1837193  0.8631710
#> 2     1 -0.14516082    0.6164339 -0.2354848  0.8253975

map_dfr() (or some function in broom package?) may help.

x %>% map_dfr(as.data.frame, .id = "id")
#>   id    Estimate Std. Error    t value  Pr(>|t|)
#> 1  0 -1.14290763  0.4655154 -2.4551444 0.1334761
#> 2  0 -0.47259892  0.3970154 -1.1903793 0.3560344
#> 3  1  0.09920347  0.5399731  0.1837193 0.8631710
#> 4  1 -0.14516082  0.6164339 -0.2354848 0.8253975