R dataframe cell contains dataframe/tibble - how to access or visualize?

Question

The excellent package inspectdf produces tibble output where one column contains either another tibble (or list). Sample data, which is in the dput at the bottom of this question, looks like this:

> test
# A tibble: 4 x 10
  col_name     min    q1 median    mean    q3     max       sd pcnt_na hist             
  <chr>      <dbl> <dbl>  <dbl>   <dbl> <dbl>   <dbl>    <dbl>   <dbl> <list>           
1 contacts    0       1      1     4.14    2     480     14.9    1.10  <tibble [26 x 2]>
2 pop         0.02  735.  2225. 4651.   4722. 289825. 10590.     0.082 <tibble [17 x 2]>
3 limit       0     750   1000  5994.   3500  500000  33355.     0.055 <tibble [27 x 2]>
4 code_count  1       1      1     1.72    2      43      1.50   0     <tibble [24 x 2]>

where the content of the column in question contains:

> test$hist[1]
[[1]]
# A tibble: 26 x 2
   value          prop
   <chr>         <dbl>
 1 [-Inf, 0)  0       
 2 [0, 20)    0.961   
 3 [20, 40)   0.0200  
 4 [40, 60)   0.00696 
 5 [60, 80)   0.00352 
 6 [80, 100)  0.00200 
 7 [100, 120) 0.00144 
 8 [120, 140) 0.000837
 9 [140, 160) 0.000696
10 [160, 180) 0.000459
# ... with 16 more rows

The question is, how to make this last column accessible or able to visualize from within a display mechanism? For example, when I use DT it puts [object: Object] in the rightmost column, where instead I would like some sort of useful thing, like maybe a link to view or popup the embedded data:

library(DT)
test %>% DT::datatable()

Is there a way to use sparklines, or make a link, or any other way to display the contents of that last column? All approaches are welcome (and it would be ideal if there was a way to do this within DT). Thank you for your help.

test <- structure(list(col_name = c("contacts", "pop", "limit", "code_count"
), min = c(0, 0.02, 0, 1), q1 = c(1, 735.1, 750, 1), median = c(1, 
2224.73, 1000, 1), mean = c(4.142, 4651.211, 5994.002, 1.717), 
q3 = c(2, 4721.56, 3500, 2), max = c(480, 289825.16, 5e+05, 
43), sd = c(14.926, 10590.427, 33354.671, 1.501), pcnt_na = c(1.103, 
0.082, 0.055, 0), hist = list(structure(list(value = c("[-Inf, 0)", 
"[0, 20)", "[20, 40)", "[40, 60)", "[60, 80)", "[80, 100)", 
"[100, 120)", "[120, 140)", "[140, 160)", "[160, 180)", "[180, 200)", 
"[200, 220)", "[220, 240)", "[240, 260)", "[260, 280)", "[280, 300)", 
"[300, 320)", "[320, 340)", "[340, 360)", "[360, 380)", "[380, 400)", 
"[400, 420)", "[420, 440)", "[440, 460)", "[460, 480)", "[480, Inf)"
), prop = c(0, 0.961394177780924, 0.0200167851400953, 0.00696279968047565, 
0.00352285711396706, 0.00199601605710992, 0.00143785959129195, 
0.000837234698726958, 0.000695673276236893, 0.000459063470074926, 
0.00138527963436707, 0.00129023125069517, 0, 0, 0, 0, 0, 
0, 0, 0, 0, 0, 0, 0, 2.02230603557236e-06, 0)), row.names = c(NA, 
-26L), class = c("tbl_df", "tbl", "data.frame")), structure(list(
    value = c("[-Inf, 0)", "[0, 20000)", "[20000, 40000)", 
    "[40000, 60000)", "[60000, 80000)", "[80000, 1e+05)", 
    "[1e+05, 120000)", "[120000, 140000)", "[140000, 160000)", 
    "[160000, 180000)", "[180000, 2e+05)", "[2e+05, 220000)", 
    "[220000, 240000)", "[240000, 260000)", "[260000, 280000)", 
    "[280000, 3e+05)", "[3e+05, Inf)"), prop = c(0, 0.965121399547629, 
    0.0219419924337957, 0.00583678616465502, 0.00279829460157329, 
    0.00160731800076062, 0.00102283872775676, 0.000630517023959647, 
    0.000388318421105306, 0.000296242919193739, 0.000178146079785424, 
    6.40525230689165e-05, 6.20508817230129e-05, 4.60377509557837e-05, 
    4.00328269180728e-06, 2.00164134590364e-06, 0)), row.names = c(NA, 
-17L), class = c("tbl_df", "tbl", "data.frame")), structure(list(
    value = c("[-Inf, 0)", "[0, 20000)", "[20000, 40000)", 
    "[40000, 60000)", "[60000, 80000)", "[80000, 1e+05)", 
    "[1e+05, 120000)", "[120000, 140000)", "[140000, 160000)", 
    "[160000, 180000)", "[180000, 2e+05)", "[2e+05, 220000)", 
    "[220000, 240000)", "[240000, 260000)", "[260000, 280000)", 
    "[280000, 3e+05)", "[3e+05, 320000)", "[320000, 340000)", 
    "[340000, 360000)", "[360000, 380000)", "[380000, 4e+05)", 
    "[4e+05, 420000)", "[420000, 440000)", "[440000, 460000)", 
    "[460000, 480000)", "[480000, 5e+05)", "[5e+05, Inf)"
    ), prop = c(0, 0.962615288870034, 0.00958330915327091, 
    0.0104617958348925, 0.00314974495870712, 0, 0.00967936236675118, 
    0, 5.00277153543063e-05, 0, 0, 8.80487790235791e-05, 
    0, 0.000580321498109953, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 
    0, 0.00379210082385642, 0)), row.names = c(NA, -27L), class = c("tbl_df", 
"tbl", "data.frame")), structure(list(value = c("[-Inf, 0)", 
"[0, 2)", "[2, 4)", "[4, 6)", "[6, 8)", "[8, 10)", "[10, 12)", 
"[12, 14)", "[14, 16)", "[16, 18)", "[18, 20)", "[20, 22)", 
"[22, 24)", "[24, 26)", "[26, 28)", "[28, 30)", "[30, 32)", 
"[32, 34)", "[34, 36)", "[36, 38)", "[38, 40)", "[40, 42)", 
"[42, 44)", "[44, Inf)"), prop = c(0, 0.663122, 0.251104, 
0.054242, 0.01933, 0.007962, 0.002444, 0.000946, 0.00032, 
0.00016, 8.4e-05, 4e-05, 1.4e-05, 1.2e-05, 8.4e-05, 0.000102, 
1.6e-05, 0, 4e-06, 2e-06, 2e-06, 4e-06, 6e-06, 0)), row.names = c(NA, 
-24L), class = c("tbl_df", "tbl", "data.frame")))), row.names = c(NA, 
-4L), class = c("tbl_df", "tbl", "data.frame"))

Regarding the use of `DT`, this gist (https://gist.github.com/stla/e3607e3cf87ddbe70b7134d4a1c874d9) might be interesting for you — davidnortes, Jun 07 '20 at 16:31

Waldi · Accepted Answer · 2020-06-07T20:38:47.893

You can visualize the nested tibble with the sparkline package :

library(dplyr)
library(htmltools)
library(sparkline)
library(purrr)
library(DT)

# Create the HTML tag for each tibble
hist_spark <- test$hist %>% purrr::map( function(x) { 
  as.character(htmltools::as.tags(sparkline(x$prop)))
  })

test$hist_spark <- hist_spark

# Render the tags
datatable(test,
          options  = list(
                          columnDefs = list(list(visible=FALSE, targets=c(c(10)))),
                          fnDrawCallback = htmlwidgets::JS('function(){HTMLWidgets.staticRender();}')
                          )
          ) %>% sparkline::spk_add_deps()

fnDrawCallback datatable option and spk_add_deps are necessary for this to work :

score 0 · Answer 2 · answered Jun 07 '20 at 16:24

0

You likely want unnest or unnest_wider to keep in same dataframe.

Unnest a list column directly into several columns

answered Jun 07 '20 at 16:24

Jope

176
7

i'm not looking for unnest, but for a way to provide access to the entire element (either by display or through visualization), preferably within the DT or a comparable interface. thanks in any case :) – mysteRious Jun 07 '20 at 16:47

R dataframe cell contains dataframe/tibble - how to access or visualize?

2 Answers2