updated dt_unnest with fill arg

DESCRIPTION

@@ -1,6 +1,6 @@
 Package: tidyfast
 Title: Fast Tidying of Data
-Version: 0.2.1
+Version: 0.2.2
 Authors@R: c(
     person(given = "Tyson",
            family = "Barrett",
@@ -9,6 +9,9 @@ Authors@R: c(
            comment = c(ORCID = "0000-0002-2137-1391")),
     person(given = "Mark",
            family = "Fairbanks",
+           role = "ctb"),
+    person(given = "Ivan",
+           family = "Leung",
            role = "ctb"))
 Description: Tidying functions built on 'data.table'
     to provide quick and efficient data manipulation with

R/unnest.R

@@ -4,6 +4,7 @@
 #'
 #' @param dt_ the data table to unnest
 #' @param col  the column to unnest
+#' @param fill when set to \code{TRUE}, it allows the user to unnest data.tables that have different variables in them (e.g. dt1 has x and y while dt2 has x and z). This uses the \code{fill} argument in \code{data.table::rbindlist()}.
 #' @param ... any of the other variables in the nested table that you want to keep in the unnested table. Bare variable names. If none are provided, all variables are kept.
 #'
 #' @examples
@@ -21,12 +22,12 @@
 #' @import data.table
 #'
 #' @export
-dt_unnest <- function(dt_, col, ...){
+dt_unnest <- function(dt_, col, fill = FALSE, ...){
   UseMethod("dt_unnest", dt_)
 }
 
 #' @export
-dt_unnest.default <- function(dt_, col, ...){
+dt_unnest.default <- function(dt_, col, fill = FALSE, ...){
   if (isFALSE(is.data.table(dt_)))
     dt_ <- as.data.table(dt_)
 
@@ -45,7 +46,7 @@ dt_unnest.default <- function(dt_, col, ...){
   others_dt <- others_dt[, ..keep]
   others_dt <- lapply(others_dt, rep, times = rows)
 
-  dt_[, list(as.data.table(others_dt), rbindlist(eval(col)))]
+  dt_[, list(as.data.table(others_dt), rbindlist(eval(col), fill = fill))]
 }
 
 

README.Rmd

@@ -16,7 +16,7 @@ knitr::opts_chunk$set(
 set.seed(843)
 ```
 
-# `tidyfast v0.2.1` <img src="man/figures/tidyfast_hex.png" align="right" width="30%" height="30%" />
+# `tidyfast v0.2.2` <img src="man/figures/tidyfast_hex.png" align="right" width="30%" height="30%" />
 
 
 <!-- badges: start -->

README.md

@@ -1,7 +1,7 @@
 
 <!-- README.md is generated from README.Rmd. Please edit that file -->
 
-# `tidyfast v0.2.1` <img src="man/figures/tidyfast_hex.png" align="right" width="30%" height="30%" />
+# `tidyfast v0.2.2` <img src="man/figures/tidyfast_hex.png" align="right" width="30%" height="30%" />
 
 <!-- badges: start -->
 
@@ -216,13 +216,13 @@ highlighted below. Notably, the timings are without the `nested1` and
     #> # A tibble: 2 x 3
     #>   expression   median mem_alloc
     #>   <chr>      <bch:tm> <bch:byt>
-    #> 1 dt_nest      3.31ms    2.88MB
-    #> 2 group_nest   5.73ms    2.56MB
+    #> 1 dt_nest      3.15ms    2.88MB
+    #> 2 group_nest   9.34ms     5.4MB
     #> # A tibble: 2 x 3
     #>   expression   median mem_alloc
     #>   <chr>      <bch:tm> <bch:byt>
-    #> 1 dt_unnest    4.94ms    5.48MB
-    #> 2 unnest       9.76ms    6.15MB
+    #> 1 dt_unnest    4.85ms    5.48MB
+    #> 2 unnest      17.95ms   15.31MB
 
 ## Pivoting
 
@@ -295,10 +295,10 @@ But let’s compare some basic speed and efficiency. Because of the
     #> # A tibble: 4 x 3
     #>   expression        median mem_alloc
     #>   <chr>           <bch:tm> <bch:byt>
-    #> 1 dt_pivot_longer 958.91µs  996.21KB
-    #> 2 pivot_longer      7.54ms    3.13MB
-    #> 3 dt_pivot_wider   11.11ms    1.86MB
-    #> 4 pivot_wider      10.41ms    2.05MB
+    #> 1 dt_pivot_longer    1.1ms  997.09KB
+    #> 2 pivot_longer      11.4ms    2.55MB
+    #> 3 dt_pivot_wider    10.9ms    1.86MB
+    #> 4 pivot_wider       13.3ms    1.95MB
 
 ### If Else
 
@@ -347,9 +347,9 @@ built on `data.table::fifelse()`.
     #> # A tibble: 3 x 3
     #>   expression     median mem_alloc
     #>   <chr>        <bch:tm> <bch:byt>
-    #> 1 case_when     137.5ms   148.8MB
-    #> 2 dt_case_when   36.3ms    34.3MB
-    #> 3 fifelse        36.7ms    34.3MB
+    #> 1 case_when     133.7ms   148.8MB
+    #> 2 dt_case_when   35.8ms    34.3MB
+    #> 3 fifelse        34.1ms    34.3MB
 
 ## Fill
 
@@ -457,7 +457,7 @@ marks3 <-
     #> # A tibble: 2 x 3
     #>   expression                                    median mem_alloc
     #>   <bch:expr>                                  <bch:tm> <bch:byt>
-    #> 1 tidyr::fill(dplyr::group_by(df3, id), x, y)   65.7ms    30.7MB
+    #> 1 tidyr::fill(dplyr::group_by(df3, id), x, y)   44.1ms    42.6MB
     #> 2 tidyfast::dt_fill(dt3, x, y, id = list(id))   23.8ms    29.1MB
 
 ## Separate
@@ -500,9 +500,9 @@ than `tidyr::separate()`.
     #> # A tibble: 3 x 3
     #>   expression            median mem_alloc
     #>   <chr>               <bch:tm> <bch:byt>
-    #> 1 separate               363ms    11.8MB
-    #> 2 dt_separate            122ms    30.6MB
-    #> 3 dt_separate-mutable    118ms    26.7MB
+    #> 1 separate               363ms    11.6MB
+    #> 2 dt_separate            132ms    30.6MB
+    #> 3 dt_separate-mutable    114ms    26.7MB
 
 ## Count and Uncount