don't alphabetise "mapped" targets

[kendonB]

I think these should just retain the order of whatever the mapping variable is. See below for counterintuitive behavior (numbers don't match in the third stage).

library(drake)
library(tidyverse)
drake_plan(
  data = target(
    get_data1(param),
    transform = map(param = c(1,2,3,4,5,6,7,8,9,10,11))
  ),
  results = target(
    get_results(data),
    transform = map(data)
  ),
  more = target(
    get_more(data, results),
    transform = map(data, results)
  )
) %>% 
  print(n=33)
#> # A tibble: 33 x 2
#>    target                      command                            
#>    <chr>                       <expr>                             
#>  1 data_1                      "get_data1(1)                     "
#>  2 data_2                      "get_data1(2)                     "
#>  3 data_3                      "get_data1(3)                     "
#>  4 data_4                      "get_data1(4)                     "
#>  5 data_5                      "get_data1(5)                     "
#>  6 data_6                      "get_data1(6)                     "
#>  7 data_7                      "get_data1(7)                     "
#>  8 data_8                      "get_data1(8)                     "
#>  9 data_9                      "get_data1(9)                     "
#> 10 data_10                     "get_data1(10)                    "
#> 11 data_11                     "get_data1(11)                    "
#> 12 results_data_1              "get_results(data_1)              "
#> 13 results_data_10             "get_results(data_10)             "
#> 14 results_data_11             "get_results(data_11)             "
#> 15 results_data_2              "get_results(data_2)              "
#> 16 results_data_3              "get_results(data_3)              "
#> 17 results_data_4              "get_results(data_4)              "
#> 18 results_data_5              "get_results(data_5)              "
#> 19 results_data_6              "get_results(data_6)              "
#> 20 results_data_7              "get_results(data_7)              "
#> 21 results_data_8              "get_results(data_8)              "
#> 22 results_data_9              "get_results(data_9)              "
#> 23 more_data_1_results_data_1  "get_more(data_1, results_data_1) "
#> 24 more_data_10_results_data_8 get_more(data_10, results_data_8)  
#> 25 more_data_11_results_data_9 get_more(data_11, results_data_9)  
#> 26 more_data_2_results_data_10 get_more(data_2, results_data_10)  
#> 27 more_data_3_results_data_11 get_more(data_3, results_data_11)  
#> 28 more_data_4_results_data_2  "get_more(data_4, results_data_2) "
#> 29 more_data_5_results_data_3  "get_more(data_5, results_data_3) "
#> 30 more_data_6_results_data_4  "get_more(data_6, results_data_4) "
#> 31 more_data_7_results_data_5  "get_more(data_7, results_data_5) "
#> 32 more_data_8_results_data_6  "get_more(data_8, results_data_6) "
#> 33 more_data_9_results_data_7  "get_more(data_9, results_data_7) "

Created on 2019-03-19 by the reprex package (v0.2.0.9000).

[wlandau]

Thanks, @kendonB. merge(sort = FALSE) in dsl_left_outer_join() fixed the problem.