R Under development (unstable) (2024-01-31 r85845 ucrt) -- "Unsuffered Consequences"
Copyright (C) 2024 The R Foundation for Statistical Computing
Platform: x86_64-w64-mingw32/x64

R is free software and comes with ABSOLUTELY NO WARRANTY.
You are welcome to redistribute it under certain conditions.
Type 'license()' or 'licence()' for distribution details.

R is a collaborative project with many contributors.
Type 'contributors()' for more information and
'citation()' on how to cite R or R packages in publications.

Type 'demo()' for some demos, 'help()' for on-line help, or
'help.start()' for an HTML browser interface to help.
Type 'q()' to quit R.

> library(reclin2)
Loading required package: data.table
> source("helpers.R")
> 
> 
> data("linkexample1", "linkexample2")
> linkexample1$id2 <- c(1,1,3,3,5,6)
> linkexample2$id2 <- c(2,3,3,6,7)
> 
> pairs <- pair_blocking(linkexample1, linkexample2, "postcode")
> pairs[, score := runif(nrow(pairs))]
  First data set:  6 records
  Second data set: 5 records
  Total number of pairs: 17 pairs
  Blocking on: 'postcode'

       .x    .y      score
    <int> <int>      <num>
 1:     1     1 0.27419415
 2:     1     2 0.56777076
 3:     1     3 0.17351057
 4:     2     1 0.27617001
 5:     2     2 0.95404493
 6:     2     3 0.31544355
 7:     3     1 0.76453290
 8:     3     2 0.91346322
 9:     3     3 0.45524865
10:     4     1 0.15560062
11:     4     2 0.07531252
12:     4     3 0.82084235
13:     5     1 0.94760515
14:     5     2 0.22172045
15:     5     3 0.02280451
16:     6     4 0.40029323
17:     6     5 0.77482137
> pairs[, select := score > 0.5]
  First data set:  6 records
  Second data set: 5 records
  Total number of pairs: 17 pairs
  Blocking on: 'postcode'

       .x    .y      score select
    <int> <int>      <num> <lgcl>
 1:     1     1 0.27419415  FALSE
 2:     1     2 0.56777076   TRUE
 3:     1     3 0.17351057  FALSE
 4:     2     1 0.27617001  FALSE
 5:     2     2 0.95404493   TRUE
 6:     2     3 0.31544355  FALSE
 7:     3     1 0.76453290   TRUE
 8:     3     2 0.91346322   TRUE
 9:     3     3 0.45524865  FALSE
10:     4     1 0.15560062  FALSE
11:     4     2 0.07531252  FALSE
12:     4     3 0.82084235   TRUE
13:     5     1 0.94760515   TRUE
14:     5     2 0.22172045  FALSE
15:     5     3 0.02280451  FALSE
16:     6     4 0.40029323  FALSE
17:     6     5 0.77482137   TRUE
> pairs[, select2 := FALSE]
  First data set:  6 records
  Second data set: 5 records
  Total number of pairs: 17 pairs
  Blocking on: 'postcode'

       .x    .y      score select select2
    <int> <int>      <num> <lgcl>  <lgcl>
 1:     1     1 0.27419415  FALSE   FALSE
 2:     1     2 0.56777076   TRUE   FALSE
 3:     1     3 0.17351057  FALSE   FALSE
 4:     2     1 0.27617001  FALSE   FALSE
 5:     2     2 0.95404493   TRUE   FALSE
 6:     2     3 0.31544355  FALSE   FALSE
 7:     3     1 0.76453290   TRUE   FALSE
 8:     3     2 0.91346322   TRUE   FALSE
 9:     3     3 0.45524865  FALSE   FALSE
10:     4     1 0.15560062  FALSE   FALSE
11:     4     2 0.07531252  FALSE   FALSE
12:     4     3 0.82084235   TRUE   FALSE
13:     5     1 0.94760515   TRUE   FALSE
14:     5     2 0.22172045  FALSE   FALSE
15:     5     3 0.02280451  FALSE   FALSE
16:     6     4 0.40029323  FALSE   FALSE
17:     6     5 0.77482137   TRUE   FALSE
> 
> 
> # === TESTS FOR ID_X AND ID_Y ARGUMENTS
> 
> test <- reclin2:::select_preprocess(pairs, score = "score", id_x = "id", id_y = "id")
> expect_equal(test$.x, linkexample1$id[pairs$.x])
> expect_equal(test$.y, linkexample2$id[pairs$.y])
> 
> test <- reclin2:::select_preprocess(pairs, score = "score", id_x = "id2", id_y = "id2")
> expect_equal(test$.x, linkexample1$id2[pairs$.x])
> expect_equal(test$.y, linkexample2$id2[pairs$.y])
> 
> test <- reclin2:::select_preprocess(pairs, score = "score", id_x = "id2", id_y = "id2",
+   preselect = "select")
> expect_equal(sort(test$.x), sort(linkexample1$id2[pairs$.x[pairs$select]]))
> expect_equal(sort(test$.y), sort(linkexample2$id2[pairs$.y[pairs$select]]))
> 
> test <- reclin2:::select_preprocess(pairs, score = "score", id_x = "id2", id_y = "id2",
+   preselect = "select2")
> expect_equal(nrow(test), 0)
> expect_equal(names(test), c(".x", ".y", "score", "index"))
> 
> expect_error(
+   test <- reclin2:::select_preprocess(pairs, score = "score", id_x = 1:3, id_y = 1:3,
+     preselect = "select")
+ )
> 
> proc.time()
   user  system elapsed 
   0.54    0.14    0.67