R Under development (unstable) (2024-02-08 r85876 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 0.5894881
 2:     1     2 0.1824231
 3:     1     3 0.8353371
 4:     2     1 0.6770434
 5:     2     2 0.6544106
 6:     2     3 0.8774025
 7:     3     1 0.5120988
 8:     3     2 0.6981675
 9:     3     3 0.4878957
10:     4     1 0.9716523
11:     4     2 0.8297809
12:     4     3 0.8262862
13:     5     1 0.1826654
14:     5     2 0.3685295
15:     5     3 0.4131194
16:     6     4 0.6243491
17:     6     5 0.2456658
> 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 0.5894881   TRUE
 2:     1     2 0.1824231  FALSE
 3:     1     3 0.8353371   TRUE
 4:     2     1 0.6770434   TRUE
 5:     2     2 0.6544106   TRUE
 6:     2     3 0.8774025   TRUE
 7:     3     1 0.5120988   TRUE
 8:     3     2 0.6981675   TRUE
 9:     3     3 0.4878957  FALSE
10:     4     1 0.9716523   TRUE
11:     4     2 0.8297809   TRUE
12:     4     3 0.8262862   TRUE
13:     5     1 0.1826654  FALSE
14:     5     2 0.3685295  FALSE
15:     5     3 0.4131194  FALSE
16:     6     4 0.6243491   TRUE
17:     6     5 0.2456658  FALSE
> 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 0.5894881   TRUE   FALSE
 2:     1     2 0.1824231  FALSE   FALSE
 3:     1     3 0.8353371   TRUE   FALSE
 4:     2     1 0.6770434   TRUE   FALSE
 5:     2     2 0.6544106   TRUE   FALSE
 6:     2     3 0.8774025   TRUE   FALSE
 7:     3     1 0.5120988   TRUE   FALSE
 8:     3     2 0.6981675   TRUE   FALSE
 9:     3     3 0.4878957  FALSE   FALSE
10:     4     1 0.9716523   TRUE   FALSE
11:     4     2 0.8297809   TRUE   FALSE
12:     4     3 0.8262862   TRUE   FALSE
13:     5     1 0.1826654  FALSE   FALSE
14:     5     2 0.3685295  FALSE   FALSE
15:     5     3 0.4131194  FALSE   FALSE
16:     6     4 0.6243491   TRUE   FALSE
17:     6     5 0.2456658  FALSE   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    0.10    0.62