From 9ba41a9fd79d60fca867a085504fc974a8bfbef6 Mon Sep 17 00:00:00 2001
From: Jacob Kastl <60064070+jacobvjk@users.noreply.github.com>
Date: Mon, 23 Sep 2024 12:21:40 +0200
Subject: [PATCH] use prepared abcd in matching (#153)

* use prepared abcd in matching

* Update R/run_match_prioritize.R

Co-authored-by: CJ Yetman <cj@cjyetman.com>

* Update R/run_matching.R

Co-authored-by: CJ Yetman <cj@cjyetman.com>

---------

Co-authored-by: CJ Yetman <cj@cjyetman.com>
---
 R/run_match_prioritize.R | 15 ++++++++++-----
 R/run_matching.R         | 17 ++++++++++-------
 2 files changed, 20 insertions(+), 12 deletions(-)

diff --git a/R/run_match_prioritize.R b/R/run_match_prioritize.R
index f5d6efdc..d84b3fdd 100644
--- a/R/run_match_prioritize.R
+++ b/R/run_match_prioritize.R
@@ -2,8 +2,7 @@ run_match_prioritize <- function(config) {
   config <- load_config(config)
 
   dir_matched <- get_matched_dir(config)
-  path_abcd <- get_abcd_path(config)
-  sheet_abcd <- get_abcd_sheet(config)
+  abcd_dir <- get_abcd_dir(config)
 
   match_prio_priority <- get_match_priority(config)
 
@@ -11,6 +10,10 @@ run_match_prioritize <- function(config) {
   sector_split_type_select <- get_sector_split_type(config)
 
   # validate config values----
+  stop_if_not_length(abcd_dir, 1L)
+  stop_if_not_inherits(abcd_dir, "character")
+  stop_if_dir_not_found(abcd_dir, desc = "ABCD data")
+  stop_if_file_not_found(file.path(abcd_dir, "abcd_final.csv"), desc = "ABCD final")
   stop_if_not_length(dir_matched, 1L)
   stop_if_not_inherits(dir_matched, "character")
   stop_if_dir_not_found(dir_matched, desc = "Matched loanbook")
@@ -62,9 +65,11 @@ run_match_prioritize <- function(config) {
       col_select = dplyr::all_of(col_select_companies_sector_split)
     )
 
-    # TODO: better use prepared abcd?
-    abcd <- read_abcd_raw(path_abcd, sheet_abcd)
-    stop_if_not_expected_columns(abcd, cols_abcd, desc = "ABCD")
+    abcd <- readr::read_csv(
+      file.path(abcd_dir, "abcd_final.csv"),
+      col_select = dplyr::all_of(cols_abcd),
+      col_types = col_types_abcd_final
+    )
   }
 
   # prioritize and save files----
diff --git a/R/run_matching.R b/R/run_matching.R
index a569d31b..5ad48741 100644
--- a/R/run_matching.R
+++ b/R/run_matching.R
@@ -2,8 +2,7 @@ run_matching <- function(config) {
   config <- load_config(config)
 
   dir_raw <- get_raw_dir(config)
-  path_abcd <- get_abcd_path(config)
-  sheet_abcd <- get_abcd_sheet(config)
+  abcd_dir <- get_abcd_dir(config)
   dir_matched <- get_matched_dir(config)
 
   matching_by_sector <- get_match_by_sector(config)
@@ -23,9 +22,10 @@ run_matching <- function(config) {
   stop_if_not_inherits(dir_raw, "character")
   stop_if_dir_not_found(dir_raw, desc = "Raw loanbook")
 
-  stop_if_not_length(path_abcd, 1L)
-  stop_if_not_inherits(path_abcd, "character")
-  stop_if_file_not_found(path_abcd, desc = "ABCD data")
+  stop_if_not_length(abcd_dir, 1L)
+  stop_if_not_inherits(abcd_dir, "character")
+  stop_if_dir_not_found(abcd_dir, desc = "ABCD data")
+  stop_if_file_not_found(file.path(abcd_dir, "abcd_final.csv"), desc = "ABCD final")
 
   stop_if_not_length(dir_matched, 1L)
   stop_if_not_inherits(dir_matched, "character")
@@ -64,8 +64,11 @@ run_matching <- function(config) {
   # load data----
 
   ## load abcd----
-  abcd <- read_abcd_raw(path_abcd, sheet_abcd)
-  stop_if_not_expected_columns(abcd, cols_abcd, desc = "ABCD")
+  abcd <- readr::read_csv(
+    file.path(abcd_dir, "abcd_final.csv"),
+    col_select = dplyr::all_of(cols_abcd),
+    col_types = col_types_abcd_final
+  )
 
   ## optionally load manual classification system----
   if (matching_use_manual_sector_classification) {