Update I/O paths to new information architecture (#156)

* export four user facing functions * better name to avoid cnflicts * update * update cookbook * update pkgdown * improve names of exported functions * update io paths based on new ia * rm dead code * update example.config.yml * update vignette config_yml * resolve NOTE * rm dead code * update error messages
RMI-PACTA · Sep 25, 2024 · a886d51 · a886d51
1 parent b58c728
commit a886d51
Show file tree

Hide file tree

Showing 12 changed files with 191 additions and 233 deletions.
diff --git a/R/load_config.R b/R/load_config.R
@@ -3,8 +3,15 @@ load_config <- function(config) {
   config
 }
 
-get_abcd_dir <- function(params) {
-  params[["directories"]][["dir_abcd"]]
+get_input_dir <- function(params) {
+  params[["directories"]][["dir_input"]]
+}
+
+get_loanbook_dir <- function(params) {
+  file.path(
+    params[["directories"]][["dir_input"]],
+    "loanbooks"
+  )
 }
 
 get_abcd_filename <- function(params) {
@@ -15,20 +22,36 @@ get_abcd_sheet <- function(params) {
   params[["file_names"]][["sheet_abcd"]]
 }
 
-get_matched_dir <- function(params) {
-  params[["directories"]][["dir_matched"]]
+get_output_dir <- function(params) {
+  params[["directories"]][["dir_output"]]
+}
+
+get_output_prepare_dir <- function(params) {
+  file.path(
+    params[["directories"]][["dir_output"]],
+    "prepare_abcd"
+  )
 }
 
-get_raw_dir <- function(params) {
-  params[["directories"]][["dir_raw"]]
+get_output_matched_loanbooks_dir <- function(params) {
+  file.path(
+    params[["directories"]][["dir_output"]],
+    "matched_loanbooks"
+  )
 }
 
-get_scenario_dir <- function(params) {
-  params[["directories"]][["dir_scenario"]]
+get_output_prio_diagnostics_dir <- function(params) {
+  file.path(
+    params[["directories"]][["dir_output"]],
+    "prioritized_loanbooks_and_diagnostics"
+  )
 }
 
-get_output_dir <- function(params) {
-  params[["directories"]][["dir_output"]]
+get_output_analysis_dir <- function(params) {
+  file.path(
+    params[["directories"]][["dir_output"]],
+    "analysis"
+  )
 }
 
 get_scenario_tms_filename <- function(params) {
@@ -123,38 +146,34 @@ get_use_manual_sector_classification <- function(params) {
   params[["matching"]][["manual_sector_classification"]][["use_manual_sector_classification"]]
 }
 
-get_manual_sector_classification_dir <- function(params) {
-  params[["matching"]][["manual_sector_classification"]][["dir_manual_sector_classification"]]
-}
-
 get_manual_sector_classification_filename <- function(params) {
   params[["matching"]][["manual_sector_classification"]][["filename_manual_sector_classification"]]
 }
 
 get_manual_sector_classification_path <- function(params) {
   file.path(
-    get_manual_sector_classification_dir(params),
+    get_input_dir(params),
     get_manual_sector_classification_filename(params)
   )
 }
 
 get_abcd_path <- function(config) {
   file.path(
-    get_abcd_dir(config),
+    get_input_dir(config),
     get_abcd_filename(config)
   )
 }
 
 get_sector_split_path <- function(config) {
   file.path(
-    config[["sector_split"]][["dir_split_company_id"]],
+    get_input_dir(config),
     config[["sector_split"]][["filename_split_company_id"]]
   )
 }
 
 get_advanced_company_indicators_path <- function(config) {
   file.path(
-    config[["sector_split"]][["dir_advanced_company_indicators"]],
+    get_input_dir(config),
     config[["sector_split"]][["filename_advanced_company_indicators"]]
   )
 }
@@ -165,14 +184,14 @@ get_advanced_company_indicators_sheet <- function(config) {
 
 get_scenario_tms_path <- function(config) {
   file.path(
-    get_scenario_dir(config),
+    get_input_dir(config),
     get_scenario_tms_filename(config)
   )
 }
 
 get_scenario_sda_path <- function(config) {
   file.path(
-    get_scenario_dir(config),
+    get_input_dir(config),
     get_scenario_sda_filename(config)
   )
 }
diff --git a/R/match_loanbooks.R b/R/match_loanbooks.R
@@ -21,9 +21,10 @@
 match_loanbooks <- function(config) {
   config <- load_config(config)
 
-  dir_raw <- get_raw_dir(config)
-  abcd_dir <- get_abcd_dir(config)
-  dir_matched <- get_matched_dir(config)
+  input_loanbooks_dir <- get_loanbook_dir(config)
+  output_prepare_dir <- get_output_prepare_dir(config)
+  output_matched_loanbooks_dir <- get_output_matched_loanbooks_dir(config)
+  dir.create(output_matched_loanbooks_dir, recursive = TRUE, showWarnings = FALSE)
 
   matching_by_sector <- get_match_by_sector(config)
   matching_min_score <- get_match_min_score(config)
@@ -38,18 +39,18 @@ match_loanbooks <- function(config) {
   }
 
   # validate config values----
-  stop_if_not_length(dir_raw, 1L)
-  stop_if_not_inherits(dir_raw, "character")
-  stop_if_dir_not_found(dir_raw, desc = "Raw loanbook")
+  stop_if_not_length(input_loanbooks_dir, 1L)
+  stop_if_not_inherits(input_loanbooks_dir, "character")
+  stop_if_dir_not_found(input_loanbooks_dir, desc = "Input - loanbooks")
 
-  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(output_prepare_dir, 1L)
+  stop_if_not_inherits(output_prepare_dir, "character")
+  stop_if_dir_not_found(output_prepare_dir, desc = "Output - prepare ABCD")
+  stop_if_file_not_found(file.path(output_prepare_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")
+  stop_if_not_length(output_matched_loanbooks_dir, 1L)
+  stop_if_not_inherits(output_matched_loanbooks_dir, "character")
+  stop_if_dir_not_found(output_matched_loanbooks_dir, desc = "Output - Matched loanbooks")
 
   stop_if_not_length(matching_by_sector, 1L)
   stop_if_not_inherits(matching_by_sector, "logical")
@@ -85,7 +86,7 @@ match_loanbooks <- function(config) {
 
   ## load abcd----
   abcd <- readr::read_csv(
-    file.path(abcd_dir, "abcd_final.csv"),
+    file.path(output_prepare_dir, "abcd_final.csv"),
     col_select = dplyr::all_of(cols_abcd),
     col_types = col_types_abcd_final
   )
@@ -100,11 +101,11 @@ match_loanbooks <- function(config) {
   }
 
   ## load raw loan books----
-  list_raw <- list.files(path = dir_raw, pattern = "[.]csv$")
-  stop_if_no_files_found(list_raw, dir_raw, "dir_raw", "raw loan book CSVs")
+  list_raw <- list.files(path = input_loanbooks_dir, pattern = "[.]csv$")
+  stop_if_no_files_found(list_raw, input_loanbooks_dir, "dir_input", "raw loan book CSVs")
 
   raw_lbk <- readr::read_csv(
-    file = file.path(dir_raw, list_raw),
+    file = file.path(input_loanbooks_dir, list_raw),
     col_types = col_types_raw,
     id = "group_id"
   ) %>%
@@ -154,7 +155,7 @@ match_loanbooks <- function(config) {
     ## write matched data to file----
     matched_lbk_i %>%
       readr::write_csv(
-        file = file.path(dir_matched, glue::glue("matched_lbk_{group_name}.csv")),
+        file = file.path(output_matched_loanbooks_dir, glue::glue("matched_lbk_{group_name}.csv")),
         na = ""
       )
     cli::cli_progress_update()

diff --git a/R/plot_aggregate_loanbooks.R b/R/plot_aggregate_loanbooks.R
@@ -12,9 +12,8 @@ plot_aggregate_loanbooks <- function(config) {
   config <- load_config(config)
 
   # paths
-  input_path_matched <- get_matched_dir(config)
-  output_path <- get_output_dir(config)
-  output_path_aggregated <- file.path(output_path, "aggregated")
+  output_analysis_dir <- get_output_analysis_dir(config)
+  output_analysis_aggregated_dir <- file.path(output_analysis_dir, "aggregated")
 
   # project parameters
   scenario_source_input <- get_scenario_source(config)
@@ -30,13 +29,13 @@ plot_aggregate_loanbooks <- function(config) {
 
   # if a sector split is applied, write results into a directory that states the type
   if (apply_sector_split) {
-    output_path_aggregated <- file.path(output_path, sector_split_type_select, "aggregated")
+    output_analysis_aggregated_dir <- file.path(output_analysis_dir, sector_split_type_select, "aggregated")
   }
 
   by_group <- get_by_group(config)
   by_group <- check_and_prepare_by_group(by_group)
 
-  dir.create(output_path_aggregated, recursive = TRUE, showWarnings = FALSE)
+  dir.create(output_analysis_aggregated_dir, recursive = TRUE, showWarnings = FALSE)
 
 
   # load required data----
@@ -70,7 +69,7 @@ plot_aggregate_loanbooks <- function(config) {
   company_aggregated_alignment_net  <-
     readr::read_csv(
       file = file.path(
-        output_path_aggregated,
+        output_analysis_aggregated_dir,
         glue::glue("company_exposure_net_aggregate_alignment{file_by_group}.csv")
       ),
       col_types = col_types_company_aggregated_alignment,
@@ -80,7 +79,7 @@ plot_aggregate_loanbooks <- function(config) {
   company_aggregated_alignment_bo_po <-
     readr::read_csv(
       file = file.path(
-        output_path_aggregated,
+        output_analysis_aggregated_dir,
         glue::glue("company_exposure_bo_po_aggregate_alignment{file_by_group}.csv")
       ),
       col_types = col_types_company_aggregated_alignment,
@@ -91,7 +90,7 @@ plot_aggregate_loanbooks <- function(config) {
   loanbook_exposure_aggregated_alignment_bo_po <-
     readr::read_csv(
       file = file.path(
-        output_path_aggregated,
+        output_analysis_aggregated_dir,
         glue::glue("loanbook_exposure_bo_po_aggregate_alignment{file_by_group}.csv")
       ),
       col_types = readr::cols(
@@ -111,7 +110,7 @@ plot_aggregate_loanbooks <- function(config) {
   loanbook_exposure_aggregated_alignment_net <-
     readr::read_csv(
       file = file.path(
-        output_path_aggregated,
+        output_analysis_aggregated_dir,
         glue::glue("loanbook_exposure_net_aggregate_alignment{file_by_group}.csv")
       ),
       col_types = readr::cols(
@@ -158,7 +157,7 @@ plot_aggregate_loanbooks <- function(config) {
       data_sankey_sector %>%
         readr::write_csv(
           file = file.path(
-            output_path_aggregated,
+            output_analysis_aggregated_dir,
             glue::glue("data_{output_file_sankey_sector}.csv")
           ),
           na = ""
@@ -167,7 +166,7 @@ plot_aggregate_loanbooks <- function(config) {
       plot_sankey(
         data_sankey_sector,
         group_var = by_group,
-        save_png_to = path.expand(output_path_aggregated),
+        save_png_to = path.expand(output_analysis_aggregated_dir),
         png_name = glue::glue("plot_{output_file_sankey_sector}.png"),
         nodes_order_from_data = TRUE
       )
@@ -200,7 +199,7 @@ plot_aggregate_loanbooks <- function(config) {
       data_sankey_company_sector %>%
         readr::write_csv(
           file = file.path(
-            output_path_aggregated,
+            output_analysis_aggregated_dir,
             glue::glue("data_{output_file_sankey_company_sector}.csv")
           ),
           na = ""
@@ -209,7 +208,7 @@ plot_aggregate_loanbooks <- function(config) {
       plot_sankey(
         data_sankey_company_sector,
         group_var = by_group,
-        save_png_to = path.expand(output_path_aggregated),
+        save_png_to = path.expand(output_analysis_aggregated_dir),
         png_name = glue::glue("plot_{output_file_sankey_company_sector}.png")
       )
     }
@@ -243,7 +242,7 @@ plot_aggregate_loanbooks <- function(config) {
       data_scatter_alignment_exposure %>%
         readr::write_csv(
           file = file.path(
-            output_path_aggregated,
+            output_analysis_aggregated_dir,
             glue::glue("data_{output_file_alignment_exposure}.csv")
           ),
           na = ""
@@ -259,7 +258,7 @@ plot_aggregate_loanbooks <- function(config) {
 
       ggplot2::ggsave(
         filename = glue::glue("plot_{output_file_alignment_exposure}.png"),
-        path = output_path_aggregated,
+        path = output_analysis_aggregated_dir,
         width = 8,
         height = 5,
         dpi = 300,
@@ -302,7 +301,7 @@ plot_aggregate_loanbooks <- function(config) {
       data_scatter_automotive_group %>%
         readr::write_csv(
           file = file.path(
-            output_path_aggregated,
+            output_analysis_aggregated_dir,
             glue::glue("data_{output_file_scatter_sector}.csv")
           ),
           na = ""
@@ -319,7 +318,7 @@ plot_aggregate_loanbooks <- function(config) {
       )
       ggplot2::ggsave(
         filename = glue::glue("plot_{output_file_scatter_sector}.png"),
-        path = output_path_aggregated,
+        path = output_analysis_aggregated_dir,
         width = 8,
         height = 5
       )
@@ -356,7 +355,7 @@ plot_aggregate_loanbooks <- function(config) {
       data_scatter_power_group %>%
         readr::write_csv(
           file = file.path(
-            output_path_aggregated,
+            output_analysis_aggregated_dir,
             glue::glue("data_{output_file_scatter_sector}.csv")
           ),
           na = ""
@@ -374,7 +373,7 @@ plot_aggregate_loanbooks <- function(config) {
 
       ggplot2::ggsave(
         filename = glue::glue("plot_{output_file_scatter_sector}.png"),
-        path = output_path_aggregated,
+        path = output_analysis_aggregated_dir,
         width = 8,
         height = 5
       )
@@ -399,7 +398,7 @@ plot_aggregate_loanbooks <- function(config) {
       unique()
 
     for (i in dirs_for_by_group) {
-      dir.create(file.path(output_path_aggregated, i), recursive = TRUE, showWarnings = FALSE)
+      dir.create(file.path(output_analysis_aggregated_dir, i), recursive = TRUE, showWarnings = FALSE)
     }
   }
 
@@ -441,7 +440,7 @@ plot_aggregate_loanbooks <- function(config) {
         data_scatter_automotive_company_i %>%
           readr::write_csv(
             file = file.path(
-              output_path_aggregated,
+              output_analysis_aggregated_dir,
               i,
               glue::glue("data_scatter_automotive_company_by_{by_group}_{i}.csv")
             ),
@@ -462,7 +461,7 @@ plot_aggregate_loanbooks <- function(config) {
 
         ggplot2::ggsave(
           filename = glue::glue("plot_scatter_automotive_company_by_{by_group}_{i}.png"),
-          path = file.path(output_path_aggregated, i),
+          path = file.path(output_analysis_aggregated_dir, i),
           width = 8,
           height = 5
         )
@@ -504,7 +503,7 @@ plot_aggregate_loanbooks <- function(config) {
         data_scatter_power_company_i %>%
           readr::write_csv(
             file = file.path(
-              output_path_aggregated,
+              output_analysis_aggregated_dir,
               i,
               glue::glue("data_scatter_power_company_by_{by_group}_{i}.csv")
             ),
@@ -525,7 +524,7 @@ plot_aggregate_loanbooks <- function(config) {
 
         ggplot2::ggsave(
           filename = glue::glue("plot_scatter_power_company_by_{by_group}_{i}.png"),
-          path = file.path(output_path_aggregated, i),
+          path = file.path(output_analysis_aggregated_dir, i),
           width = 8,
           height = 5
         )