Added 'get_all_subset_matches()' function to the package

NAMESPACE

@@ -3,6 +3,7 @@
 export("%>%")
 export(biblioverApp)
 export(biblioverlap)
+export(get_all_subset_matches)
 export(matching_summary_plot)
 export(merge_input_files)
 export(merge_results)

NEWS.md

@@ -7,6 +7,7 @@
 * Added package logo to plots
 * Added `merge_results`: a function to merge biblioverlap's results into a single dataframe
 * Added `merge_input_files`: a function to merge multiple files from the same source into a single file
+* Added `get_all_subset_matches`: a function to recover missing matches in subsets of data due to differences in fields from distinct bibliographic sources
 
 
 # biblioverlap 1.0.3

R/05-biblioverlap.R

@@ -286,3 +286,52 @@ merge_results <- function(db_list, filter = 'none') {
 
   return(df)
 }
+
+
+#' Get all matches from a given subset of biblioverlap's results
+#'
+#' @param subset_db_list - a subset of the results generated by [`biblioverlap`]
+#' @param db_list - the full set of results generated by [`biblioverlap`]
+#'
+#' @return - the subset data plus any other records outside the subset that have been matched to its documents
+#' @importFrom rlang .data
+#' @export
+#'
+#' @examples
+#' #Running document-level matching procedure for two datasets
+#' biblioverlap_results <- biblioverlap(ufrj_bio_0122[1:2])$db_list
+#'
+#' #Change the document type of one of the datasets from 'journal article' to
+#' #'article' to emulate bibliographical source differences
+#' biblioverlap_results[[2]][['Publication Type']] <- gsub('journal article',
+#' 'article',
+#' biblioverlap_results[[2]][['Publication Type']])
+#'
+#' #Generating venn diagram for the entire dataset
+#' venn_plot(biblioverlap_results)
+#'
+#' #Obtaining only the subset of records with publication type 'article'
+#' biblioverlap_results_subset <- lapply(biblioverlap_results, function(db) {
+#' db[db$'Publication Type' == "article", ] })
+#'
+#' #Generating venn diagram for data subset
+#' #Returns us how many documents categorized as 'article' are unique to a given
+#' #dataset and how many find a match against other documents in the subset
+#' #(i.e. that are also categorized as 'article', in this example)
+#' venn_plot(biblioverlap_results_subset)
+#'
+#' #Recovering missing matches due to bibliographical source differences
+#' #in the subsetting process
+#' subset_all_matches <- get_all_subset_matches(biblioverlap_results_subset,
+#' biblioverlap_results)
+#'
+#' #Generating venn diagram for data subset plus all its matches
+#' #Returns us how many documents categorized as 'article' are unique to a given
+#' #dataset and how many find a match against any other document
+#' venn_plot(subset_all_matches)
+
+get_all_subset_matches <- function(subset_db_list, db_list) {
+  filtered_uuids <- unique(unlist(lapply(subset_db_list, function(db) db$UUID )))
+  no_missing <- lapply(db_list, function(db) dplyr::filter(db, .data$UUID %in% filtered_uuids))
+  return(no_missing)
+}