Screen datapapers candidates and build the dataset

data-raw/apply_datapapers_decisions.R

@@ -0,0 +1,60 @@
+# Merge decisions exported from the review app (issue #28) into the committed
+# screening sheet. Reads data-raw/datapapers_decisions.csv (doi, include,
+# reason; produced by the Export button in datapapers_review.html) and fills
+# include/reason in data-raw/datapapers_screening.csv for those DOIs.
+#
+# The decisions file is the reviewer's own choices, so it overwrites any
+# earlier value for the same DOI; rows not in the decisions file are left
+# untouched. Re-running is idempotent.
+#
+# Run from the package root:
+#   Rscript data-raw/apply_datapapers_decisions.R
+
+library(dplyr)
+library(readr)
+
+decisions_path <- "data-raw/datapapers_decisions.csv"
+screening_path <- "data-raw/datapapers_screening.csv"
+
+if (!file.exists(decisions_path)) {
+  stop("Missing ", decisions_path, ". Export decisions from ",
+       "data-raw/datapapers_review.html first (Export CSV button) and save ",
+       "the download there.", call. = FALSE)
+}
+
+decisions <- read_csv(
+  decisions_path,
+  col_types = cols(doi = col_character(), include = col_logical(),
+                   reason = col_character())
+)
+stopifnot(!any(duplicated(decisions$doi)), !any(is.na(decisions$include)))
+
+screening <- read_csv(
+  screening_path,
+  col_types = cols(
+    doi = col_character(), title = col_character(), journal = col_character(),
+    published_year = col_integer(), auto_relevant = col_logical(),
+    include = col_logical(), reason = col_character()
+  )
+)
+
+unknown <- setdiff(decisions$doi, screening$doi)
+if (length(unknown) > 0) {
+  stop(length(unknown), " decision DOI(s) not present in the screening sheet, ",
+       "first: ", unknown[1], call. = FALSE)
+}
+
+updated <- screening |>
+  left_join(decisions, by = "doi", suffix = c("", "_new")) |>
+  mutate(
+    include = coalesce(include_new, include),
+    reason = if_else(!is.na(include_new), reason_new, reason),
+    include_new = NULL, reason_new = NULL
+  )
+
+write_csv(updated, screening_path, na = "")
+
+message(nrow(decisions), " decisions applied to ", screening_path, ": now ",
+        sum(updated$include %in% TRUE), " included, ",
+        sum(updated$include %in% FALSE), " excluded, ",
+        sum(is.na(updated$include)), " pending.")

data-raw/datapapers_country_fixes.csv

@@ -1 +1,2 @@
 doi,first_author_affiliation_country
+10.46471/gigabyte.167,Cabo Verde

data-raw/datapapers_decisions.csv

@@ -0,0 +1,46 @@
+doi,include,reason
+10.3390/data2030024,FALSE,
+10.3390/data3040050,FALSE,
+10.3390/data6080079,FALSE,
+10.3390/data8060103,TRUE,
+10.3390/data8030048,FALSE,
+10.3390/data8090141,TRUE,
+10.3390/data8110162,FALSE,
+10.3390/data10100155,FALSE,
+10.3390/data10040051,FALSE,
+10.46471/gigabyte.167,TRUE,
+10.1093/gigascience/gix101,FALSE,
+10.1093/gigascience/giaa125,FALSE,
+10.1093/gigascience/giae051,FALSE,
+10.1038/sdata.2017.98,FALSE,
+10.1038/sdata.2018.108,TRUE,
+10.1038/s41597-019-0316-y,FALSE,
+10.1038/s41597-020-00648-2,FALSE,
+10.1038/s41597-020-0562-z,FALSE,
+10.1038/s41597-020-0496-5,FALSE,
+10.1038/s41597-020-0384-z,FALSE,
+10.1038/s41597-020-00596-x,FALSE,
+10.1038/s41597-021-01002-w,FALSE,
+10.1038/s41597-021-00856-4,FALSE,
+10.1038/s41597-022-01686-8,TRUE,
+10.1038/s41597-022-01749-w,TRUE,
+10.1038/s41597-022-01439-7,TRUE,
+10.1038/s41597-022-01358-7,FALSE,
+10.1038/s41597-022-01788-3,FALSE,
+10.1038/s41597-023-02732-9,FALSE,
+10.1038/s41597-023-01973-y,FALSE,
+10.1038/s41597-023-02297-7,FALSE,
+10.1038/s41597-023-02277-x,FALSE,
+10.1038/s41597-024-03298-w,FALSE,
+10.1038/s41597-024-03453-3,FALSE,
+10.1038/s41597-023-02886-6,FALSE,
+10.1038/s41597-024-03601-9,FALSE,
+10.1038/s41597-025-06059-5,FALSE,
+10.1038/s41597-025-05459-x,FALSE,
+10.1038/s41597-025-05600-w,FALSE,
+10.1038/s41597-025-06468-6,FALSE,
+10.1038/s41597-025-05625-1,FALSE,
+10.1038/s41597-025-04537-4,TRUE,
+10.1038/s41597-025-05100-x,FALSE,
+10.1038/s41597-026-06698-2,FALSE,
+10.1038/s41597-026-07352-7,FALSE,