Download new taxdmp.zip as part of setup_gb

fernphy · Nov 1, 2024 · 097d85c · 097d85c
1 parent 20f6874
commit 097d85c
Show file tree

Hide file tree

Showing 2 changed files with 17 additions and 0 deletions.
diff --git a/R/setup_gb.R b/R/setup_gb.R
@@ -147,6 +147,22 @@ archive::archive_write_files(
   filter = "gzip"
 )
 
+# Also download new taxdump file,
+# store previous one as 'bak'
+# This needs to be done *after* GenBank data are downloaded so that all
+# GenBank taxonomic data is up-to-date
+taxdmp_path <- "_targets/user/data_raw/taxdmp.zip"
+
+if (file_exists(taxdmp_path)) {
+  file_move(
+    taxdmp_path,
+    "_targets/user/data_raw/taxdmp.zip.bak"
+  )
+}
+
+download.file(
+  "https://ftp.ncbi.nlm.nih.gov/pub/taxonomy/taxdmp.zip", taxdmp_path)
+
 # Cleanup ----
 
 # Download done, so delete "running" file

diff --git a/_targets.R b/_targets.R
@@ -370,6 +370,7 @@ tar_plan(
   ),
   # Load taxdump file downloaded from
   # https://ftp.ncbi.nlm.nih.gov/pub/taxonomy/taxdmp.zip
+  # This is automatically downloaded by setup_gb.R for each new GenBank release
   tar_file(
     taxdump_zip_file,
     path(data_raw, "taxdmp.zip")