Forgotten mets (#2647)

* add script to push METS records mentioned in a CSV

* Apply auto-formatting rules

* add source id script

---------

Co-authored-by: Buildkite on behalf of Wellcome Collection <[email protected]>

mets_adapter/populate_mets/populate_mets_from_csv.py

@@ -0,0 +1,25 @@
+import csv
+import sys
+from populate_mets import specific
+
+
+# Push each Born Digital bag mentioned in a CSV through the adapter.
+# The CSV should have a row corresponding to each Bag (Item) thus:
+# Item, MY/BAG/1/2/3
+# Item, MY/BAG/4/5/6
+# More columns can be added, and rows starting with other values are ignored.
+
+
+def main(csv_path):
+    with open(csv_path, "r") as csv_file:
+        specific(extract_bag_ids(csv.reader(csv_file)))
+
+
+def extract_bag_ids(csv_reader):
+    for row in csv_reader:
+        if row[0] == "Item":
+            yield f"born-digital/{row[1]}"
+
+
+if __name__ == "__main__":
+    main(sys.argv[1])

scripts/es_source_identifier.py

@@ -0,0 +1,12 @@
+# Given the identifier and source system name, return a GET request that can be
+# used in the Elasticsearch console to fetch the record from the works-source index.
+
+# It's basically just URL Encoding, but it can be a bit of a faff compared to ids in other indices
+import sys
+from urllib.parse import quote_plus
+
+date = sys.argv[1]
+source_system = sys.argv[2]
+work_id = sys.argv[3]
+doc_id = quote_plus(f"Work[{source_system}/{work_id.lower()}]")
+print(f"GET works-source-{date}/_doc/{doc_id}")