Fix vault check script, get latest metadata file.

tools/metadata/vault-check-orcid-format.r

@@ -33,41 +33,52 @@ def main(rule_args, callback, rei):
     schemas_to_be_checked = ['core-1', 'default-1', 'default-2', 'hptlab-1', 'teclab-1', 'dag-0', 'vollmer-0']
 
     for schema in schemas_to_be_checked:
+
         callback.writeLine("stdout", "")
         callback.writeLine("stdout", "SCHEMA: {}".format(schema))
-        packages = genquery.row_iterator(
-            "COLL_NAME, META_COLL_ATTR_VALUE, DATA_NAME",
+
+        data_packages = genquery.row_iterator(
+            "COLL_NAME",
             "META_COLL_ATTR_NAME = 'href' AND META_COLL_ATTR_VALUE like '%/{}/metadata.json' "
             "AND COLL_NAME not like '%/original' AND COLL_NAME like '/%/home/vault-%' "
             "AND DATA_NAME like 'yoda-metadata%.json'".format(schema),
             genquery.AS_TUPLE,
             callback)
 
-        for (coll, href, data) in packages:
-            # New package so when reporting orcid errors must be mentioned only once
+        metadata_files = genquery.row_iterator(
+            "COLL_NAME, ORDER_DESC(DATA_NAME)",
+            "DATA_NAME like 'yoda-metadata[%].json' "
+            "AND COLL_NAME not like '%/original'",
+            genquery.AS_TUPLE,
+            callback)
+
+        for coll in data_packages:
+            json_file = None
+
+            for (coll_, metadata_file) in metadata_files:
+                if coll == coll_:
+                    json_file = metadata_file
+                    break
+
             wrote_package_line = False
 
-            md = get_metadata_as_dict(callback, coll + '/' + data)
+            md = get_metadata_as_dict(callback, coll + '/' + json_file)
 
             for pi_holder in ['Creator', 'Contributor']:
                 if pi_holder in md:
                     for holder in md[pi_holder]:
                         for pi in holder.get('Person_Identifier', []):
                             if pi.get('Name_Identifier_Scheme', None) == 'ORCID':
-                                if not re.search("^(https://orcid.org/)[0-9]{4}-[0-9]{4}-[0-9]{4}-[0-9]{3}[0-9xX]$", pi['Name_Identifier']):
+                                if not re.search("^(https://orcid.org/)[0-9]{4}-[0-9]{4}-[0-9]{4}-[0-9]{3}[0-9X]$", pi['Name_Identifier']):
                                     if not wrote_package_line:
                                         # Only write this line once
+                                        callback.writeLine("stdout", '----------------------------------')
                                         callback.writeLine("stdout", "Package: {}".format(coll))
                                         wrote_package_line = True
 
-                                    try:
-                                        callback.writeLine("stdout", "{}: {} {}".format(pi_holder, holder['Name']['Given_Name'], holder['Name']['Family_Name']))
-                                    except TypeError:
-                                        callback.writeLine("stdout", "{}: {}".format(pi_holder, holder['Name'].encode('utf-8')))
-                                    except UnicodeEncodeError:
-                                        callback.writeLine("stdout", "Warning: could not process " + coll + " due to encoding error.")
-                                    callback.writeLine("stdout", "      Erroneous ORCID: \"{}\"".format(pi['Name_Identifier']))
-                                    callback.writeLine("stdout", '----------------------------------')
+                                    callback.writeLine("stdout", "Invalid ORCID: \"{}\"".format(pi['Name_Identifier']))
+            if wrote_package_line:
+                callback.writeLine("stdout", '----------------------------------')
 
 
 INPUT null