Almost working name script

Author: tmorrell

caltechdata_api/customize_schema.py

@@ -5,6 +5,21 @@
 from datetime import date
 import yaml
 from pathlib import Path
+import requests
+
+
+def grid_to_ror(grid):
+    # Temporary until InvenioRDM stops spitting out GRIDS
+    # We manually handle some incorrect/redundant GRID Ids
+    if grid == "grid.451078.f":
+        ror = "https://ror.org/00hm6j694"
+    elif grid == 'grid.5805.8':
+        ror = "https://ror.org/02en5vm52"
+    else:
+        url = f"https://api.ror.org/organizations?query.advanced=external_ids.GRID.all:{grid}"
+        results = requests.get(url)
+        ror = results.json()["items"][0]["id"]
+    return ror
 
 
 def get_vocabularies():
@@ -320,6 +335,11 @@ def customize_schema_rdm(json_record):
                         ror = ror.split("ror.org/")[1]
                     funder["id"] = ror
                     fund.pop("funderIdentifierType")
+                elif fund["funderIdentifierType"] == "GRID":
+                    #We need this temporarily to round-trip data
+                    ror = grid_to_ror(fund.pop("funderIdentifier"))
+                    funder["id"] = ror
+                    fund.pop("funderIdentifierType")
                 else:
                     print(f'Unknown Type mapping {fund["funderIdentifierType"]}')
             if "awardTitle" in fund:

fix_names.py

@@ -4,8 +4,7 @@
 from caltechdata_api import caltechdata_edit
 
 
-def fix_name(metadata):
-    fixed = False
+def fix_name(metadata, fixed):
     for name in metadata:
         if name["nameType"] == "Personal":
             if "givenName" not in name:
@@ -15,37 +14,40 @@ def fix_name(metadata):
     return metadata, fixed
 
 
-url = "https://data.caltech.edu/api/records"
+url = 'https://data.caltech.edu/api/records?q=-metadata.related_identifiers.identifier%3A"10.25989%2Fes8t-kswe"'
 
 headers = {
     "accept": "application/vnd.datacite.datacite+json",
 }
 
-response = requests.get(f"{url}?search_type=scan&scroll=5m")
+response = requests.get(f"{url}&search_type=scan&scroll=5m")
 
 total = response.json()["hits"]["total"]
 pages = math.ceil(int(total) / 1000)
 hits = []  # [{'id':'a7f64-a8k10'}]
 print(total)
-for c in progressbar(range(1, 2)):  # , pages + 1)):
-    chunkurl = f"{url}?&sort=newest&size=1000&page={c}"
+for c in progressbar(range(1, pages + 1)):
+    chunkurl = f"{url}&sort=newest&size=1000&page={c}"
     response = requests.get(chunkurl)
     response = response.json()
-
     hits += response["hits"]["hits"]
 
+
+url = 'https://data.caltech.edu/api/records'
+
 for h in progressbar(hits):
     idv = str(h["id"])
+    
     response = requests.get(f"{url}/{idv}", headers=headers)
     if response.status_code != 200:
         print(response.text)
         exit()
     else:
+        fixed = False
         metadata = response.json()
-        metadata["creators"], fixed = fix_name(metadata["creators"])
+        metadata["creators"], fixed = fix_name(metadata["creators"], fixed)
         if "contributors" in metadata:
-            metadata["contributors"] = fix_name(metadata["contributors"])
+            metadata["contributors"], fixed = fix_name(metadata["contributors"], fixed)
         if fixed:
             print(idv)
             caltechdata_edit(idv, metadata, production=True, publish=True)
-            exit()