verify LFX handles against the GitHub API in merge_lfx_contributors

scripts/gc-elections/README.md

@@ -10,7 +10,7 @@ Two environment variables are used across these scripts to configure output file
 The `generate-voters-roll.py` script generates a CSV file with the GitHub logins and contributions of the eligible voters for the upcoming elections. It works in two steps:
 
 1. Queries the PostgreSQL Data Source backing `opentelemetry.devstats.cncf.io` to get the list of contributors with at least 20 contributions in the last year. It then uses the GitHub REST API to get the GitHub login (capitalized) for each user.
-2. Fetches the full contributor list from [LFX Insights](https://insights.linuxfoundation.org/project/opentelemetry/contributors) and adds any contributors that devstats missed. Only contributors with at least 50 contributions are considered. For a contributor with multiple GitHub handles, if any of them is already on the list the contributor is skipped; otherwise only the first handle is added.
+2. Fetches the full contributor list from [LFX Insights](https://insights.linuxfoundation.org/project/opentelemetry/contributors) and adds any contributors that devstats missed. Only contributors with at least 50 contributions are considered. For a contributor with multiple GitHub handles, if any of them is already on the list the contributor is skipped; otherwise the handles are resolved through the GitHub REST API and the login of the first one that resolves is added. A contributor whose handles do not resolve to any GitHub account is skipped.
 
 The merged list is written sorted case-insensitively by GitHub login.
 

scripts/gc-elections/generate-voters-roll.py

@@ -119,12 +119,28 @@ def get_lfx_contributors():
     return contributors
 
 
+# Resolve an LFX contributor to a canonical GitHub login.
+# LFX handle arrays are not validated against GitHub and are not ordered by
+# recency, so the first entry is often a retired or renamed account. Return the
+# first handle the API resolves, or None if none of them is a live account.
+def resolve_lfx_handle(handles):
+    for handle in handles:
+        handle = handle.strip()
+        if not handle:
+            continue
+        login = get_github_login(handle)
+        if login:
+            return login
+    return None
+
+
 # Merge LFX contributors into the devstats rows.
 # For each LFX contributor:
 #   - skip if their contribution count is below the threshold (50)
 #   - skip if they have no GitHub handle at all
 #   - skip if any of their handles is already present (case-insensitive)
-#   - otherwise add only the first handle in the array
+#   - skip if none of their handles resolves to a GitHub account
+#   - otherwise add the canonical login of the first handle that resolves
 def merge_lfx_contributors(rows, lfx_contributors):
     existing = {row[0].lower() for row in rows}
     added = 0
@@ -135,11 +151,16 @@ def merge_lfx_contributors(rows, lfx_contributors):
         handles = c.get("githubHandleArray") or []
         if not handles:
             continue
-        if any(h.lower() in existing for h in handles):
+        if any(h.strip().lower() in existing for h in handles):
+            continue
+        login = resolve_lfx_handle(handles)
+        if not login:
+            print(f"Skipping LFX contributor with no live GitHub account: {handles}")
+            continue
+        if login.lower() in existing:
             continue
-        primary = handles[0]
-        rows.append([primary, c["contributions"]])
-        existing.add(primary.lower())
+        rows.append([login, c["contributions"]])
+        existing.add(login.lower())
         added += 1
 
     print(f"Added {added} new contributors from LFX")