refine

msr.py

@@ -433,7 +433,11 @@ def fetch_all_pr_metadata_single_query(conn, identifiers, start_date, end_date):
     # Build identifier list for IN clause
     identifier_list = ', '.join([f"'{id}'" for id in identifiers])
 
-    # Build comprehensive query with CTEs using parameterized file lists (JSON.gz format)
+    # Convert file patterns to SQL array format for direct interpolation
+    review_patterns_sql = str(review_patterns).replace("'", "'")
+    status_patterns_sql = str(status_patterns).replace("'", "'")
+
+    # Build comprehensive query with CTEs using direct SQL array format (JSON.gz format)
     # Optimized: Single file scan + ROW_NUMBER() deduplication (no DISTINCT)
     query = f"""
     WITH all_review_events AS (
@@ -444,7 +448,7 @@ def fetch_all_pr_metadata_single_query(conn, identifiers, start_date, end_date):
         TRY_CAST(json_extract_string(repo, '$.name') AS VARCHAR) as repo_name,
         payload,
         created_at
-      FROM read_json($review_patterns, union_by_name=true, filename=true, compression='gzip', format='newline_delimited', ignore_errors=true, maximum_object_size=2147483648)
+      FROM read_json({review_patterns_sql}, union_by_name=true, filename=true, compression='gzip', format='newline_delimited', ignore_errors=true, maximum_object_size=2147483648)
       WHERE
         TRY_CAST(type AS VARCHAR) IN ('PullRequestReviewEvent', 'IssueCommentEvent', 'PullRequestReviewCommentEvent')
         AND TRY_CAST(json_extract_string(actor, '$.login') AS VARCHAR) IN ({identifier_list})
@@ -490,9 +494,9 @@ def fetch_all_pr_metadata_single_query(conn, identifiers, start_date, end_date):
         TRY_CAST(json_extract_string(payload, '$.pull_request.closed_at') AS VARCHAR) as closed_at,
         created_at,
         ROW_NUMBER() OVER (PARTITION BY json_extract_string(payload, '$.pull_request.html_url') ORDER BY created_at DESC) as rn
-      FROM read_json($status_patterns, union_by_name=true, filename=true, compression='gzip', format='newline_delimited', ignore_errors=true, maximum_object_size=2147483648)
+      FROM read_json({status_patterns_sql}, union_by_name=true, filename=true, compression='gzip', format='newline_delimited', ignore_errors=true, maximum_object_size=2147483648)
       WHERE
-        TRY_CAST(type AS VARCHAR) = 'PullRequestEvent'
+        type = 'PullRequestEvent'
         AND TRY_CAST(json_extract_string(payload, '$.action') AS VARCHAR) = 'closed'
         AND json_extract_string(payload, '$.pull_request.html_url') IS NOT NULL
         AND json_extract_string(payload, '$.pull_request.html_url') IN (
@@ -545,8 +549,8 @@ def fetch_all_pr_metadata_single_query(conn, identifiers, start_date, end_date):
                 WHERE reviewer IN ({identifier_list})
             """).fetchall()
         else:
-            # Execute query with parameters
-            results = conn.execute(query, {'review_patterns': review_patterns, 'status_patterns': status_patterns}).fetchall()
+            # Execute query using f-string interpolation
+            results = conn.execute(query).fetchall()
 
             # Cache the complete results for all future queries in this date range
             if len(results) > 0: