initial commit

peek_live.py

@@ -0,0 +1,15 @@
+import duckdb
+import json
+
+def peek_data():
+    conn = duckdb.connect()
+    conn.execute("INSTALL httpfs; LOAD httpfs;")
+    conn.execute("SET s3_region='sgp1';")
+    conn.execute("SET s3_endpoint='sgp1.digitaloceanspaces.com';")
+    conn.execute("SET s3_url_style='path';")
+    
+    row = conn.execute("SELECT * FROM read_parquet('s3://nearle/parquet/deliveries/*.parquet') LIMIT 1").df().to_dict('records')[0]
+    print(json.dumps(row, indent=2, default=str))
+
+if __name__ == "__main__":
+    peek_data()