tasks/offline-compute/PySpark/pyspark_010/init/init_db.py

#!/usr/bin/env python3
"""pyspark_010 database initialization: create tables + load seed data"""
from pyspark.sql import SparkSession

spark = SparkSession.builder \
    .appName('dataclaw_eval_init_pyspark_010') \
    .enableHiveSupport() \
    .config('spark.sql.warehouse.dir', '/tmp/hive_warehouse') \
    .getOrCreate()

spark.sql('CREATE DATABASE IF NOT EXISTS internal_platform_db')

# Create input table
spark.sql('''
CREATE TABLE IF NOT EXISTS internal_platform_db.ods_pmtool_v_workitem_changes_df_pyspark_245 (
  `id` STRING COMMENT '-',
  `workspace_id` BIGINT COMMENT '-',
  `workitem_id` BIGINT COMMENT '-',
  `workitem_type_id` BIGINT COMMENT '-',
  `creator` STRING COMMENT '-',
  `created` STRING COMMENT '-',
  `change_summary` STRING COMMENT '-',
  `comment` STRING COMMENT '-',
  `changes` STRING COMMENT '-',
  `entity_type` STRING COMMENT '-',
  `change_type` STRING COMMENT '-',
  `_srcinstanceid` STRING COMMENT '-',
  `_srcdatabasename` STRING COMMENT '-'
)
PARTITIONED BY (`databus_imp_date` STRING COMMENT '-')
STORED AS ORC
''')

# Insert seed data - inline VALUES
spark.sql('''
INSERT INTO TABLE internal_platform_db.ods_pmtool_v_workitem_changes_df_pyspark_245 PARTITION (databus_imp_date='20260513')
VALUES
('1001', 100, 2001, 3001, 'alice', '2026-05-13 10:00:00', 'status changed', 'comment1', '{"f":"status","o":"open","n":"done"}', 'story', 'update', 'src1', 'db1'),
('1001', 100, 2001, 3001, 'alice', '2026-05-13 10:05:00', 'status changed again', 'comment1b', '{"f":"status","o":"done","n":"closed"}', 'story', 'update', 'src1', 'db1'),
('1002', 101, 2002, 3002, 'bob', '2026-05-13 11:00:00', 'priority changed', 'comment2', '{"f":"priority","o":"low","n":"high"}', 'bug', 'modify', 'src2', 'db2'),
('1003', 102, 2003, 3003, 'carol', '2026-05-13 12:00:00', 'owner changed', 'comment3', '{"f":"owner","o":"carol","n":"dave"}', 'task', 'reassign', 'src3', 'db3'),
('1004', 103, 2004, 3004, 'dave', '2026-05-13 13:00:00', 'field added', 'comment4', '{"f":"desc","o":"","n":"new desc"}', 'story', 'create', 'src4', 'db4'),
('1004', 103, 2004, 3004, 'dave', '2026-05-13 13:30:00', 'field modified', 'comment4b', '{"f":"desc","o":"new desc","n":"updated desc"}', 'story', 'update', 'src4', 'db4')
''')

print('Database initialization complete')
spark.stop()