tasks/offline-compute/PrestoSQL/prestosql_008_en/task.md

metadata

id: offline-compute_PrestoSQL_prestosql_008
name: Report WUID Detail Filtering
category: offline-compute/PrestoSQL
timeout_seconds: 900
modality: pure-text
engine: prestosql

Prompt

Task Objective: From the ad report WUID detail wide table, filter target ad delivery records based on multi-dimensional filter conditions, extract key fields, and write the results to the output table.

Input Table:

• internal_platform_db.t_report_wuid_detail_d_prestosql_008 (report WUID detail table)
  • wuid STRING — user identifier
  • product_id STRING — product ID
  • adgroup_id BIGINT — ad group ID
  • action_time BIGINT — action time
  • crm_id BIGINT — CRM ID
  • creative_id BIGINT — creative ID
  • ocpx_conversion_cnt BIGINT — oCpx conversion count
  • optimization_goal INT — optimization goal
  • second_optimization_goal INT — second-level optimization goal
  • roi_goal INT — ROI goal
  • product_type INT — product type
  • site_set INT — site set
  • crm_advertiser_industry_id BIGINT — CRM advertiser industry ID
  • label INT — label
  • partition_time BIGINT — partition time
  • ds BIGINT — partition field

Filter Conditions:

• partition_time = 20260608
• ocpx_conversion_cnt = 0
• optimization_goal IN (105, 603)
• second_optimization_goal = 0
• roi_goal = 7
• product_type = 46
• site_set = 21
• product_id IS NOT NULL
• crm_advertiser_industry_id != 0
• creative_id != 0

Output Requirements:

• Target table: internal_platform_db.sdk_sample_pay_neg_mini_cand_prestosql_008
• Output fields and order: wuid STRING, product_id STRING, adgroup_id BIGINT, action_time BIGINT, crm_id BIGINT, creative_id BIGINT, label INT, product_type INT, optimization_goal INT, roi_goal INT, partition_time BIGINT
• If the target table does not exist, first create the table, then write the data
• Use Presto/Trino SQL syntax; do not use Hive/Spark SQL dialects

Environment and Execution Notes:

• Presto is running, connected via the Hive catalog
• Execute SQL: presto-cli --catalog hive --schema internal_platform_db -f /tmp_workspace/result.sql
• After writing result.sql, you must execute it yourself to verify that it runs successfully and produces correct data