tasks/online-compute/FlinkSQL/flinksql_017_en/task.md

metadata

id: online-compute_FlinkSQL_flinksql_017
name: Row-Level TopN (Windowless)
category: online-compute/FlinkSQL
timeout_seconds: 600
modality: pure-text
engine: flink-sql

Prompt

I need you to write a Flink SQL that uses the built-in datagen connector to simulate a spending stream, retrieves the top 3 records by spending amount grouped by country_id without any window, and outputs the results directly to the console.

Business Background and Objective: Generate user spending records using a datagen built-in table at a rate of 1000 rows per second. Fields include country_id (country ID, 1–30), user_id (user ID, 1–100,000), and cost_money (spending amount, 0–10,000). No event time or watermark is required, since no window operations are involved.

Use the ROW_NUMBER() window function, partitioned by country_id and sorted by cost_money in descending order, keeping only the top 3 records per partition (row_num <= 3), and output the results directly to the console table. This is a continuous TopN query where each country's top 3 dynamically updates as new data flows in.

Source Table Definition:

• spend_source (datagen connector):
  • country_id INT: Country ID, randomly generated in the range 1–30
  • user_id INT: User ID, randomly generated in the range 1–100,000
  • cost_money DOUBLE: Spending amount, randomly generated in the range 0–10,000
  • Generation rate: rows-per-second = 1000

Output Table Definition:

• console_output (print connector):
  • country_id INT: Country ID
  • user_id INT: User ID
  • cost_money DOUBLE: Spending amount
  • row_num BIGINT: Row number

TopN Logic:

• Use ROW_NUMBER() OVER (PARTITION BY country_id ORDER BY cost_money DESC) to generate row numbers
• Apply WHERE row_num <= 3 to keep the top 3 per country
• No window operation is required; no event time or watermark needs to be defined

Output Requirements:

• Use INSERT INTO console_output to output the results.
• Output field order: country_id, user_id, cost_money, row_num