Comprehensive Overview of SQL and Apache Flink SQL Features with Practical Code Examples

SQL (Structured Query Language) originated in the 1970s at IBM and has evolved through multiple ANSI/ISO standards, becoming the dominant language for relational data manipulation and retrieval.

The article explains core SQL operations such as SELECT with scalar functions and aliases, WHERE filtering with logical operators, GROUP BY aggregation, set operations UNION ALL and UNION, and various JOIN types (INNER, LEFT, RIGHT, FULL), illustrating each with clear examples and result tables.

It then introduces Apache Flink's support for both batch and streaming SQL, covering Over Windows (bounded ROWS and RANGE) and Group Windows (TUMBLE, HOP, SESSION) with syntax, semantics, and sample queries on test datasets such as item_tab and pageAccess_tab. The article also describes user‑defined functions (UDF, UDTF, UDAF) and shows how to implement them in Scala.

Finally, a complete end‑to‑end Flink SQL job is presented: a custom SourceFunction emitting event‑time data with watermarks, a StreamTableSource definition, a CSV sink, and a main program that registers the source and sink, executes a TUMBLE window query to compute page‑view counts per region, and writes the results to a file.

All code snippets are preserved in pre blocks to maintain their original formatting.