slinkydeveloper commented on a change in pull request #18134: URL: https://github.com/apache/flink/pull/18134#discussion_r773735973
########## File path: flink-table/README.md ########## @@ -0,0 +1,62 @@ +# Table API & SQL + +Apache Flink features two relational APIs - the Table API and SQL - for unified stream and batch processing. +The Table API is a language-integrated query API for Java, Scala, and Python that allows the composition of queries from relational operators such as selection, filter, and join in a very intuitive way. + +For more details on how to use it, check out the [documentation](https://nightlies.apache.org/flink/flink-docs-master/docs/dev/table/overview/). + +## Modules + +### Common + +* `flink-table-common`: + * Type system definition and UDF stack + * Internal data types definitions + * `Factory` definitions for catalogs, formats, connectors + * Other core APIs such as `Schema` + * Utilities to deal with type system, internal data types and printing + * When implementing a format, you usually need to depend only on this module + +### API + +* `flink-table-api-java`: + * Java APIs for Table API and SQL + * Package `org.apache.flink.table.delegation`, which serves as entrypoint for all planner capabilities +* `flink-table-api-scala`: Scala APIs for Table API and SQL +* `flink-table-api-bridge-base`: Base classes for APIs to bridge between Table API and DataStream API +* `flink-table-api-java-bridge`: + * Java APIs to bridge between Table API and DataStream API + * When implementing a connector, you usually need to depend only on this module, in order to bridge your connector implementation developed with DataStream to Table API +* `flink-table-api-scala-bridge`: Scala APIs to bridge between Table API and DataStream API +* `flink-table-api-uber`: Uber JAR bundling `flink-table-common` and all the Java API modules, including 3rd party dependencies. + +### Runtime + +* `flink-table-code-splitter`: Tool to split generated Java code so that each method does not exceed the limit of 64KB. +* `flink-table-runtime`: + * Operator implementations + * Built-in functions implementations + * Type system implementation, including readers/writers, converters and utilities + * Raw format + * The produced jar includes all the classes from this module and `flink-table-code-splitter`, including 3rd party dependencies + +### Parser and planner + +* `flink-sql-parser`: Default ANSI SQL parser implementation +* `flink-sql-parser-hive`: Hive SQL dialect parser implementation +* `flink-table-planner`: + * AST and Semantic tree + * SQL validator + * Planner and rules implementation + * Two jars are produced: one doesn't have any classifier and bundles all the classes from this module together with the two parsers, including 3rd party dependencies, while the other jar, classified as `loader-bundle`, extends the first jar including scala dependencies. +* `flink-table-planner-loader`: Loader for `flink-table-planner` that loads the planner in a separate classpath, isolating the Scala version used to compile the planner. + +### SQL client + +* `flink-sql-client`: CLI tool to submit queries to a Flink cluster + +### Notes + +No module except `flink-table-planner` should depend on `flink-table-runtime` in production classpath, Review comment: For testing you need to depend on table-planner-loader, not on table-planner. I added a sentence about testing. -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: issues-unsubscr...@flink.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org
