Uploaded byMariaDB plc
Introducing the R2DBC async Java connector
The document introduces the R2DBC (Reactive Relational Database Connectivity) connector for reactive programming with MariaDB, emphasizing its non-blocking nature and the importance of reactive programming in developing high-efficiency applications. It outlines the principles and goals of R2DBC, including seamless integration into reactive JVM platforms and vendor-neutral access to features, and explains the service-provider interface (SPI) designed to overcome limitations of JDBC. The document also provides a brief overview of setting up connections and executing SQL commands using R2DBC.
Introducing the R2DBC async Java connector
- 1. Introducing the R2DBC connector ReactiveProgramming with MariaDB
- 2. Who is thisguy? Rob Hedgpeth Developer Evangelist robh@mariadb.com @probablyrealrob rhedgpeth
- 3. Agenda ● Reactive Programming ○What is it? ○ Why is it important? ● R2DBC ○ What is it? ○ How can you use it? ● Demo Non-blocking Database Euphoria
- 4.
- 5.
- 6. @mariadb reactive programming /rēˈaktiv/ /ˈprōˌgramiNG/ Adeclarative programming paradigm concerned with data streams and the propagation of change.
- 7. Data Streams A B Streamof Data C @mariadb
- 8. What is astream? @mariadb Variables changes in data Starting event initiates stream of data Error disruption in data Completion all data is processed Time
- 9. Reactive Streams @mariadb ● Aninitiative started in 2013 by several companies (e.g. Netflix, Pivotal, Lightbend and many others) ● A specification ○ API types ○ Technology Compatibility Kit (TCK) ● The goal was to provide a standard for asynchronous stream processing with non-blocking back pressure.
- 10.
- 11. Back Pressure @mariadb A B Pushstream of N data items Request N data items Publisher Subscriber
- 12. Back Pressure @mariadb Publisher Subscriber Request(10) Send Request (1) Send Send Request (3)
- 13. Reactive Streams API @mariadb PublisherSubscriber Subscription onSubscribe request(n) cancel() onNext* (onError | onComplete)
- 14. Reactive Streams API @mariadb publicinterface Publisher<T> { public void subscribe(Subscriber<? super T> s); } public interface Subscriber<T> { public void onSubscribe(Subscription s); public void onNext(T t); public void onError(Throwable t); public void onComplete(); } public interface Subscription { public void request(long n); public void cancel(); } public interface Processor<T, R> extends Subscriber<T>, Publisher<R> { }
- 15. Why is itimportant?
- 16. Reactive Programming @mariadb ● Thenext frontier in Java (and the JVM) for high-efficiency applications ● Fundamentally non-blocking ○ Often paired with asynchronous behaviors ○ ...but is completely agnostic to sync/async ● The key takeaway - back pressure Async != Reactive
- 17.
- 18.
- 19.
- 20. Reactive Relational DatabaseConnectivity R2 DB C
- 21. Design principles* @mariadb 1. Becompletely non-blocking, all the way to the database 2. Utilize Reactive Streams Types and Patterns 3. Provide a minimal set of operations that are implementation specific 4. Enable “humane” APIs to be built on top of the driver * Ben Hale, Creator of R2DBC
- 22. Goals @mariadb ✓ Fit seamlesslyinto Reactive JVM platforms ✓ Offer vendor-neutral access to standard features ✓ Embrace vendor-specific features ✓ Keep the focus on SQL ✓ Keep it simple ○ Provide a foundation for tools and higher-level API’s ○ Compliance should be unambiguous and easy to identify
- 23.
- 24. Why a service-providerinterface (SPI)? @mariadb ● One of JDBC’s biggest failings was that the same API had to serve as both humane API for users as well as an inhumane API for alternative clients like JPA, Jdbi, etc. ○ API that users didn’t like using ○ Driver authors didn’t like implementing ● It also lead to drivers duplicating effort ○ ? binding ○ URL parsing
- 25. URL Parsing @mariadb r2dbc:a-driver:pipes://localhost:3306/my_database?locale=en_US scheme driverprotocol authority path query
- 26. R2DBC SPI Compliance @mariadb io.r2dbc.spi.ConnectionFactory io.r2dbc.spi.ConnectionFactoryMetadata io.r2dbc.spi.ConnectionFactoryProvider io.r2dbc.spi.Result io.r2dbc.spi.Row io.r2dbc.spi.RowMetadata io.r2dbc.spi.Batch FullySupport Partially Support io.r2dbc.spi.Connection io.r2dbc.spi.Statement io.r2dbc.spi.ColumnMetaData https://github.com/r2dbc/r2dbc-spi io.r2dbc.spi.ConnectionFactory io.r2dbc.spi.ConnectionFactoryMetadata io.r2dbc.spi.ConnectionFactoryProvider io.r2dbc.spi.Result io.r2dbc.spi.Row io.r2dbc.spi.RowMetadata io.r2dbc.spi.Batch io.r2dbc.spi.Connection io.r2dbc.spi.Statement io.r2dbc.spi.ColumnMetaData
- 27. SPI - ConnectionFactory @mariadb package io.r2dbc.spi; import org.reactivestreams.Publisher; public interface ConnectionFactory { Publisher<? extends Connection> create(); ConnectionFactoryMetadata getMetadata(); }
- 28. SPI - Connection @mariadb packageio.r2dbc.spi; import org.reactivestreams.Publisher; public interface Connection { Publisher<Void> beginTransaction(); Publisher<Void> close(); Publisher<Void> commitTransaction(); Batch createBatch(); Statement createStatement(String sql); ConnectionMetadata getMetadata(); ... } Lots of other methods
- 29. SPI - Statement @mariadb packageio.r2dbc.spi; import org.reactivestreams.Publisher; public interface Statement { Publisher<? extends Result> execute(); Statement add(); Statement bind(int index, Object value); Statement bind(String name, Object value); Statement bindNull(int index, Class<?> type); Statement bindNull(String name, Class<?> type); }
- 30. SPI - Result @mariadb packageio.r2dbc.spi; import org.reactivestreams.Publisher; import java.util.function.BiFunction; public interface Result { Publisher<Integer> getRowsUpdated(); <T> Publisher<T> map(BiFunction<Row, RowMetadata, ? extends T> mappingFunction); }
- 31. SPI - Row @mariadb packageio.r2dbc.spi; import java.util.function.BiFunction; public interface Row { @Nullable <T> T get(int index, Class<T> type); @Nullable <T> T get(String name, Class<T> type); ... } A couple default methods
- 32. How can youuse it?
- 33. Establishing a connection @mariadb //Configuration the Connection config = MariadbConnectionConfiguration.builder() .host(props.getProperty("host")) .port(Integer.parseInt(props.getProperty("port"))) .username(props.getProperty("username")) .password(props.getProperty("password")) .build();
- 34. Establishing a connection @mariadb //Instantiate Connection Factory connFactory = new MariadbConnectionFactory(config); // Instantiate Connection conn = connFactory.create().block();
- 35. SELECT @mariadb Statement select =conn.createStatement("select * from todo.tasks"); Flux<Task> = Flux.from(select.execute()) .flatMap( res -> res.map( (row, metadata) -> { int id = row.get(0, Integer.class); String description = row.get(1, String.class); Boolean completed = row.get(2, Boolean.class); return new Task(id,description,completed); }) );
- 36. INSERT (Prepared) @mariadb Statement insert= conn.createStatement("insert into tasks (description) values (?)"); insert.bind(0, task.getDescription()); Mono.from(insert.execute()).subscribe();
- 37.
- 38. Demo A hands onlook at the new MariaDB R2DBC Connector
- 39.
- 40. Roadmap @mariadb ● Performance ● Fastbatch using MariaDB bulk (processing) ● GeoJSON data type ● Pluggable types for MariaDB 10.5 (JSON, INET4, INET6, BOOLEAN, ...)
- 41. https://mariadb.com/products/skysql/get-started/ $500 credit toget started @mariadb
- 42. developers@mariadb.com @mariadb mariadb-corporation Thankyou! https://github.com/mariadb-corporation/mariadb-connector-r2dbc https://github.com/mariadb-corporation/developer-examples Open Source Developer Examples