v2.5.3

Changelog

• (bugfix) Fix signed error when type is nullable (#305) Thanks @liuneng1994

v2.5.2

Changelog

• (bugfix) Fix ResultSetMetaData.isSigned (#281 #283)
• (bugfix) Fix LoggerFactory#getLogger (#287)
• (enhance) Set ClickHouseSQLException errorCode correctly (#290) Thanks @baibaichen
• (feature) Rework stuffs related to time api (#288) Thanks @sbouchex

v2.5.1

Changelog

• (bugfix) Fix Negative Decimal serde (#279)

v2.5.0

Highlight

• The project static website is available at Home Page
• The China mirror repo is available at Gitee
• Implemented BalancedClickhouseDataSource to support connect to clickhouse cluster
• Now the JDBC driver works with DataGrip and DBeaver
• Introduced a log facade which prefer to bind slf4j if it exists in classpath, otherwise fallback to JDK Logger, and added logs for tracing, debugging and warning

This release contains lots of internal refactors, and currently, this driver guarantee none stable API except JDBC API itself, please be careful if you use the internal API which not belongs to JDBC standard interfaces.

Changelog

• (feature) Implement BalancedClickhouseDataSource to support multi instances (#205)
• (bugfix) Fix invalid package clickhouse-integration-spark_2.12 (#206)
• (feature) Implement ClickHouseConnection catalog and schema method (#208)
• (enhance) Enhance Number process
• (build) Disable deploy examples module
• (test) Integration tests with connection pool (#212)
• (feature) Support charset parameter (#213)
• (feature) Implement ClickHousePreparedStatment #setBytes and #getBytes (#213)
• (enhance) Trim space before parsing SQL
• (feature) Support tcp_keep_alive parameter
• (bugfix) Fix DateTime64 nanos and scale (#217)
• (bugfix) Reset connection STATE when recreate connection (#221)
• (bugfix) Skip remaining response when ping (#224)
• (refactor) Decouple RequestOrResponse and rename PhysicalConnection to NativeClient (#225)
• (feature) Introduce log classes to avoid hard dependence on slf4j (#231)
• (build) Add debug Maven profile (#235)
• (enhance) Add LRU cache for lexer IDataType (#236)
• (feature) Refactor and implement ClickHouseDatabaseMetadata (#233)
• (feature) Support type alias (#240)
• (bugfix) Explicit upcast ByteBuffer to Buffer (#243)
• (deps) Bump ini from 1.3.5 to 1.3.8 (#250)
• (feature) Support Decimal256 && Decimal128 (#239)
• (refactor) Refactor and fixup (#246)
• (deps) bump alibaba druid 1.2.4
• (deps) bump slf4j 1.7.30
• (deps) bump mockito 3.6.28
• (deps) bump jmh 1.27
• (bugfix) Cast correctly numbers to get a boolean (#256)
• (bugfix) Added check against null data on ClickHouseResultSet#getXXX (#258) (#259)
• (feature) Handle Nothing datatype (#261)
• (bugfix) ClickHouseConnection#connect should return null for unacceptable url (#267)
• (enhance) BalancedClickhouseDataSource support optional database url
• (test) Added user & password for ClickHouse unit tests (#264)
• (feature) Support arbitrary settings (#268)
• (refactor) Rewrite StringView implements CharSequence (#270)
• (workflow) Auto report benchmark report.

v2.4.4

Changelog

• (backport) cast correctly numbers to get a boolean (#256)
• (backport) enhance null check on ClickHouseResultSet#getXXX (#259)

v2.4.3

Changelog

• (backport) validate connection by ping instead of SELECT 1
• (backport) explicit upcast ByteBuffer to Buffer (#243)
• (backport) fix Connection#getCatalog (#249)
• (backport) fix ClickHouseStatement#getUpdateCount
• (backport) fix ClickHouseStatement#setMaxRows
• (backport) ClickHouseResultSet#getString support all types
• (backport) implement ClickHouseResultSet#getBoolean
• (backport) implement ClickHouseResultSet#isBeforeFirst, #isAfterLast, #isFirst etc.
• ClickHouseConnection#getMetaData return null instead of throw exception

v2.4.2

Changelog

• Trim space before parsing SQL
• Fix DateTime64 nanos and scale (#217)
• Reset connection STATE when recreate connection (#221)
• Skip remaining responses when consume PingResponse (#224)
• Minor tunes

v2.4.1

Changelog

• Fix invalid package clickhouse-integration-spark_2.12 (#206)

v2.4.0

Highlight

• Since this release v2.4.0, we switch to semantic versioning, and the bug fix version would be available quickly.

• We introduce a new module clickhouse-integration-spark base on Spark Jdbc DataSource API for integrating with Spark, check detail at README

• In the past few weeks, we were more focus on code quality, fixed all LGTM alerts and got A+ score.

Changelog

• Implement clickhouse-integration-spark base on Spark Jdbc DataSource API (#170 #184)
• Batch insert support nullable types in arrays and nested types (#144 #194)
• Support Boolean (#196)
• Fix timezone issue (#195)
• Fix potential concurrence issue (#191)
• Code refactor, fixed all LGTM alerts and got A+ score.
• Refactor ClickHouseStatement inheritance tree (#201)
• Migrate to mockito3 (#175)
• Migrate to org.lz4:lz4-java (#174)
• Migrate to Junit5 (#171)
• Bump yandex/clickhouse-jdbc 0.2.4
• Bump jmh 1.26
• Enable checkstyle (#172)
• Enable code coverage (#190)
• Improve Docs and Readme
• Add missing license headers

v2.3

Highlight

• Provide shaded version clickhouse-native-jdbc-shaded since this release, see detail at README.md

Changelog

• Fixed SQLFeatureNotSupportedException in several scenarios (#142 thanks @dcastanier) (thanks @sundy-li)
• Fixed return value for executeBatch in PreparedInsertStatement (#145 thanks @tauitdnmd)
• Fixed jdbc url parse (#148 thanks @tauitdnmd)
• Support zoned datetime DataType, i.e. DateTime(Asia/Shanghai) (#158 thanks @sundy-li)
• Refactor project to multi modules (#156 thanks @pan3793)
• Remove joda-time, migrate joda-time and legacy java.util.Date to java.time (#164 thanks @pan3793)
• Integration Test against Java 8, Java 11 (thanks @sundy-li)
• Integration Test with Spark 2.4.7&Scala 2.11 (thanks @sundy-li)
• Bump junit from 4.11 to 4.13.1