Skip to content

Latest commit

 

History

History
285 lines (197 loc) · 10.3 KB

CHANGELOG.md

File metadata and controls

285 lines (197 loc) · 10.3 KB
Raw

CHANGELOG

8.1.1

8.1.0

  • Fixed Quoting Issues raised in Issue #118. Thanks to @krisse7 for PR #120!

8.0.0

  • Moved entire code into a single project, remove multi modules.

7.0.1

  • Re-added support for primtive data types, but used method names:
    • AbstractMapping#mapBooleanPrimitive
    • AbstractMapping#mapBytePrimitive
    • AbstractMapping#mapShortPrimtive
    • AbstractMapping#mapIntegerPrimitive
    • AbstractMapping#mapFloatPrimitive
    • AbstractMapping#mapDoublePrimitive

7.0.0

  • The JpaMapping<> has been dropped from the library.

6.0.1

  • Adding SimpleRow#setTime for mapping LocalTime

6.0.0

This release saw major improvements and some breaking changes. The release is a dramatic improvement of the library, because it adds:

  • Java11 support
  • Improved Type Safety
  • Internal Refactorings to improve maintainability
  • Build Pipelines
  • Code Coverage
  • Dependency updates

All of this great work was done by @jonfreedman.

Packages

There is a major change in how the library is released. The project now targets both Java8 and Java11. You are able to use the JDK8 version by using the jdk8 classifier.

So for Java8 you have to use the following dependencies:

<dependency>
	<groupId>de.bytefish.pgbulkinsert</groupId>
	<artifactId>pgbulkinsert-core</artifactId>
	<version>6.0.0</version>
	<classifier>jdk8</classifier>
</dependency>
<dependency>
	<groupId>de.bytefish.pgbulkinsert</groupId>
	<artifactId>pgbulkinsert-rowwriter</artifactId>
	<version>6.0.0</version>
	<classifier>jdk8</classifier>
</dependency>
<dependency>
	<groupId>de.bytefish.pgbulkinsert</groupId>
	<artifactId>pgbulkinsert-jpa</artifactId>
	<version>6.0.0</version>
	<classifier>jdk8</classifier>
</dependency>

For Java11 you have to use the following dependencies:

<dependency>
	<groupId>de.bytefish.pgbulkinsert</groupId>
	<artifactId>pgbulkinsert-core</artifactId>
	<version>6.0.0</version>
</dependency>
<dependency>
	<groupId>de.bytefish.pgbulkinsert</groupId>
	<artifactId>pgbulkinsert-rowwriter</artifactId>
	<version>6.0.0</version>
</dependency>
<dependency>
	<groupId>de.bytefish.pgbulkinsert</groupId>
	<artifactId>pgbulkinsert-jpa</artifactId>
	<version>6.0.0</version>
</dependency>

Breaking Changes

  • PgBinaryWriter#open removed and now opened in constructor
  • SimpleRowWriter#open removed and now opened in constructor
  • SimpleRowWriter now implements the AutoClosable interface, so it can be used in try-with-resources blocks.
  • PostgreSqlUtils#tryGetPGConnection now returns an Optional
  • BulkProcessor#cancel static method removed

Additional Information

  • Static Analysis has been added to the project using Error Prone
  • Integration Tests are now run on every commit using Github Pipelines and the Postgres image.
  • There have been a lot of refactorings internally, so the overall maintainability is improved.
  • The project now has a Code Coverage report, so you get a feeling about the test coverage.
  • The README now has badges, so it is easier to see the latest stable release.

Project development

  • The project has been moved to an organization structure, so it easier to assign rights and collaborate.

5.1.0

This release saw additions and refactorings of the JPA module. It now supports a lot more data types and is easier to configure by using annotations. If you want control over how the data is written to the database, you can use the @PostgresDataType annotation, which overrides the default values.

Here is an example for mapping a short value to an int4 Postgres data type.

@Entity
@Table(name = "unit_test", schema = "public")
public static class SampleEntity {

    // ...

    @Column(name = "some_field_name")
    @PostgresDataType(columnName = "some_field_name", dataType = DataType.Int4)
    private Short shortField;

    public Short getShortField() {
        return shortField;
    }

    public void setShortField(Short shortField) {
        this.shortField = shortField;
    }
}

5.0.0

A lot of thanks to the great efforts of user @cheffe in this release!

This release sees some major and breaking changes:

  • Introducing Semantic Versioning with major.minor.bugfix starting with 5.0.0.
  • Split the project into multiple modules to provide a stable core and build modules around it:
    • Moving the SimpleRowWriter to module pgbulkinsert-rowwriter
    • Moving the BulkProcessor to module pgbulkinsert-bulkprocessor
    • Creating a new module pgbulkinsert-jpa to better integrate with existing JPA mappings.
  • Added a new mapping JpaMapping to module pgbulkinsert-jpa, which provides a better JPA integration:
  • Made it easier to run the tests for the projects by introducing a properties file.

From now on the groupId is de.bytefish.pgbulkinsert and all dependencies can be installed from Central Maven Repositories as:

<dependency>
	<groupId>de.bytefish.pgbulkinsert</groupId>
	<artifactId>pgbulkinsert-core</artifactId>
	<version>5.0.0</version>
</dependency>
<dependency>
	<groupId>de.bytefish.pgbulkinsert</groupId>
	<artifactId>pgbulkinsert-rowwriter</artifactId>
	<version>5.0.0</version>
</dependency>
<dependency>
	<groupId>de.bytefish.pgbulkinsert</groupId>
	<artifactId>pgbulkinsert-jpa</artifactId>
	<version>5.0.0</version>
</dependency>

4.1

  • Simplified Quoting identifiers for the SimpleRowWriter

4.0

Added support for Range Types:

  • int4range
  • int8range
  • numrange
  • tsrange
  • tstzrange
  • daterange

Thanks to @csimplestring for help and valuable feedback!

4.0-alpha-2

  • Bugfixes for the Range Type

4.0-alpha-1

  • Added support for Ranges, such as tstzrange. Thanks to @csimplestring for the Feature request!
    • Use mapTsTzRange if you are using the AbstractMapping
    • Use setTsTzRange if you are using the SimpleRowWriter

3.8

  • Many improvements added by @tangyibo!
    • Closing the Stream of the SimpleRowWriter, so it doesn't block anymore.
    • Improved error logging to get to the underlying cause of the error in the JDBC driver.
    • Added Null Character Handling, so Null Bytes (0x00) do not crash the COPY

You can enable the Null Character Handling like this. It removes all 0x00 occurences and replaces them with an empty string:

// Create the Table Definition:
SimpleRowWriter.Table table = new SimpleRowWriter.Table(schema, tableName, columnNames);

// Create the Writer:
SimpleRowWriter writer = new SimpleRowWriter(table);

// Enable the Null Character Handler:
writer.enableNullCharacterHandler();

If you need to customize the Null Character Handling, then you can use the setNullCharacterHandler(Function<String, String> nullCharacterHandler) function.

3.5

  • Updates the PostgreSQL JDBC driver to Version 42.2.9.

3.4

  • Improving the ZonedDateTime handling.

3.3

  • @robsonbittencourt reported a severe bug in the Numeric Value handling. Everyone should update to 3.3!

3.2

3.1

  • Thanks to @kowalczm for adding support for primitive types (byte, int, float, double, ...), see Pull Request #27. This makes boxing primitive types unnecessary now and will make it easier to integrate the library into existing applications.

3.0

  • The library now uses the standard functional interfaces of Java, see Pull Request #24. The ByteArrayValueHandler now uses primitive Byte Arrays (byte[]), which makes boxing unnecessary. This was a breaking change in the API and as such, the major revision was increased to 3.0.
  • An overloaded saveAll method has been added by @kowalczm in Pull Request #25. You can now save a Collection instead of being forced to use streams.

2.2

  • Thanks to @momania for fixing Issue #20, which was a severe bug with the mapping procedures. This led to improvements for the AbstractMapping API now exposing almost all types as arrays.

2.1

  • Issue #16 updates the PostgreSQL JDBC driver to Version 42.2.2. A configurable buffer size was added to the API for improving the throughput to PostgreSQL. Thanks to @kowalczm for both improvements.

2.0

  • Pull Request #19 was a major refactoring of the API by creating an AbstractMapping. This now separates the Mapping and the actual saving, making it possible (and easier) to reuse the PgBulkInsert Mapping API as a standalone implementation. As a major breaking change, the major revision has been set to 2.0. Thanks to @The-Alchemist for raising the issue and making improvements to the library.

1.4

  • Issue #11 added Microsecond resolution for Timestamps. Thanks to @zach-m for raising the issue.

1.3

  • Issue #10 added support for the Numeric Data Type to the API. Thanks to @li-xiangdong for raising the issue.

1.2

  • Commit b4543db restructured the Value Handler API, which was an internal implementation detail, so the major revision wasn't increased.

1.1

  • Added Geometric Data Types, JSONB.

1.0

  • Initial Release.