Skip to content

Latest commit

 

History

History
126 lines (93 loc) · 3.98 KB

CHANGELOG.md

File metadata and controls

126 lines (93 loc) · 3.98 KB

CHANGELOG

1.3.1 (2020-03-03)

  • Correctly decode strings that are between 157 and 288 bytes long. 1.3.0 introduced a regression when decoding these due to using a signed byte as an unsigned value. Reported by Dongmin Yu. GitHub #181 in maxmind/GeoIP2-java.
  • Update jackson-databind dependency.

1.3.0 (2019-12-13)

  • IMPORTANT: Java 7 is now required. If you need Java 6 support, please continue using 1.2.2 or earlier.
  • The method getRecord was added to com.maxmind.db.Reader. This method returns a com.maxmind.db.Record object that includes the data for the record as well as the network associated with the record.

1.2.2 (2017-02-22)

  • Remove the version range. As of today, jackson-databind is no longer resolved correctly when a range is used. GitHub #28.

1.2.1 (2016-04-15)

  • Specify a hard minimum dependency for jackson-databind. This API will not work with versions earlier than 2.7.0, and Maven's nearest-first resolution rule often pulled in older versions.

1.2.0 (2016-01-13)

  • JsonNode containers returned by the get(ip) are now backed by unmodifiable collections. Any mutation done to them will fail with an UnsupportedOperationException exception. This allows safe caching of the nodes to be done without doing a deep copy of the cached data. Pull request by Viktor Szathmáry. GitHub #24.

1.1.0 (2016-01-04)

  • The reader now supports pluggable caching of the decoded data. By default, no caching is performed. Please see the README.md file or the API docs for information on how to enable caching. Pull requests by Viktor Szathmáry. GitHub #21.
  • This release also includes several additional performance enhancements as well as code cleanup from Viktor Szathmáry. GitHub #18, #19, #20, #22,and #23.

1.0.1 (2015-12-17)

  • Several optimizations have been made to reduce allocations when decoding a record. Pull requests by Viktor Szathmáry. GitHub #16 & #17.

1.0.0 (2014-09-29)

  • First production release.

0.4.0 (2014-09-23)

  • Made com.maxmind.db.Metadata public and added public getters for most of the interesting metadata. This is accessible through the getMetadata() method on a Reader object.

0.3.4 (2014-08-27)

  • Previously the Reader would hold onto the underlying file and FileChannel, not closing them until the Reader was closed. This was unnecessary; they are now closed immediately after they are used. Fix by Andrew Snare; GitHub issue #7.
  • The Reader now discards the reference to the underlying buffer when close() is called. This is done to help ensure that the buffer is garbage collected sooner, which may mitigate file locking issues that some users have experienced on Windows when updating the database. Patch by Andrew Snare; GitHub issue #8.

0.3.3 (2014-06-02)

  • A potential (small) resource leak when using this library with a thread pool was fixed.

0.3.2 (2014-04-02)

  • Added tests and documentation for multi-threaded use.

0.3.1 (2013-11-05)

  • An InputStream constructor was added to the Reader class. This reads the stream into memory as if it was using FileMode.MEMORY. Patch by Matthew Daniel.
  • The source code is now attached during packaging. Patch by Matthew Daniel.
  • The artifact ID was changed to maxmind-db in order to increase naming consistency.

0.3.0 (2013-10-17)

  • IMPORTANT: The package name was changed to com.maxmind.db. The MaxMindDbReader class was renamed to Reader.
  • Improved error handling and test coverage.
  • Performance improvements.

0.2.0 (2013-07-08)

  • The reader and database format now uses IEEE 754 doubles and floats.
  • FileMode.IN_MEMORY was renamed to FileMode.MEMORY.
  • Cache Type enum values array.

0.1.0 (2013-06-14)

  • Initial release