Releases: snowplow/snowplow-s3-loader
Version 2.1.4
Version 2.1.3
Version 2.1.2
Bug fixes
Bump amazon-kinesis-client to 1.14.5 (#245)
Version 2.1.1
Version 2.1.1-rc2
Exlude transitive dependencies of hadoop (close #234)
Version 2.1.1-rc1
Exlude transitive dependencies of hadoop (close #234)
Version 2.1.0
This version adds a feature to give complete control over the partitioning of S3 directories via the configuration hocon. It also fixes a major bug that could cause the loader to hang during kinesis scaling events.
Bug fixes and performance improvements
- Optimise fromEnriched function (#227)
- Fix dateFormat partitioning in output path (#236)
- Fix premature shutdown of HTTP connection pool (#231)
- Fix duplicate statsd metrics when loading lzo files (#229)
Under the hood
Version 2.0.0
A new major releasing bringing StatsD observability, Sentry and config refactoring.
New features
- Report metrics to StatsD (#216)
- Integrate Sentry (close #218)
- Use AdoptOpenJDK 11 as docker base image (#224)
- Attach jar files to Github releases (#197)
API changes
Under the hood
- Use snowplow-badrows (#215)
- Use sbt-tpolecat (#222)
- Harmonize module structure (#210)
- Extend copyright notice to 2021 (#219)
- Apply automated code-formatting (#221)
- Point to Snowplow roadmap in the README (#220)
- Replace joda-time by java-time (#202)
- Replace scopt by decline (#201)
- Remove json4s dependency (#200)
Dependency bumps
Version 1.0.0 (2020-12-22)
This version comes with minor bug fixes (#180 and #192) and 2 new features:
- #173: files on S3 now contain the time in the prefix, making it possible to partition data in Athena.
- #193: it's now possible to disable KCL metrics that are automatically sent to CloudWatch, in order to save money. To do so, this parameter needs to be added to the config.
Documentation for this component can be found on our docs website.
Bug fixes / new features
- Handle case where partitionRecords is empty (#192)
- Use random partition key for bad rows (#180)
- Add sink timestamp section into file names (#173)
- Make it possible to disable KCL CloudWatch metrics (#193)
Misc
- Set log level of records serialized to DEBUG (#174)
- Migrate from Travis to GH actions (#181)
- Use hadoop-lzo 0.4.20 from Snowplow Bintray maven (#183)
- Add coveralls integration (#195)
- Use log4j-over-slf4j for log4j logs (#194)
- Use sbt-scoverage plugin (#164)
- Use sbt-scalafmt plugin (#163)
- Bump sbt to 1.4.4 (#167)
- Bump specs2-core to 4.10.5 (#189)
- Bump iglu-core-json4s to 1.0.0 (#188)
- Bump pureconfig to 0.14.0 (#187)
- Bump json4s-jackson to 3.6.10 (#190)
- Bump cats-core to 2.3.0 (#186)
- Bump scopt to 4.0.0 (#185)
- Bump Scala tracker to 0.7.0 (#184)
- Bump Scala to 2.13 (#182)
Version 0.7.0 (2020-02-24)
An initial release of data partitioning by schema
Features
Add partitioning for self-describing types (#146)
Remove colons from NSQ filenames (#125)
Allow configurable and dynamic s3 path (#134)
Allow override of kinesis and s3 endpoints in configuration (#131)
Update flags to resemble ones used in GCS loader config (#157)
Add sbt-native-packager and Docker config (#152)
Remove curly braces from transformed directory pattern (#137)
Document that maxTimeout is in milliseconds (#129)
Dependencies
Bump sbt-assembly to 0.14.10 (#156)
Replace scalaz by cats (#148)
Bump SBT to 1.2.8 (#151)
Misc
Extend copyright notice to 2019 (#154)
Change Travis distribution to Trusty (#155)
Add Docker Hub credentials to .travis.yml (#153)
Move Twitter repo to https (#127)
Remove vagrant setup (#140)