0.7.25 - 2024-12-06
- without-validation option in read-bw-raw.
- Bump beam to 2.61.0.
0.7.24 - 2024-10-30
- BQ: temp-dataset and temp-project options for reading.
- Bump beam to 2.60.0.
0.7.23 - 2024-09-16
- Bump beam to 2.59.0. Downgrade kafka-clients to match beam's version.
0.7.22 - 2024-07-01
- Build compression-type map from Compression enum.
- Bump beam to 2.57.0.
0.7.21 - 2024-05-07
- Bump beam to 2.56.0
- update of BQ schema spec types to now allow
:json
and:range
- add support to set collation for STRING type
0.7.20 - 2024-03-16
- Bump beam to 2.54.0
- Bump charred to 1.034
- Bump math.combinatronics
- Bump clojure
- Downgrade slf4j to 1.7.x as v2 is not compatible with beam > 2.53.0
- Replace :dev profile logback dep with slj4j-simple
0.7.19 - 2023-12-15
- Bump beam to 2.52.0
- Add taoensso/encore required dependency for nippy.
v0.7.18 - 2023-09-29
- Bump beam to 2.50.0
- Bump charred to 1.033
- Bump superstring to 3.2.0
v0.7.17 - 2023-04-07
- add ds/distinct-by, ds/keep
- normalize BQ schema
- support for maxLength, precision, scale fields in BQ tables
- new PAssert testing utilities
- clean up aot files to reduce size
- BQ specs for schema-validation
- new exported clj-kondo config
- bump charred to 1.028, beam sdk to 2.46.0
v0.7.16 - 2023-01-27
- Bump charred to 1.019, beam to 2.44.0
v0.7.15 - 2022-12-15
- Bump charred to 1.018
v0.7.14 - 2022-12-02
- adapt charred options to avoid throwing on empty lines
- Bump charred to 1.016, beam to 2.43.0
v0.7.13 - 2022-10-26
- Add BQ raw read
:query-location
option. - Add support for Pcoll set operations.
- Bump beam to 2.42.0
- Bump charred to 1.0.14
v0.7.12 - 2022-09-16
- handle collection for generating pipeline args
- Bump beam to 2.41.0
- Bump charred to 1.0.12
v0.7.11 - 2022-08-23
- Add BQ clustering & optimized-write options in write-bq-table.
- Bump beam to 2.40.0.
- Bump charred to 1.0.11.
- Bump nippy to 3.2.0.
v0.7.10 - 2022-06-24
- Bump nippy to fix serializing of boolean values in map collections.
- Bump Beam to 2.39.0.
v0.7.9 - 2022-05-24
- Bump charred to fix more white-space issues. Enable options to match previous output.
- Add write-json-file-test.
v0.7.8 - 2022-05-16
- More complete tests, again.
- Add ability to specify a teardown function for a
DoFn
- Invoke
start-bundle
,finish-bundle
andteardown-fn
when specified - Reuse no-op function in
dofn
- switch from cheshire to com.cnuernber/charred.
v0.7.7 - 2022-05-13
- Enforce target jdk as 1.8
- Bump clojure to 1.11.1
- more complete tests
v0.7.6 - 2022-04-23
- don't require bq, pubsub, datastore in api.clj - Issue #94
- Bump beam to 2.38.0
v0.7.5 - 2022-04-09
- replace cheshire usage with jsonista by @domparry
- Bump beam to 2.37.0; bump other deps
v0.7.4 - 2021-12-15
- Datastore fix by @domparry for breaking change introduced by nippy bump in 0.7.3
- Bump beam dep (2.34.0)
- Bump tools.logging (1.2.2)
- Clean up
v0.7.3 - 2021-08-14
- Resolve various java compiler warnings
- Bump dependencies including beam (2.31.0)
- Bump nippy to 3.1.1, adding alter-var-root of
nippy/*thaw-serializable-allowlist*
to allow serializable org.apache.beam.sdk.values.KV - Fix probems with Id keys in Google Datastore storage
v0.7.2 - 2021-01-07
- fix locking problem with requires in safe-exec => Issue #97
- Update Apache Beam to 2.26.0
v0.7.1 - 2020-10-06
- Update Apache Beam to 2.24.0
v0.7.0 - 2020-09-07
- Add Kafka IO support
- Add delete-datastore-raw to datastore features
- add :schema-update-options for BQ write
- Update Apache Beam to 2.22.0
- Performance improvements
- Fixed require bug in
partition-by
v0.6.6 - 2019-09-11
- Fix
cogroup-by
for numbers of pcoll > 10 => Issue #83 - Fix
:many-files
option inread-text-file
=> Issue #75
v0.6.5 - 2019-08-23
- Add FileIO Read to read a PCollection of files
- Add ElasticSearchIO
- Upgrade to Apache Beam 2.15.0
- Upgrade dependencies
v0.6.4 - 2019-01-16
- Upgrade to Apache Beam 2.9.0
- Upgrade dependencies
v0.6.3 - 2018-11-19
- Add json-schema in bq write
- Combine-fn accepts now map as arguments
- Remove deprecated write-by
v0.6.2 - 2018-11-15
- Upgrade to Apache Beam 2.8.0
- Support dynamic file writes
- Support
:deflate
and:zip
compressions bq/->time-partitioning
now accepts:field
and:require-partition-filter
- Improve docs
v0.6.1 - 2018-03-20
- Upgrade to Dataflow 2.3.0
- Support time-partitioned Big Query tables (thanks to @neuromantik33)
- Fix broken input from BigQuery
- Fix missing safe-exec from filename policy
- Fix broken serializing after group-by in some cases
- Fix rare (hopefully) bug introduced by being lazy. with the vals after a group-by. Now we are greedy.
v0.6.0 - 2018-02-13
- Compatibility with Dataflow 2.2.0
- Compatibility with Clojure 1.9 and specs (some code had to be ported to Java shims because
proxy
is notSerializable
anymore in Clojure 1.9)
- Removed superfluous logging,
safe-exec
call and type hints
v0.5.3 - 2017-12-18
- Fix options on PipelineWithOptions to have it work again on Dataflow (oops, sorry!)
v0.5.2 - 2017-12-14
- Fixed
defoptions
again and made it useful for interop with Beam - Added
:checkpoint
option, will be made more useful later - Switched to new PubSub API
- Add support for custom filename-policy
- Fixed a possible bug causing StackOverflows on
combine
v0.5.1 - 2017-10-05
- Added
wait-pipeline-result
- Fix a bug with incorrect call to
seq
in combine - Add a missing safe-exec in
write-bq-table-raw
- Update to v2.x Dataflow version (Apache Beam)
- Add new streaming features (custom timestamp in PubSub #30 (thanks to @RoIT), BigQuery streaming writes)
- Add compressed TextIO output #31 (thanks to @RoIT)
- Add FileNamePolicy compatibility for TextIO
- Handle recursion in RECORD schema generation for Big Query #9 (thanks to @torbjornvatn)
- Add windows and trigger support #13 (thanks to @RoIT)
- Bump Dataflow SDK to 1.9.0
- Add support for standard SQL syntax for Big Query #11 (thanks to @torbjornvatn)
- Read data from pubsub topics #10 (thanks to @torbjornvatn)
v0.4.0 - 2016-11-12
- Update Datastore API to v1, add wrappers for creating Entities from and converting to Clojure maps. Some breakage in the datasplash.datastore namespace ensues.
- Add pubsub namespace #6 (Thanks to @MartinSahlen)
- Put slf4j deps in :dev profiles #8 (Thanks to @torbjornvatn)