Skip to content
This repository has been archived by the owner on Mar 24, 2021. It is now read-only.

Releases: cloudant-labs/spark-cloudant

Cloudant Spark connector v2.0.0 for Spark 2.0

23 Sep 11:40
Compare
Choose a tag to compare

Contains fixes for tickets:

BugzId: 74074 - Upgrade to Spark 2.0

Cloudant Spark connector v1.6.4

07 Sep 00:25
Compare
Choose a tag to compare

Contains fixes for tickets:

BugzId: 73249 - Update libraries, upgrade version to 1.6.4
BugzId: 68888 - Make user-agent to be equal spark-cloudant
BugzId: 65662 - Improve spark streaming
BugzId: 67396 - Correct errors during save operation
BugzId: 64055 - Create database on save
BugzId: 63503 - Make schemaSampleSize by default to be equal -1

Cloudant Spark connector v1.6.3

23 Mar 19:53
Compare
Choose a tag to compare

Contains fixes for tickets:

add supporting in passing configuration through spark-submit --conf
revise default bulk size
better throughput: during save
revise dispatcher
Improve save parallelism, sharing actorSystem for the duration of the job
BugzId: 62934 - Make views to use N docs for schema discovery
BugzId: 62617 - Fixed the issue of the column order
BugzId: 61745 - Integrate Cloudant with Spark Streaming
BugzID: 61793 - Set up bulkSize on the level of dataframe

Cloudant Spark connector v1.6.2

10 Mar 13:49
Compare
Choose a tag to compare

Contains fixes for tickets:

BugzID: 53022 - Register cloudant-spark with Spark Packages community index
BugzId: 61792 - Message for saving data with 0 records
BugzId: 60775 - Remove all providers accept DefaultSource
BugzId: 59451- Cache documents retrieved for schema when schemaSampleSize = -1
BugzId: 60785 - Adding connector version

BugzId: 60495 - Loading data from views
BugzId: 59902 - Make a protocol http or https configurable
BugzId: 60263 - Increase timeouts in order to be able to load big dbs into Spark

Based on release v1.6.1

Cloudant Spark connector v1.6.1

03 Feb 16:34
Compare
Choose a tag to compare

Contains fixes for tickets:

BugzId: 59832 - fix for nested doc objects
BugzId: 59432 - added test cases to validate schemaSampleSize option
BugzId: 59236 - implement schemaSampleSize option to support schema discovery for N documents
BugzId: 57609 - support for chunked responses and large databases
BugzId: 57464 - automate test database generation

Based on pre-relase v1.4.1.5

Bug fixes for v1.4.1

09 Nov 16:49
Compare
Choose a tag to compare
Bug fixes for v1.4.1 Pre-release
Pre-release

BugzId: 52947 - Connect to Cloudant using https - not http
BugzId: 55905 - Remove FilterUtil to avoid type cast problems for JSON and Scala types
BugzId: 56197 - Raise an exception if save() to a missing database fails
BugzId: 56458 - Support predicates on id columns
BugzId: 56722 - Remove support for Riak
BugzId: 56450 - Remove design documents from default query

v1.4.1.4 for Spark Service on Bluemix

02 Nov 13:56
Compare
Choose a tag to compare
Pre-release

Version has been compiled with Spark 1.4.1

Assembly uses a merge strategy where includeScala = false

That allows it to work properly with the Spark Service deployed on Bluemix and the connector can be used as pre-loaded dependency for iPython using both, Scala and Python runtimes.

Spark 1.4.1 with bulk save

08 Sep 18:34
Compare
Choose a tag to compare
Pre-release
v1.4.1.3

bulk save

Spark 1.4.1 with more fixes

03 Sep 16:47
Compare
Choose a tag to compare
Pre-release

Introduce requestTimeout and concurrentSave configuration for large volume data save.

Spark 1.4.1 with fixes

20 Aug 18:42
Compare
Choose a tag to compare
Pre-release

Enable cloudant configuration passed in to DataFrame using option