Skip to content

Releases: buoyant-data/oxbow

v1.8.2

12 Jan 19:24

Choose a tag to compare

Upgrade to delta-rs 0.30.1 and improve performance with deferred tabl…

v1.8.1

05 Jan 21:36

Choose a tag to compare

upgrade to delta-rs 0.30.0

This release incluhdes some performance improvements in the underlying
libraries which are most noticable on tables with very large transaction
logs or file stats entries

v1.8.0

24 Oct 21:16

Choose a tag to compare

Upgrade to delta-rs 0.29.2 and pull CommitProperties through for writers

Closes #39

Signed-off-by: R. Tyler Croy <rtyler@buoyantdata.com>

v1.7.2

17 Oct 22:20

Choose a tag to compare

Missed a write on the README for parquet-concat

v1.7.1

13 Oct 18:15

Choose a tag to compare

bump to 1.7.1 for a release of more performance improvements

Signed-off-by: R. Tyler Croy <rtyler@buoyantdata.com>

v1.7.0

09 Oct 23:24

Choose a tag to compare

Prepare the parquet-concat lambda for release with 1.7.0

The minor changes are imply removing the collect on some Vec<T> to keep
memory consumption down

Signed-off-by: R. Tyler Croy <rtyler@buoyantdata.com>

v1.6.3

23 Sep 19:39

Choose a tag to compare

Revert "Upgrade oxbow to deltalake 0.28"

This reverts commit d8f7b683d7ff1498d1c2eea96a2642d8f5b490c4.

The 0.27 and 0.28 versions of delta-rs incorporate kernel-based log
reply and we're seeing pretty dramatic performance degredation

v1.6.2

20 Sep 01:05

Choose a tag to compare

Properly iterate through record batches on deserialization

Testing this in an automated fashion proved to be majority of the
challenge since the aws-sdk-s3 crate can produce an AsyncBufRead
implementation for asynchronously reading through the S3 input file, but
chaining that together within an abstracted function for execution.

This change also includes the introduction of stats_alloc to try to do a
better job tracking memory as files are loaded.

v1.6.1

12 Sep 00:35

Choose a tag to compare

This is a minor release with a big notable upgrade of delta-rs to 0.28

No functionality changes are expected however

v1.6.0

10 Sep 16:56

Choose a tag to compare

Wrap unit test calls to set_var in unsafe {}

Newer rustc's complain about it, previous ones didn't :shrug: