Releases: linkedin/iceberg
Releases · linkedin/iceberg
Backport upstream fixes for lock issue and unexpected metadata deletion
What's Changed
- reverting commits 2c59857 and f362aed by @shenodaguirguis in #88
- Backport apache/iceberg#2328 and its prerequisites by @autumnust in #89
Full Changelog: 0.11.0.19...0.11.0.22
Release-r's Note:
This release contains a few patches to fix bugs that we saw.
Details are contained in #89
New union to struct schema (Trino style) support
New union to struct schema (Trino style) support
New union to struct schema (Trino style) support
Fix spark avro reader reading union schema data
Fix spark avro reader reading union schema data (#83) * Fix spark avro reader to read correctly structured nested data values * Make sure field-id mapping is correctly maintained given arbitrary nested schema that contains union
Move to Avro 1.10.2
Move to Avro 1.10.2, and disable Avro schema defaults validation, for backward compatibility
Fix ORC schema visitors to support reading ORC files with deeply nested union type schema
0.11.0.16 Fix ORC schema visitors to support reading ORC files with deeply nest…
0.11.0.15: Accomodate for HiveConf overwriting defaults in copy constructor
Test Plan: wrote a unit test but turns out that it's dependent on the HiveConf.ConfVars you have on the classpath so it was difficult to get it right. So, ended up testing on the azkaban cluster manually
Support avro.schema.literal/hive union types in Hive legacy table to Iceberg conversion
Support non-null Default Values for AVRO
This release adds support for non-null default values to AVRO file format. The default-values are supported for primitive and complex data types (i.e., struct, map and array). APIs in the Iceberg Schema's NestedField. The default value is now propagated during schema mappings (to and from AVRO) and are used by the DeSer.