3.3.0-rc2

Discovery features

Enhancement

Enhance displaying the resources associated with the item in the workspace (#2167)
Add re-ingestion setting with the file (#2239)
Disable roll-up mode in case of including geo type column (#2299)
Data exploration based on popularity, updated-date, catalog, tags using metadata (#2131)
Display and download metadata's sample data(#2172)
Improve Java version(11 or 12) compatibility (#1704)
Documentation for application properties (#2246)

Bug Fix

Back button toward dataflow list acts like browser back button (#2232)
'Series Percent' needs to be modified so that the sum of the bar charts is 100 percent. (#2184)
The user-defined field is not working in the prediction line (#2227)
In dataflows no other join type besides an inner join works (#2241)
Validation is disabled when a column dictionary of type timestamp is connected. (#2242)
If a permission schema is deleted, workspaces which are linked in the permission schema is not changed (#2214)
If the measurement is null, it must be able to be replaced with a negative value in the data source ingestion process. (#2120)
Unexpected retry occurs when a query fails. (#2272)
Fix error during processing DB connection in workbench (#2250)
A snapshot that contains join rules looks like empty. (#2216)
Treat null values and empty strings the same way. (#2243)

Druid features

You need to upgrade your Druid according to the Discovery version.
Use druid-metatron-3.2.5.tar.gz from this link

Full features

Type of PostAggregator is not valid (#2281)
Support wasb scheme for deep storage (#2278)
Allow numeric null from text file (#2274)
Minor log improvement of QueryResource (#2270)
Add licenses to LICENSE (#2268)
Remove unnecessary wrapping in hash join on single key (#2265)
Keep collation information of join for further processing (#2262)
Exclude NaN from covariance query results (#2261)
Append task stats in task status (#2260)
Lessen gc pressure of coordinator with a lot of segments (#2258)
Trivialize timeout exception on scheduleNow() (#2257)
Support ordering pushdown for stream query (#2256)
Change license headers (#2252)
Make option for bush join optimizer (#2238)
Fix some regressions from #2133 (#2236)
Support heuristic join reordering in SQL (#2233)
Add prepare-metric.sh example for engine monitor (#2231)
Add option for dumping details of query planning in SQL (#2226)
Keep some silly legacy convention for pivot (#2207)
Amplify cost reduction of scan projection (#2020)
Simplify state machine of query generator in SQL (#2194)
Support windowing functions in SQL (#1282)
Support windowing in stream query (#2133)
Improve general type resolution (#2171)