We’re well underway on our 20.1 releases for SQL Developer, Modeler, SQLcl, ORDS, and SQL Developer Web.
In fact, we’ve recently went into code (feature) freeze, so I have a REALLY good idea of what users will notice when version 20.1 of everything is released at the end of the calendar quarter.
Here’s a couple of features that has me most excited.
We’ll lead with something VERY simple. But, I think this will make a big impact on all of our SQL performance gear heads and scientists out there.
When looking at AutoTrace and Execution Plans, you’ll be able to search across all of the plan steps, notes, metadata, etc.
How many NESTED LOOPS does this plan have?
In 20.1, you’ll see a new SEARCH button in the PLAN toolbars. Once you hit that, the search buttons you see in the graphic above will appear.
May all your plans be so simple you never even wondered about a search feature. Unfortunately, I’ve seen your queries, and I know better than to expect this.
Importing JSON, XML, Parquet data to new relational tables
We’re building this in SQL Developer Web, which has been distributed as a feature in Oracle REST Data Services (ORDS) since version 19.4.
Version 20.1 will let you create a new table by simply pointing to a JSON file.
‘Flat’ structures will come in a rows and columns. A nested JSON attribute will come in as a CLOB, which you can then also say ‘IS JSON.’
Once Oracle Database 20c is officially released, we’ll of course look at upgrading that to use the new native JSON data type.
To have a go at this, I downloaded ALL the TWITTER data for the @oraclesqldev account.
So once I point to my file, in this case a JSON file,
Advance the wizard a step for the juicy TABLE DDL implementation –
And then I clicked ahead to FINISH and started to explore my data.
So instead of perusing 4MB of JSON data, I can now use the power of SQL to start answering some interesting questions.