August 2023

Initial Iceberg support

Available in: GlareDB@v0.3.0, GlareDB Cloud

We’ve added initial support for reading Iceberg tables with iceberg_scan, iceberg_snapshots and iceberg_data_files SQL functions. See https://github.com/GlareDB/glaredb/issues/1448 for more details on what’s to come in future releases.

SQL workspace improvements

Available in: GlareDB Cloud

  • Tutorials were added to demonstrate connecting and reading external sources. The tutorials can be accessed by clicking the info icon on the top right of the results panel. To switch back to results view, click the table icon.

    tutorials tutorial-toggle

  • Double-clicking a table in the explorer generates a SELECT query
  • The explorer added Views and distinguishes external from default databases
  • The results view was redesigned. In a multi-query statement, all results are now shown as tabs, even if one of the statements returned nothing.

Native tables: UPDATE, DELETE and CREATE AS

Available in: GlareDB@v0.3.0, GlareDB Cloud

A table can be created from a query:

CREATE TABLE t1 AS SELECT * FROM generate_series(1, 5, 2);

Available in: GlareDB@v0.4.0, GlareDB Cloud

Native tables now support UPDATE and DELETE statements:

CREATE TABLE public.users ( name text );
INSERT INTO public.users VALUES ('Eldon');
UPDATE public.users SET name = 'Eldon Tyrell' WHERE name = 'Eldon';
DELETE FROM public.users;

Improvements to SQL functions

Available in: GlareDB@v0.3.0, GlareDB Cloud

  • You can now glob (** and *) to scan multiple files and directories. See csv_scan, for example.
  • delta_scan function was added to query delta tables from local paths, s3 and gcs
  • All *_scan functions received support for passing multiple URLs

Available in: GlareDB@v0.4.0, GlareDB Cloud

Exclude columns in SELECT

Available in: GlareDB@v0.3.0, GlareDB Cloud

SELECT now accepts an EXCLUDE clause to exclude specific columns from output:

SELECT * EXCLUDE (id) FROM users;

Misc updates and fixes

Available in: GlareDB@v0.3.0, GlareDB Cloud

  • Fixed CREATE SCHEMA IF NOT EXISTS failing if schema existed

Available in: GlareDB@v0.3.0

  • glaredb local now has prettier output
  • Python bindings now include show() with pretty formatting and close() for gracefully closing connections.

Available in: GlareDB@v0.4.0, GlareDB Cloud

  • Fixed now() returning the incorrect time

Available in: GlareDB@v0.4.0

  • You can now run glaredb locally without specifying the local subcommand:
./glaredb

Available in: GlareDB Cloud

  • Fixed an issue where the button for adding an SSH tunnel was not displaying
  • Organization billing now contains historic billing data
  • The sign in page was redesigned
  • The SQL workspace is now the first page users are loaded into when they sign in