Looker 3.28 Release Notes

  • 20 August 2015
  • 2 replies

Userlevel 4

Release Date: August 18, 2015

End of Cycle Date: September 4, 2015

See All Release Notes

Visualizations and Dashboards

  • New “Table” visualization type. Learn More.

  • Support for measure-versus-measure scatterplots. Learn More.

  • Ability to filter visualizations based on results of yesno fields. Learn More.

  • Dashboard PDF filename now includes timestamp

  • Visualization Configuration Option Updates:

    • Provide more informative field names when “Show Full Field Name” is disabled

    • New single value visualization size option, “Auto Fit”

    • Donut multiple labels are now on by default

    • X-axis defaults to type “time” if stacking isn’t “grouped”

    • Default min and max y-axis values are based on values in the chart instead of being symmetric

  • Performance Improvements:

    • Improved default visualization type for each query

    • Improved performance when rendering cartesian charts

    • Improved efficiency of rendering large result sets


  • NULL values are now treated as 0 for measures of type sum

  • Table Calculation Updates:

    • New pivot_index function allows referencing pivot columns individually. Learn More.

    • Suggestions now appear after space, enter and delete (in addition to after regular text entry)

    • Table calculation columns will be present in the data table while query is running

  • Updated “Save to dashboard” modal.

  • Performance Improvements

    • Filter suggestions are now cached for 6 hours (instead of 5 minutes)

    • Improved caching for queries accessed via public URLs

    • “No Results” now shown when query returns zero rows

LookML & Development

  • Support for datatype: yyyymmdd time fields in all dialects

  • one_to_one joins can now be specified without required_joins

  • Deprecated map_point, map_line, and locations field types

Dialects and Databases

  • Beta Support for Spark SQL:

    • Persistent derived tables

    • In-database pivoting/column limits

    • list field type

  • Basic support for EXASOL, including symmetric aggregates (no support for PDTs or time zones yet)

  • Improved support for additional JDBC parameters across dialects

  • Dialect-Specific Improvements:

    • Oracle field names are now required to be lowercase

    • Microsoft SQL Server now has SSL support

    • Redshift queries are now cancelled in a way that doesn’t produce errors in log files

Administration Panel

  • Users can now be added to the “Admin” role in bulk

  • Warnings about misconfigured proxies in front of Looker are presented on the “General” admin page

Powered By Looker

  • Option to display normal login screen to unauthenticated users (when SSO authentication is not being used)

Authentication and Security

  • Beta support for group-to-role mappings with SAML

2 replies

Userlevel 3

Great new features!

Userlevel 1

Nice! Fewer coalesce() statements in PDTs!