Looker 4.16 Release Notes


Userlevel 5
Badge

Anticipated Deployment Dates


Release Rollout Begins: June 11, 2017

Release Final & Download Available: June 26, 2017






Release Highlights



In addition to general tweaks and enhancements, this release comes with new and improved features in the following categories. Read on for more detail.




  • Downloading from Table visualizations

  • Smarter query caching

  • Improved SAML integrations

  • Explore searching and Instant Insight

  • Dashboard usability improvements

  • Better timeframe handling



Preparing for Release


Please take notice of items marked with a ⚡ as they indicate changes to existing functionality and may require your attention. For more information see the Legacy Feature Updates and Features by Section below.




  • Introduced the “User-based dev mode” Legacy Feature

  • Data exports to Google Sheets and using the lookerFetchHtmlTable() should be replaced to use lookerFetchData()

  • Dashboard Element rendering order now prioritizes Elements in the current viewport.

  • Introduced an on-Element toolbar for editing individual Dashboard Elements.

  • The model name associated with Looks and Dashboards is now only displayed if there is more than one model used in that Space.

  • Beta API changes.

  • On the Edit Group page, limited the displayed Roles to those related to a given Group.

  • Now perform case insensitive filtering by applying lower() and performing an equality match in BigQuery.

  • Where possible, date_trunc() will be used for timeframes rather than date_format() in MemSQL.

  • Where possible, cast() will be used for timeframes rather than formattimestamp() in DataVirtuality.




Notable Features


Export from Table visualizations


Ever wanted to rename field headers before exporting data from Looker? How about hide columns that were included to help you build a Table Calculation? You can now export data from a Table visualization rather than from the results table. This will allow you to apply most of your Table’s configurations to downloads, schedules, and Public Looks. If you are exporting data to Google Sheets and using the lookerFetchHtmlTable() function, you should replace it with lookerFetchData(). Conditional formatting and custom colors are currently not applied. Images and HTML applied to table values will not carry through.



Smarter Query Caching


Using datagroups you can tie query caching closer to your ETL timing. No need to estimate when your data load will finish and set a cache length that will hopefully not result in yesterday’s data. A datagroup will let you say “just don’t use a query’s result if it’s old data.”





Features by Section


Scheduling and Downloading




  • Apply Table visualization configurations to scheduled and exported content. Table visualization options including renamed series, hidden columns, and totals are now available in downloads, schedules, and public Looks. You can still download from the data table if you’d prefer, although the visualization is the default option. Learn more about downloading. Learn more about scheduling.

  • Conditional formatting and custom colors are currently not applied. Images and HTML applied to table values will not carry through.


  • ⚡ If you are exporting to Google Sheets and using the lookerFetchHtmlTable() function, you should replace it with lookerFetchData(). Learn more.

  • Scheduled Look and Dashboard delivery format choices were reorganized for better clarity. Learn more.

  • Added the option to select TLS and SSL versions if enabled on the SMTP admin panel.


Dashboards




  • Introduced an on-Element toolbar for editing individual Dashboard Elements. Learn more..


  • ⚡ When loading a dashboard, the Dashboard Element rendering order now prioritizes Elements in the current viewport. Learn more.


Explore




  • Instant Insight. Introduced the ability to Explore directly from the SQL Runner. Learn more..


  • Added a search bar to the Explore menu. Learn more.

  • Added the ability to drill on null values.


Content Management and Discoverability




  • Reintroduced pagination for Spaces. Looks and Dashboards are now capped at 30 per page, with the option to page forward to see additional content. This does not apply to the LookML Dashboards Spaces.


  • ⚡ The model name associated with Looks and Dashboards is now only displayed if there is more than one model used in that Space. Learn more.


Looker Platform and Administration




  • Smarter Query Caching. datagroups will allow you to set LookML triggers and intervals to refresh your query cache, as well as define a maximum potential cache age. Learn more.

  • SAML integration improvements.

  • Added the ability to map User Attributes directly to SAML Attributes rather than requiring User Attributes map to SAML Groups. Learn more.

  • Introduced the ability to set default Groups for SAML and LDAP integrations. Learn more.

  • Added a Group Finder Strategy to set Groups as individual SAML attributes with a membership value. Learn more.


  • Beta API changes.

  • The scheduled_plans_for_space() route now returns both disabled and enabled plans. Learn more.

  • Project file routes now include deleted files.

  • Added an update_session() endpoint to switch to and from development mode. Learn more.


  • ⚡ Introduced the “User-based dev mode” Legacy Feature


  • ⚡ On the Edit Group page, limited the displayed Roles to those related to a given Group.

  • License Keys are now hidden by default on the Admin panel. Click Show to display the full key.


LookML and Development Environment




  • Added the tags parameter, which allows you to provide arbitrary metadata to any LookML field. Learn more.


  • Project Importing. This experimental Labs feature enables extensions across projects. Learn more.


Dialects




  • Google BigQuery. Now perform case insensitive filtering by applying lower() and performing an equality match.

  • MemSQL


  • ⚡ Where possible, date_trunc() will be used for timeframes rather than date_format().

  • Updated to the new driver to allow for SSL connections.


  • DataVirtuality. Where possible, cast() will be used for timeframes rather than formattimestamp()


  • SAP HANA. added support for Symmetric Aggregates and Persistent Derived Tables.


  • Athena. Updated the driver to 1.1.0.


  • Impala. Added the ability to authenticate with a user and password when using Cloudera native drivers.


General Tweaks and Bug Fixes



  • Table Calculations and Custom Filters now autocorrect case sensitivity issues and provide better warnings in the event of improper constant usage (Yes vs yes, for example).

  • Fixed an issue where Table Calculations were not displayed in a visualization when all other fields were hidden.

  • Fixed an issue where conditional formatting did not color small decimal values.

  • Fixed an issue where the background color of embedded Funnel visualizations was not fully applied.

  • Fixed an issue where text Dashboard Elements did not properly wrap on Internet Explorer.

  • Fixed an issue where group_labels did not properly wrap.

  • Fixed an issue where users with no Roles could not load the Spaces page.

  • Addressed an issue where new sub-personal Spaces did not appear until the page was refreshed.

  • Fixed an issue where the LookML Dashboards Space was not visible in certain cases.

  • Fixed an API issue where sql_table_name was not included in the lookml_model_explore() output.

  • Fixed an issue where dimension fill would re-enable when dimensions were added or removed.

  • Addressed an issue where dimension fill would warn when no results were returned.

  • Addressed issues related to datatype: date on several SQL dialects.


15 replies

Userlevel 3

A fantastic step forward, we can now seriously looking at reporting use-cases. Thanks guys.

Will there be an option to trigger the datagroup cache refresh via the API?

Userlevel 5
Badge

Sure will! There’s an update_datagroup() route that will let you set {"stale_before": now()}. We’ll be sure to have the API documentation updated before release rolls.

Great stuff @cutler. This feature is a big leap forward for us.

Userlevel 5
Badge

Glad to hear it!

Hi Jon,

The API docs for datagroups is now posted on the website here.


-Tig

Userlevel 2

Aw, rats. Don’t bring back paging! Infinite scrolling would be so much better, like you do in the save modals. Paging is painfully slow to work with.


But, definitely looking forward to the save viz feature!

Thank you for table viz download option! My team will be so happy to know about this update.

Awesome! Glad to hear it @ruchir3 🙂

Amazing work Looker!


One question, what is User-based dev mode? Thanks?


Userlevel 2

Hey @quindici,


Hope you are doing fine. Basically User-based dev mode is a legacy feature:


When enabled, browser and API sessions using the same user’s credentials share the same dev_mode state. Changes made to project files by an API call will affect the currently logged in user in dev mode. When disabled, API sessions using the same user’s credentials as a logged-in user operate independently of the logged in user in dev mode. API changes to a project in a session workspace will not affect the logged-in user in dev mode.


Also please note: Deprecated in 4.16

Userlevel 1

I think optimally, we could control the options available here so that client users who are downloading via embedded dashboards cannot choose “As displayed in the data table.”

Userlevel 5
Badge

an interesting idea! I’ll talk to folks about it.


Can we do this for scheduled dashboards / looks as well? i.e. say “don’t send this email until the nightly job is finished”

Userlevel 5
Badge

Nothing is currently built around that, but I’ll make sure it’s on the radar

Reply