Looker Connect Training

Help Center

Documentation

Community

Cloud Certifications

Looker 4.10 Release Notes

cutler
Rank
Userlevel 5
Badge

Release Rollout Begins: March 19, 2017

Release Final & Download Available: March 30, 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.




  • Dashboards and Visualizations

  • Administration

  • Dialects



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 “Implicit Dashboard Default Timezone” Legacy Feature.

  • The “Legacy Drill Behavior” Legacy Feature is disabled by default.

  • The “YAML-based LookML for modeling” Legacy Feature is disabled by default.

  • Addressed inconsistent Look and Dashboard presentation in Spaces for users with access_data and see_user_dashboards and users with just see_user_dashboards.

  • The Save button on the Explore page will now return you to the Explore Page instead of taking you to the Look.

  • Filter values rather than filter names are now displayed in bold when the filters bar is collapsed.

  • Deleted Looks will no longer appear in the Content Validator.



Notable Features


Improved Y-Axes Controls


The y-axis configuration settings has been revamped, allowing Cartesian charts to be much more flexible. You can now combine any number of y-axes (or chose not to!), scale and format them independently, and split pivoted series across axes. Happy creating!



Control Access Filters with User Attributes


The existing Access Filter mechanism has been beefed up, allowing you to govern the whole system with User Attributes. You can now organize and manage filters by logical association rather than individuals and Groups’ Access Filters can be given default values—filter in bulk!





Legacy Feature Updates


The following feature has been deprecated and will only be available by enabling the associated legacy feature:





Upon updating to Looker 4.10 the following legacy features will be disabled by default. To reenable these features, check the associated box in the Legacy Features section of the admin panel.





Details on how to migrate off legacy features can be found here.




Features by Section


Dashboards and Visualizations




  • Selective assignment of series to y-axes. Create or combine any number of y-axes. Learn more.


  • Introduced a Funnel visualization type. Learn more.


  • Added Visual Drilling to the Labs menu, allowing it to be disabled if so desired. Learn more.

  • Table Visualizations in PDFs now show their full width.


Administration



Dialects




  • Google BigQuery. Introduced a Max Billing Gigabytes connection parameter which can be used to limit the size of any single query. It can also be tied to a User Attribute value.

  • Introducing support for Google Cloud PosgreSQL.

  • Introducing initial support for Google Spanner.


  • Snowflake. Updated to the latest driver.


General Tweaks and Bug Fixes



  • LookML and Development Environment.

  • Added missing quick help docs when the cursor is placed at the start of LookML blocks.

  • Fixed an issue with multi-line pasting in the new IDE.

  • Fixed an issue related to LookML validation when extension: required is used.

  • Fixed an issue related to from and view_name explore inheritance.

  • Fixed an issue where saving could hang.

  • Dialects.


  • Amazon Athena. Fixed an issue where the LookML generator did not see more than 100 tables across all databases.


  • Google BigQuery. Addressed an issue related to casting dimensions with datatype: date.


  • ⚡ Addressed inconsistent Look and Dashboard presentation in Spaces for users with access_data and see_user_dashboards and users with just see_user_dashboards. Users’ ability to query reports has not changed. Learn more.

  • Fixed an issue where the edit modal would occasionally not return data.

  • Looks & Dashboards can now be moved into personal Spaces.


  • ⚡ The Save button on the Explore page will now return you to the Explore Page instead of taking you to the Look.


  • ⚡ Filter values rather than filter names are now displayed in bold when the filters bar is collapsed.


  • ⚡ Deleted Looks will no longer appear in the Content Validator.

  • Hovering over relative timestamps now displays the exact time.

  • Added the ability to copy values from SQL Runner Tables & Drill Overlays.

  • Improved error messages related to malformed tiers and views without primary_key dimensions.

  • Addressed an issue where unchecking a default role on a user did not remove that role.

  • Fixed an issue where single value visualizations would be cut off in scheduled inline dashboards.

  • Increased the size of click targets on mobile Dashboards and Visualizations.

  • Fixed an issue related to MandrillApp and custom SMTP settings that occasionally caused scheduled Dashboard elements to render empty.

  • Fixed an issue where downloaded files could not be renamed.

  • Fixed an issue where Dashboards could not be renamed.

  • Fixed an issue where Looks and Dashboards could be missing from Spaces if a Space was moved into another.

  • Fixed an issue where sorts were not respected in Explores with a single, pivoted dimension.

  • Fixed an issue where Looker-generated SQL incorrectly escaped regex special characters in regex string comparison.

  • Fixed an issue where the Rebuild Derived Tables & Run option was hidden when persist_for: 0 seconds was declared.

13 replies

J
Userlevel 1

Any additional details about how they’re improved?

J
Userlevel 1

Is this link working? It says I don’t have access to the topic.

cutler
Rank
Userlevel 5
Badge

Sorry about that, @Jay_Stricks. Should be publicly available now. My fault.


We’re also working on getting some documentation on how user-specific timezones are changing, but the highlights are that you’ll have a way to control timezones on Dashboard elements by setting a Dashboard-level default. We’ll also be removing the different timezone behavior between embedded and non-embedded dashboards and forcing a query’s timezone to be hard coded to something like “America/Los Angeles” rather than a variable like “application_time_zone”.


All that is, of course, only going to apply if you are using user-specific timezones. Timezones are really hard, hopefully these changes will make them make more sense.

D

Could you alaborate on this change? I don’t see a difference in the Table output of SQL Runner.


Thanks.

sam
Rank
Userlevel 5
Badge

@Dominic_Ricard: When you are in SQL Runner, you can now click the gear on any column header to copy the values of that column into your clipboard. The interface looks like this:

D

Thanks. Is this going to be extended even more? It would be nice that a Cog appear on the upper left corner to copy the entire table, or allow to download results in table similar to the way it is possible in Explores.

sam
Rank
Userlevel 5
Badge

@Dominic_Ricard we do have the ability to Download using the gear in the upper right corner - are you looking for additional functionality?

D

Yes. The Modal only offer to Download ALL, causing the executed query to be relaunched.



The Modal is missing the option to Download “Results in Table”, similar to the Explore Download modal:


sam
Rank
Userlevel 5
Badge

Got it, makes sense. We don’t have anything planned here in the short term, but I’ll certainly bring your feedback to the product team for you!

J
Userlevel 1

Does this work only in the “Single Column PDF” download?

cutler
Rank
Userlevel 5
Badge

Yeah, those and the inline versions.

J
Userlevel 1

I do still see wide tables cut off when I download the PDF using this link. This is the main link available. Is there a plan to fix the tables in these PDFs?


Download Button


CutOff PDF on the Right

colin
Rank
Userlevel 4

Hi Jay,


At the moment, the adjusted-width tables will only apply the unstacked/single-column PDFs. We don’t have any immediate plans to address composed/as-seen PDFs, as the work would require dissecting and reassembling the dashboard in a way that would require some fairly significant re-architecture.


Hopefully something we can address in the future, but our focus is first and foremost on making sure that disassembled PDFs can handle these much larger data sets first.


Colin

Reply


Terms of Service