Looker's product roadmap is developed from our own vision and experience as data analysts and in response to customer feedback. We strive to build an analytics platform that turns any organization into a data-driven organization. Looker v2.2 is our latest release and adds important new capabilities for:
A Looker implementation consists of a 3-step process. The first step is to create a connection to a SQL dialect database. The second step is to run the model generator to create the model files. The final step is to further develop the generated model to reflect the analytical needs of your business. All of these tasks are done in the browser, using Looker’s IDE. With v2.2 we’ve made it even more intuitive and robust than before.
Please enjoy a walkthrough of the new IDE in detail.
Looker automatically manages the workflows for committing, pulling, and pushing changes in model files. Looker’s IDE is integrated with Git for version control and production staging. You can edit a private copy of the project files in developer mode in Looker before pushing changes to the production environment. Looker v2.2 makes it easy to connect your project to a new GitHub repo. The overall git interface has been streamlined, added git controls to easily run git commands on both the project and file levels.
LookML is a human-readable language for modeling data structure, which allows Looker to automatically generate SQL queries. The new IDE makes it even easier to quickly create new LookML files of all types by preloading the files with the necessary templated syntax to customize.
In the development lifecycle it’s useful to look at existing LookML patterns as a reference for building new model elements. The new global search functionality provides an easy way to search across your repository and find syntax examples or a particular custom metric.
The replace functionality allows you to easily update field names throughout your project. You can easily select all changes or use the granular controls to limit the replacement logic. Regular expressions are also supported for creating more complex search and replace patterns.
Version 2.2 makes modeling joins even easier without writing SQL, by adding the foreign_key parameter. With foreign_key, you specify the Looker dimension you want to join to another table, and Looker handles the rest. For most needs, foreign_key is now the primary method for declaring joins (you can still do any kind of custom join if you need to). The following declaration is all it takes to join, for example, an inventory items view into an order-items base view:
Other join methods (sql_on and sql parameters) are still supported for modeling more complex relationships.
Simple plotting of data on fixed maps is very useful, but our customers asked for more flexibility as geo data has become so important to their analysis. The enhanced mapping capabilities have expanded beyond basic US and World maps to include custom map visualizations. Custom maps can be created by specifying a TopoJSON file to provide the custom geo shape. The TopoJSON library can be used to create graphically rich maps like a county circle map, a custom colored world map or a topographical map.
Join us for Office Hours: Looker What’s New on Tuesday, July 1 at 10am PST / 12pm CT / 1pm ET
Hear from our product team about our newest release v2.2. We’ll cover the following topics:
We’ll open up the call to take your questions.