Analytics extensions connections, Connection Dialog v2 for Connector SDK, and Hyper API monthly updates
As part of the Tableau Developer Program, we host monthly Sprint Demos, where members of the engineering team demo what they have been working on for our developer community. It is a chance for you to meet the team, be informed and inspired by upcoming features, ask questions, and give feedback. This month, we had product managers and engineers from the Advanced Analytics, Connector SDK, and Hyper API teams.
Multiple connections for analytics extensions: All on the same server
In Tableau 2020.2, we are adding multiple connections for the Analytics Extensions API, which was formerly known as "External Services." The existing Python, R, and MATLAB External Services for Tableau will now be known as “analytics extensions.” Currently, Tableau users can add one connection for each instance of Tableau Server. In this release, users will be able to add one connection for every site, with the ability to connect to an unlimited number of sites. We are also adding the ability to enable or disable analytics extensions at the Server level, but also at the Site level by going to the Settings menu. (By default, analytics extensions are disabled.) Tableau Server admins can configure analytics extensions in the Site settings and Server settings user interface in Tableau Server. You can also enable and configure the connection in the Tableau REST API. Please be aware that Tableau Services Manager (TSM) will no longer be used to configure or enable analytics extensions for workbooks, and all related TSM commands will be deprecated, beginning with Tableau 2020.2.
We have some additional exciting news: R, Python, and Analytics Extensions API users will be able to share the same Tableau Server!
Connector SDK: Connection Dialog v2
Last year, we released the Connector SDK that allows developers to build native-looking connectors like the Databricks Connector. In Tableau 2020.2, we are adding v2 of the Connection Dialog, a new feature which enables a more data-driven dialog for connector plugins. Currently, developers are limited when building the connection dialog, which prompts a person to enter details that will be used to set the connection. With the latest version of Connection Dialog, developers can easily add new custom fields to the dialog or ask for any other information they need to help people connect to a data source. For example, you can find out a user's database region by adding this field to the connection dialog. Additionally, custom fields provide some control over the metadata hierarchy elements—Database, Schema, and Table—both in the connection dialog and in the schema viewer, which someone would see after a connection is established. Here is one use case: People using a connector will be able to see the fields for username and password—only if they selected Username and Password in the Authentication dropdown.
To preview v2 of the Connection Dialog, go to the manifest.xml and replace
Hyper API updates
Let’s take a look back at the Hyper API journey. Six months ago, we released the Hyper API to the public. In the most recent updates, we added new features to make it easier to insert spatial data, giving into Hyper extract, and giving developers more control over new settings. Let’s start with the ability to insert data into columns based on SQL expressions. This feature is especially useful for developers who want to insert spatial data into their Hyper extracts. With previous versions of the API, while this was technically possible, it was complicated and required the use of temporary tables. No need for this workaround anymore, with the latest version of the Hyper API providing a faster and more straightforward way to insert spatial data into Hyper databases. This feature is also exciting because it enables developers to push down calculations to Hyper. Now, you can use SQL calculations to specify the columns to be evaluated directly in Hyper when data is inserted. Functions built into Hyper, like interpreting well-known text for geometry data, become especially useful to developers, while also creating the potential to improve performance. We are also introducing new settings that can be passed to the Hyper Process and Connection constructors. These settings give you control over:
- Internal communication – Switch the communication protocol between Hyper and the Hyper API to be TCP or to configure which directory to use for domain sockets.
- Logging behavior – Control how Hyper writes its activity logs. For example, logs can be disabled or limited in size.
- Date and time parsing – Control how Hyper handles date and time. For example, developers can specify the time zone to be used by Hyper or what format to expect when interpreting date strings.
Check out the details on our Help page. This Hyper API release also comes with some quality of life improvements that you can find on the release page. Upgrade today to the latest Hyper API version to take advantage of all these new and improved features.
#DataDev Site Challenges
During this month’s Sprint Demos, we also announced the #DataDev Site Challenge. One of the biggest benefits of joining the Tableau Developer Program is a free, personal developer site. With this site, you can learn and test Tableau APIs on various projects. We provide developers a safe space to experiment and test, so you don't have to stress about deleting various tests on your own production sites. The #DataDev Site Challenge is a way to introduce and encourage our developer community to do more with their personal site, discovering ways to use it and having a fun along the way. Obviously, we want to reward developers for achieving site challenges with exclusive #DataDev swag packs. Over the course of three months, there will be 18 mini-challenges total, with all challenges divided into three skill levels. All the challenges are organized into one of three categories: Extensibility, Automation/Integration, and Embedding. You'll have two weeks to solve one set of mini-challenges or more, if you have time. At the end of the two weeks, we will host a how-to webinar to spotlight what one of you from the #DataDev community has built before publishing the next set of challenges. We officially start the mini-challenges on April 6, 2020. All the challenges will be published on the main page. Join the Tableau Developer Program to participate in the site challenge and shape the future of Tableau. We can't wait to see what you create.