Variables are a powerful way to customize your dashboards' filters (and much more).
You can use it for advanced filtering. For example, you can use it to create a table displaying sales of a given year N and the year before N-1, with a filter to let the user choose year N.
You can also use it to create related dashboards. For example, a global dashboard can have a table showing all the sales per business unit, and links in each row to open a dashboard focused on a single business unit.
If you need to create not-so-simple dashboards, it's a must-read.
When you use Storages in a production environment, once your project is set up, it's important to lock the models of these Storages.
When a Storage model is locked, if an operation tries to modify a column's type (e.g. change its type from datetime to string), it will be refused. Let's imagine you import CSV files in your workflow, and one day, someone types the text "this date is missing" in a datetime column. If your models are not locked, this may break your dashboards.
With the new version, in the datasources list, a new icon shows if a Storage model is locked or not, so you can quickly check that all your production models are locked:
There is a new automation to import files from S3 in a Serenytics storage. You can use it import a single CSV file from a bucket, a single parquet file or a multi-files parquet.
For the last months, we have been working on a new engine in our software. Our goal was to have a more modular application and to use a more modern tech-stack (we moved from AngularJS to React/Redux).
In the near future, this will also allow us to deploy new functionalities more easily (this new version does not bring any new functionality yet, just a new engine).
We just deployed this new version ! You should see... nothing new. But under the hood, the new engine will render your dashboards.
We have hundreds of tests running to ensure this new version has not broken anything. But as this is a very large evolution, let us know if you have any issue/feedback (firstname.lastname@example.org).
Until now, the formulas defined on joined datasources were not available within the join itself. You had to clone these formulas to reproduce them in the join. Because of these duplicates, it was painful to maintain large projects with many joins.
When the new option "include datasources formulas" is enabled (in a join datasource configuration), a formula created in a joined datasource is available in the join datasource. You can use it within new formulas, directly in your dashboards, or in your data-preparation steps.
This option is not enabled by default on existing joins as it might create name collisions between your existing formulas on the join and the joined datasources formulas.
You can now add buttons to your dashboard in a simple way. Buttons are a new type of widget. Each button can trigger either an automation or a full-datasource download.
If you trigger a Python script from a dashboard button, all the dashboard variables (including the new input fields) are passed to the script. With this option, you can create rich web applications.