This content is also available at learn.palantir.com ↗ and is presented here for accessibility purposes.
When the data in your backing dataset updates (if it is, for example, the target of a build schedule), the changes are automatically synchronized to the object storage service. New rows become new objects in the registered object storage service table and updated values are transmitted.
However if the schema for the backing dataset updates then additional work may be required to keep the object storage service synchronized with the dataset. This is because the schema of the backing dataset will no longer match the schema of the table that is registered with the object storage service. The remedial actions necessary will vary depending on the type of update, but we will explore these as we go through this exercise.
It's important to note though that most updates to a backing dataset will not necessitate any additional work. The most common type of dataset update by far is one that updates the data but not the schema; datasets building regularly on a schedule will most likely update in this manner. But do bear in mind that conditional formatting rules on properties may still need to be updated. For example, if the values in the "priority" column for our flight alert dataset changed from ["Low", "Medium", "High"] to ["low", "medium", "high"] then we would need to update our conditional formatting rules for the Priority property of our Flight Alert object type to match the new lowercase values in the backing dataset.
In summary, data value and row additions don't usually require much action beyond updating any metadata configurations. Now let's take a look at how to handle some more complex updates.