I am trying to understand Lakehouse and Spark Managed and Unmanaged tables as well as Shortcut Tables (is there even a real name for these). I have tried following the wonderful tutorial of data mozart: https://data-mozart.com/working-with-tables-in-microsoft-fabric-lakehouse-everything-you-need-to-kno.... But some things just dont add up:
1. When I click on a folder under /Files and say ‘Create Table’ a new table gets created this table is always managed. However when I delete the table the folder used to create it still remains. How was the table managed then? Does it mean that when i create a managed table from a folder in files all data is copied?
2. Using pyspark or pysql I was easily able to create a table in the /Tables section which is unmanaged/external. This table still resides in the /Tables section, I thought this section is supposed to be reserved for managed entitys exclusively.
3. If I click on a shortcut folder I can also create a managed table, how is this possible? Does the data get copied?
4. Is there any relation between building a table on a shortcut and the definitions of managed and unmanaged? so far it seems i was able to create both types of tables of a shortcut.
5. What is the use of an external table? It seems like i can always create a managed table off any type of data why would i ever create an unmanaged one? These freshly created managed tables can be deleted and the original data source seems to still remain.
