TMDL View

One of the most useful applications of TMDL is copying a table from one PBIX file to another. While it is possible to copy the query or M code directly, that approach does not carry over calculated columns, column formatting, or other semantic model metadata. TMDL preserves these elements, making it much easier to replicate tables across models.

It can also be used to change the storage mode of a table from import to Direct Query although you have to ensure that none of the applied steps breaks query folding as the switch will not take affect.

TMDL is also useful for creating a backup of a semantic model's state. Before making changes, the entire semantic model can be scripted to TMDL. If measures, columns, or other semantic model objects are accidentally deleted, the previously generated TMDL script can be reapplied to restore them.

However, TMDL only affects the semantic layer. It cannot recover deleted report pages and visuals.

Hi powerbiexpert22 

TMDL View is useful when you need to manage and modify semantic model metadata at scale. Instead of manually creating or editing measures, relationships and calculation groups you can define them as code using TMDL. 

Common scenarios include version control with Git, deploying model changes across environments, editing measures, collaborating with developers, and tracking model changes in source control.

https://youtu.be/WGMIoteHMs4?si=VTXOG_5mW30Zn5bU

https://youtu.be/lnuaD4gIBe4?si=UYVx71DbgIT35bsi

https://youtube.com/playlist?list=PLL5m0piY28pjojhP-nKKXaXieiHSwchvp&si=2nZszEGGC