Maximize Your Workspace Flexibility: Introducing Dynamic Workflows
Have you ever wished you could avoid tweaking your workspace every time your source dataset’s schema changed? Or save time by using a single workspace on multiple datasets with varying schemas? Or perhaps even perform the same operation on dozens of input feature types with one workspace? Well dream no more, because your hopes have turned into reality!
FME 2010 introduces dynamic workflows, launching you into a whole new realm of flexibility. First previewed in the Generic Writer last year, dynamic schema capabilities are now available across all formats, helping you break your dependence on source and destination schemas.
Dynamic workflows allow you to use the source schema or a schema template to determine the destination schema at runtime, creating workspaces that are schema-independent. This means that you can design workspaces to perform translations on any dataset, regardless of its schema, and also include transformations that you wish to perform on your source data.
This is a huge time saver in a wide variety of scenarios. For example, if your source dataset’s schema changes, you no longer need to make any modifications to your workspace before running it. You can also create a single workspace to perform quick translations and even transformations on multiple source datasets whose schema is unpredictable. Additionally, you can build one workspace to perform the same translation or transformation task on multiple feature input types, such as tables within a database.
In all of these scenarios, building your workspace with a dynamic workflow saves you time in both the design and the maintenance of your work. Even better, it cuts down on the number of workspaces you need to create in the first place.
Plus, you can gain further control within your dynamic workspace by setting the Feature Types to Read parameter. This restricts which feature types are read into the workspace, regardless of any other changes to the source data’s schema. Additionally, you can un-pair the destination schema from the source schema and instead use a reader resource to reference a data model template or a table defining a data model which the destination’s schema will adhere to.
The possibilities are nearly limitless, especially when you think about dynamic workflows within the FME Server environment. Want to learn more about this flexible combination? Check out the article on the next page. Want to dig deeper into dynamic workflows? Check out the tech brief at safe.local/DynamicWorkflows.
