# Workflows with a Dynamic Destination Path

Workflows are designed to orchestrate an automated and iterative execution of ordered tasks. When designing a workflow to orchestrate a process, a source object is required, that can provide access to a directory containing a bulk load of source files of the same layout and format. A workflow is designed, such that each source file in that particular directory goes through the same data processing.

This requires a source file to be replaced in the run-time until all source files (of the same layout and format) within a directory are processed and loaded to respective destination files. Now, this further involves the associated destination files to be replaced in the run-time as well, hence making it essential to provide a dynamic destination path. For example, a source directory containing 100 Excel source files in this workflow will create 100 destination files after going through the process.

In this document, we will learn to create workflows with a dynamic destination path that directs Astera to create a unique destination file for each source file.

In Astera, a dynamic destination path is achieved in two steps:

1. Parameterizing the dataflow through a *Variables* object.
2. Providing a run-time destination path through an [*Expression Transformation*](https://documentation.astera.com/astera-data-stack-v7/dataflows/transformations/expression-transformation) object in the workflow.

## **Creating a Workflow**

In this case, we are orchestrating a simple dataflow that contains customer account details coming in through an [*Excel Workbook Source*](https://documentation.astera.com/v/astera-data-stack-v7/dataflows/sources/excel-workbook-source). The source data is passed through an [*Aggregate Transformation*](https://documentation.astera.com/v/astera-data-stack-v7/dataflows/transformations/aggregate-transformation) and loaded to an[ *Excel Workbook Destination*](https://documentation.astera.com/v/astera-data-stack-v7/dataflows/destinations/excel-workbook-destination).

![](https://627607815-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2F6xzBT0roYJkfVS5klkLl%2Fuploads%2FYviSrbwBWzkPucIWBgN7%2F0.png?alt=media)

Now, there are 100 Excel source files in a directory containing customer account details of 100 customers. Each of these *Customer\_Accounts* files must go through the same aggregate transformation and be loaded to a distinctive *Excel Workbook Destination* file.

For this purpose, we will create a workflow.

1. Open a new workflow by going to *File > New > Workflow*.

![](https://627607815-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2F6xzBT0roYJkfVS5klkLl%2Fuploads%2FhC7SX47Ohn3SV9eRpdYB%2F1.png?alt=media)

2. To provide access to the local source directory containing 100 customer account files, go to *Toolbox > Sources > File System Item Source* and drag and drop the *File System Item Source* object onto the workflow designer.
3. Right-click on the object’s header and select *Properties* from the context menu. A properties window will open.

![](https://627607815-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2F6xzBT0roYJkfVS5klkLl%2Fuploads%2FB1gS6vMGTSFs2bzOxOc3%2F2.png?alt=media)

4. Provide the local directory path in *File Location*. Apply a filter .xls and click *OK*.

![](https://627607815-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2F6xzBT0roYJkfVS5klkLl%2Fuploads%2FiqQek8lwUGf9SrfokVh1%2F3.png?alt=media)

5. Right-click on the *File System Item Source* object’s header and select *Loop* from the context menu. This will direct Astera to run the task in a loop until all the source files have been processed through it.

![](https://627607815-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2F6xzBT0roYJkfVS5klkLl%2Fuploads%2FGOF7sjALwLnn02qocwxa%2F4.png?alt=media)

6. Go to *Toolbox > Workflow Tasks > Run Dataflow* and drag and drop the *Run Dataflow* object onto the workflow designer. Map the header output port of the *File System Item Source* onto the header input port of the *Run Dataflow* object.

![](https://627607815-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2F6xzBT0roYJkfVS5klkLl%2Fuploads%2FeXE8KBZI8RJKGU6xpp8f%2F5.png?alt=media)

7. Right-click on the *Run Dataflow* object’s header and select *Properties* from the context menu. Here, provide the *Job Info* path for the dataflow that is being orchestrated and click *OK*.

![](https://627607815-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2F6xzBT0roYJkfVS5klkLl%2Fuploads%2Fz5FAMSRZQDimWg38hizI%2F6.png?alt=media)

8. Go to *Toolbox > Transformation > Constant Value*, and drag and drop the *Constant Value* transformation object onto the workflow designer. Double-click on its header, a *Constant Value Map Properties* window will open.

![](https://627607815-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2F6xzBT0roYJkfVS5klkLl%2Fuploads%2FNSMT0sXorsOTQ0HRsp23%2F7.png?alt=media)

9. Paste the path to the folder where Astera will load the destination files, in the *Constant Value* box as shown below. Click *OK*.\\

![](https://627607815-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2F6xzBT0roYJkfVS5klkLl%2Fuploads%2Fbuzd1KfmRKlqfIWUlRoy%2F8.png?alt=media)

10. Go to *Toolbox > Transformation > Expression*, and drag and drop the *Expression* transformation object onto the workflow designer. Double-click on its header and a *Layout Builder* screen will open.

![](https://627607815-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2F6xzBT0roYJkfVS5klkLl%2Fuploads%2Fy4VxKaDeDVUniqq9Mo3L%2F9.png?alt=media)

11. On this *Layout Builder* screen, create four new fields and set the data type accordingly.
    * *FullPath*
    * *DestinationDirectory*
    * *FileName*
    * *DestinationFilePath*

![](https://627607815-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2F6xzBT0roYJkfVS5klkLl%2Fuploads%2FzLDRQxFiv4xq7pJUso3Q%2F10.png?alt=media)

Click *OK.*

12. Define field mappings as follows:
    * *FullPath* field under *FileSystem* source object onto the *FullPath* field under *Expression* transformation object.
    * *FullPath* field under *FileSystem* source objects onto the *Excel\_Source* field under *RunDataflow* task object.
    * *Value* field under *Constant Value* transformation object onto the *DestinationDirectory* field under *Expression* transformation object.
    * *DestinationFilePath* field under *Expression* transformation object onto the *Excel\_Destination* field under *RunDataflow* task object.

![](https://627607815-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2F6xzBT0roYJkfVS5klkLl%2Fuploads%2FZxcUU49mHJtxEkk6DNYp%2F11.png?alt=media)

13. Double-click on the *Expression* transformation object’s header, to open its *Layout Builder* screen. Write the following expressions for these fields:
    * *FileName*: LSplit(RSplit(FullPath,2,”\”,0),2,”.”,0)
    * *DestinationFilePath*: DestinationDirectory+FileName+”.xls”

![](https://627607815-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2F6xzBT0roYJkfVS5klkLl%2Fuploads%2Fay56JhxiQn4QVfH2v1rQ%2F12.png?alt=media)

Click *OK.* A dynamic destination file path has been created.

14. Click on the *Start Workflow* icon located in the toolbar at the top and execute this workflow.

![](https://627607815-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2F6xzBT0roYJkfVS5klkLl%2Fuploads%2FVMEGV1SP4cR8PRYnvZgy%2F13.png?alt=media)

This will create a unique destination file for each source file.