Question

How can I batch a FME workspace without a workspace-runner?

  • 30 August 2021
  • 2 replies
  • 10 views

Dear FME community,

I have a question concerning the FME workbench. I have a pretty complex workspace where I need two different inputs, which have to be compared with each other. Both inputs require several hundred files. Now it is important that input a starts with file 1 (not with any other file) and input b also starts with the first file. Both inputs have to select one file after another in the correct order. The files are named after a certain scheme so that corresponding have the same naming parts.

Manually I could just select the files, but since there are hundreds of processes, I would prefer a batch method. I tried a workspace runner, but this only can handle one input.

Is there any process that can handle this?

 

I would appreciate any hint concerning this matter.

Thank you so much in advance.


2 replies

Badge +22

I'm confused about your workspace Runner comment.

 

The workspaceRunner can handle as many inputs as you have published parameters on your child workspace.

 

The parent workspace can use the Directory and File Pathnames reader to create the trigger features. You can either point it at one input and then use FilenamePartExtractor and AttributeCreator to build the corresponding file in input two. Or you can use two readers and use a featureMerger or equivalent to join the two input streams prior to the WorkspaceRunner.

 

Badge +20

You should Join/Merge the data that you want to compare so that one input trigger (of WorkspaceRunner) has both the filenames in two different attributes. Then use WorkspaceRunner to sent the two attributes to two Parameters in the workspace that is being run. Then each Parameter triggers a different Reader.

You give very little detail as to how the workspaces work, but there are many ways you can do a comparison.

Reply