You are here: Workbench Basics > Running and Debugging Workspaces > Running a Batch Translation

Running a Batch Translation

The FME Workbench interface supports batch execution and batch script creation. For example, you can apply an existing workspace transformation to a large number of input files, and produce separate outputs for each.

Open your existing workspace and select Batch Deploy from the Run menu. The Create Batch Translation Wizard launches. Click Next to proceed through the following dialogs:

Note: If the Run > Batch Deploy option is disabled, it is because there is currently no source or destination dataset defined in the workspace.  

Select Source Data

Click the file browser button to select the dataset(s) to use for the batch translation.

To add multiple files or folders, click the advanced browser button .

Add Files: Opens the file browser. You can select individual files, or Ctrl + click to select multiple files.

Add Folders: Opens the file browser so you can select entire folders to add to the reader. All files that are in the specified format in those folders will be included.

Remove: Removes the highlighted selection.

Subfolders: If there are subfolders below the initial dataset location, check the box to include them.

Identical Schema Check this box if you know that all the files have the same schema. This is a time-saving function: there will be no difference in the workspace results. If you know the files have the same schema, FME will not have to perform an initial scan of all the files to determine their schema. Instead, FME will take the first file as being representative of the data model.

Set Destination Location For:

If you have specified more than one output destination, this pane appears for each destination.

Batch deploy this writer

Check this box, and enter or browse to the destination location.

Retain Source dataset basename in destination path

The output dataset name is always taken from the source name. However, this may not always make sense when using folder-based datasets. For example:

C:\myData\ShapeFiles\roads.shp

converted to MIF/MID and pointed to a MIFFiles folder would become

C:\myData\MIFFiles\ShapeFiles\roads.mif

This is because the "ShapeFiles" portion is treated by FME as the name of the source dataset, and is therefore included in the output.

To drop the source dataset name, uncheck this box. In the above example, the destination would now be:

C:\myData\MIFFiles\roads.mif

Note: Before using this option, you should be aware of the difference between file- and folder-based datasets.

If checked, you have the option to Type the optional suffix to append to your destination dataset (for example, a file-type suffix).

Batch Translation Run Options

Select when you want to run the batch translation:

  • Run Now: Execute the batch translation immediately, from within Workbench.
  • Save scripts to run outside of Workbench: Type or browse for the location of the batch translation file. Note that batch files consist of both a .tcl file and a .bat file. Both files are needed to run the batch translation.

    • To run the scripts outside of Workbench, go to the Windows DOS prompt and:

    • Type the name of the .bat file; or
    • Type fme tclfilename.tcl

Run/Create the Batch Translation

Click the Finish button to run or create the batch translation.

See Also