I have one input and only one output like below

enter image description here

There doesn't have any complication to run but I have nearly 1k+ repeated tasks.

I created batch file but in that it showing only FME WB path but I need input output paths.

  • I think we need a bit more information here. How did you create the batch file and how are you trying to run it? If there are more tasks than expected, it sounds like you might be running the batch once per incoming feature. On the command line you can specify input and output paths using --SourceDataset_XXXX "D:\RPR\......" - where XXXX is usually the short format name. But you could look in Workbench User Parameters to find what those names are exactly. Jul 7 at 21:02
  • exporting batch file from run>batch deploy option. but in that i'm getting only WB path not parameterst. Jul 8 at 3:53

OK, so I think there are a few things we can help with here. I mention the Batch Deploy option first because it's what you asked about, but I think there are better methods that I can mention afterwards.

With the Batch Deploy option, you get a bat file and a tcl file. You run the bat file and it runs the tcl file in FME, triggering multiple translations. The trick is to pick all of your source datasets when you create the batch deploy files:

enter image description here

When you do that the tcl will include all of them:

enter image description here

...and the script should run as you want. You can check this box to get a separate output file per input file:

enter image description here

...but, it still depends on the source dataset. If the source is Shapefile (for example) then each shp file counts as a layer in a larger dataset (the folder) so in my test I got one GML file with multiple layers (one for each shp). With your Geopackage, you should get a new output for each input geopackage. With the output being a Geodatabase it becomes slightly more uncertain, but I think it should still work.

Of course, sometimes you want to use all the files in a folder, without having to define them in advance. So instead of picking all of the source files when you create the Batch Deploy setup, you pick a folder instead:

enter image description here

Now FME will read all the files of a certain type that exist in a folder (or subfolder) without you having to pick them separately.

But as I said, the Batch Deploy isn't really the method of choice any more. Instead you would go for a two-workspace setup.

The first one of these is your workspace for processing the data as normal. The second one use a WorkspaceRunner transformer to run that workspace:

enter image description here

As you can see, that allows you to pick the published parameters.

Now you just have to send that WorkspaceRunner transformer a single feature for each input dataset. Some users have a list of files to translate in a CSV file or Excel spreadsheet. If you add a reader for that and connect it to the WorkspaceRunner you can then set the file to translate in the published parameter from whatever CSV/Excel attribute stores that information.

The other way of doing this is to use a Directory and File Pathnames reader:

enter image description here

Here I'm reading a list of all of the DWG files in a folder. I will send that list to the WorkspaceRunner and it will run the sub-workspace once per input.

So that's how we generally recommend this is done now. I hope you find this all useful!

  • 1
    thank you so much mark ireland. Jul 14 at 7:02
  • 1
    @RakeshReddyPosham accept answers with the checkbox
    – BERA
    Sep 16 at 5:18

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.