Running the CSFE standalone - ralfw/csfe GitHub Wiki

To run the CSFE standalone as a program it's easiest to co-locate it with the flow. Put all the CSFE's files into the flow directory. Currently that's mainly:

/charcount
  csfe.exe
  CLAP.dll		
  System.ValueTuple.dll

  /split
    ...
  ...

Also put the data flow description in the flow directory, e.g.

/charcount
  flow.txt

  csfe.exe
  CLAP.dll		
  System.ValueTuple.dll

  /split
    ...
  ...

The input to be processes by a flow can reside anywhere, but if you like, you can put it in the flow's /input directory:

/charcount
  flow.txt

  csfe.exe
  CLAP.dll		
  System.ValueTuple.dll

  /input
    poem.txt

  /split
    ...
  ...

Now you're ready to run the CSFE in standalone mode by calling it on the command line. Open a terminal window on the flow directory and run this command:

mono csfe.exe -s=flow.txt -i=input

Usage

The CSFE is a .NET/Mono application, hence you need to start it with mono as the executable on Linux/Mac OSX; on Windows you can start csfe.exe directly. That also means you need to ensure the .NET/Mono framework is installed on any machine where the CSFE is supposed to be executed. Download Mono here for Linux/OSX or the .NET Framework here for Windows.

Please note: Input files will not be deleted/moved by the CSFE. Existing output files when starting the engine will also not be deleted.

Command line parameters

• -s or -source: Name of file containing the flow description.
• -i or -input: Name of file to process or of directory containing files to process.
• -p or -path: Path to data flow root directory. This is an optional parameter. If it's missing the current directory is assumed. The flow root directory is the directory where the CSFE looks for services and places the flow's input/output directories.