This tiny DataWeave lib has the only purpose to generate RAML datatypes by any data samples, written on different DSLs like JSON, XML, CSV, YML.

Unfortunately, for now, the author didn`t find a way to create an executable file in order to run it from CLI, therefore we will use Visual Code for this purpose.

1. Install Visual Studio Code and its DataWeave extension. You can find out how to do it here: https://developer.mulesoft.com/tutorials-and-howtos/dataweave/dataweave-extension-vscode-getting-started/
2. Import generate-raml-lib folder into Visual Studio Code.
3. Run mvn clean install
4. Open src/main/dw/Main.dwl file. There you can see the variable inpt that store file data by its path. This file has the structure you want to convert to RAML datatype. So there could be link to any file, but do not forget to change second parameter of the function readUrl according to the format you use to proceed with reading the file properly.
5. Function convert2raml(inpt, isOrNull, isRequired) has 3 parameters:
   • inpt depics data structure we want to convert to RAML datatype
   • isOrNull is responsible for making type of fields strict or united with nil (by default False)
   • isRequired is responsible for making fields required or not (by default True)
6. Click on the run button.

content of file input.json

{
    "mess1": "mes",
    "mess2": {
        "sub_mess": 2,
        "sub_mess_arr": [
            {
            "arr_elem": true
            }],
        "sub_mess_arr_prim": [ "dsds", "SDSF"]
    }
}

content of file Main.hs

%dw 2.0
output application/yaml
import convert2raml from GenRamlModule

var inpt = readUrl("classpath://input.json", "application/json")
---
convert2raml(inpt, true, false)

output

%YAML 1.2
---
type: object
properties:
 mess1?: string?
 mess2?:
   type: object
   properties:
     sub_mess?: number?
     sub_mess_arr?:
       type: array
       items:
         type: object
         properties:
           arr_elem?: boolean?
     sub_mess_arr_prim?:
       type: array
       items: string?