RPA.Dialogs

The Dialogs library provides a way to ask for user input during executions through HTML forms. Form elements can be built with library keywords or they can be defined in a static JSON file.

How the library works

The main keyword of the library is Request Response which works as follows:

  1. It starts an HTTP server in the background
  2. The HTML form is generated either according to a JSON file or the keywords called during the task
  3. It opens a browser and shows the created form (The browser is opened with the Open Available Browser keyword from the RPA.Browser.Selenium library)
  4. Once the form is filled and submitted by the user, the server will process the response and extract the field values, which in turn are returned by the keyword
  5. In the end, the browser is closed and the HTTP server is stopped

Request Response can be invoked in two ways:

  1. Without any parameters. This means that form shown is the one created by other library keywords. If no form elements have been added with keywords then the form will contain just one submit button. Form building must be started with the keyword Create Form.
  2. Giving a path to a JSON file (using the parameter formspec) which specifies the elements that form should include.

The keyword has optional parameters to specify form window width and height. The default size is 600px wide and 1000px high.

Setting library arguments

Library has arguments server_port and stylesheet. The server_port argument takes integer value, which defines port where HTTP server will be run. By default port is 8105. The stylesheet can be used to point CSS file, which will be used to modify style of form, which is shown to the user. Defaults to built-in Robocorp stylesheet.

Supported element types

As a bare minimum, the form is displayed with a submit button when the Request Response keyword is called.

The supported input elements and their corresponding HTML tags are:

  • form (<form>)
  • title (<h3>)
  • text (<p>)
  • radiobutton (<input type='radio'>)
  • checkbox (<input type='checkbox'>)
  • dropdown (<select>)
  • textarea (<textarea>)
  • textinput (<input type='text'>)
  • fileinput (<input type='file'>)
  • hiddeninput (<input type='hidden'>)
  • submit (<input type='submit'>)

About file types

The Add File Input keyword has parameter filetypes. Parameter sets filter for file types that can be uploaded via element. Parameter can be set to filetypes=${EMPTY} to accept all files. Multiple types are separated with comma filetypes=image/jpeg,image/png.

Some common filetypes:

  • image/* (all image types)
  • audio/* (all audio types)
  • video/* (all video types)
  • application/pdf (PDFs)
  • application/vnd.ms-excel (.xls, .xlsx)

The list of all possible MIME-types.

Examples

Robot Framework

Examples of creating forms through keywords and a JSON file:

*** Settings ***
Library    RPA.Dialogs

*** Keywords ***
Ask Question From User By Form Built With Keywords
    Create Form     questions
    Add Text Input  label=What is your name?  name=username
    &{response}=    Request Response
    Log             Username is "${response}[username]"

Ask Question From User By Form Specified With JSON
    &{response}=    Request Response  /path/to/myform.json
    Log             Username is "${response}[username]"

Python

The library can also be used inside Python:

from RPA.Dialogs import Dialogs

def ask_question_from_user(question, attribute):
    d = Dialogs()
    d.create_form('questions')
    d.add_text_input(label=question, name=attribute)
    response = request_response()
    return response

response = ask_question_from_user('What is your name ?', 'username')
print(f"Username is '{response['username']}'")