A free text input for the user.

questionary.text("What's your first name").ask()

A free text input for the user where the input is not shown but replaced with ***.

questionary.password("What's your secret?").ask()

A yes or no question. The user can either confirm or deny.

questionary.confirm("Are you amazed?").ask()

A list of items to select a choice from. The user can pick one option and confirm it.

questionary.select(
    "What do you want to do?",
    choices=[
        "Order a pizza",
        "Make a reservation",
        "Ask for opening hours"
    ]).ask()

A list of items to select a choice from. The user can pick one option using shortcuts and confirm it.

questionary.rawselect(
    "What do you want to do?",
    choices=[
        "Order a pizza",
        "Make a reservation",
        "Ask for opening hours"
    ]).ask()

A list of items to select multiple choices from. The user can pick none, one or multiple options and confirm the selection.

questionary.checkbox(
    'Select toppings',
    choices=[
        "foo",
        "bar",
        "bazz"
    ]).ask()

Text input with autocomplete help.

questionary.autocomplete(
    'Choose ant specie',
    choices=[
         'Camponotus pennsylvanicus',
         'Linepithema humile',
         'Eciton burchellii',
         "Atta colombica",
         'Polyergus lucidus',
         'Polyergus rufescens',
    ]).ask()

Sometimes it is helpfull to e.g. provide a command line flag to your app to skip any prompts, to avoid the need for an if around any question you can pass that flag when you create the question:

DISABLED = True

response = questionary.confirm("Are you amazed?").skip_if(DISABLED, default=True).ask()

If the condition (in this case DISABLED) is True, the question will be skipped and the default value gets returned, otherwise the user will be prompted as usual and the default value will be ignored.

Instead of creating questions using the python functions, you can also create them using a configuration dictionary.

questions = [
    {
        'type': 'text',
        'name': 'phone',
        'message': "What's your phone number",
    },
    {
        'type': 'confirm',
        'message': 'Do you want to continue?',
        'name': 'continue',
        'default': True,
    }
]

answers = prompt(questions)

The returned answers will be a dict containing the responses, e.g. {"phone": "0123123", "continue": False, ""}. The questions will be prompted one after another and prompt will return once all of them are answered.

You can customize all the colors used for the prompts. Every part of the prompt has an identifier, which you can use to style it. Let's create our own custom style:

from prompt_toolkit.styles import Style

custom_style_fancy = Style([
    ('qmark', 'fg:#673ab7 bold'),       # token in front of the question
    ('question', 'bold'),               # question text
    ('answer', 'fg:#f44336 bold'),      # submitted answer text behind the question
    ('pointer', 'fg:#673ab7 bold'),     # pointer used in select and checkbox prompts
    ('highlighted', 'fg:#673ab7 bold'), # pointed-at choice in select and checkbox prompts
    ('selected', 'fg:#cc5454'),         # style for a selected item of a checkbox
    ('separator', 'fg:#cc5454'),        # separator in lists
    ('instruction', ''),                # user instructions for select, rawselect, checkbox
    ('text', ''),                       # plain text
    ('disabled', 'fg:#858585 italic')   # disabled choices for select and checkbox prompts
])

To use our custom style, we need to pass it to the question type:

questionary.text("What's your phone number", style=custom_style_fancy).ask()

It is also possible to use a list of token tuples as a Choice title. This example assumes there is a style token named bold in the custom style you are using:

Choice(
    title=[
        ('class:text', 'plain text '),
        ('class:bold', 'bold text')
    ]
)

As you can see it is possible to use custom style tokens for this purpose as well. Note that Choices with token tuple titles will not be styled by the selected or highlighted tokens. If not provided, the value of the Choice will be the text concatenated ('plain text bold text' in the above example).

Bug fix release.

• Fixed .ask_async for questions on prompt_toolkit==2.*. Added tests for it.

Feature release.

• Added support for prompt_toolkit 3
• All tests will be run against prompt_toolkit 2 and 3
• Removed support for python 3.5 (prompt_toolkit 3 does not support that anymore)

Feature release.

• Added additional question type autocomplete
• Allow pointer and highlight in select question type

Feature release.

• Add additional options to style checkboxe and select prompts https://github.com/tmbo/questionary/pull/14

Bug fix release.

• Fixed compatibility with python 3.5.2 by removing Type annotation (this time for real)

Feature release.

• Allow a user to pass in a validator as an instance https://github.com/tmbo/questionary/pull/10

Bug fix release.

• Fixed compatibility with python 3.5.2 by removing Type annotation

Feature release.

• Added skip_if to questions to allow skipping questions using a flag

Bug fix release.

• Fixed odd behaviour if select is created without providing any choices instead, we will raise a ValueError now. (#6)

Bug fix release, adding some convenience shortcuts.

• Added shortcut keys j (move down^ the list) and k (move up) to the prompts select and checkbox (fixes #2)
• Fixed unclosed file handle in setup.py
• Fixed unecessary empty lines moving selections to far down (fixes #3)

Initial public release of the library

• Added python interface
• Added dict style question creation
• Improved the documentation
• More tests and automatic travis test execution