Home

Awesome

Latest PyPI version Supported Python versions Total downloads

Travis Build Status Appveyor Build Status

AutoPy Introduction and Tutorial

Introduction

AutoPy is a simple, cross-platform GUI automation library for Python. It includes functions for controlling the keyboard and mouse, finding colors and bitmaps on-screen, and displaying alerts.

Currently supported on macOS, Windows, and X11 with the XTest extension.

Getting Started

Requirements

Installation

First, see if a binary wheel is available for your machine by running:

$ pip install -U autopy

If that fails, install rustup and then run:

$ rustup default nightly-2019-10-05
$ pip install -U setuptools-rust
$ pip install -U autopy

Another option is to build from the latest source on the GitHub repository:

$ git clone git://github.com/autopilot-rs/autopy-rs.git
$ cd autopy
$ make
$ make install

Note: AutoPy currently requires the 2019-10-05 Rust nightly in order to build from source. This is to maintain compatibility with an older version of PyO3, as the latest version has dropped Python 2 support. Python 2 support will likely be dropped from AutoPy as well sometime later this year, depending on how necessary it is to upgrade to a more recent version of PyO3 or Rust. In the meantime, it may be necessary to install the required nightly via the following when building locally:

rustup install nightly 2019-10-05 --force

This is due to rustup complaining that it doesn't include certain components such as rustfmt.

Additional instructions for installing from source on Windows are available here.

Hello World

The following is the source for a "hello world" script in autopy. Running this code will cause an alert dialog to appear on every major platform:

import autopy


def hello_world():
    autopy.alert.alert("Hello, world")
hello_world()

Cross platform alerts

Tutorials

Controlling the Mouse

AutoPy includes a number of functions for controlling the mouse. For a full list, consult the API Reference. E.g., to immediately "teleport" the mouse to the top left corner of the screen:

>>> import autopy
>>> autopy.mouse.move(0, 0)

To move the mouse a bit more realistically, we could use:

>>> import autopy
>>> autopy.mouse.smooth_move(0, 0)

Even better, we could write our own function to move the mouse across the screen as a sine wave:

import autopy
import math
import time
import random
import sys

TWO_PI = math.pi * 2.0


def sine_mouse_wave():
    """
    Moves the mouse in a sine wave from the left edge of
    the screen to the right.
    """
    width, height = autopy.screen.size()
    height /= 2
    height -= 10  # Stay in the screen bounds.

    for x in range(int(width)):
        y = int(height * math.sin((TWO_PI * x) / width) + height)
        autopy.mouse.move(x, y)
        time.sleep(random.uniform(0.001, 0.003))


sine_mouse_wave()

<a href="https://www.autopy.org/documentation/sine-wave"><img src="https://github.com/autopilot-rs/autopy/raw/gh-pages/sine-move-mouse-thumbnail.jpg" alt="Demonstration video"/></a>

Controlling the Keyboard

The following will enter the keys from the string "Hello, world!" in the currently focused input at 100 WPM:

import autopy


autopy.key.type_string("Hello, world!", wpm=100)

Alternatively, individual keys can be entered using the following:

import autopy


autopy.key.tap(autopy.key.Code.TAB, [autopy.key.Modifier.META])
autopy.key.tap("w", [autopy.key.Modifier.META])

Working with Bitmaps

All of autopy's bitmap routines can be found in the module autopy.bitmap. A useful way to explore autopy is to use Python's built-in help() function, for example in help(autopy.bitmap.Bitmap). AutoPy's functions are documented with descriptive docstrings, so this should show a nice overview.

>>> import autopy
>>> autopy.bitmap.capture_screen()
<Bitmap object at 0x12278>

This takes a screenshot of the main screen, copies it to a bitmap, displays its memory address, and then immediately destroys it. Let's do something more useful, like look at its pixel data:

>>> import autopy
>>> autopy.bitmap.capture_screen().get_color(0, 0)
15921906

AutoPy uses a coordinate system with its origin starting at the top-left, so this should return the color of pixel at the top-left corner of the screen. The number shown looks a bit unrecognizable, but we can format it with Python's built-in hex function:

>>> import autopy
>>> hex(autopy.bitmap.capture_screen().get_color(0, 0))
'0xF2F2F2'

Alternatively, we can use:

>>> import autopy
>>> autopy.color.hex_to_rgb(autopy.screen.get_color(0, 0))
(242, 242, 242)

which converts that hex value to a tuple of (r, g, b) values. (Note that autopy.screen.get_color(), used here, is merely a more convenient and efficient version of autopy.bitmap.capture_screen().get_color().)

To save the screen capture to a file, we can use:

>>> import autopy
>>> autopy.bitmap.capture_screen().save('screengrab.png')

The filetype is either parsed automatically from the filename, or given as an optional parameter. Currently only jpeg and png files are supported.

>>> import autopy
>>> autopy.bitmap.Bitmap.open('needle.png')
<Bitmap object at 0x1001d5378>

Aside from analyzing a bitmap's pixel data, the main use for loading a bitmap is finding it on the screen or inside another bitmap. For example, the following prints the coordinates of the first image found in a bitmap (scanned from left to right, top to bottom):

import autopy


def find_image_example():
    needle = autopy.bitmap.Bitmap.open('needle.png')
    haystack = autopy.bitmap.Bitmap.open('haystack.png')

    pos = haystack.find_bitmap(needle)
    if pos:
        print("Found needle at: %s" % str(pos))

find_image_example()

It's also possible to do a bounded search by passing a tuple ((x, y), (width, height)):

haystack.find_bitmap(needle, rect=((10, 10), (100, 100)))

Projects using AutoPy

API Reference

Hope you enjoy using autopy! For a more in depth overview, see the API Reference.

Contributing

If you are interested in this project, please consider contributing. Here are a few ways you can help:

License

This project is licensed under either the Apache-2.0 or MIT license, at your option.

Unless you explicitly state otherwise, any contribution intentionally submitted for inclusion in the work by you, as defined in the Apache-2.0 license, shall be dual licensed as above, without any additional terms or conditions.