Getting started
Welcome to Richard!, a Python library that translates human language into Python code. It allows you to query a database, create commands to make your computer do things, and learn the computer new words and rules.
Richard! adheres to the principle of progressive disclosure of complexity: simple systems can be built using little code, complex systems can be built as well, but require a lot more code.
The system is still very much in development, so don’t expect it to work flawlessly, and the code you develop will probably not work with the next release of the library. Also, the documentation is not up-to-date. But you will be able to experiment with natural language understanding.
Demos
You’re probably interested to see some demos of the system. See the introduction to the demos for how to find them and how to run them.
Start your own project
To start your own project, you can copy the “integration test” hello_world, and change it into the system you want.
Grow your project by extending a dialog
When looking at the demos you’ll notice that they all use a demo dialog to test all interactions. Such a dialog serves as a means to automatically test your system, to make sure that previous constructions still work after you change your modules.
Start with a single, simple test, make it work, then add another one. From simple to complex.
The library comes with a testing framework that shows you the intermediate results of your request, to help debugging. To use it, run your test through this code:
logger = Logger()
logger.log_all_tests()
logger.log_products()
logger.log_stats()
tester = DialogTester(self, tests, pipeline, logger)
tester.run()
print(logger)
The logger class has several options. Turn them all on at first. When you’re completely done with the system, change the settings to logger.log_no_tests() to skip the logs and just perform the tests.
Test your code in the command line, like this:
python3 -m unittest tests/integration/hello_world/test.py
If you’re using Bash, the intermediate results show up with colored headings.
Documentation
Now you know how to get started, read the documentation to find out more about the system:
Introduction teaches you the basics
Language shows you how to turn language constructs into semantic constructs
Modules provide information about the available modules