blqs: Building Blocks for Domain Specific Languages

Blqs is a framework for building (internal) domain specific language that can be written in Python. It was inspired by TensorFlow's autograph library, and motivated by the state of the art (circa 2021) in quantum programming frameworks like Cirq and Qiskit.

In short blqs let you define domain specific languages and gives you access to use native Python features like if, while, or for in these languages. This fills in a missing feature for Python, which while they do have operator overloading, does not allow for overloading these built in constructions.

Example:

import blqs
H, M, CX = blqs.Op('H'), blqs.Op('M'), blqs.Op('CX')

@blqs.build
def hello_blqs():
    a = blqs.Register('a')
    H(0)
    M(0, 'a')
    if a:
        CX(0, 1)
    else:
        CX(1, 0)

Then, if we call this method, we will produce a blqs.Program that includes both statements, like H(0) but also the if and else statements.

program = hello_blqs()
for s in program:
    print(type(s))
> prints
> <class 'blqs.instruction.Instruction'>
> <class 'blqs.instruction.Instruction'>
> <class 'blqs.conditional.If'>

Where the last statement contains blocks that hold the CX statements.

Installation

To install blqs one can simply pip install the appropriate package

pip install blqs

See requirements.txt for the dependencies that blqs will pull in.

Documentation

A good place to get started is to read the introduction to blqs. After that, the ideas and abstractions in blqs are described in the concepts guide. To learn about support for capturing native python code, see refer to the protocols section of the docs. For a quick intro via a jupyter notebook, see hello blqs.