Generating data shapes with Hypothesis

"I couldn't get Hypothesis to generate usable data for my test. I wanted to assert that two equal data items would hash equally, but Hypothesis was finding pairs like[0] and [False]. These are equal but hash differently because the hash takes the types into account."

"I don't want a fixed schema for the data Hasher would accept, but tests to compare data generated from the same schema. It shouldn't compare a list of ints to a list of bools. Hypothesis is good at generating things randomly. Usually it generates data randomly, but we can also use it to generate schemas randomly!"

Hypothesis strategies produce random values and can be composed to build complex schemas that generate nested and constrained data. Tests use strategies with @given to run many examples and assert properties across those examples. Hash-based comparisons fail when logically equal values of different types (for example 0 and False) are compared because the hash incorporates type information. Generating a random strategy first, then sampling multiple values from that single strategy, ensures paired test inputs share schema, shape, and types. Steering Hypothesis this way yields realistic, schema-consistent test pairs suitable for testing functions that hash nested data.

Read at Nedbatchelder