如何调整Hypothesis中Flakiness的重试次数？

Absolutely! You can totally tweak how many times Hypothesis retries a test that fails on its first execution but passes on subsequent runs. Here's how to do it:

Per-test configuration #

Use Hypothesis's @settings decorator to set the retries parameter directly on individual tests. This lets you customize behavior for specific flaky tests without affecting others.

Example: Increase retries to 5 #

from hypothesis import given, settings
from hypothesis.strategies import integers

@given(integers())
@settings(retries=5)
def test_my_flaky_calculation(x):
    # Your test logic here (e.g., a function that occasionally fails)
    assert x + 1 > x  # Trivial example, replace with your actual test

Example: Disable retries entirely #

If you don't want Hypothesis to retry a test at all, set retries=0:

@given(integers())
@settings(retries=0)
def test_no_retry_allowed(x):
    assert x > 0

Global configuration #

If you want to apply the same retry count to all your Hypothesis tests, you can register a custom settings profile and load it:

from hypothesis import settings

# Create a profile with your desired retry count
settings.register_profile("custom_retries", retries=4)
# Load the profile to apply it globally
settings.load_profile("custom_retries")

When Hypothesis detects a flaky test (fails first run, passes on retries), you'll still see that familiar message:

Flaky: Hypothesis ... produces unreliable results: Falsified on the first call but did not on a subsequent one

This adjustment works exactly as you'd expect—changing the retries value controls how many times Hypothesis will re-run the failed test before marking it as flaky.

内容的提问来源于stack exchange，提问作者curious_george