[renovate]

This PR contains the following updates:

Package	Type	Update	Change	Pending
hypothesis (source)	feature-tests	minor	>=6.168.1 → >=6.170.12	6.170.15 (+2)
hypothesis (source)	feature-lint	minor	>=6.168.1 → >=6.170.12	6.170.15 (+2)

---

Release Notes

HypothesisWorks/hypothesis (hypothesis)

v6.170.12: Hypothesis version 6.170.12

Compare Source

This patch reduces Hypothesis' overhead for each test case. For simple
properties and strategies, generating test cases takes up to a third
less time and shrinking failing examples is up to twice as fast; most
tests will see a smaller improvement. "import hypothesis" is also
about 20% faster.

The canonical version of these notes (with links) is on readthedocs.

v6.170.11: Hypothesis version 6.170.11

Compare Source

This patch makes writing observability data to ".hypothesis/observed"
cheaper when tests run in multiple threads, and defers cleaning up old
observation files from import time until observations are first
written.

The canonical version of these notes (with links) is on readthedocs.

v6.170.10: Hypothesis version 6.170.10

This patch fixes an internal error from "lists()" with "unique=True",
when the elements are a filtered "sampled_from()" with too few allowed
values to reach "min_size". It also ensures that "domains()" always
generates at least one subdomain, even with a small "max_length".

We also fixed some internal bookkeeping that could go wrong when an
exception was raised and caught during generation, for example by
"recursive()", which could make shrinking less effective.

The canonical version of these notes (with links) is on readthedocs.

v6.170.5: Hypothesis version 6.170.5

Compare Source

This patch fixes a bug for users of our "PrimitiveProvider" interface,
where "draw_integer" could incorrectly reuse weights from an earlier
draw (issue #​4908).

The canonical version of these notes (with links) is on readthedocs.

v6.170.4: Hypothesis version 6.170.4

This fills in the missing type annotations for public-facing
attributes of "settings", such as "settings.max_examples" and
"settings.database".

The canonical version of these notes (with links) is on readthedocs.

v6.170.2: Hypothesis version 6.170.2

Compare Source

Add support for Python 3.16, which is still in alpha.

The canonical version of these notes (with links) is on readthedocs.

v6.170.1: Hypothesis version 6.170.1

Compare Source

This patch makes generating unique collections, such as "sets()" or
"lists()" with "unique=True", considerably more efficient (issue
#​4458). Hypothesis no longer mutates test cases by copying one element
of a unique collection over another element of the same collection,
and no longer probes for minimal examples by drawing every element at
its simplest value - both of which were certain to be rejected, and
could waste several times as many draws from the element strategy as
were actually returned.

The canonical version of these notes (with links) is on readthedocs.

v6.170.0: Hypothesis version 6.170.0

Compare Source

This release substantially strengthens "run_conformance_test()", which
authors of alternative backends can use to test their provider. It can
now test providers with required constructor arguments, runs many test
cases against each provider, treats running out of entropy as expected
control flow, and checks that providers are able to generate a
representative range of values.

These new checks catch two bugs in the provider underlying
fuzz_one_input, which this release also fixes: integer draws could
never generate negative values, and string draws from an empty
alphabet consumed the whole buffer instead of returning the empty
string.

Verbose output for stateful tests now reads "Test case: " rather than trailing off after the colon.

Thanks to @​reachsridhard for reporting the integers bug and
prototyping these improvements.

The canonical version of these notes (with links) is on readthedocs.

v6.169.4: Hypothesis version 6.169.4

Compare Source

This release fixes an error when applying "@​given" on Python 3.14 to
functions with annotations referring to types that are only available
during type checking, when explicit strategies are supplied (issue
#​4897).

Thanks to Arjun Aravind for this fix!

The canonical version of these notes (with links) is on readthedocs.

v6.169.3: Hypothesis version 6.169.3

The ".seed()" method of "randoms()" now accepts the same arguments as
"random.Random.seed()", such as "a=" and "version=". Previously, it
required exactly one positional argument. Calling ".seed()" without a
seed now raises a clearer error.

The canonical version of these notes (with links) is on readthedocs.

v6.169.1: Hypothesis version 6.169.1

Compare Source

Calling ".map()" or ".flatmap()" on "data()" with a keyword argument
now raises the intended error explaining that this is not supported,
instead of a generic error about an unexpected keyword argument.

The canonical version of these notes (with links) is on readthedocs.

v6.169.0: Hypothesis version 6.169.0

Compare Source

This release drops support for Python 3.10, which reached end of life
in October 2026.

The canonical version of these notes (with links) is on readthedocs.

v6.168.5: Hypothesis version 6.168.5

Compare Source

This patch fixes the type annotation of "from_field()", which claimed
to return a strategy for instances of the field type rather than for
values of the field.

The canonical version of these notes (with links) is on readthedocs.

v6.168.4: Hypothesis version 6.168.4

Compare Source

This patch adds support for PyPy 3.12.

The canonical version of these notes (with links) is on readthedocs.

v6.168.3: Hypothesis version 6.168.3

Compare Source

This patch fixes quadratic-time behaviour when summarising the
statistics of a test run, which Hypothesis does for every test when
running under pytest. This avoids an apparent hang after the last test
case with a high number of test cases (>100k).

The canonical version of these notes (with links) is on readthedocs.

v6.168.2: Hypothesis version 6.168.2

Compare Source

This patch makes writes to the cache of constants collected from local
source files atomic. Previously, a concurrent process sharing the same
".hypothesis" directory, such as another pytest-xdist worker, could
read a partially written cache file, which silently changed the data
generated for a given "@​seed" (issue #​4885).

The canonical version of these notes (with links) is on readthedocs.

---

Configuration

📅 Schedule: (UTC)

• Branch creation
  • On day 10 of the month (* * 10 * *)
• Automerge
  • At any time (no schedule defined)

🚦 Automerge: Enabled.

♻ Rebasing: Whenever PR is behind base branch, or you tick the rebase/retry checkbox.

🔕 Ignore: Close this PR and you won't be reminded about these updates again.

---

• If you want to rebase/retry this PR, check this box

---

This PR was generated by Mend Renovate. View the repository job log.