schemathesis 3.39.7__py3-none-any.whl → 4.0.0a2__py3-none-any.whl

schemathesis/runner/impl/core.py

@@ -1,1246 +0,0 @@
-from __future__ import annotations
-
-import functools
-import logging
-import operator
-import re
-import threading
-import time
-import unittest
-import uuid
-import warnings
-from contextlib import contextmanager
-from dataclasses import dataclass, field
-from typing import TYPE_CHECKING, Any, Callable, Generator, Iterable, List, Literal, cast
-from warnings import WarningMessage, catch_warnings
-
-import hypothesis
-import requests
-from _pytest.logging import LogCaptureHandler, catching_logs
-from hypothesis.errors import HypothesisException, InvalidArgument
-from hypothesis_jsonschema._canonicalise import HypothesisRefResolutionError
-from jsonschema.exceptions import SchemaError as JsonSchemaError
-from jsonschema.exceptions import ValidationError
-from requests.structures import CaseInsensitiveDict
-from urllib3.exceptions import InsecureRequestWarning
-
-from ... import experimental, failures, hooks
-from ..._compat import MultipleFailures
-from ..._hypothesis import (
-    get_invalid_example_headers_mark,
-    get_invalid_regex_mark,
-    get_non_serializable_mark,
-    has_unsatisfied_example_mark,
-)
-from ...auths import unregister as unregister_auth
-from ...checks import _make_max_response_time_failure_message
-from ...constants import (
-    DEFAULT_STATEFUL_RECURSION_LIMIT,
-    RECURSIVE_REFERENCE_ERROR_MESSAGE,
-    SERIALIZERS_SUGGESTION_MESSAGE,
-    USER_AGENT,
-)
-from ...exceptions import (
-    CheckFailed,
-    DeadlineExceeded,
-    InternalError,
-    InvalidHeadersExample,
-    InvalidRegularExpression,
-    NonCheckError,
-    OperationSchemaError,
-    RecursiveReferenceError,
-    SerializationNotPossible,
-    SkipTest,
-    format_exception,
-    get_grouped_exception,
-    maybe_set_assertion_message,
-)
-from ...generation import DataGenerationMethod, GenerationConfig
-from ...hooks import HookContext, get_all_by_name
-from ...internal.checks import CheckConfig, CheckContext
-from ...internal.datetime import current_datetime
-from ...internal.result import Err, Ok, Result
-from ...models import APIOperation, Case, Check, Status, TestResult
-from ...runner import events
-from ...service import extensions
-from ...service.models import AnalysisResult, AnalysisSuccess
-from ...specs.openapi import formats
-from ...stateful import Feedback, Stateful
-from ...stateful import events as stateful_events
-from ...stateful import runner as stateful_runner
-from ...targets import Target, TargetContext
-from ...transports import RequestConfig, RequestsTransport
-from ...transports.auth import get_requests_auth, prepare_wsgi_headers
-from ...utils import capture_hypothesis_output
-from .. import probes
-from ..serialization import SerializedTestResult
-from .context import RunnerContext
-
-if TYPE_CHECKING:
-    from types import TracebackType
-
-    from requests.auth import HTTPDigestAuth
-
-    from ..._override import CaseOverride
-    from ...internal.checks import CheckFunction
-    from ...schemas import BaseSchema
-    from ...service.client import ServiceClient
-    from ...transports.responses import GenericResponse, WSGIResponse
-    from ...types import RawAuth
-
-
-def _should_count_towards_stop(event: events.ExecutionEvent) -> bool:
-    return isinstance(event, events.AfterExecution) and event.status in (Status.error, Status.failure)
-
-
-@dataclass
-class BaseRunner:
-    schema: BaseSchema
-    checks: Iterable[CheckFunction]
-    max_response_time: int | None
-    targets: Iterable[Target]
-    hypothesis_settings: hypothesis.settings
-    generation_config: GenerationConfig | None
-    probe_config: probes.ProbeConfig
-    checks_config: CheckConfig
-    request_config: RequestConfig = field(default_factory=RequestConfig)
-    override: CaseOverride | None = None
-    auth: RawAuth | None = None
-    auth_type: str | None = None
-    headers: dict[str, Any] | None = None
-    store_interactions: bool = False
-    seed: int | None = None
-    exit_first: bool = False
-    no_failfast: bool = False
-    max_failures: int | None = None
-    started_at: str = field(default_factory=current_datetime)
-    unique_data: bool = False
-    dry_run: bool = False
-    stateful: Stateful | None = None
-    stateful_recursion_limit: int = DEFAULT_STATEFUL_RECURSION_LIMIT
-    count_operations: bool = True
-    count_links: bool = True
-    service_client: ServiceClient | None = None
-    _failures_counter: int = 0
-    _is_stopping_due_to_failure_limit: bool = False
-
-    def execute(self) -> EventStream:
-        """Common logic for all runners."""
-        event = threading.Event()
-        return EventStream(self._generate_events(event), event)
-
-    def _generate_events(self, stop_event: threading.Event) -> Generator[events.ExecutionEvent, None, None]:
-        # If auth is explicitly provided, then the global provider is ignored
-        if self.auth is not None:
-            unregister_auth()
-        ctx = RunnerContext(
-            auth=self.auth,
-            seed=self.seed,
-            stop_event=stop_event,
-            unique_data=self.unique_data,
-            checks_config=self.checks_config,
-            override=self.override,
-            no_failfast=self.no_failfast,
-        )
-        start_time = time.monotonic()
-        initialized = None
-        __probes = None
-        __analysis: Result[AnalysisResult, Exception] | None = None
-
-        def _initialize() -> events.Initialized:
-            nonlocal initialized
-            initialized = events.Initialized.from_schema(
-                schema=self.schema,
-                count_operations=self.count_operations,
-                count_links=self.count_links,
-                seed=ctx.seed,
-                start_time=start_time,
-            )
-            return initialized
-
-        def _finish() -> events.Finished:
-            if ctx.has_all_not_found:
-                ctx.add_warning(ALL_NOT_FOUND_WARNING_MESSAGE)
-            return events.Finished.from_results(results=ctx.data, running_time=time.monotonic() - start_time)
-
-        def _before_probes() -> events.BeforeProbing:
-            return events.BeforeProbing()
-
-        def _run_probes() -> None:
-            if not self.dry_run:
-                nonlocal __probes
-
-                __probes = run_probes(self.schema, self.probe_config)
-
-        def _after_probes() -> events.AfterProbing:
-            _probes = cast(List[probes.ProbeRun], __probes)
-            return events.AfterProbing(probes=_probes)
-
-        def _before_analysis() -> events.BeforeAnalysis:
-            return events.BeforeAnalysis()
-
-        def _run_analysis() -> None:
-            nonlocal __analysis, __probes
-
-            if self.service_client is not None:
-                try:
-                    _probes = cast(List[probes.ProbeRun], __probes)
-                    result = self.service_client.analyze_schema(_probes, self.schema.raw_schema)
-                    if isinstance(result, AnalysisSuccess):
-                        extensions.apply(result.extensions, self.schema)
-                    __analysis = Ok(result)
-                except Exception as exc:
-                    __analysis = Err(exc)
-
-        def _after_analysis() -> events.AfterAnalysis:
-            return events.AfterAnalysis(analysis=__analysis)
-
-        if ctx.is_stopped:
-            yield _finish()
-            return
-
-        for event_factory in (
-            _initialize,
-            _before_probes,
-            _run_probes,
-            _after_probes,
-            _before_analysis,
-            _run_analysis,
-            _after_analysis,
-        ):
-            event = event_factory()
-            if event is not None:
-                yield event
-            if ctx.is_stopped:
-                yield _finish()  # type: ignore[unreachable]
-                return
-
-        try:
-            warnings.simplefilter("ignore", InsecureRequestWarning)
-            if not experimental.STATEFUL_ONLY.is_enabled:
-                yield from self._execute(ctx)
-            if not self._is_stopping_due_to_failure_limit:
-                yield from self._run_stateful_tests(ctx)
-        except KeyboardInterrupt:
-            yield events.Interrupted()
-
-        yield _finish()
-
-    def _should_stop(self, event: events.ExecutionEvent) -> bool:
-        result = self.__should_stop(event)
-        if result:
-            self._is_stopping_due_to_failure_limit = True
-        return result
-
-    def __should_stop(self, event: events.ExecutionEvent) -> bool:
-        if _should_count_towards_stop(event):
-            if self.exit_first:
-                return True
-            if self.max_failures is not None:
-                self._failures_counter += 1
-                return self._failures_counter >= self.max_failures
-        return False
-
-    def _execute(self, ctx: RunnerContext) -> Generator[events.ExecutionEvent, None, None]:
-        raise NotImplementedError
-
-    def _run_stateful_tests(self, ctx: RunnerContext) -> Generator[events.ExecutionEvent, None, None]:
-        # Run new-style stateful tests
-        if self.stateful is not None and experimental.STATEFUL_TEST_RUNNER.is_enabled and self.schema.links_count > 0:
-            result = TestResult(
-                method="",
-                path="",
-                verbose_name="Stateful tests",
-                seed=ctx.seed,
-                data_generation_method=self.schema.data_generation_methods,
-            )
-            headers = self.headers or {}
-            if isinstance(self.schema.transport, RequestsTransport):
-                auth = get_requests_auth(self.auth, self.auth_type)
-            else:
-                auth = None
-                headers = prepare_wsgi_headers(headers, self.auth, self.auth_type)
-            config = stateful_runner.StatefulTestRunnerConfig(
-                checks=tuple(self.checks),
-                headers=headers,
-                hypothesis_settings=self.hypothesis_settings,
-                exit_first=self.exit_first,
-                max_failures=None if self.max_failures is None else self.max_failures - self._failures_counter,
-                request=self.request_config,
-                auth=auth,
-                seed=ctx.seed,
-                override=self.override,
-            )
-            state_machine = self.schema.as_state_machine()
-            runner = state_machine.runner(config=config)
-            status = Status.success
-
-            def from_step_status(step_status: stateful_events.StepStatus) -> Status:
-                return {
-                    stateful_events.StepStatus.SUCCESS: Status.success,
-                    stateful_events.StepStatus.FAILURE: Status.failure,
-                    stateful_events.StepStatus.ERROR: Status.error,
-                    stateful_events.StepStatus.INTERRUPTED: Status.error,
-                }[step_status]
-
-            if self.store_interactions:
-                if isinstance(state_machine.schema.transport, RequestsTransport):
-
-                    def on_step_finished(event: stateful_events.StepFinished) -> None:
-                        if event.response is not None and event.status is not None:
-                            response = cast(requests.Response, event.response)
-                            result.store_requests_response(
-                                status=from_step_status(event.status),
-                                case=event.case,
-                                response=response,
-                                checks=event.checks,
-                                headers=headers,
-                                session=None,
-                            )
-
-                else:
-
-                    def on_step_finished(event: stateful_events.StepFinished) -> None:
-                        from ...transports.responses import WSGIResponse
-
-                        if event.response is not None and event.status is not None:
-                            response = cast(WSGIResponse, event.response)
-                            result.store_wsgi_response(
-                                status=from_step_status(event.status),
-                                case=event.case,
-                                response=response,
-                                headers=headers,
-                                elapsed=response.elapsed.total_seconds(),
-                                checks=event.checks,
-                            )
-            else:
-
-                def on_step_finished(event: stateful_events.StepFinished) -> None:
-                    return None
-
-            test_start_time: float | None = None
-            test_elapsed_time: float | None = None
-
-            for stateful_event in runner.execute():
-                if isinstance(stateful_event, stateful_events.SuiteFinished):
-                    if stateful_event.failures and status != Status.error:
-                        status = Status.failure
-                elif isinstance(stateful_event, stateful_events.RunStarted):
-                    test_start_time = stateful_event.timestamp
-                elif isinstance(stateful_event, stateful_events.RunFinished):
-                    test_elapsed_time = stateful_event.timestamp - cast(float, test_start_time)
-                elif isinstance(stateful_event, stateful_events.StepFinished):
-                    result.checks.extend(stateful_event.checks)
-                    on_step_finished(stateful_event)
-                elif isinstance(stateful_event, stateful_events.Errored):
-                    status = Status.error
-                    result.add_error(stateful_event.exception)
-                yield events.StatefulEvent(data=stateful_event)
-            ctx.add_result(result)
-            yield events.AfterStatefulExecution(
-                status=status,
-                result=SerializedTestResult.from_test_result(result),
-                elapsed_time=cast(float, test_elapsed_time),
-                data_generation_method=self.schema.data_generation_methods,
-            )
-
-    def _run_tests(
-        self,
-        maker: Callable,
-        test_func: Callable,
-        settings: hypothesis.settings,
-        generation_config: GenerationConfig | None,
-        ctx: RunnerContext,
-        recursion_level: int = 0,
-        headers: dict[str, Any] | None = None,
-        **kwargs: Any,
-    ) -> Generator[events.ExecutionEvent, None, None]:
-        """Run tests and recursively run additional tests."""
-        if recursion_level > self.stateful_recursion_limit:
-            return
-
-        def as_strategy_kwargs(_operation: APIOperation) -> dict[str, Any]:
-            kw = {}
-            if self.override is not None:
-                for location, entry in self.override.for_operation(_operation).items():
-                    if entry:
-                        kw[location] = entry
-            if headers:
-                kw["headers"] = {key: value for key, value in headers.items() if key.lower() != "user-agent"}
-            return kw
-
-        for result in maker(
-            test_func,
-            settings=settings,
-            generation_config=generation_config,
-            seed=ctx.seed,
-            as_strategy_kwargs=as_strategy_kwargs,
-        ):
-            if isinstance(result, Ok):
-                operation, test = result.ok()
-                if self.stateful is not None and not experimental.STATEFUL_TEST_RUNNER.is_enabled:
-                    feedback = Feedback(self.stateful, operation)
-                else:
-                    feedback = None
-                # Track whether `BeforeExecution` was already emitted.
-                # Schema error may happen before / after `BeforeExecution`, but it should be emitted only once
-                # and the `AfterExecution` event should have the same correlation id as previous `BeforeExecution`
-                before_execution_correlation_id = None
-                try:
-                    for event in run_test(
-                        operation,
-                        test,
-                        ctx=ctx,
-                        feedback=feedback,
-                        recursion_level=recursion_level,
-                        data_generation_methods=self.schema.data_generation_methods,
-                        headers=headers,
-                        **kwargs,
-                    ):
-                        yield event
-                        if isinstance(event, events.BeforeExecution):
-                            before_execution_correlation_id = event.correlation_id
-                        if isinstance(event, events.Interrupted):
-                            return
-                    # Additional tests, generated via the `feedback` instance
-                    if feedback is not None:
-                        yield from self._run_tests(
-                            feedback.get_stateful_tests,
-                            test_func,
-                            settings=settings,
-                            generation_config=generation_config,
-                            recursion_level=recursion_level + 1,
-                            ctx=ctx,
-                            headers=headers,
-                            **kwargs,
-                        )
-                except OperationSchemaError as exc:
-                    yield from handle_schema_error(
-                        exc,
-                        ctx,
-                        self.schema.data_generation_methods,
-                        recursion_level,
-                        before_execution_correlation_id=before_execution_correlation_id,
-                    )
-            else:
-                # Schema errors
-                yield from handle_schema_error(result.err(), ctx, self.schema.data_generation_methods, recursion_level)
-
-
-def run_probes(schema: BaseSchema, config: probes.ProbeConfig) -> list[probes.ProbeRun]:
-    """Discover capabilities of the tested app."""
-    results = probes.run(schema, config)
-    for result in results:
-        if isinstance(result.probe, probes.NullByteInHeader) and result.is_failure:
-            from ...specs.openapi.formats import HEADER_FORMAT, header_values
-
-            formats.register(HEADER_FORMAT, header_values(blacklist_characters="\n\r\x00"))
-    return results
-
-
-@dataclass
-class EventStream:
-    """Schemathesis event stream.
-
-    Provides an API to control the execution flow.
-    """
-
-    generator: Generator[events.ExecutionEvent, None, None]
-    stop_event: threading.Event
-
-    def __next__(self) -> events.ExecutionEvent:
-        return next(self.generator)
-
-    def __iter__(self) -> Generator[events.ExecutionEvent, None, None]:
-        return self.generator
-
-    def stop(self) -> None:
-        """Stop the event stream.
-
-        Its next value will be the last one (Finished).
-        """
-        self.stop_event.set()
-
-    def finish(self) -> events.ExecutionEvent:
-        """Stop the event stream & return the last event."""
-        self.stop()
-        return next(self)
-
-
-def handle_schema_error(
-    error: OperationSchemaError,
-    ctx: RunnerContext,
-    data_generation_methods: Iterable[DataGenerationMethod],
-    recursion_level: int,
-    *,
-    before_execution_correlation_id: str | None = None,
-) -> Generator[events.ExecutionEvent, None, None]:
-    if error.method is not None:
-        assert error.path is not None
-        assert error.full_path is not None
-        data_generation_methods = list(data_generation_methods)
-        method = error.method.upper()
-        verbose_name = f"{method} {error.full_path}"
-        result = TestResult(
-            method=method,
-            path=error.full_path,
-            verbose_name=verbose_name,
-            data_generation_method=data_generation_methods,
-        )
-        result.add_error(error)
-        # It might be already emitted - reuse its correlation id
-        if before_execution_correlation_id is not None:
-            correlation_id = before_execution_correlation_id
-        else:
-            correlation_id = uuid.uuid4().hex
-            yield events.BeforeExecution(
-                method=method,
-                path=error.full_path,
-                verbose_name=verbose_name,
-                relative_path=error.path,
-                recursion_level=recursion_level,
-                data_generation_method=data_generation_methods,
-                correlation_id=correlation_id,
-            )
-        yield events.AfterExecution(
-            method=method,
-            path=error.full_path,
-            relative_path=error.path,
-            verbose_name=verbose_name,
-            status=Status.error,
-            result=SerializedTestResult.from_test_result(result),
-            data_generation_method=data_generation_methods,
-            elapsed_time=0.0,
-            hypothesis_output=[],
-            correlation_id=correlation_id,
-        )
-        ctx.add_result(result)
-    else:
-        # When there is no `method`, then the schema error may cover multiple operations, and we can't display it in
-        # the progress bar
-        ctx.add_generic_error(error)
-
-
-def run_test(
-    operation: APIOperation,
-    test: Callable,
-    checks: Iterable[CheckFunction],
-    data_generation_methods: Iterable[DataGenerationMethod],
-    targets: Iterable[Target],
-    ctx: RunnerContext,
-    headers: dict[str, Any] | None,
-    recursion_level: int,
-    **kwargs: Any,
-) -> Generator[events.ExecutionEvent, None, None]:
-    """A single test run with all error handling needed."""
-    data_generation_methods = list(data_generation_methods)
-    result = TestResult(
-        method=operation.method.upper(),
-        path=operation.full_path,
-        verbose_name=operation.verbose_name,
-        data_generation_method=data_generation_methods,
-    )
-    # To simplify connecting `before` and `after` events in external systems
-    correlation_id = uuid.uuid4().hex
-    yield events.BeforeExecution.from_operation(
-        operation=operation,
-        recursion_level=recursion_level,
-        data_generation_method=data_generation_methods,
-        correlation_id=correlation_id,
-    )
-    hypothesis_output: list[str] = []
-    errors: list[Exception] = []
-    test_start_time = time.monotonic()
-    setup_hypothesis_database_key(test, operation)
-
-    def _on_flaky(exc: Exception) -> Status:
-        if isinstance(exc.__cause__, hypothesis.errors.DeadlineExceeded):
-            status = Status.error
-            result.add_error(DeadlineExceeded.from_exc(exc.__cause__))
-        elif (
-            hasattr(hypothesis.errors, "FlakyFailure")
-            and isinstance(exc, hypothesis.errors.FlakyFailure)
-            and any(isinstance(subexc, hypothesis.errors.DeadlineExceeded) for subexc in exc.exceptions)
-        ):
-            for sub_exc in exc.exceptions:
-                if isinstance(sub_exc, hypothesis.errors.DeadlineExceeded):
-                    result.add_error(DeadlineExceeded.from_exc(sub_exc))
-            status = Status.error
-        elif errors:
-            status = Status.error
-            add_errors(result, errors)
-        else:
-            status = Status.failure
-            result.mark_flaky()
-        return status
-
-    try:
-        with catch_warnings(record=True) as warnings, capture_hypothesis_output() as hypothesis_output:
-            test(
-                ctx=ctx,
-                checks=checks,
-                targets=targets,
-                result=result,
-                errors=errors,
-                headers=headers,
-                data_generation_methods=data_generation_methods,
-                **kwargs,
-            )
-        # Test body was not executed at all - Hypothesis did not generate any tests, but there is no error
-        if not result.is_executed:
-            status = Status.skip
-            result.mark_skipped(None)
-        else:
-            status = Status.success
-    except unittest.case.SkipTest as exc:
-        # Newer Hypothesis versions raise this exception if no tests were executed
-        status = Status.skip
-        result.mark_skipped(exc)
-    except CheckFailed:
-        status = Status.failure
-    except NonCheckError:
-        # It could be an error in user-defined extensions, network errors or internal Schemathesis errors
-        status = Status.error
-        result.mark_errored()
-        for error in deduplicate_errors(errors):
-            result.add_error(error)
-    except hypothesis.errors.Flaky as exc:
-        status = _on_flaky(exc)
-    except MultipleFailures:
-        # Schemathesis may detect multiple errors that come from different check results
-        # They raise different "grouped" exceptions
-        if errors:
-            status = Status.error
-            add_errors(result, errors)
-        else:
-            status = Status.failure
-    except hypothesis.errors.Unsatisfiable:
-        # We need more clear error message here
-        status = Status.error
-        result.add_error(hypothesis.errors.Unsatisfiable("Failed to generate test cases for this API operation"))
-    except KeyboardInterrupt:
-        yield events.Interrupted()
-        return
-    except SkipTest as exc:
-        status = Status.skip
-        result.mark_skipped(exc)
-    except AssertionError as exc:  # May come from `hypothesis-jsonschema` or `hypothesis`
-        status = Status.error
-        try:
-            operation.schema.validate()
-            msg = "Unexpected error during testing of this API operation"
-            exc_msg = str(exc)
-            if exc_msg:
-                msg += f": {exc_msg}"
-            try:
-                raise InternalError(msg) from exc
-            except InternalError as exc:
-                error = exc
-        except ValidationError as exc:
-            error = OperationSchemaError.from_jsonschema_error(
-                exc,
-                path=operation.path,
-                method=operation.method,
-                full_path=operation.schema.get_full_path(operation.path),
-            )
-        result.add_error(error)
-    except HypothesisRefResolutionError:
-        status = Status.error
-        result.add_error(RecursiveReferenceError(RECURSIVE_REFERENCE_ERROR_MESSAGE))
-    except InvalidArgument as error:
-        status = Status.error
-        message = get_invalid_regular_expression_message(warnings)
-        if message:
-            # `hypothesis-jsonschema` emits a warning on invalid regular expression syntax
-            result.add_error(InvalidRegularExpression.from_hypothesis_jsonschema_message(message))
-        else:
-            result.add_error(error)
-    except hypothesis.errors.DeadlineExceeded as error:
-        status = Status.error
-        result.add_error(DeadlineExceeded.from_exc(error))
-    except JsonSchemaError as error:
-        status = Status.error
-        result.add_error(InvalidRegularExpression.from_schema_error(error, from_examples=False))
-    except Exception as error:
-        status = Status.error
-        # Likely a YAML parsing issue. E.g. `00:00:00.00` (without quotes) is parsed as float `0.0`
-        if str(error) == "first argument must be string or compiled pattern":
-            result.add_error(
-                InvalidRegularExpression(
-                    "Invalid `pattern` value: expected a string. "
-                    "If your schema is in YAML, ensure `pattern` values are quoted",
-                    is_valid_type=False,
-                )
-            )
-        else:
-            result.add_error(error)
-    if status == Status.success and ctx.no_failfast and any(check.value == Status.failure for check in result.checks):
-        status = Status.failure
-    if has_unsatisfied_example_mark(test):
-        status = Status.error
-        result.add_error(
-            hypothesis.errors.Unsatisfiable("Failed to generate test cases from examples for this API operation")
-        )
-    non_serializable = get_non_serializable_mark(test)
-    if non_serializable is not None and status != Status.error:
-        status = Status.error
-        media_types = ", ".join(non_serializable.media_types)
-        result.add_error(
-            SerializationNotPossible(
-                "Failed to generate test cases from examples for this API operation because of"
-                f" unsupported payload media types: {media_types}\n{SERIALIZERS_SUGGESTION_MESSAGE}",
-                media_types=non_serializable.media_types,
-            )
-        )
-    invalid_regex = get_invalid_regex_mark(test)
-    if invalid_regex is not None and status != Status.error:
-        status = Status.error
-        result.add_error(InvalidRegularExpression.from_schema_error(invalid_regex, from_examples=True))
-    invalid_headers = get_invalid_example_headers_mark(test)
-    if invalid_headers:
-        status = Status.error
-        result.add_error(InvalidHeadersExample.from_headers(invalid_headers))
-    test_elapsed_time = time.monotonic() - test_start_time
-    # DEPRECATED: Seed is the same per test run
-    # Fetch seed value, hypothesis generates it during test execution
-    # It may be `None` if the `derandomize` config option is set to `True`
-    result.seed = getattr(test, "_hypothesis_internal_use_seed", None) or getattr(
-        test, "_hypothesis_internal_use_generated_seed", None
-    )
-    ctx.add_result(result)
-    for status_code in (401, 403):
-        if has_too_many_responses_with_status(result, status_code):
-            ctx.add_warning(TOO_MANY_RESPONSES_WARNING_TEMPLATE.format(f"`{operation.verbose_name}`", status_code))
-    yield events.AfterExecution.from_result(
-        result=result,
-        status=status,
-        elapsed_time=test_elapsed_time,
-        hypothesis_output=hypothesis_output,
-        operation=operation,
-        data_generation_method=data_generation_methods,
-        correlation_id=correlation_id,
-    )
-
-
-TOO_MANY_RESPONSES_WARNING_TEMPLATE = (
-    "Most of the responses from {} have a {} status code. Did you specify proper API credentials?"
-)
-TOO_MANY_RESPONSES_THRESHOLD = 0.9
-
-
-def has_too_many_responses_with_status(result: TestResult, status_code: int) -> bool:
-    # It is faster than creating an intermediate list
-    unauthorized_count = 0
-    total = 0
-    for check in result.checks:
-        if check.response is not None:
-            if check.response.status_code == status_code:
-                unauthorized_count += 1
-            total += 1
-    if not total:
-        return False
-    return unauthorized_count / total >= TOO_MANY_RESPONSES_THRESHOLD
-
-
-ALL_NOT_FOUND_WARNING_MESSAGE = "All API responses have a 404 status code. Did you specify the proper API location?"
-
-
-def setup_hypothesis_database_key(test: Callable, operation: APIOperation) -> None:
-    """Make Hypothesis use separate database entries for every API operation.
-
-    It increases the effectiveness of the Hypothesis database in the CLI.
-    """
-    # Hypothesis's function digest depends on the test function signature. To reflect it for the web API case,
-    # we use all API operation parameters in the digest.
-    extra = operation.verbose_name.encode("utf8")
-    for parameter in operation.iter_parameters():
-        extra += parameter.serialize(operation).encode("utf8")
-    test.hypothesis.inner_test._hypothesis_internal_add_digest = extra  # type: ignore
-
-
-def get_invalid_regular_expression_message(warnings: list[WarningMessage]) -> str | None:
-    for warning in warnings:
-        message = str(warning.message)
-        if "is not valid syntax for a Python regular expression" in message:
-            return message
-    return None
-
-
-MEMORY_ADDRESS_RE = re.compile("0x[0-9a-fA-F]+")
-URL_IN_ERROR_MESSAGE_RE = re.compile(r"Max retries exceeded with url: .*? \(Caused by")
-
-
-def add_errors(result: TestResult, errors: list[Exception]) -> None:
-    group_errors(errors)
-    for error in deduplicate_errors(errors):
-        result.add_error(error)
-
-
-def group_errors(errors: list[Exception]) -> None:
-    """Group errors of the same kind info a single one, avoiding duplicate error messages."""
-    serialization_errors = [error for error in errors if isinstance(error, SerializationNotPossible)]
-    if len(serialization_errors) > 1:
-        errors[:] = [error for error in errors if not isinstance(error, SerializationNotPossible)]
-        media_types: list[str] = functools.reduce(
-            operator.iadd, (entry.media_types for entry in serialization_errors), []
-        )
-        errors.append(SerializationNotPossible.from_media_types(*media_types))
-
-
-def canonicalize_error_message(error: Exception, include_traceback: bool = True) -> str:
-    message = format_exception(error, include_traceback)
-    # Replace memory addresses with a fixed string
-    message = MEMORY_ADDRESS_RE.sub("0xbaaaaaaaaaad", message)
-    return URL_IN_ERROR_MESSAGE_RE.sub("", message)
-
-
-def deduplicate_errors(errors: list[Exception]) -> Generator[Exception, None, None]:
-    """Deduplicate errors by their messages + tracebacks."""
-    seen = set()
-    for error in errors:
-        message = canonicalize_error_message(error)
-        if message in seen:
-            continue
-        seen.add(message)
-        yield error
-
-
-def run_checks(
-    *,
-    case: Case,
-    ctx: CheckContext,
-    checks: Iterable[CheckFunction],
-    check_results: list[Check],
-    result: TestResult,
-    response: GenericResponse,
-    elapsed_time: float,
-    max_response_time: int | None = None,
-    no_failfast: bool,
-) -> None:
-    errors = []
-
-    def add_single_failure(error: AssertionError) -> None:
-        msg = maybe_set_assertion_message(error, check_name)
-        errors.append(error)
-        if isinstance(error, CheckFailed):
-            context = error.context
-        else:
-            context = None
-        check_results.append(result.add_failure(check_name, copied_case, response, elapsed_time, msg, context))
-
-    for check in checks:
-        check_name = check.__name__
-        copied_case = case.partial_deepcopy()
-        try:
-            skip_check = check(ctx, response, copied_case)
-            if not skip_check:
-                check_result = result.add_success(check_name, copied_case, response, elapsed_time)
-                check_results.append(check_result)
-        except AssertionError as exc:
-            add_single_failure(exc)
-        except MultipleFailures as exc:
-            for exception in exc.exceptions:
-                add_single_failure(exception)
-
-    if max_response_time:
-        if elapsed_time > max_response_time:
-            message = _make_max_response_time_failure_message(elapsed_time, max_response_time)
-            errors.append(AssertionError(message))
-            result.add_failure(
-                "max_response_time",
-                case,
-                response,
-                elapsed_time,
-                message,
-                failures.ResponseTimeExceeded(message=message, elapsed=elapsed_time, deadline=max_response_time),
-            )
-        else:
-            result.add_success("max_response_time", case, response, elapsed_time)
-
-    if errors and not no_failfast:
-        raise get_grouped_exception(case.operation.verbose_name, *errors)(causes=tuple(errors))
-
-
-def run_targets(targets: Iterable[Callable], context: TargetContext) -> None:
-    for target in targets:
-        value = target(context)
-        hypothesis.target(value, label=target.__name__)
-
-
-def add_cases(case: Case, response: GenericResponse, test: Callable, *args: Any) -> None:
-    context = HookContext(case.operation)
-    for case_hook in get_all_by_name("add_case"):
-        _case = case_hook(context, case.partial_deepcopy(), response)
-        # run additional test if _case is not an empty value
-        if _case:
-            test(_case, *args)
-
-
-@dataclass
-class ErrorCollector:
-    """Collect exceptions that are not related to failed checks.
-
-    Such exceptions may be considered as multiple failures or flakiness by Hypothesis. In both cases, Hypothesis hides
-    exception information that, in our case, is helpful for the end-user. It either indicates errors in user-defined
-    extensions, network-related errors, or internal Schemathesis errors. In all cases, this information is useful for
-    debugging.
-
-    To mitigate this, we gather all exceptions manually via this context manager to avoid interfering with the test
-    function signatures, which are used by Hypothesis.
-    """
-
-    errors: list[Exception]
-
-    def __enter__(self) -> ErrorCollector:
-        return self
-
-    def __exit__(
-        self, exc_type: type[BaseException] | None, exc_val: BaseException | None, exc_tb: TracebackType | None
-    ) -> Literal[False]:
-        # Don't do anything special if:
-        #   - Tests are successful
-        #   - Checks failed
-        #   - The testing process is interrupted
-        if not exc_type or issubclass(exc_type, CheckFailed) or not issubclass(exc_type, Exception):
-            return False
-        # These exceptions are needed for control flow on the Hypothesis side. E.g. rejecting unsatisfiable examples
-        if isinstance(exc_val, HypothesisException):
-            raise
-        # Exception value is not `None` and is a subclass of `Exception` at this point
-        exc_val = cast(Exception, exc_val)
-        self.errors.append(exc_val.with_traceback(exc_tb))
-        raise NonCheckError from None
-
-
-def _force_data_generation_method(values: list[DataGenerationMethod], case: Case) -> None:
-    # Set data generation method to the one that actually used
-    data_generation_method = cast(DataGenerationMethod, case.data_generation_method)
-    values[:] = [data_generation_method]
-
-
-def cached_test_func(f: Callable) -> Callable:
-    def wrapped(*, ctx: RunnerContext, case: Case, **kwargs: Any) -> None:
-        if ctx.unique_data:
-            cached = ctx.get_cached_outcome(case)
-            if isinstance(cached, BaseException):
-                raise cached
-            elif cached is None:
-                return None
-            try:
-                f(ctx=ctx, case=case, **kwargs)
-            except BaseException as exc:
-                ctx.cache_outcome(case, exc)
-                raise
-            else:
-                ctx.cache_outcome(case, None)
-        else:
-            f(ctx=ctx, case=case, **kwargs)
-
-    wrapped.__name__ = f.__name__
-
-    return wrapped
-
-
-@cached_test_func
-def network_test(
-    *,
-    ctx: RunnerContext,
-    case: Case,
-    checks: Iterable[CheckFunction],
-    targets: Iterable[Target],
-    result: TestResult,
-    session: requests.Session,
-    request_config: RequestConfig,
-    store_interactions: bool,
-    headers: dict[str, Any] | None,
-    feedback: Feedback | None,
-    max_response_time: int | None,
-    data_generation_methods: list[DataGenerationMethod],
-    dry_run: bool,
-    errors: list[Exception],
-) -> None:
-    """A single test body will be executed against the target."""
-    with ErrorCollector(errors):
-        _force_data_generation_method(data_generation_methods, case)
-        result.mark_executed()
-        headers = headers or {}
-        if "user-agent" not in {header.lower() for header in headers}:
-            headers["User-Agent"] = USER_AGENT
-        if not dry_run:
-            args = (
-                ctx,
-                checks,
-                targets,
-                result,
-                session,
-                request_config,
-                store_interactions,
-                headers,
-                feedback,
-                max_response_time,
-            )
-            response = _network_test(case, *args)
-            add_cases(case, response, _network_test, *args)
-        elif store_interactions:
-            result.store_requests_response(case, None, Status.skip, [], headers=headers, session=session)
-
-
-def _network_test(
-    case: Case,
-    ctx: RunnerContext,
-    checks: Iterable[CheckFunction],
-    targets: Iterable[Target],
-    result: TestResult,
-    session: requests.Session,
-    request_config: RequestConfig,
-    store_interactions: bool,
-    headers: dict[str, Any] | None,
-    feedback: Feedback | None,
-    max_response_time: int | None,
-) -> requests.Response:
-    check_results: list[Check] = []
-    hook_context = HookContext(operation=case.operation)
-    kwargs: dict[str, Any] = {
-        "session": session,
-        "headers": headers,
-        "timeout": request_config.prepared_timeout,
-        "verify": request_config.tls_verify,
-        "cert": request_config.cert,
-    }
-    if request_config.proxy is not None:
-        kwargs["proxies"] = {"all": request_config.proxy}
-    hooks.dispatch("process_call_kwargs", hook_context, case, kwargs)
-    try:
-        response = case.call(**kwargs)
-    except CheckFailed as exc:
-        check_name = "request_timeout"
-        requests_kwargs = RequestsTransport().serialize_case(case, base_url=case.get_full_base_url(), headers=headers)
-        request = requests.Request(**requests_kwargs).prepare()
-        elapsed = cast(
-            float, request_config.prepared_timeout
-        )  # It is defined and not empty, since the exception happened
-        check_result = result.add_failure(
-            check_name, case, None, elapsed, f"Response timed out after {1000 * elapsed:.2f}ms", exc.context, request
-        )
-        check_results.append(check_result)
-        if store_interactions:
-            result.store_requests_response(case, None, Status.failure, [check_result], headers=headers, session=session)
-        raise exc
-    context = TargetContext(case=case, response=response, response_time=response.elapsed.total_seconds())
-    run_targets(targets, context)
-    status = Status.success
-
-    check_ctx = CheckContext(
-        override=ctx.override,
-        auth=ctx.auth,
-        headers=CaseInsensitiveDict(headers) if headers else None,
-        config=ctx.checks_config,
-        transport_kwargs=kwargs,
-    )
-    try:
-        run_checks(
-            case=case,
-            ctx=check_ctx,
-            checks=checks,
-            check_results=check_results,
-            result=result,
-            response=response,
-            elapsed_time=context.response_time * 1000,
-            max_response_time=max_response_time,
-            no_failfast=ctx.no_failfast,
-        )
-    except CheckFailed:
-        status = Status.failure
-        raise
-    finally:
-        if feedback is not None:
-            feedback.add_test_case(case, response)
-        if store_interactions:
-            result.store_requests_response(case, response, status, check_results, headers=headers, session=session)
-    return response
-
-
-@contextmanager
-def get_session(auth: HTTPDigestAuth | RawAuth | None = None) -> Generator[requests.Session, None, None]:
-    with requests.Session() as session:
-        if auth is not None:
-            session.auth = auth
-        yield session
-
-
-@cached_test_func
-def wsgi_test(
-    ctx: RunnerContext,
-    case: Case,
-    checks: Iterable[CheckFunction],
-    targets: Iterable[Target],
-    result: TestResult,
-    auth: RawAuth | None,
-    auth_type: str | None,
-    headers: dict[str, Any] | None,
-    store_interactions: bool,
-    feedback: Feedback | None,
-    max_response_time: int | None,
-    data_generation_methods: list[DataGenerationMethod],
-    dry_run: bool,
-    errors: list[Exception],
-) -> None:
-    with ErrorCollector(errors):
-        _force_data_generation_method(data_generation_methods, case)
-        result.mark_executed()
-        headers = prepare_wsgi_headers(headers, auth, auth_type)
-        if not dry_run:
-            args = (
-                ctx,
-                checks,
-                targets,
-                result,
-                headers,
-                store_interactions,
-                feedback,
-                max_response_time,
-            )
-            response = _wsgi_test(case, *args)
-            add_cases(case, response, _wsgi_test, *args)
-        elif store_interactions:
-            result.store_wsgi_response(case, None, headers, None, Status.skip, [])
-
-
-def _wsgi_test(
-    case: Case,
-    ctx: RunnerContext,
-    checks: Iterable[CheckFunction],
-    targets: Iterable[Target],
-    result: TestResult,
-    headers: dict[str, Any],
-    store_interactions: bool,
-    feedback: Feedback | None,
-    max_response_time: int | None,
-) -> WSGIResponse:
-    from ...transports.responses import WSGIResponse
-
-    with catching_logs(LogCaptureHandler(), level=logging.DEBUG) as recorded:
-        hook_context = HookContext(operation=case.operation)
-        kwargs: dict[str, Any] = {"headers": headers}
-        hooks.dispatch("process_call_kwargs", hook_context, case, kwargs)
-        response = cast(WSGIResponse, case.call(**kwargs))
-    context = TargetContext(case=case, response=response, response_time=response.elapsed.total_seconds())
-    run_targets(targets, context)
-    result.logs.extend(recorded.records)
-    status = Status.success
-    check_results: list[Check] = []
-    check_ctx = CheckContext(
-        override=ctx.override,
-        auth=ctx.auth,
-        headers=CaseInsensitiveDict(headers) if headers else None,
-        config=ctx.checks_config,
-        transport_kwargs=kwargs,
-    )
-    try:
-        run_checks(
-            case=case,
-            ctx=check_ctx,
-            checks=checks,
-            check_results=check_results,
-            result=result,
-            response=response,
-            elapsed_time=context.response_time * 1000,
-            max_response_time=max_response_time,
-            no_failfast=ctx.no_failfast,
-        )
-    except CheckFailed:
-        status = Status.failure
-        raise
-    finally:
-        if feedback is not None:
-            feedback.add_test_case(case, response)
-        if store_interactions:
-            result.store_wsgi_response(case, response, headers, response.elapsed.total_seconds(), status, check_results)
-    return response
-
-
-@cached_test_func
-def asgi_test(
-    ctx: RunnerContext,
-    case: Case,
-    checks: Iterable[CheckFunction],
-    targets: Iterable[Target],
-    result: TestResult,
-    store_interactions: bool,
-    headers: dict[str, Any] | None,
-    feedback: Feedback | None,
-    max_response_time: int | None,
-    data_generation_methods: list[DataGenerationMethod],
-    dry_run: bool,
-    errors: list[Exception],
-) -> None:
-    """A single test body will be executed against the target."""
-    with ErrorCollector(errors):
-        _force_data_generation_method(data_generation_methods, case)
-        result.mark_executed()
-        headers = headers or {}
-
-        if not dry_run:
-            args = (
-                ctx,
-                checks,
-                targets,
-                result,
-                store_interactions,
-                headers,
-                feedback,
-                max_response_time,
-            )
-            response = _asgi_test(case, *args)
-            add_cases(case, response, _asgi_test, *args)
-        elif store_interactions:
-            result.store_requests_response(case, None, Status.skip, [], headers=headers, session=None)
-
-
-def _asgi_test(
-    case: Case,
-    ctx: RunnerContext,
-    checks: Iterable[CheckFunction],
-    targets: Iterable[Target],
-    result: TestResult,
-    store_interactions: bool,
-    headers: dict[str, Any] | None,
-    feedback: Feedback | None,
-    max_response_time: int | None,
-) -> requests.Response:
-    hook_context = HookContext(operation=case.operation)
-    kwargs: dict[str, Any] = {"headers": headers}
-    hooks.dispatch("process_call_kwargs", hook_context, case, kwargs)
-    response = case.call(**kwargs)
-    context = TargetContext(case=case, response=response, response_time=response.elapsed.total_seconds())
-    run_targets(targets, context)
-    status = Status.success
-    check_results: list[Check] = []
-    check_ctx = CheckContext(
-        override=ctx.override,
-        auth=ctx.auth,
-        headers=CaseInsensitiveDict(headers) if headers else None,
-        config=ctx.checks_config,
-        transport_kwargs=kwargs,
-    )
-    try:
-        run_checks(
-            case=case,
-            ctx=check_ctx,
-            checks=checks,
-            check_results=check_results,
-            result=result,
-            response=response,
-            elapsed_time=context.response_time * 1000,
-            max_response_time=max_response_time,
-            no_failfast=ctx.no_failfast,
-        )
-    except CheckFailed:
-        status = Status.failure
-        raise
-    finally:
-        if feedback is not None:
-            feedback.add_test_case(case, response)
-        if store_interactions:
-            result.store_requests_response(case, response, status, check_results, headers, session=None)
-    return response