TracebackEntry needs the excinfo for the `__tracebackhide__ = callback`
functionality, where `callback` accepts the excinfo.
Currently it achieves this by storing a weakref to the excinfo which
created it. I think this is not great, mixing layers and bloating the
objects.
Instead, have `ishidden` (and transitively, `Traceback.filter()`) take
the excinfo as a parameter.
* [pre-commit.ci] pre-commit autoupdate
updates:
- [github.com/psf/black: 22.12.0 → 23.1.0](https://github.com/psf/black/compare/22.12.0...23.1.0)
- [github.com/PyCQA/autoflake: v2.0.0 → v2.0.1](https://github.com/PyCQA/autoflake/compare/v2.0.0...v2.0.1)
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* Update .pre-commit-config.yaml
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
---------
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: Bruno Oliveira <nicoddemus@gmail.com>
Since pytest now requires Python>=3.7, we can use the stdlib attrs
clone, dataclasses, instead of the OG package.
attrs is still somewhat nicer than dataclasses and has some extra
functionality, but for pytest usage there's not really a justification
IMO to impose the extra dependency on users when a standard alternative
exists.
The main motivation for this change is to simplify the type shown in
code editors -- `Sequence[str]` is easier to follow than
`Union[list[str], tuple[str, ...]]`.
It also permits using other types if desired. It might lead to problems
if someone uses some oddball sequence type, but hopefully they won't do
that.
In the following
@pytest.mark.parametrize(..., ids=[val])
the ID values are only allowed to be `str`, `float`, `int` or `bool`.
In the following
@pytest.mark.parametrize(..., [val])
@pytest.mark.parametrize(..., [pytest.param(..., id=val])
a different code path is used, which also allows `bytes`, `complex`,
`re.Pattern`, `Enum` and anything with a `__name__`.
In the interest of consistency, use the latter code path for all cases.
This commit only refactors, it does not change or add functionality yet. Public
API is retained. Reason or refactoring:
User provided parameter IDs (e.g. Metafunc.parametrize(ids=...)) had so far
only been used to calculate a unique test ID for each test invocation. That
test ID was a joined string where each parameter contributed some partial ID.
We're soon going to reuse functionality to generate parameter keys for
reorder_items and FixtureDef cache. We will be interested in the partial
IDs, and only if they originate from explicit user information. Refactoring
makes logic and data accessible for reuse, and increases cohesion in general.
The docstring (and function name itself) described things as if IDs are
being assigned to the argnames, but actually they're assigned to the
parameter sets.
If we do the `update`s in the right order, we can avoid the `mark.name
not in self.keywords` check, since `self.keywords` starts out clean and
`update` will override previously set keywords.
By my analysis, this deleted code block has no effect:
1. `self.keywords` is `update`d with `callspec.marks`.
2. `self.own_markers` is `update`d with `callspec.marks`.
3. `self.keywords` is `update`d with `self.own_markers`.
So together steps 2+3 completely undo step 1.
This does have a slight semantic change: in a node hierarchy parent ->
child, if parent has a marker applied, then child is constructed, then
`parent.themarker = "overridden"`, previously
`child.keywords['themarker']` would return `True`, now it returns
`"overridden"`. But that's actually what I would have expected so I see
it as more of a bugfix.
(except `Instance`)
Currently, `Function` does this manually, but other node types don't get
their markers added to their `keywords`, but they should, if only for
consistency.
Marks are added to keywords in three places:
- `Node.add_marker`: name -> `Mark`
- `Function.__init__(callspec)`: name -> `Mark`
- `Function.__init__ iter_markers`: name -> True
I think it should be consistent, which will also help with some upcoming
code cleaning. The `Mark` seems more useful than just a `True`, so
switch to that.
Since commit 89f0b5b5a2 cases as in the
added test started to fail, like they do for the standard pytest names
(`setup_module` etc). But the name `setup` in particular is way too
common for us to start taking it over more aggressively, so restore the
previous behavior which required the object to be callable.
Fix#9391.
The `pytest_pycollector_makeitem` argument `collector` is currently
annotated with type `PyCollector`. As part of #7469, that would have
required us to expose it in the public API. But really it's an
implementation detail, not something we want to expose. So replace the
annotation with the concrete python collector types that are passed.
Strictly speaking, `pytest_pycollector_makeitem` is called from
`PyCollector.collect()`, so the new type annotation is incorrect if
another type subclasses `PyCollector`. But the set of python collectors
is closed (mapping to language constructs), and the type is private, so
there shouldn't be any other deriving classes, and we can consider it
effectively sealed (unfortunately Python does not provide a way to
express this - yet?).
* Improve reference and path/fspath docs
Closes#9283
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* fixups
* Add explanation
* Update wording after #9363
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
* Rename pytest_ignore_collect fspath parameter to collection_path
* Rename pytest_collect_file fspath parameter to file_path
* Rename pytest_pycollect_makemodule fspath parameter to module_path
* Rename pytest_report_header startpath parameter to start_path
* Rename pytest_report_collectionfinish startpath parameter to start_path
* Update docs with the renamed parameters
* Use pytest-flakes fork temporarily to prove it works
* Use pytest-flakes 4.0.5
* porting pytest.skip() to use reason=, adding tests
* avoid adding **kwargs, it breaks other functionality, use optional msg= instead
* deprecation of `pytest.fail(msg=...)`
* fix bug with not capturing the returned reason value
* pass reason= in acceptance async tests instead of msg=
* finalising deprecations of `msg` in `pytest.skip()` and `pytest.fail()`
* Update doc/en/deprecations.rst
Co-authored-by: Bruno Oliveira <nicoddemus@gmail.com>
* Update doc/en/deprecations.rst
Co-authored-by: Bruno Oliveira <nicoddemus@gmail.com>
* fix failing test after upstream merge
* adding deprecation to `pytest.exit(msg=...)`
* add docs for pytest.exit deprecations
* finalising deprecation of msg for pytest.skip, pytest.exit and pytest.fail
* hold a reference to the Scope instance to please mypy
Co-authored-by: Bruno Oliveira <nicoddemus@gmail.com>
Fixes#9272.
Fixing the issue directly in the plugin is somewhat hard, so do it in
core. Since the plugin is going to be deprecated, I figure it's OK to
cheat a bit.
`reportinfo()` is the last remaining py.path-only code path in pytest,
i.e. the last piece holding back py.path deprecation. The problem with
it is that plugins/users use it from both sides -- implementing it
(returning the value) and using it (using the return value). Dealing
with implementers is easy enough -- allow to return `os.PathLike[str]`.
But for callers who expect strictly `py.path` this will break and
there's not really a good way to provide backward compat for this.
From analyzing a corpus of 680 pytest plugins, the vast majority of
`reportinfo` appearances are implementations, and the few callers don't
actually access the path part of the return tuple.
As for test suites that might access `reportinfo` (e.g. using
`request.node.reportinfo()` or other ways), that is much harder to
survey, but from the ones I searched, I only found case
(`pytest_teamcity`, but even then it uses `str(fspath)` so is unlikely
to be affected in practice). They are better served with using
`node.location` or `node.path` directly.
Therefore, just break it and change the return type to
`str|os.PathLike[str]`.
Refs #7259.
This type is semi-private; not documented but many plugins access it
through `item.callspec`. However, plugins access the public fields and
almost none try to construct or monkeypatch it. So we should be allowed
to clean it up some.
- Convert to attrs, add slots and frozen
- Instead of doing `new = old.copy(); new.setmulti2()`, do `new =
old.setmulti()`. This is cleaner and faster.
- Remove the `metafunc` attribute. This causes a reference cycle
(multifunc._calls -> callspec -> multifunc) for no good reason --
neither pytest itself or plugins access this attribute, so let's not
keep the Metafunc objects alive past their due.
- Some comments.
I would have also like to make the dicts and lists themselves immutable,
however some plugins mess with those so that should be done separately,
if at all.
* add feature to view fixture source location in invocations with --fixtures-per-test option
* remove unrelated changes to show_fixtures_per_test::test_doctest_items
* eshew the extraneous else in _show_fixtures_per_test.write_fixture
* enable the accommodation of multi-line docstring with --fixtures-per-test option
* add feature to view fixture source location in invocations with --fixtures
* add colour encoding to fixture location paths
* add changelog for #8606 fixing
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
* [pre-commit.ci] pre-commit autoupdate
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: Ran Benita <ran@unusedvar.com>
Similarly to #7143, at work we have a project with a custom pytest.Class
subclass, adding an additional argument to the constructor.
All from_parent implementations in pytest accept and forward *kw, except
Class (before this change) and DoctestItem - since I'm not familiar with
doctest support, I've left the latter as-is.
as PyObjMixin is always supposed to be mixed in the mro
before nodes.Node the behavior doesn't change,
but all the typing information carry over to help mypy.
extracted from #8037
* [pre-commit.ci] pre-commit autoupdate
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* manual fixes after configuration update
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: Anthony Sottile <asottile@umich.edu>
In order to allow users to type annotate fixtures they request, the
types need to be imported from the `pytest` namespace. They are/were
always available to import from the `_pytest` namespace, but that is
not guaranteed to be stable.
These types are only exported for the purpose of typing. Specifically,
the following are *not* public:
- Construction (`__init__`)
- Subclassing
- staticmethods and classmethods
We try to combat them being used anyway by:
- Marking the classes as `@final` when possible (already done).
- Not documenting private stuff in the API Reference.
- Using `_`-prefixed names or marking as `:meta private:` for private
stuff.
- Adding a keyword-only `_ispytest=False` to private constructors,
warning if False, and changing pytest itself to pass True. In the
future it will (hopefully) become a hard error.
Hopefully that will be enough.
Since commit 0f918b1a9d pytest uses auto-generated autouse
pytest fixtures for the xunit fixtures
{setup,teardown}_{module,class,method,function}. All of these fixtures
were given the same name.
Unfortunately, pytest fixture lookup for a name works by grabbing all of
the fixtures globally declared with a name and filtering to only those
which match the specific node. So each xunit-using item iterates over a
list (of fixturedefs) of a size of all previous same-xunit-using items,
i.e. quadratic.
Fixing this properly to use a better data structure is likely to take
some effort, but we can avoid the immediate problem by just using
a different name for each item's autouse fixture, so it only matches
itself.
A benchmark is added to demonstrate the issue. It is still way too slow
after the fix and possibly still quadratic, but for a different reason
which is another matter.
Running --collect-only, before (snipped):
202533232 function calls (201902604 primitive calls) in 86.379 seconds
ncalls tottime percall cumtime percall filename:lineno(function)
1 0.000 0.000 85.688 85.688 main.py:320(pytest_collection)
1 0.000 0.000 85.688 85.688 main.py:567(perform_collect)
80557/556 0.021 0.000 85.050 0.153 {method 'extend' of 'list' objects}
85001/15001 0.166 0.000 85.045 0.006 main.py:785(genitems)
10002 0.050 0.000 84.219 0.008 runner.py:455(collect_one_node)
10002 0.049 0.000 83.763 0.008 runner.py:340(pytest_make_collect_report)
10002 0.079 0.000 83.668 0.008 runner.py:298(from_call)
10002 0.019 0.000 83.541 0.008 runner.py:341(<lambda>)
5001 0.184 0.000 81.922 0.016 python.py:412(collect)
5000 0.020 0.000 81.072 0.016 python.py:842(collect)
30003 0.118 0.000 78.478 0.003 python.py:218(pytest_pycollect_makeitem)
30000 0.190 0.000 77.957 0.003 python.py:450(_genfunctions)
40001 0.081 0.000 76.664 0.002 nodes.py:183(from_parent)
30000 0.087 0.000 76.629 0.003 python.py:1595(from_parent)
40002 0.092 0.000 76.583 0.002 nodes.py:102(_create)
30000 0.305 0.000 76.404 0.003 python.py:1533(__init__)
15000 0.132 0.000 74.765 0.005 fixtures.py:1439(getfixtureinfo)
15000 0.165 0.000 73.664 0.005 fixtures.py:1492(getfixtureclosure)
15000 0.044 0.000 57.584 0.004 fixtures.py:1653(getfixturedefs)
30000 18.840 0.001 57.540 0.002 fixtures.py:1668(_matchfactories)
37507500 31.352 0.000 38.700 0.000 nodes.py:76(ischildnode)
15000 10.464 0.001 15.806 0.001 fixtures.py:1479(_getautousenames)
112930587/112910019 7.333 0.000 7.339 0.000 {built-in method builtins.len}
After:
51890333 function calls (51259706 primitive calls) in 27.306 seconds
ncalls tottime percall cumtime percall filename:lineno(function)
1 0.000 0.000 26.783 26.783 main.py:320(pytest_collection)
1 0.000 0.000 26.783 26.783 main.py:567(perform_collect)
80557/556 0.020 0.000 26.108 0.047 {method 'extend' of 'list' objects}
85001/15001 0.151 0.000 26.103 0.002 main.py:785(genitems)
10002 0.047 0.000 25.324 0.003 runner.py:455(collect_one_node)
10002 0.045 0.000 24.888 0.002 runner.py:340(pytest_make_collect_report)
10002 0.069 0.000 24.805 0.002 runner.py:298(from_call)
10002 0.017 0.000 24.690 0.002 runner.py:341(<lambda>)
5001 0.168 0.000 23.150 0.005 python.py:412(collect)
5000 0.019 0.000 22.223 0.004 python.py:858(collect)
30003 0.101 0.000 19.818 0.001 python.py:218(pytest_pycollect_makeitem)
30000 0.161 0.000 19.368 0.001 python.py:450(_genfunctions)
30000 0.302 0.000 18.236 0.001 python.py:1611(from_parent)
40001 0.084 0.000 18.051 0.000 nodes.py:183(from_parent)
40002 0.116 0.000 17.967 0.000 nodes.py:102(_create)
30000 0.308 0.000 17.770 0.001 python.py:1549(__init__)
15000 0.117 0.000 16.111 0.001 fixtures.py:1439(getfixtureinfo)
15000 0.134 0.000 15.135 0.001 fixtures.py:1492(getfixtureclosure)
15000 9.320 0.001 14.738 0.001 fixtures.py:1479(_getautousenames)
This indicates at least for people using type checkers that these
classes are not designed for inheritance and we make no stability
guarantees regarding inheritance of them.
Currently this doesn't show up in the docs. Sphinx does actually support
`@final`, however it only works when imported directly from `typing`,
while we import from `_pytest.compat`.
In the future there might also be a `@sealed` decorator which would
cover some more cases.
Inline `_makeitem()` so that `self.ihook` (which is moderately
expensive) can be called only once.
Note: the removed test "test_makeitem_non_underscore" comes from an old
behavior of skipping names that start with `_` which has since been
generalized, making the test no longer relevant.
When a Python object (module/class/instance) is collected, for each name
in `obj.__dict__` (and up its MRO) the pytest_pycollect_makeitem hook is
called for potentially creating a node for it.
These Python objects have a bunch of builtin attributes that are
extremely unlikely to be collected. But due to their pervasiveness,
dispatching the hook for them ends up being mildly expensive and also
pollutes PYTEST_DEBUG=1 output and such.
Let's just ignore these attributes.
On the pandas test suite commit 04e9e0afd476b1b8bed930e47bf60e,
collect only, irrelevant lines snipped, about 5% improvement:
Before:
```
51195095 function calls (48844352 primitive calls) in 39.089 seconds
ncalls tottime percall cumtime percall filename:lineno(function)
226602/54 0.145 0.000 38.940 0.721 manager.py:90(_hookexec)
72227 0.285 0.000 20.146 0.000 python.py:424(_makeitem)
72227 0.171 0.000 16.678 0.000 python.py:218(pytest_pycollect_makeitem)
```
After:
```
48410921 function calls (46240870 primitive calls) in 36.950 seconds
ncalls tottime percall cumtime percall filename:lineno(function)
181429/54 0.113 0.000 36.777 0.681 manager.py:90(_hookexec)
27054 0.130 0.000 17.755 0.001 python.py:465(_makeitem)
27054 0.121 0.000 16.219 0.001 python.py:218(pytest_pycollect_makeitem)
```
In ff8b7884e8 NOTSET was changed to a
singleton enum, which ended up unexpectedly triggering a code path in ID
generation which checks for `isinstance(Enum)`.
Add an explicit case for it, which is not too bad anyway.
This reverts commit f10ab021e2.
The commit was good in that it removed a non-trivial amount of code
duplication. However it was done in the wrong layer (nodes.py) and split
up a major part of the collection (the filesystem traversal) to a
separate class making it harder to understand.
We should try to reduce the duplication, but in a more appropriate
manner.
This fixes an issue where pylint complains about missing implementations
of abstract methods in subclasses of `File` which only override
`collect()` (as they should).
It is also cleaner and makes sense, these methods really don't need to
be overridden.
The previous methods defined directly on `FSCollector` and `Package` are
deprecated, to be removed in pytest 7.
See commits e2934c3f8c and
f10ab021e2 for reference.
This makes mypy raise an error whenever it detects code which is
statically unreachable, e.g.
x: int
if isinstance(x, str):
... # Statement is unreachable [unreachable]
This is really neat and finds quite a few logic and typing bugs.
Sometimes the code is intentionally unreachable in terms of types, e.g.
raising TypeError when a function is given an argument with a wrong
type. In these cases a `type: ignore[unreachable]` is needed, but I
think it's a nice code hint.
This prevents referring to a generic type without filling in its generic
type parameters.
The FixtureDef typing might need some more refining in the future.
In Python, if module A defines a name `name`, and module B does `import
name from A`, then another module C can `import name from B`.
Sometimes it is intentional -- module B is meant to "reexport" `name`.
But sometimes it is just confusion/inconsistency on where `name` should
be imported from.
mypy has a flag `--no-implicit-reexport` which puts some order into
this. A name can only be imported from a module if
1. The module defines the name
2. The module's `__all__` includes the name
3. The module imports the name as `from ... import .. as name`.
This flag is included in mypy's `--strict` flag.
I like this flag, but I realize it is a bit controversial, and in
particular item 3 above is a bit unfriendly to contributors who don't
know about it. So I didn't intend to add it to pytest.
But while investigating issue 7589 I came upon mypy issue 8754 which
causes `--no-implicit-reexport` to leak into installed libraries and
causes some unexpected typing differences *in pytest* if the user uses
this flag.
Since the diff mostly makes sense, let's just conform to it.
Ran Benita
pre-commit-ci[bot]
Teejay
Marko Pacak
Simon K
Hugo van Kemenade
Cheuk Ting Ho
Bruno Oliveira
Tobias Deiminger
Anthony Sottile
Kian Meng, Ang
Florian Bruhin
Naveen-Pratap
Graeme Smecher
Rahul Kumaresan
Ronny Pfannschmidt
Tadeu Manoel
Alexandros Tzannes
antonblr
Sorin Sbarnea