## Summary
Theoretically this should be slightly more performant, since the
`class.is_known()` calls each do a separate Salsa lookup, which we can
avoid if we do a single `match` on the value of `class.known()`. It also
ends up being two lines less code overall!
## Test Plan
`cargo test -p red_knot_python_semantic`
## Summary
Fixes#16566, fixes#16575
The semantics of `Type::class_member` changed in
https://github.com/astral-sh/ruff/pull/16416, but the property-test
infrastructure was not updated. That means that the property tests were
panicking on the second `expect_type` call here:
0361021863/crates/red_knot_python_semantic/src/types/property_tests.rs (L151-L158)
With the somewhat unhelpful message:
```
Expected a (possibly unbound) type, not an unbound symbol
```
Applying this patch, and then running `QUICKCHECK_TESTS=1000000 cargo
test --release -p red_knot_python_semantic -- --ignored
types::property_tests::stable::equivalent_to_is_reflexive` showed
clearly that it was no longer able to find _any_ methods on _any_
classes due to the change in semantics of `Type::class_member`:
```diff
--- a/crates/red_knot_python_semantic/src/types/property_tests.rs
+++ b/crates/red_knot_python_semantic/src/types/property_tests.rs
@@ -27,7 +27,7 @@
use std::sync::{Arc, Mutex, MutexGuard, OnceLock};
use crate::db::tests::{setup_db, TestDb};
-use crate::symbol::{builtins_symbol, known_module_symbol};
+use crate::symbol::{builtins_symbol, known_module_symbol, Symbol};
use crate::types::{
BoundMethodType, CallableType, IntersectionBuilder, KnownClass, KnownInstanceType,
SubclassOfType, TupleType, Type, UnionType,
@@ -150,10 +150,11 @@ impl Ty {
Ty::BuiltinsFunction(name) => builtins_symbol(db, name).symbol.expect_type(),
Ty::BuiltinsBoundMethod { class, method } => {
let builtins_class = builtins_symbol(db, class).symbol.expect_type();
- let function = builtins_class
- .class_member(db, method.into())
- .symbol
- .expect_type();
+ let Symbol::Type(function, ..) =
+ builtins_class.class_member(db, method.into()).symbol
+ else {
+ panic!("no method `{method}` on class `{class}`");
+ };
create_bound_method(db, function, builtins_class)
}
```
This PR updates the property-test infrastructure to use `Type::member`
rather than `Type::class_member`.
## Test Plan
- Ran `QUICKCHECK_TESTS=1000000 cargo test --release -p
red_knot_python_semantic -- --ignored types::property_tests::stable`
successfully
- Checked that there were no remaining uses of `Type::class_member` in
`property_tests.rs`
## Summary
Fixes a small nit of mine -- we are currently inconsistent in our
spelling between "metaclass" and "meta class", and between "meta type"
and "meta-type". This PR means that we consistently use "metaclass" and
"meta-type".
## Test Plan
`uvx pre-commit run -a`
## Summary
Part of https://github.com/astral-sh/ruff/issues/15382
This PR implements a general callable type that wraps around a
`Signature` and it uses that new type to represent `typing.Callable`.
It also implements `Display` support for `Callable`. The format is as:
```
([<arg name>][: <arg type>][ = <default type>], ...) -> <return type>
```
The `/` and `*` separators are added at the correct boundary for
positional-only and keyword-only parameters. Now, as `typing.Callable`
only has positional-only parameters, the rendered signature would be:
```py
Callable[[int, str], None]
# (int, str, /) -> None
```
The `/` separator represents that all the arguments are positional-only.
The relationship methods that check assignability, subtype relationship,
etc. are not yet implemented and will be done so as a follow-up.
## Test Plan
Add test cases for display support for `Signature` and various mdtest
for `typing.Callable`.
## Summary
Resolves#16365
Add support for unpacking `with` statement targets.
## Test Plan
Added some test cases, alike the ones added by #15058.
---------
Co-authored-by: Carl Meyer <carl@astral.sh>
## Summary
* Attributes/method are now properly looked up on metaclasses, when
called on class objects
* We properly distinguish between data descriptors and non-data
descriptors (but we do not yet support them in store-context, i.e.
`obj.data_descr = …`)
* The descriptor protocol is now implemented in a single unified place
for instances, classes and dunder-calls. Unions and possibly-unbound
symbols are supported in all possible stages of the process by creating
union types as results.
* In general, the handling of "possibly-unbound" symbols has been
improved in a lot of places: meta-class attributes, attributes,
descriptors with possibly-unbound `__get__` methods, instance
attributes, …
* We keep track of type qualifiers in a lot more places. I anticipate
that this will be useful if we import e.g. `Final` symbols from other
modules (see relevant change to typing spec:
https://github.com/python/typing/pull/1937).
* Detection and special-casing of the `typing.Protocol` special form in
order to avoid lots of changes in the test suite due to new `@Todo`
types when looking up attributes on builtin types which have `Protocol`
in their MRO. We previously
looked up attributes in a wrong way, which is why this didn't come up
before.
closes#16367closes#15966
## Context
The way attribute lookup in `Type::member` worked before was simply
wrong (mostly my own fault). The whole instance-attribute lookup should
probably never have been integrated into `Type::member`. And the
`Type::static_member` function that I introduced in my last descriptor
PR was the wrong abstraction. It's kind of fascinating how far this
approach took us, but I am pretty confident that the new approach
proposed here is what we need to model this correctly.
There are three key pieces that are required to implement attribute
lookups:
- **`Type::class_member`**/**`Type::find_in_mro`**: The
`Type::find_in_mro` method that can look up attributes on class bodies
(and corresponding bases). This is a partial function on types, as it
can not be called on instance types like`Type::Instance(…)` or
`Type::IntLiteral(…)`. For this reason, we usually call it through
`Type::class_member`, which is essentially just
`type.to_meta_type().find_in_mro(…)` plus union/intersection handling.
- **`Type::instance_member`**: This new function is basically the
type-level equivalent to `obj.__dict__[name]` when called on
`Type::Instance(…)`. We use this to discover instance attributes such as
those that we see as declarations on class bodies or as (annotated)
assignments to `self.attr` in methods of a class.
- The implementation of the descriptor protocol. It works slightly
different for instances and for class objects, but it can be described
by the general framework:
- Call `type.class_member("attribute")` to look up "attribute" in the
MRO of the meta type of `type`. Call the resulting `Symbol` `meta_attr`
(even if it's unbound).
- Use `meta_attr.class_member("__get__")` to look up `__get__` on the
*meta type* of `meta_attr`. Call it with `__get__(meta_attr, self,
self.to_meta_type())`. If this fails (either the lookup or the call),
just proceed with `meta_attr`. Otherwise, replace `meta_attr` in the
following with the return type of `__get__`. In this step, we also probe
if a `__set__` or `__delete__` method exists and store it in
`meta_attr_kind` (can be either "data descriptor" or "normal attribute
or non-data descriptor").
- Compute a `fallback` type.
- For instances, we use `self.instance_member("attribute")`
- For class objects, we use `class_attr =
self.find_in_mro("attribute")`, and then try to invoke the descriptor
protocol on `class_attr`, i.e. we look up `__get__` on the meta type of
`class_attr` and call it with `__get__(class_attr, None, self)`. This
additional invocation of the descriptor protocol on the fallback type is
one major asymmetry in the otherwise universal descriptor protocol
implementation.
- Finally, we look at `meta_attr`, `meta_attr_kind` and `fallback`, and
handle various cases of (possible) unboundness of these symbols.
- If `meta_attr` is bound and a data descriptor, just return `meta_attr`
- If `meta_attr` is not a data descriptor, and `fallback` is bound, just
return `fallback`
- If `meta_attr` is not a data descriptor, and `fallback` is unbound,
return `meta_attr`
- Return unions of these three possibilities for partially-bound
symbols.
This allows us to handle class objects and instances within the same
framework. There is a minor additional detail where for instances, we do
not allow the fallback type (the instance attribute) to completely
shadow the non-data descriptor. We do this because we (currently) don't
want to pretend that we can statically infer that an instance attribute
is always set.
Dunder method calls can also be embedded into this framework. The only
thing that changes is that *there is no fallback type*. If a dunder
method is called on an instance, we do not fall back to instance
variables. If a dunder method is called on a class object, we only look
it up on the meta class, never on the class itself.
## Test Plan
New Markdown tests.
## Summary
This PR closes#15199.
The change I just made is to set all variables to type `Unknown` if
unpacking fails, but in some cases this may be excessive.
For example:
```py
a, b, c = "ab"
reveal_type(a) # Unknown, but it would be reasonable to think of it as LiteralString
reveal_type(c) # Unknown
```
```py
# Failed to unpack before the starred expression
(a, b, *c, d, e) = (1,)
reveal_type(a) # Unknown
reveal_type(b) # Unknown
...
# Failed to unpack after the starred expression
(a, b, *c, d, e) = (1, 2, 3)
reveal_type(a) # Unknown, but should it be Literal[1]?
reveal_type(b) # Unknown, but should it be Literal[2]?
reveal_type(c) # Todo
reveal_type(d) # Unknown
reveal_type(e) # Unknown
```
I will modify it if you think it would be better to make it a different
type than just `Unknown`.
## Test Plan
I have made appropriate modifications to the test cases affected by this
change, and also added some more test cases.
## Summary
This should give us better coverage for the unsupported syntax error
features and
increases our confidence that the formatter doesn't accidentially
introduce new unsupported
syntax errors.
A feature like this would have been very useful when working on f-string
formatting
where it took a lot of iteration to find all Python 3.11 or older
incompatibilities.
## Test Plan
I applied my changes on top of
https://github.com/astral-sh/ruff/pull/16523 and
removed the target version check in the with-statement formatting code.
As expected,
the integration tests now failed
<!--
Thank you for contributing to Ruff! To help us out with reviewing,
please consider the following:
- Does this pull request include a summary of the change? (See below.)
- Does this pull request include a descriptive title?
- Does this pull request include references to any relevant issues?
-->
## Summary
<!-- What's the purpose of the change? What does it do, and why? -->
If an mdtest fails, the error output will include an example command
that can be run to re-run just the failing test, e.g
```
To rerun this specific test, set the environment variable: MDTEST_TEST_FILTER="sync.md - With statements - Context manager with non-callable `__exit__` attribute"
MDTEST_TEST_FILTER="sync.md - With statements - Context manager with non-callable `__exit__` attribute" cargo test -p red_knot_python_semantic --test mdtest -- mdtest__with_sync
```
This is very helpful, but because we're printing the envvar value
surrounded in double-quotes, the bits between backticks in this example
get interpreted as a shell interpolation. When running this in zsh, for
example, I see
```console
❯ MDTEST_TEST_FILTER="sync.md - With statements - Context manager with non-callable `__exit__` attribute" cargo test -p red_knot_python_semantic --test mdtest -- mdtest__with_sync
zsh: command not found: __exit__
Compiling red_knot_python_semantic v0.0.0 (/home/ericmarkmartin/Development/ruff/crates/red_knot_python_semantic)
Compiling red_knot_test v0.0.0 (/home/ericmarkmartin/Development/ruff/crates/red_knot_test)
Finished `test` profile [unoptimized + debuginfo] target(s) in 6.09s
Running tests/mdtest.rs (target/debug/deps/mdtest-149b8f9d937e36bc)
running 1 test
test mdtest__with_sync ... ok
```
[^1]
This is a minor annoyance which we can solve by using single-quotes
instead of double-quotes for this string. To do so safely, we also
escape single-quotes possibly contained within the string.
There is a [shell-quote](https://github.com/allenap/shell-quote) crate,
which seems to handle all this escaping stuff for you but fixing this
issue perfectly isn't a big deal (if there are more things to escape we
can deal with it then), so adding a new dependency (even a dev one)
seemed overkill.
[^1]: The filter does still work---it turns out that the filter
`MDTEST_TEST_FILTER="sync.md - With statements - Context manager with
non-callable attribute"` (what you get after the failed interpolation)
is still good enough
## Test Plan
<!-- How was it tested? -->
I broke the ``## Context manager with non-callable `__exit__`
attribute`` test by deleting the error assertion, then successfully ran
the new command it printed out.
Summary
--
Unlike the other syntax errors detected so far, parenthesized keyword
arguments are only allowed *before* 3.8. It sounds like they were only
accidentally allowed before that [^1].
As an aside, you get a pretty confusing error from Python for this, so
it's nice that we can catch it:
```pycon
>>> def f(**kwargs): ...
... f((a)=1)
...
File "<python-input-0>", line 2
f((a)=1)
^^^
SyntaxError: expression cannot contain assignment, perhaps you meant "=="?
>>>
```
Test Plan
--
Inline tests.
[^1]: https://github.com/python/cpython/issues/78822
Summary
--
Checks for tuple unpacking in `return` and `yield` statements before
Python 3.8, as described [here].
Test Plan
--
Inline tests.
[here]: https://github.com/python/cpython/issues/76298
## Summary
- `Never` is callable
- `Never` is iterable
- Arbitrary attributes can be accessed on `Never`
Split out from #16416 that is going to be required.
## Test Plan
Tests for all properties above.
## Summary
This came up in https://github.com/astral-sh/ruff/issues/16477
It's not obvious from the D417 rule's documentation that it only checks
docstrings
with an arguments section. Functions without such a section aren't
checked.
This PR tries to make this clearer in the documentation.
## Summary
This PR introduces a new mdtest option `system` that can either be
`in-memory` or `os`
where `in-memory` is the default.
The motivation for supporting `os` is so that we can write OS/system
specific tests
with mdtests. Specifically, I want to write mdtests for the module
resolver,
testing that module resolution is case sensitive.
## Test Plan
I tested that the case-sensitive module resolver test start failing when
setting `system = "os"`
## Summary
Python's module resolver is case sensitive.
This PR adds mdtests that assert that our module resolution is case
sensitive.
The tests currently all pass because our in memory file system is case
sensitive.
I'll add support for using the real file system to the mdtest framework
in a separate PR.
This PR also adds support for specifying extra search paths to the
mdtest framework.
## Test Plan
The tests fail when running them using the real file system.
To kick off the work of supporting generics, this adds many new
(currently failing) tests, showing the behavior we plan to support.
This is still missing a lot! Not included:
- typevar tuples
- param specs
- variance
- `Self`
But it's a good start! We can add more failing tests for those once we
tackle these.
---------
Co-authored-by: Carl Meyer <carl@astral.sh>
This is split out of https://github.com/astral-sh/ruff/pull/14029, to
reduce the size of that PR, and to validate that this "fallback type"
support in `TypeInference` doesn't come with a performance cost. It also
improves the reliability and debuggability of our current (temporary)
cycle handling.
In order to recover from a cycle, we have to be able to construct a
"default" `TypeInference` where all expressions and definitions have
some "default" type. In our current cycle handling, this "default" type
is just unknown or a todo type. With fixpoint iteration, the "default"
type will be `Type::Never`, which is the "bottom" type that fixpoint
iteration starts from.
Since it would be costly (both in space and time) to actually enumerate
all expressions and definitions in a scope, just to insert the same
default type for all of them, instead we add an optional "missing type"
fallback to `TypeInference`, which (if set) is the fallback type for any
expression or definition which doesn't have an explicit type set.
With this change, cycles can no longer result in the dreaded "Missing
key" errors looking up the type of some expression.
... with supporting types. This is meant to give us a base to work with
in terms of our new diagnostic data model. I expect the representations
to be tweaked over time, but I think this is a decent start.
I would also like to add doctest examples, but I think it's better if we
wait until an initial version of the renderer is done for that.
This puts them out of the way so that they can hopefully be removed more
easily in the (near) future, and so that they don't get in the way of
the new types. This also makes the intent of the migration a bit clearer
in the code and hopefully results in less confusion.
This trait should eventually go away, so we rename it (and supporting
types) to make room for a new concrete `Diagnostic` type.
This commit is just the rename. In the next commit, we'll move it to a
different module.
Summary
--
Another simple one, just detect type parameter lists in functions
and classes. Like pyright, we don't emit a second diagnostic for
`type` alias statements, which were also introduced in 3.12.
Test Plan
--
Inline tests.
## Summary
This PR does a small refactor to avoid double
`symbol_table(...).symbol(...)` call to check for `__slots__` and
`TYPE_CHECKING`. It merges them into a single call.
I noticed this while looking at
https://github.com/astral-sh/ruff/pull/16468.
## Summary
This PR adds more features to #16468.
* Adds a new error rule `invalid-type-checking-constant`, which occurs
when we try to assign a value other than `False` to a user-defined
`TYPE_CHECKING` variable (it is possible to assign `...` in a stub
file).
* Allows annotated assignment to `TYPE_CHECKING`. Only types that
`False` can be assigned to are allowed. However, the type of
`TYPE_CHECKING` will be inferred to be `Literal[True]` regardless of
what the type is specified.
## Test plan
I ran the tests with `cargo test -p red_knot_python_semantic` and
confirmed that all tests passed.
---------
Co-authored-by: Carl Meyer <carl@astral.sh>
## Summary
Fixes https://github.com/astral-sh/ruff/issues/16476fixes: #11453
We format notebooks cell by cell. That means, that offsets in parse
errors are relative
to the cell and not the entire document. We didn't account for this fact
when emitting syntax errors for notebooks in the formatter.
This PR ensures that we correctly offset parse errors by the cell
location.
## Test Plan
Added test (it panicked before)
Summary
--
Detects the presence of a [PEP 696] type parameter default before Python
3.13.
Test Plan
--
New inline parser tests for type aliases, generic functions and generic
classes.
[PEP 696]: https://peps.python.org/pep-0696/#grammar-changes
Summary
--
This is a follow-up to #16446 to fix the diagnostic range to point to
the `*` like `pyright` does
(https://github.com/astral-sh/ruff/pull/16446#discussion_r1976900643).
Storing the range in the `ExceptClauseKind::Star` variant feels slightly
awkward, but we don't store the star itself anywhere on the
`ExceptHandler`. And we can't just take `ExceptHandler.start() +
"except".text_len()` because this code appears to be valid:
```python
try: ...
except * Error: ...
```
Test Plan
--
Existing tests.
## Summary
This PR closes#15722.
The change is that if the variable `TYPE_CHECKING` is defined/imported,
the type of the variable is interpreted as `Literal[True]` regardless of
what the value is.
This is compatible with the behavior of other type checkers (e.g. mypy,
pyright).
## Test Plan
I ran the tests with `cargo test -p red_knot_python_semantic` and
confirmed that all tests passed.
---------
Co-authored-by: Carl Meyer <carl@astral.sh>
Summary
--
This is a follow up addressing the comments on #16425. As @dhruvmanila
pointed out, the naming is a bit tricky. I went with `has_no_errors` to
try to differentiate it from `is_valid`. It actually ends up negated in
most uses, so it would be more convenient to have `has_any_errors` or
`has_errors`, but I thought it would sound too much like the opposite of
`is_valid` in that case. I'm definitely open to suggestions here.
Test Plan
--
Existing tests.
## Summary
Resolves#16445.
`UP028` is now no longer always fixable: it will not offer a fix when at
least one `ExprName` target is bound to either a `global` or a
`nonlocal` declaration.
## Test Plan
`cargo nextest run` and `cargo insta test`.
## Summary
Fixes#9381. This PR fixes errors like
```
Cause: error parsing glob '/Users/me/project/{{cookiecutter.project_dirname}}/__pycache__': nested alternate groups are not allowed
```
caused by glob special characters in filenames like
`{{cookiecutter.project_dirname}}`. When the user is matching that
directory exactly, they can use the workaround given by
https://github.com/astral-sh/ruff/issues/7959#issuecomment-1764751734,
but that doesn't work for a nested config file with relative paths. For
example, the directory tree in the reproduction repo linked
[here](https://github.com/astral-sh/ruff/issues/9381#issuecomment-2677696408):
```
.
├── README.md
├── hello.py
├── pyproject.toml
├── uv.lock
└── {{cookiecutter.repo_name}}
├── main.py
├── pyproject.toml
└── tests
└── maintest.py
```
where the inner `pyproject.toml` contains a relative glob:
```toml
[tool.ruff.lint.per-file-ignores]
"tests/*" = ["F811"]
```
## Test Plan
A new CLI test in both the linter and formatter. The formatter test may
not be necessary because I didn't have to modify any additional code to
pass it, but the original report mentioned both `check` and `format`, so
I wanted to be sure both were fixed.
The PR addresses issue #16396 .
Specifically:
- If the exit statement contains a code keyword argument, it is
converted into a positional argument.
- If retrieving the code from the exit statement is not possible, a
violation is raised without suggesting a fix.
---------
Co-authored-by: Brent Westbrook <36778786+ntBre@users.noreply.github.com>
## Summary
This PR adds support for an optional list of paths that should be
checked to `knot check`.
E.g. to only check the `src` directory
```sh
knot check src
```
The default is to check all files in the project but users can reduce
the included files by specifying one or multiple optional paths.
The main two challenges with adding this feature were:
* We now need to show an error when one of the provided paths doesn't
exist. That's why this PR now collects errors from the project file
indexing phase and adds them to the output diagnostics. The diagnostic
looks similar to ruffs (see CLI test)
* The CLI should pick up new files added to included folders. For
example, `knot check src --watch` should pick up new files that are
added to the `src` folder. This requires that we now filter the files
before adding them to the project. This is a good first step to
supporting `include` and `exclude`.
The PR makes two simplifications:
1. I didn't test the changes with case-insensitive file systems. We may
need to do some extra path normalization to support those well. See
https://github.com/astral-sh/ruff/issues/16400
2. Ideally, we'd accumulate the IO errors from the initial indexing
phase and subsequent incremental indexing operations. For example, we
should preserve the IO diagnostic for a non existing `test.py` if it was
specified as an explicit CLI argument until the file gets created and we
should show it again when the file gets deleted. However, this is
somewhat complicated because we'd need to track which files we revisited
(or were removed because the entire directory is gone). I considered
this too low a priority as it's worth dealing with right now.
The implementation doesn't support symlinks within the project but that
is the same as Ruff and is unchanged from before this PR.
Closes https://github.com/astral-sh/ruff/issues/14193
## Test Plan
Added CLI and file watching integration tests. Manually testing.
Split from F841 following discussion in #8884.
Fixes#8884.
<!--
Thank you for contributing to Ruff! To help us out with reviewing,
please consider the following:
- Does this pull request include a summary of the change? (See below.)
- Does this pull request include a descriptive title?
- Does this pull request include references to any relevant issues?
-->
## Summary
<!-- What's the purpose of the change? What does it do, and why? -->
Add a new rule for unused assignments in tuples. Remove similar behavior
from F841.
## Test Plan
Adapt F841 tests and move them over to the new rule.
<!-- How was it tested? -->
---------
Co-authored-by: Micha Reiser <micha@reiser.io>
## Summary
This PR is the first in a series derived from
https://github.com/astral-sh/ruff/pull/16308, each of which add support
for detecting one version-related syntax error from
https://github.com/astral-sh/ruff/issues/6591. This one should be
the largest because it also includes the addition of the
`Parser::add_unsupported_syntax_error` method
Otherwise I think the general structure will be the same for each syntax
error:
* Detecting the error in the parser
* Inline parser tests for the new error
* New ruff CLI tests for the new error
## Test Plan
As noted above, there are new inline parser tests, as well as new ruff
CLI
tests. Once https://github.com/astral-sh/ruff/pull/16379 is resolved,
there should also be new mdtests for red-knot,
but this PR does not currently include those.
Regardless of whether #16408 and #16311 pan out, this part is worth
pulling out as a separate PR.
Before, you had to define a new `IndexVec` index type for each type of
association list you wanted to create. Now there's a single index type
that's internal to the alist implementation, and you use `List<K, V>` to
store a handle to a particular list.
This also adds some property tests for the alist implementation.
