## Summary
closes#14279
### Limitations of the Current Implementation
#### Incorrect Error Propagation
In the current implementation of lexicographic comparisons, if the
result of an Eq operation is Ambiguous, the comparison stops
immediately, returning a bool instance. While this may yield correct
inferences, it fails to capture unsupported-operation errors that might
occur in subsequent comparisons.
```py
class A: ...
(int_instance(), A()) < (int_instance(), A()) # should error
```
#### Weak Inference in Specific Cases
> Example: `(int_instance(), "foo") == (int_instance(), "bar")`
> Current result: `bool`
> Expected result: `Literal[False]`
`Eq` and `NotEq` have unique behavior in lexicographic comparisons
compared to other operators. Specifically:
- For `Eq`, if any non-equal pair exists within the tuples being
compared, we can immediately conclude that the tuples are not equal.
- For `NotEq`, if any equal pair exists, we can conclude that the tuples
are unequal.
```py
a = (str_instance(), int_instance(), "foo")
reveal_type(a == a) # revealed: bool
reveal_type(a != a) # revealed: bool
b = (str_instance(), int_instance(), "bar")
reveal_type(a == b) # revealed: bool # should be Literal[False]
reveal_type(a != b) # revealed: bool # should be Literal[True]
```
#### Incorrect Support for Non-Boolean Rich Comparisons
In CPython, aside from `==` and `!=`, tuple comparisons return a
non-boolean result as-is. Tuples do not convert the value into `bool`.
Note: If all pairwise `==` comparisons between elements in the tuples
return Truthy, the comparison then considers the tuples' lengths.
Regardless of the return type of the dunder methods, the final result
can still be a boolean.
```py
from __future__ import annotations
class A:
def __eq__(self, o: object) -> str:
return "hello"
def __ne__(self, o: object) -> bytes:
return b"world"
def __lt__(self, o: A) -> float:
return 3.14
a = (A(), A())
reveal_type(a == a) # revealed: bool
reveal_type(a != a) # revealed: bool
reveal_type(a < a) # revealed: bool # should be: `float | Literal[False]`
```
### Key Changes
One of the major changes is that comparisons no longer end with a `bool`
result when a pairwise `Eq` result is `Ambiguous`. Instead, the function
attempts to infer all possible cases and unions the results. This
improvement allows for more robust type inference and better error
detection.
Additionally, as the function is now optimized for tuple comparisons,
the name has been changed from the more general
`infer_lexicographic_comparison` to `infer_tuple_rich_comparison`.
## Test Plan
mdtest included
## Summary
Previously, we panicked on expressions like `f"{v:{f'0.2f'}}"` because
we did not infer types for expressions nested inside format spec
elements.
## Test Plan
```
cargo nextest run -p red_knot_workspace -- --ignored linter_af linter_gz
```
## Summary
Add type narrowing for `type(x) is C` conditions (and `else` clauses of
`type(x) is not C` conditionals):
```py
if type(x) is A:
reveal_type(x) # revealed: A
else:
reveal_type(x) # revealed: A | B
```
closes: #14431, part of: #13694
## Test Plan
New Markdown-based tests.
## Summary
This patches up various missing paths where sub-expressions of type
annotations previously had no type attached. Examples include:
```py
tuple[int, str]
# ~~~~~~~~
type[MyClass]
# ~~~~~~~
Literal["foo"]
# ~~~~~
Literal["foo", Literal[1, 2]]
# ~~~~~~~~~~~~~
Literal[1, "a", random.illegal(sub[expr + ession])]
# ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
```
## Test Plan
```
cargo nextest run -p red_knot_workspace -- --ignored linter_af linter_gz
```
## Summary
Follow-up to #14371, this PR simplifies the visitor logic for list
expressions to remove the state management. We just need to make sure
that we visit the nested expressions using the `QuoteAnnotator` and not
the `Generator`. This is similar to what's being done for binary
expressions.
As per the
[grammar](https://typing.readthedocs.io/en/latest/spec/annotations.html#grammar-token-expression-grammar-annotation_expression),
list expressions can be present which can contain other type expressions
(`Callable`):
```
| <Callable> '[' <Concatenate> '[' (type_expression ',')+
(name | '...') ']' ',' type_expression ']'
(where name must be a valid in-scope ParamSpec)
| <Callable> '[' '[' maybe_unpacked (',' maybe_unpacked)*
']' ',' type_expression ']'
```
## Test Plan
`cargo insta test`
## Summary
Resolves#12616.
## Test Plan
`cargo nextest run` and `cargo insta test`.
---------
Co-authored-by: Charlie Marsh <charlie.r.marsh@gmail.com>
## Summary
Resolves#14378.
## Test Plan
`cargo nextest run` and `cargo insta test`.
---------
Co-authored-by: Charlie Marsh <charlie.r.marsh@gmail.com>
## Summary
Disable the no-panic tests for the linter corpus, as there are too many
problems right now, requiring linter-contributors to add their test
files to the allow-list.
We can still run the tests using `cargo test -p red_knot_workspace --
--ignored linter_af linter_gz`. This is also why I left the
`crates/ruff_linter/` entries in the allow list for now, even if they
will get out of sync. But let me know if I should rather remove them.
## Summary
Implements `redundant-bool-literal`
## Test Plan
<!-- How was it tested? -->
`cargo test`
The ecosystem results are all correct, but for `Airflow` the rule is not
relevant due to the use of overloading (and is marked as unsafe
correctly).
---------
Co-authored-by: Charlie Marsh <charlie.r.marsh@gmail.com>
## Summary
This PR splits the corpus tests into smaller chunks because running all
of them takes 8s on my windows machine and it's by far the longest test
in `red_knot_workspace`.
Splitting the tests has the advantage that they run in parallel. This PR
brings down the wall time from 8s to 4s.
This PR also limits the glob for the linter tests because it's common to
clone cpython into the `ruff_linter/resources/test` folder for
benchmarks (because that's what's written in the contributing guides)
## Test Plan
`cargo test`
## Summary
This PR adds autofix for `redundant-numeric-union` (`PYI041`)
There are some comments below to explain the reasoning behind some
choices that might help review.
<!-- What's the purpose of the change? What does it do, and why? -->
Resolves part of https://github.com/astral-sh/ruff/issues/14185.
## Test Plan
<!-- How was it tested? -->
---------
Co-authored-by: Micha Reiser <micha@reiser.io>
Co-authored-by: Charlie Marsh <charlie.r.marsh@gmail.com>
This PR adds corrected handling of list expressions to the `Visitor`
implementation of `QuotedAnnotator` in `flake8_type_checking::helpers`.
Closes#14368
## Summary
- Add 383 files from `crates/ruff_python_parser/resources` to the test
corpus
- Add 1296 files from `crates/ruff_linter/resources` to the test corpus
- Use in-memory file system for tests
- Improve test isolation by cleaning the test environment between checks
- Add a mechanism for "known failures". Mark ~80 files as known
failures.
- The corpus test is now a lot slower (6 seconds).
Note:
While `red_knot` as a command line tool can run over all of these
files without panicking, we still have a lot of test failures caused by
explicitly "pulling" all types.
## Test Plan
Run `cargo test -p red_knot_workspace` while making sure that
- Introducing code that is known to lead to a panic fails the test
- Removing code that is known to lead to a panic from
`KNOWN_FAILURES`-files also fails the test
Fix: #13934
## Summary
Current implementation has a bug when the current annotation contains a
string with single and double quotes.
TL;DR: I think these cases happen less than other use cases of Literal.
So instead of fixing them we skip the fix in those cases.
One of the problematic cases:
```
from typing import Literal
from third_party import Type
def error(self, type1: Type[Literal["'"]]):
pass
```
The outcome is:
```
- def error(self, type1: Type[Literal["'"]]):
+ def error(self, type1: "Type[Literal[''']]"):
```
While it should be:
```
"Type[Literal['\'']"
```
The solution in this case is that we check if there’s any quotes same as
the quote style we want to use for this Literal parameter then escape
that same quote used in the string.
Also this case is not uncommon to have:
<https://grep.app/search?current=2&q=Literal["'>
But this can get more complicated for example in case of:
```
- def error(self, type1: Type[Literal["\'"]]):
+ def error(self, type1: "Type[Literal[''']]"):
```
Here we escaped the inner quote but in the generated annotation it gets
removed. Then we flip the quote style of the Literal paramter and the
formatting is wrong.
In this case the solution is more complicated.
1. When generating the string of the source code preserve the backslash.
2. After we have the annotation check if there isn’t any escaped quote
of the same type we want to use for the Literal parameter. In this case
check if we have any `’` without `\` before them. This can get more
complicated since there can be multiple backslashes so checking for only
`\’` won’t be enough.
Another problem is when the string contains `\n`. In case of
`Type[Literal["\n"]]` we generate `'Type[Literal["\n"]]'` and both
pyright and mypy reject this annotation.
https://pyright-play.net/?code=GYJw9gtgBALgngBwJYDsDmUkQWEMoAySMApiAIYA2AUAMaXkDOjUAKoiQNqsC6AXFAB0w6tQAmJYLBKMYAfQCOAVzCk5tMChjlUjOQCNytANaMGjABYAKRiUrAANLA4BGAQHJ2CLkVIVKnABEADoogTw87gCUfNRQ8VAITIyiElKksooqahpaOih6hiZmTNa29k7w3m5sHJy%2BZFRBoeE8MXEJScxAA
## Test Plan
I added test cases for the original code in the reported issue and two
more cases for backslash and new line.
---------
Co-authored-by: Dhruv Manilawala <dhruvmanila@gmail.com>
## Summary
This PR fixes a bug in the Ruff language server where the
editor-specified configuration was resolved relative to the
configuration directory and not the current working directory.
The existing behavior is confusing given that this config file is
specified by the user and is not _discovered_ by Ruff itself. The
behavior of resolving this configuration file should be similar to that
of the `--config` flag on the command-line which uses the current
working directory:
3210f1a23b/crates/ruff/src/resolve.rs (L34-L48)
This creates problems where certain configuration options doesn't work
because the paths resolved in that case are relative to the
configuration directory and not the current working directory in which
the editor is expected to be in. For example, the
`lint.per-file-ignores` doesn't work as mentioned in the linked issue
along with `exclude`, `extend-exclude`, etc.
fixes: #14282
## Test Plan
Using the following directory tree structure:
```
.
├── .config
│ └── ruff.toml
└── src
└── migrations
└── versions
└── a.py
```
where, the `ruff.toml` is:
```toml
# 1. Comment this out to test `per-file-ignores`
extend-exclude = ["**/versions/*.py"]
[lint]
select = ["D"]
# 2. Comment this out to test `extend-exclude`
[lint.per-file-ignores]
"**/versions/*.py" = ["D"]
# 3. Comment both `per-file-ignores` and `extend-exclude` to test selection works
```
And, the content of `a.py`:
```py
"""Test"""
```
And, the VS Code settings:
```jsonc
{
"ruff.nativeServer": "on",
"ruff.path": ["/Users/dhruv/work/astral/ruff/target/debug/ruff"],
// For single-file mode where current working directory is `/`
// "ruff.configuration": "/tmp/ruff-repro/.config/ruff.toml",
// When a workspace is opened containing this path
"ruff.configuration": "./.config/ruff.toml",
"ruff.trace.server": "messages",
"ruff.logLevel": "trace"
}
```
I also tested out just opening the file in single-file mode where the
current working directory is `/` in VS Code. Here, the
`ruff.configuration` needs to be updated to use absolute path as shown
in the above VS Code settings.
## Summary
This PR adds support for parsing and inferring types within string
annotations.
### Implementation (attempt 1)
This is preserved in
6217f48924.
The implementation here would separate the inference of string
annotations in the deferred query. This requires the following:
* Two ways of evaluating the deferred definitions - lazily and eagerly.
* An eager evaluation occurs right outside the definition query which in
this case would be in `binding_ty` and `declaration_ty`.
* A lazy evaluation occurs on demand like using the
`definition_expression_ty` to determine the function return type and
class bases.
* The above point means that when trying to get the binding type for a
variable in an annotated assignment, the definition query won't include
the type. So, it'll require going through the deferred query to get the
type.
This has the following limitations:
* Nested string annotations, although not necessarily a useful feature,
is difficult to implement unless we convert the implementation in an
infinite loop
* Partial string annotations require complex layout because inferring
the types for stringified and non-stringified parts of the annotation
are done in separate queries. This means we need to maintain additional
information
### Implementation (attempt 2)
This is the final diff in this PR.
The implementation here does the complete inference of string annotation
in the same definition query by maintaining certain state while trying
to infer different parts of an expression and take decisions
accordingly. These are:
* Allow names that are part of a string annotation to not exists in the
symbol table. For example, in `x: "Foo"`, if the "Foo" symbol is not
defined then it won't exists in the symbol table even though it's being
used. This is an invariant which is being allowed only for symbols in a
string annotation.
* Similarly, lookup name is updated to do the same and if the symbol
doesn't exists, then it's not bounded.
* Store the final type of a string annotation on the string expression
itself and not for any of the sub-expressions that are created after
parsing. This is because those sub-expressions won't exists in the
semantic index.
Design document:
https://www.notion.so/astral-sh/String-Annotations-12148797e1ca801197a9f146641e5b71?pvs=4Closes: #13796
## Test Plan
* Add various test cases in our markdown framework
* Run `red_knot` on LibCST (contains a lot of string annotations,
specifically
https://github.com/Instagram/LibCST/blob/main/libcst/matchers/_matcher_base.py),
FastAPI (good amount of annotated code including `typing.Literal`) and
compare against the `main` branch output
## Summary
Add a typed representation of function signatures (parameters and return
type) and infer it correctly from a function.
Convert existing usage of function return types to use the signature
representation.
This does not yet add inferred types for parameters within function body
scopes based on the annotations, but it should be easy to add as a next
step.
Part of #14161 and #13693.
## Test Plan
Added tests.
Follow-up to #14287 : when checking that `name` is the same as `as_name`
in `import name as as_name`, we do not need to first do an early return
if `'.'` is found in `name`.
This PR handles a panic that occurs when applying unsafe fixes if a user
inserts a required import (I002) that has a "useless alias" in it, like
`import numpy as numpy`, and also selects PLC0414 (useless-import-alias)
In this case, the fixes alternate between adding the required import
statement, then removing the alias, until the recursion limit is
reached. See linked issue for an example.
Closes#14283
---------
Co-authored-by: Charlie Marsh <charlie.r.marsh@gmail.com>
