## Summary
This PR adds support for named expressions when analyzing `__all__`
assignments, as per https://github.com/astral-sh/ruff/issues/7672. It
also loosens the enforcement around assignments like: `__all__ =
list(some_other_expression)`. We shouldn't flag these as invalid, even
though we can't analyze the members, since we _know_ they evaluate to a
`list`.
Closes https://github.com/astral-sh/ruff/issues/7672.
## Test Plan
`cargo test`
## Summary
This is a follow-up to #7469 that attempts to achieve similar gains, but
without introducing malachite. Instead, this PR removes the `BigInt`
type altogether, instead opting for a simple enum that allows us to
store small integers directly and only allocate for values greater than
`i64`:
```rust
/// A Python integer literal. Represents both small (fits in an `i64`) and large integers.
#[derive(Clone, PartialEq, Eq, Hash)]
pub struct Int(Number);
#[derive(Debug, Clone, PartialEq, Eq, Hash)]
pub enum Number {
/// A "small" number that can be represented as an `i64`.
Small(i64),
/// A "large" number that cannot be represented as an `i64`.
Big(Box<str>),
}
impl std::fmt::Display for Number {
fn fmt(&self, f: &mut std::fmt::Formatter<'_>) -> std::fmt::Result {
match self {
Number::Small(value) => write!(f, "{value}"),
Number::Big(value) => write!(f, "{value}"),
}
}
}
```
We typically don't care about numbers greater than `isize` -- our only
uses are comparisons against small constants (like `1`, `2`, `3`, etc.),
so there's no real loss of information, except in one or two rules where
we're now a little more conservative (with the worst-case being that we
don't flag, e.g., an `itertools.pairwise` that uses an extremely large
value for the slice start constant). For simplicity, a few diagnostics
now show a dedicated message when they see integers that are out of the
supported range (e.g., `outdated-version-block`).
An additional benefit here is that we get to remove a few dependencies,
especially `num-bigint`.
## Test Plan
`cargo test`
## Summary
This is only used for the `level` field in relative imports (e.g., `from
..foo import bar`). It seems unnecessary to use a wrapper here, so this
PR changes to a `u32` directly.
## Summary
If a function has no parameters (and no comments within the parameters'
`()`), we're supposed to wrap the return annotation _whenever_ it
breaks. However, our `empty_parameters` test didn't properly account for
the case in which the parameters include a newline (but no other
content), like:
```python
def get_dashboards_hierarchy(
) -> Dict[Type['BaseDashboard'], List[Type['BaseDashboard']]]:
"""Get hierarchy of dashboards classes.
Returns:
Dict of dashboards classes.
"""
dashboards_hierarchy = {}
```
This PR fixes that detection. Instead of lexing, it now checks if the
parameters itself is empty (or if it contains comments).
Closes https://github.com/astral-sh/ruff/issues/7457.
## Summary
The tokenizer was split into a forward and a backwards tokenizer. The
backwards tokenizer uses the same names as the forwards ones (e.g.
`next_token`). The backwards tokenizer gets the comment ranges that we
already built to skip comments.
---------
Co-authored-by: Micha Reiser <micha@reiser.io>
`ComparableExpr` includes the `ExprContext` field on an expression, so,
e.g., the two tuples in `(a, b) = (a, b)` won't be considered equal.
Similarly, the tuples in `[(a, b) for (a, b) in c]` _also_ wouldn't be
considered equal. I find this behavior surprising, since
`ComparableExpr` is intended to allow you to compare two ASTs, but
`ExprContext` is really encoding information about the broader context
for the expression.
## Motivation
The `ast::Arguments` for call argument are split into positional
arguments (args) and keywords arguments (keywords). We currently assume
that call consists of first args and then keywords, which is generally
the case, but not always:
```python
f(*args, a=2, *args2, **kwargs)
class A(*args, a=2, *args2, **kwargs):
pass
```
The consequence is accidentally reordering arguments
(https://github.com/astral-sh/ruff/pull/7268).
## Summary
`Arguments::args_and_keywords` returns an iterator of an `ArgOrKeyword`
enum that yields args and keywords in the correct order. I've fixed the
obvious `args` and `keywords` usages, but there might be some cases with
wrong assumptions remaining.
## Test Plan
The generator got new test cases, otherwise the stacked PR
(https://github.com/astral-sh/ruff/pull/7268) which uncovered this.
## Summary
This PR attempts to address a problem in the parser related to the
range's of `WithItem` nodes in certain contexts -- specifically,
`WithItem` nodes in parentheses that do not have an `as` token after
them.
For example,
[here](https://play.ruff.rs/71be2d0b-2a04-4c7e-9082-e72bff152679):
```python
with (a, b):
pass
```
The range of the `WithItem` `a` is set to the range of `(a, b)`, as is
the range of the `WithItem` `b`. In other words, when we have this kind
of sequence, we use the range of the entire parenthesized context,
rather than the ranges of the items themselves.
Note that this also applies to cases
[like](https://play.ruff.rs/c551e8e9-c3db-4b74-8cc6-7c4e3bf3713a):
```python
with (a, b, c as d):
pass
```
You can see the issue in the parser here:
```rust
#[inline]
WithItemsNoAs: Vec<ast::WithItem> = {
<location:@L> <all:OneOrMore<Test<"all">>> <end_location:@R> => {
all.into_iter().map(|context_expr| ast::WithItem { context_expr, optional_vars: None, range: (location..end_location).into() }).collect()
},
}
```
Fixing this issue is... very tricky. The naive approach is to use the
range of the `context_expr` as the range for the `WithItem`, but that
range will be incorrect when the `context_expr` is itself parenthesized.
For example, _that_ solution would fail here, since the range of the
first `WithItem` would be that of `a`, rather than `(a)`:
```python
with ((a), b):
pass
```
The `with` parsing in general is highly precarious due to ambiguities in
the grammar. Changing it in _any_ way seems to lead to an ambiguous
grammar that LALRPOP fails to translate. Consensus seems to be that we
don't really understand _why_ the current grammar works (i.e., _how_ it
avoids these ambiguities as-is).
The solution implemented here is to avoid changing the grammar itself,
and instead change the shape of the nodes returned by various rules in
the grammar. Specifically, everywhere that we return `Expr`, we instead
return `ParenthesizedExpr`, which includes a parenthesized range and the
underlying `Expr` itself. (If an `Expr` isn't parenthesized, the ranges
will be equivalent.) In `WithItemsNoAs`, we can then use the
parenthesized range as the range for the `WithItem`.
## Summary
This PR adds a higher-level enum (`SourceType`) around `PySourceType` to
allow us to use the same detection path to handle TOML files. Right now,
we have ad hoc `is_pyproject_toml` checks littered around, and some
codepaths are omitting that logic altogether (like `add_noqa`). Instead,
we should always be required to check the source type and handle TOML
files as appropriate.
This PR will also help with our pre-commit capabilities. If we add
`toml` to pre-commit (to support `pyproject.toml`), pre-commit will
start to pass _other_ files to Ruff (along with `poetry.lock` and
`Pipfile` -- see
[identify](b59996304f/identify/extensions.py (L355))).
By detecting those files and handling those cases, we avoid attempting
to parse them as Python files, which would lead to pre-commit errors.
(We tried to add `toml` to pre-commit here
(https://github.com/astral-sh/ruff-pre-commit/pull/44), but had to
revert here (https://github.com/astral-sh/ruff-pre-commit/pull/45) as it
led to the pre-commit hook attempting to parse `poetry.lock` files as
Python files.)
## Summary
The motivation here is that this enables us to implement `Ranged` in
crates that don't depend on `ruff_python_ast`.
Largely a mechanical refactor with a lot of regex, Clippy help, and
manual fixups.
## Test Plan
`cargo test`
## Summary
This PR introduces two new AST nodes to improve the representation of
`PatternMatchClass`. As a reminder, `PatternMatchClass` looks like this:
```python
case Point2D(0, 0, x=1, y=2):
...
```
Historically, this was represented as a vector of patterns (for the `0,
0` portion) and parallel vectors of keyword names (for `x` and `y`) and
values (for `1` and `2`). This introduces a bunch of challenges for the
formatter, but importantly, it's also really different from how we
represent similar nodes, like arguments (`func(0, 0, x=1, y=2)`) or
parameters (`def func(x, y)`).
So, firstly, we now use a single node (`PatternArguments`) for the
entire parenthesized region, making it much more consistent with our
other nodes. So, above, `PatternArguments` would be `(0, 0, x=1, y=2)`.
Secondly, we now have a `PatternKeyword` node for `x=1` and `y=2`. This
is much more similar to the how `Keyword` is represented within
`Arguments` for call expressions.
Closes https://github.com/astral-sh/ruff/issues/6866.
Closes https://github.com/astral-sh/ruff/issues/6880.
## Summary
Another drive-by change to remove unnecessary custom lexing. We just
need to know the parenthesized range, so we can use...
`parenthesized_range`. I've also updated `parenthesized_range` to
support nested parentheses.
## Test Plan
`cargo test`
## Summary
If a lambda doesn't contain any parameters, or any parameter _tokens_
(like `*`), we can use `None` for the parameters. This feels like a
better representation to me, since, e.g., what should the `TextRange` be
for a non-existent set of parameters? It also allows us to remove
several sites where we check if the `Parameters` is empty by seeing if
it contains any arguments, so semantically, we're already trying to
detect and model around this elsewhere.
Changing this also fixes a number of issues with dangling comments in
parameter-less lambdas, since those comments are now automatically
marked as dangling on the lambda. (As-is, we were also doing something
not-great whereby the lambda was responsible for formatting dangling
comments on the parameters, which has been removed.)
Closes https://github.com/astral-sh/ruff/issues/6646.
Closes https://github.com/astral-sh/ruff/issues/6647.
## Test Plan
`cargo test`
## Summary
This PR exposes our `is_expression_parenthesized` logic such that we can
use it to expand expressions when autofixing to include their
parenthesized ranges.
This solution has a few drawbacks: (1) we need to compute parenthesized
ranges in more places, which also relies on backwards lexing; and (2) we
need to make use of this in any relevant fixes.
However, I still think it's worth pursuing. On (1), the implementation
is very contained, so IMO we can easily swap this out for a more
performant solution in the future if needed. On (2), this improves
correctness and fixes some bad syntax errors detected by fuzzing, which
means it has value even if it's not as robust as an _actual_
`ParenthesizedExpression` node in the AST itself.
Closes https://github.com/astral-sh/ruff/issues/4925.
## Test Plan
`cargo test` with new cases that previously failed the fuzzer.
## Summary
I noticed some inconsistencies around uses of `.range.start()`, structs
that have a `TextRange` field but don't implement `Ranged`, etc.
## Test Plan
`cargo test`
## Summary
Instead, we set an `is_star` flag on `Stmt::Try`. This is similar to the
pattern we've migrated towards for `Stmt::For` (removing
`Stmt::AsyncFor`) and friends. While these are significant differences
for an interpreter, we tend to handle these cases identically or nearly
identically.
## Test Plan
`cargo test`
## Summary
In https://github.com/astral-sh/ruff/pull/6512, we added a flag to the
AST to mark implicitly-concatenated string expressions. This PR makes
use of that flag to remove the `is_implicit_concatenation` method.
## Test Plan
`cargo test`
## Summary
Per the discussion in
https://github.com/astral-sh/ruff/discussions/6183, this PR adds an
`implicit_concatenated` flag to the string and bytes constant variants.
It's not actually _used_ anywhere as of this PR, but it is covered by
the tests.
Specifically, we now use a struct for the string and bytes cases, along
with the `Expr::FString` node. That struct holds the value, plus the
flag:
```rust
#[derive(Clone, Debug, PartialEq, is_macro::Is)]
pub enum Constant {
Str(StringConstant),
Bytes(BytesConstant),
...
}
#[derive(Clone, Debug, PartialEq, Eq)]
pub struct StringConstant {
/// The string value as resolved by the parser (i.e., without quotes, or escape sequences, or
/// implicit concatenations).
pub value: String,
/// Whether the string contains multiple string tokens that were implicitly concatenated.
pub implicit_concatenated: bool,
}
impl Deref for StringConstant {
type Target = str;
fn deref(&self) -> &Self::Target {
self.value.as_str()
}
}
#[derive(Clone, Debug, PartialEq, Eq)]
pub struct BytesConstant {
/// The bytes value as resolved by the parser (i.e., without quotes, or escape sequences, or
/// implicit concatenations).
pub value: Vec<u8>,
/// Whether the string contains multiple string tokens that were implicitly concatenated.
pub implicit_concatenated: bool,
}
impl Deref for BytesConstant {
type Target = [u8];
fn deref(&self) -> &Self::Target {
self.value.as_slice()
}
}
```
## Test Plan
`cargo test`
<!--
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
This PR adds the `AnyNodeRef.visit_preorder` method. I'll need this method to mark all comments of a suppressed node's children as formatted (in debug builds).
I'm not super happy with this because it now requires a double-dispatch where the `walk_*` methods call into `node.visit_preorder` and the `visit_preorder` then calls back into the visitor. Meaning,
the new implementation now probably results in way more function calls. The other downside is that `AnyNodeRef` now contains code that is difficult to auto-generate. This could be mitigated by extracting the `visit_preorder` method into its own `VisitPreorder` trait.
Anyway, this approach solves the need and avoids duplicating the visiting code once more.
<!-- What's the purpose of the change? What does it do, and why? -->
## Test Plan
`cargo test`
<!-- How was it tested? -->
## Summary
This PR renames the `MagicCommand` token to `IpyEscapeCommand` token and
`MagicKind` to `IpyEscapeKind` type to better reflect the purpose of the
token and type. Similarly, it renames the AST nodes from `LineMagic` to
`IpyEscapeCommand` prefixed with `Stmt`/`Expr` wherever necessary.
It also makes renames from using `jupyter_magic` to
`ipython_escape_commands` in various function names.
The mode value is still `Mode::Jupyter` because the escape commands are
part of the IPython syntax but the lexing/parsing is done for a Jupyter
notebook.
### Motivation behind the rename:
* IPython codebase defines it as "EscapeCommand" / "Escape Sequences":
* Escape Sequences:
292e3a2345/IPython/core/inputtransformer2.py (L329-L333)
* Escape command:
292e3a2345/IPython/core/inputtransformer2.py (L410-L411)
* The word "magic" is used mainly for the actual magic commands i.e.,
the ones starting with `%`/`%%`
(https://ipython.readthedocs.io/en/stable/interactive/reference.html#magic-command-system).
So, this avoids any confusion between the Magic token (`%`, `%%`) and
the escape command itself.
## Test Plan
* `cargo test` to make sure all renames are done correctly.
* `grep` for `jupyter_escape`/`magic` to make sure all renames are done
correctly.
## Summary
This PR leverages the unified function definition node to add precise
AST node types to `MemberKind`, which is used to power our docstring
definition tracking (e.g., classes and functions, whether they're
methods or functions or nested functions and so on, whether they have a
docstring, etc.). It was painful to do this in the past because the
function variants needed to support a union anyway, but storing precise
nodes removes like a dozen panics.
No behavior changes -- purely a refactor.
## Test Plan
`cargo test`
## Summary
Per the suggestion in
https://github.com/astral-sh/ruff/discussions/6183, this PR removes
`AsyncWith`, `AsyncFor`, and `AsyncFunctionDef`, replacing them with an
`is_async` field on the non-async variants of those structs. Unlike an
interpreter, we _generally_ have identical handling for these nodes, so
separating them into distinct variants adds complexity from which we
don't really benefit. This can be seen below, where we get to remove a
_ton_ of code related to adding generic `Any*` wrappers, and a ton of
duplicate branches for these cases.
## Test Plan
`cargo test` is unchanged, apart from parser snapshots.
## Summary
See discussion in
https://github.com/astral-sh/ruff/pull/6351#discussion_r1284996979. We
can remove `RefEquality` entirely and instead use a text offset for
statement keys, since no two statements can start at the same text
offset.
## Test Plan
`cargo test`
## Summary
This PR adds support for help end escape command in the lexer.
### What are "help end escape commands"?
First, the escape commands are special IPython syntax which enhances the
functionality for the IPython REPL. There are 9 types of escape kinds
which are recognized by the tokens which are present at the start of the
command (`?`, `??`, `!`, `!!`, etc.).
Here, the help command is using either the `?` or `??` token at the
start (`?str.replace` for example). Those 2 tokens are also supported
when they're at the end of the command (`str.replace?`), but the other
tokens aren't supported in that position.
There are mainly two types of help end escape commands:
1. Ending with either `?` or `??`, but it also starts with one of the
escape tokens (`%matplotlib?`)
2. On the other hand, there's a stricter version for (1) which doesn't
start with any escape tokens (`str.replace?`)
This PR adds support for (1) while (2) will be supported in the parser.
### Priority
Now, if the command starts and ends with an escape token, how do we
decide the kind of this command? This is where priority comes into
picture. This is simple as there's only one priority where `?`/`??` at
the end takes priority over any other escape token and all of the other
tokens are at the same priority. Remember that only `?`/`??` at the end
is considered valid.
This is mainly useful in the case where someone would want to invoke the
help command on the magic command itself. For example, in `%matplotlib?`
the help command takes priority which means that we want help for the
`matplotlib` magic function instead of calling the magic function
itself.
### Specification
Here's where things get a bit tricky. What if there are question mark
tokens at both ends. How do we decide if it's `Help` (`?`) kind or
`Help2` (`??`) kind?
| | Magic | Value | Kind |
| --- | --- | --- | --- |
| 1 | `?foo?` | `foo` | `Help` |
| 2 | `??foo?` | `foo` | `Help` |
| 3 | `?foo??` | `foo` | `Help2` |
| 4 | `??foo??` | `foo` | `Help2` |
| 5 | `???foo??` | `foo` | `Help2` |
| 6 | `??foo???` | `foo???` | `Help2` |
| 7 | `???foo???` | `?foo???` | `Help2` |
Looking at the above table:
- The question mark tokens on the right takes priority over the ones on
the left but only if the number of question mark on the right is 1 or 2.
- If there are more than 2 question mark tokens on the right side, then
the left side is used to determine the same.
- If the right side is used to determine the kind, then all of the
question marks and whitespaces on the left side are ignored in the
`value`, but if it’s the other way around, then all of the extra
question marks are part of the `value`.
### References
- IPython implementation using the regex:
292e3a2345/IPython/core/inputtransformer2.py (L454-L462)
- Priorities:
292e3a2345/IPython/core/inputtransformer2.py (L466-L469)
## Test Plan
Add a bunch of test cases for the lexer and verify that it matches the
behavior of
IPython transformer.
resolves: #6357
## Summary
Historically, we've stored "qualified names" on our
`BindingKind::Import`, `BindingKind::SubmoduleImport`, and
`BindingKind::ImportFrom` structs. In Ruff, a "qualified name" is a
dot-separated path to a symbol. For example, given `import foo.bar`, the
"qualified name" would be `"foo.bar"`; and given `from foo.bar import
baz`, the "qualified name" would be `foo.bar.baz`.
This PR modifies the `BindingKind` structs to instead store _call paths_
rather than qualified names. So in the examples above, we'd store
`["foo", "bar"]` and `["foo", "bar", "baz"]`. It turns out that this
more efficient given our data access patterns. Namely, we frequently
need to convert the qualified name to a call path (whenever we call
`resolve_call_path`), and it turns out that we do this operation enough
that those conversations show up on benchmarks.
There are a few other advantages to using call paths, rather than
qualified names:
1. The size of `BindingKind` is reduced from 32 to 24 bytes, since we no
longer need to store a `String` (only a boxed slice).
2. All three import types are more consistent, since they now all store
a boxed slice, rather than some storing an `&str` and some storing a
`String` (for `BindingKind::ImportFrom`, we needed to allocate a
`String` to create the qualified name, but the call path is a slice of
static elements that don't require that allocation).
3. A lot of code gets simpler, in part because we now do call path
resolution "earlier". Most notably, for relative imports (`from .foo
import bar`), we store the _resolved_ call path rather than the relative
call path, so the semantic model doesn't have to deal with that
resolution. (See that `resolve_call_path` is simpler, fewer branches,
etc.)
In my testing, this change improves the all-rules benchmark by another
4-5% on top of the improvements mentioned in #6047.
## Summary
Update `F841` autofix to not remove line magic expr
## Test Plan
Added test case for assignment statement with and without type
annotation
fixes: #6116
**Summary** This adds the information whether we're in a .py python
source file or in a .pyi stub file to enable people working on #5822 and
related issues.
I'm not completely happy with `Default` for something that depends on
the input.
**Test Plan** None, this is currently unused, i'm leaving this to first
implementation of stub file specific formatting.
---------
Co-authored-by: Micha Reiser <micha@reiser.io>
Part of #5062
Closes https://github.com/astral-sh/ruff/issues/5931
Implements formatting of a sequence of type parameters in a dedicated
struct for reuse by classes, functions, and type aliases (preparing for
#5929). Adds formatting of type parameters in class and function
definitions — previously, they were just elided.
## Summary
Similar to #6279, moving some helpers onto the struct in the name of
reducing the number of random undiscoverable utilities we have in
`helpers.rs`.
Most of the churn is migrating rules to take `ast::ExprCall` instead of
the spread call arguments.
## Test Plan
`cargo test`
## Summary
This PR removes a now-unnecessary abstraction from `helper.rs`
(`CallArguments`), in favor of adding methods to `Arguments` directly,
which helps with discoverability.
## Summary
This PR boxes the `TypeParams` and `Arguments` fields on the class
definition node. These fields are optional and often emitted, and given
that class definition is our largest enum variant, we pay the cost of
including them for every statement in the AST. Boxing these types
reduces the statement size by 40 bytes, which seems like a good tradeoff
given how infrequently these are accessed.
## Test Plan
Need to benchmark, but no behavior changes.
## Summary
This PR leverages the `Arguments` AST node introduced in #6259 in the
formatter, which ensures that we correctly handle trailing comments in
calls, like:
```python
f(
1,
# comment
)
pass
```
(Previously, this was treated as a leading comment on `pass`.)
This also allows us to unify the argument handling across calls and
class definitions.
## Test Plan
A bunch of new fixture tests, plus improved Black compatibility.
