(This is not possible to actually use until
https://github.com/astral-sh/ruff/pull/8854 is merged.)
ruff_python_formatter: add reStructuredText docstring formatting support
This commit makes use of the refactoring done in prior commits to slot
in reStructuredText support. Essentially, we add a new type of code
example and look for *both* literal blocks and code block directives.
Literal blocks are treated as Python by default because it seems to be a
[common
practice](https://github.com/adamchainz/blacken-docs/issues/195).
That is, literal blocks like this:
```
def example():
"""
Here's an example::
foo( 1 )
All done.
"""
pass
```
Will get reformatted. And code blocks (via reStructuredText directives)
will also get reformatted:
```
def example():
"""
Here's an example:
.. code-block:: python
foo( 1 )
All done.
"""
pass
```
When looking for a code block, it is possible for it to become invalid.
In which case, we back out of looking for a code example and print the
lines out as they are. As with doctest formatting, if reformatting the
code would result in invalid Python or if the code collected from the
block is invalid, then formatting is also skipped.
A number of tests have been added to check both the formatting and
resetting behavior. Mixed indentation is also tested a fair bit, since
one of my initial attempts at dealing with mixed indentation ended up
not working.
I recommend working through this PR commit-by-commit. There is in
particular a somewhat gnarly refactoring before reST support is added.
Closes#8859
## Summary
We should avoid inlining the ellipsis in:
```python
def h():
...
# bye
```
Just as we omit the ellipsis in:
```python
def h():
# bye
...
```
Closes https://github.com/astral-sh/ruff/issues/8905.
In the source of working on #8859, I made a number of smallish refactors
to how code snippet formatting works. Most or all of these were
motivated by writing in support for reStructuredText blocks. They have
some fundamentally different requirements than doctests, and there are a
lot more ways for reStructuredText blocks to become invalid.
(Commit-by-commit review is recommended as the commit messages provide
further context on each change. I split this off from ongoing work to
make review more manageable.)
This PR removes several uses of `unsafe`. I generally limited myself to
low hanging fruit that I could see. There are still a few remaining uses
of `unsafe` that looked a bit more difficult to remove (if possible at
all). But this gets rid of a good chunk of them.
I put each `unsafe` removal into its own commit with a justification for
why I did it. So I would encourage reviewing this PR commit-by-commit.
That way, we can legislate them independently. It's no problem to drop a
commit if we feel the `unsafe` should stay in that case.
This turns `string` into a parent module with a `docstring` sub-module.
I arranged things this way because there are parts of the `string`
module that the `docstring` module wants to know about (such as a
`NormalizedString`). The alternative I think would be to make
`docstring` a sibling module and expose more of `string`'s internals.
I think I overall like this change because it gives docstring handling a
bit more room to breath. It has grown quite a bit with the addition of
code snippet formatting.
[This was suggested by
@charliermarsh.](https://github.com/astral-sh/ruff/pull/8811#discussion_r1401169531)
## Summary
This PR adds opt-in support for formatting doctests in docstrings. This
reflects initial support and it is intended to add support for Markdown
and reStructuredText Python code blocks in the future. But I believe
this PR lays the groundwork, and future additions for Markdown and reST
should be less costly to add.
It's strongly recommended to review this PR commit-by-commit. The last
few commits in particular implement the bulk of the work here and
represent the denser portions.
Some things worth mentioning:
* The formatter is itself not perfect, and it is possible for it to
produce invalid Python code. Because of this, reformatted code snippets
are checked for Python validity. If they aren't valid, then we
(unfortunately silently) bail on formatting that code snippet.
* There are a couple places where it would be nice to at least warn the
user that doctest formatting failed, but it wasn't clear to me what the
best way to do that is.
* I haven't yet run this in anger on a real world code base. I think
that should happen before merging.
Closes#7146
## Test Plan
* [x] Pass the local test suite.
* [x] Scrutinize ecosystem changes.
* [x] Run this formatter on extant code and scrutinize the results.
(e.g., CPython, numpy.)
A follow-up to auto-generate the `FormatNodeRule` implementation for the
string part nodes. This is just a dummy implementation that is
unreachable because it's handled by the parent nodes.
## Summary
This PR updates the string nodes (`ExprStringLiteral`,
`ExprBytesLiteral`, and `ExprFString`) to account for implicit string
concatenation.
### Motivation
In Python, implicit string concatenation are joined while parsing
because the interpreter doesn't require the information for each part.
While that's feasible for an interpreter, it falls short for a static
analysis tool where having such information is more useful. Currently,
various parts of the code uses the lexer to get the individual string
parts.
One of the main challenge this solves is that of string formatting.
Currently, the formatter relies on the lexer to get the individual
string parts, and formats them including the comments accordingly. But,
with PEP 701, f-string can also contain comments. Without this change,
it becomes very difficult to add support for f-string formatting.
### Implementation
The initial proposal was made in this discussion:
https://github.com/astral-sh/ruff/discussions/6183#discussioncomment-6591993.
There were various AST designs which were explored for this task which
are available in the linked internal document[^1].
The selected variant was the one where the nodes were kept as it is
except that the `implicit_concatenated` field was removed and instead a
new struct was added to the `Expr*` struct. This would be a private
struct would contain the actual implementation of how the AST is
designed for both single and implicitly concatenated strings.
This implementation is achieved through an enum with two variants:
`Single` and `Concatenated` to avoid allocating a vector even for single
strings. There are various public methods available on the value struct
to query certain information regarding the node.
The nodes are structured in the following way:
```
ExprStringLiteral - "foo" "bar"
|- StringLiteral - "foo"
|- StringLiteral - "bar"
ExprBytesLiteral - b"foo" b"bar"
|- BytesLiteral - b"foo"
|- BytesLiteral - b"bar"
ExprFString - "foo" f"bar {x}"
|- FStringPart::Literal - "foo"
|- FStringPart::FString - f"bar {x}"
|- StringLiteral - "bar "
|- FormattedValue - "x"
```
[^1]: Internal document:
https://www.notion.so/astral-sh/Implicit-String-Concatenation-e036345dc48943f89e416c087bf6f6d9?pvs=4
#### Visitor
The way the nodes are structured is that the entire string, including
all the parts that are implicitly concatenation, is a single node
containing individual nodes for the parts. The previous section has a
representation of that tree for all the string nodes. This means that
new visitor methods are added to visit the individual parts of string,
bytes, and f-strings for `Visitor`, `PreorderVisitor`, and
`Transformer`.
## Test Plan
- `cargo insta test --workspace --all-features --unreferenced reject`
- Verify that the ecosystem results are unchanged
Fix an instability where await was followed by a breaking fluent style
expression:
```python
test_data = await (
Stream.from_async(async_data)
.flat_map_async()
.map()
.filter_async(is_valid_data)
.to_list()
)
```
Note that this technically a minor style change (see ecosystem check)
Update to [Rust
1.74](https://blog.rust-lang.org/2023/11/16/Rust-1.74.0.html) and use
the new clippy lints table.
The update itself introduced a new clippy lint about superfluous hashes
in raw strings, which got removed.
I moved our lint config from `rustflags` to the newly stabilized
[workspace.lints](https://doc.rust-lang.org/stable/cargo/reference/workspaces.html#the-lints-table).
One consequence is that we have to `unsafe_code = "warn"` instead of
"forbid" because the latter now actually bans unsafe code:
```
error[E0453]: allow(unsafe_code) incompatible with previous forbid
--> crates/ruff_source_file/src/newlines.rs:62:17
|
62 | #[allow(unsafe_code)]
| ^^^^^^^^^^^ overruled by previous forbid
|
= note: `forbid` lint level was set on command line
```
---------
Co-authored-by: Charlie Marsh <charlie.r.marsh@gmail.com>
Testing the compatibility with the future stable black style, i realized
the `ruff_python_formatter` dev main was lacking the
`--skip-magic-trailing-comma` option. This does not affect `ruff
format`.
Usage:
```shell
cargo run --bin ruff_python_formatter -p ruff_python_formatter -- --skip-magic-trailing-comma --emit stdout scratch.py
```
## Summary
This PR implements validation in the formatter tests to ensure that we
don't modify the AST during formatting. Black has similar logic.
In implementing this, I learned that Black actually _does_ modify the
AST, and their test infrastructure normalizes the AST to wipe away those
differences. Specifically, Black changes the indentation of docstrings,
which _does_ modify the AST; and it also inserts parentheses in `del`
statements, which changes the AST too.
Ruff also does both these things, so we _also_ implement the same
normalization using a new visitor that allows for modifying the AST.
Closes https://github.com/astral-sh/ruff/issues/8184.
## Test Plan
`cargo test`
It seems as though using `include!(...)` to avoid the source code copy
breaks rust-analzer. Namely, it treats the included file as unlinked,
and so any part of analysis (e.g., goto-definition) that needs that file
to reason about the code ends up failing.
<!--
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? -->
## Test Plan
<!-- How was it tested? -->
## Summary
This PR updates the formatter to preserve trailing semicolon for Jupyter
Notebooks.
The motivation behind the change is that semicolons in notebooks are
typically used to hide the output, for example when plotting. This is
highlighted in the linked issue.
The conditions required as to when the trailing semicolon should be
preserved are:
1. It should be a top-level statement which is last in the module.
2. For statement, it can be either assignment, annotated assignment, or
augmented assignment. Here, the target should only be a single
identifier i.e., multiple assignments or tuple unpacking isn't
considered.
3. For expression, it can be any.
## Test Plan
Add a new integration test in `ruff_cli`. The test notebook basically
acts as a document as to which trailing semicolons are to be preserved.
fixes: #8254
This ensures the python label is used for all python code blocks for
consistency.
## Test Plan
Visual inspection of all changes via git client ensuring no other
changes were made in error.
## Summary
This commit adds some additional error checking to the parser such that
assignments that are invalid syntax are rejected. This covers the
obvious cases like `5 = 3` and some not so obvious cases like `x + y =
42`.
This does add an additional recursive call to the parser for the cases
handling assignments. I had initially been concerned about doing this,
but `set_context` is already doing recursion during assignments, so I
didn't feel as though this was changing any fundamental performance
characteristics of the parser. (Also, in practice, I would expect any
such recursion here to be quite shallow since the recursion is done on
the target of an assignment. Such things are rarely nested much in
practice.)
Fixes#6895
## Test Plan
I've added unit tests covering every case that is detected as invalid on
an `Expr`.
## Summary
This PR fixes a bug in our formatter where a multiline lambda expression
statement was formatted over multiple lines without adding parentheses.
The PR "fixes" the problem by not splitting the lambda parameters if it
is not parenthesized
## Test Plan
Added test
**Summary** Previously, own line comment following after a docstring
followed by newline(s) before the first content statement were treated
as trailing on the docstring and we didn't insert a newline after the
docstring as black would.
Before:
```python
class ModuleBrowser:
"""Browse module classes and functions in IDLE."""
# This class is also the base class for pathbrowser.PathBrowser.
def __init__(self, master, path, *, _htest=False, _utest=False):
pass
```
After:
```python
class ModuleBrowser:
"""Browse module classes and functions in IDLE."""
# This class is also the base class for pathbrowser.PathBrowser.
def __init__(self, master, path, *, _htest=False, _utest=False):
pass
```
I'm not entirely happy about hijacking
`handle_own_line_comment_between_statements`, but i don't know a better
spot to put it.
Fixes#7948
**Test Plan** Fixtures
We previously incorrectly treated byte strings in docstring position as
docstrings because black does so
(https://github.com/astral-sh/ruff/pull/8283#discussion_r1375682931,
https://github.com/psf/black/issues/4002), even CPython doesn't
recognize them:
```console
$ python3.12
Python 3.12.0 (main, Oct 6 2023, 17:57:44) [GCC 11.4.0] on linux
Type "help", "copyright", "credits" or "license" for more information.
>>> def f():
... b""" a"""
...
>>> print(str(f.__doc__))
None
```
<!--
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 inlines the formatting logic for `ExprNumberLiteral` and removes
the need of having dedicated `Format*` struct for each number type.
## Test Plan
`cargo test`
## Summary
This PR splits the `Constant` enum as individual literal nodes. It
introduces the following new nodes for each variant:
* `ExprStringLiteral`
* `ExprBytesLiteral`
* `ExprNumberLiteral`
* `ExprBooleanLiteral`
* `ExprNoneLiteral`
* `ExprEllipsisLiteral`
The main motivation behind this refactor is to introduce the new AST
node for implicit string concatenation in the coming PR. The elements of
that node will be either a string literal, bytes literal or a f-string
which can be implemented using an enum. This means that a string or
bytes literal cannot be represented by `Constant::Str` /
`Constant::Bytes` which creates an inconsistency.
This PR avoids that inconsistency by splitting the constant nodes into
it's own literal nodes, literal being the more appropriate naming
convention from a static analysis tool perspective.
This also makes working with literals in the linter and formatter much
more ergonomic like, for example, if one would want to check if this is
a string literal, it can be done easily using
`Expr::is_string_literal_expr` or matching against `Expr::StringLiteral`
as oppose to matching against the `ExprConstant` and enum `Constant`. A
few AST helper methods can be simplified as well which will be done in a
follow-up PR.
This introduces a new `Expr::is_literal_expr` method which is the same
as `Expr::is_constant_expr`. There are also intermediary changes related
to implicit string concatenation which are quiet less. This is done so
as to avoid having a huge PR which this already is.
## Test Plan
1. Verify and update all of the existing snapshots (parser, visitor)
2. Verify that the ecosystem check output remains **unchanged** for both
the linter and formatter
### Formatter ecosystem check
#### `main`
| project | similarity index | total files | changed files |
|----------------|------------------:|------------------:|------------------:|
| cpython | 0.75803 | 1799 | 1647 |
| django | 0.99983 | 2772 | 34 |
| home-assistant | 0.99953 | 10596 | 186 |
| poetry | 0.99891 | 317 | 17 |
| transformers | 0.99966 | 2657 | 330 |
| twine | 1.00000 | 33 | 0 |
| typeshed | 0.99978 | 3669 | 20 |
| warehouse | 0.99977 | 654 | 13 |
| zulip | 0.99970 | 1459 | 22 |
#### `dhruv/constant-to-literal`
| project | similarity index | total files | changed files |
|----------------|------------------:|------------------:|------------------:|
| cpython | 0.75803 | 1799 | 1647 |
| django | 0.99983 | 2772 | 34 |
| home-assistant | 0.99953 | 10596 | 186 |
| poetry | 0.99891 | 317 | 17 |
| transformers | 0.99966 | 2657 | 330 |
| twine | 1.00000 | 33 | 0 |
| typeshed | 0.99978 | 3669 | 20 |
| warehouse | 0.99977 | 654 | 13 |
| zulip | 0.99970 | 1459 | 22 |
## Summary
This PR adds a new `Singleton` enum for the `PatternMatchSingleton`
node.
Earlier the node was using the `Constant` enum but the value for this
pattern can only be either `None`, `True` or `False`. With the coming PR
to remove the `Constant`, this node required a new type to fill in.
This also has the benefit of narrowing the type down to only the
possible values for the node as evident by the removal of `unreachable`.
## Test Plan
Update the AST snapshots and run `cargo test`.
Change
```python
"""Test docstring"""
a = 1
```
to
```python
"""Test docstring"""
a = 1
```
in preview style, but don't touch the docstring otherwise.
Do we want to ask black to also format the content of module level
docstrings? Seems inconsistent to me that we change function and class
docstring indentation/contents but not module docstrings.
Fixes https://github.com/astral-sh/ruff/issues/7995
**Summary** Prepare for the black preview style becoming the black
stable style at the end of the year.
This adds a new test file to compare stable and preview on some relevant
preview options in black, and makes `format_dev` understand the black
preview flag. I've added poetry as a project that uses preview.
I've implemented one specific deviation (collapsing of stub
implementation in non-stub files) which showed up in poetry for testing.
This also improves poetry compatibility from 0.99891 to 0.99919.
Fixes#7440
New compatibility stats:
| project | similarity index | total files | changed files |
|----------------|------------------:|------------------:|------------------:|
| cpython | 0.75803 | 1799 | 1647 |
| django | 0.99983 | 2772 | 35 |
| home-assistant | 0.99953 | 10596 | 189 |
| poetry | 0.99919 | 317 | 12 |
| transformers | 0.99963 | 2657 | 332 |
| twine | 1.00000 | 33 | 0 |
| typeshed | 0.99978 | 3669 | 20 |
| warehouse | 0.99969 | 654 | 15 |
| zulip | 0.99970 | 1459 | 22 |
## Summary
Given:
```python
# comment
class A:
def foo(self):
pass
```
We need to insert an additional newline between `# comment` and `class
A`. We were missing this handling for the case in which `# comment` is a
leading comment on `class A`, as opposed to a trailing comment of some
preceding statement.
In practice, I think this only applies to the specific case in which a
class or function is the first statement in a module, and there's a
single empty line between a leading comment and that class or function.
If there are no empty lines, then the comment "sticks" to the
definition; if there are two or more, then `leading_comments` will
truncate appropriately. If the class or function is nested, then we only
need one empty line anyway.
Closes https://github.com/astral-sh/ruff/issues/8215.
## Test Plan
No change in similarity.
Before:
| project | similarity index | total files | changed files |
|----------------|------------------:|------------------:|------------------:|
| cpython | 0.75803 | 1799 | 1647 |
| django | 0.99983 | 2772 | 34 |
| home-assistant | 0.99953 | 10596 | 186 |
| poetry | 0.99891 | 317 | 17 |
| transformers | 0.99966 | 2657 | 330 |
| twine | 1.00000 | 33 | 0 |
| typeshed | 0.99978 | 3669 | 20 |
| warehouse | 0.99977 | 654 | 13 |
| zulip | 0.99970 | 1459 | 22 |
After:
| project | similarity index | total files | changed files |
|----------------|------------------:|------------------:|------------------:|
| cpython | 0.75803 | 1799 | 1648 |
| django | 0.99983 | 2772 | 34 |
| home-assistant | 0.99953 | 10596 | 186 |
| poetry | 0.99891 | 317 | 17 |
| transformers | 0.99966 | 2657 | 330 |
| twine | 1.00000 | 33 | 0 |
| typeshed | 0.99978 | 3669 | 20 |
| warehouse | 0.99977 | 654 | 13 |
| zulip | 0.99970 | 1459 | 22 |
## Summary
This PR removes the `todo!()` around `IpyEscapeCommand` in the
formatter.
The `NeedsParentheses` trait needs to be implemented which always return
`Never`. The reason being that if an escape command is parenthesized,
then that's not parsed as an escape command. IOW, the parentheses
shouldn't be present around an escape command.
In the similar way, the `CanSkipOptionalParenthesesVisitor` will skip
this node.
## Test Plan
Updated the `unformatted.ipynb` fixture with new cells containing
IPython escape commands and the corresponding snapshot was verified.
Also, tested it out in a few open source repositories containing
notebooks (`openai/openai-cookbook`, `huggingface/notebooks`).
#### New cells in `unformatted.ipynb`
**Cell 2**
```markdown
A markdown cell
```
**Cell 3**
```python
def some_function(foo, bar):
pass
%matplotlib inline
```
**Cell 4**
```python
foo = %pwd
def some_function(foo,bar,):
foo = %pwd
print(foo
)
```
fixes: #8204
## Summary
This PR fixes the bug where if a Notebook contained IPython syntax, then
the format command would fail. This was because the correct mode was not
being used while parsing through the formatter code path.
## Test Plan
This PR isn't the only requirement for Notebook formatting to start
working with IPython escape commands. The following PR in the stack is
required as well.
## Summary
This was just a bug in the parser ranges, probably since it was
initially implemented. Given `match n % 3, n % 5: ...`, the "subject"
(i.e., the tuple of two binary operators) was using the entire range of
the `match` statement.
Closes https://github.com/astral-sh/ruff/issues/8091.
## Test Plan
`cargo test`
## Summary
This PR updates our documentation for the upcoming formatter release.
Broadly, the documentation is now structured as follows:
- Overview
- Tutorial
- Installing Ruff
- The Ruff Linter
- Overview
- `ruff check`
- Rule selection
- Error suppression
- Exit codes
- The Ruff Formatter
- Overview
- `ruff format`
- Philosophy
- Configuration
- Format suppression
- Exit codes
- Black compatibility
- Known deviations
- Configuring Ruff
- pyproject.toml
- File discovery
- Configuration discovery
- CLI
- Shell autocompletion
- Preview
- Rules
- Settings
- Integrations
- `pre-commit`
- VS Code
- LSP
- PyCharm
- GitHub Actions
- FAQ
- Contributing
The major changes include:
- Removing the "Usage" section from the docs, and instead folding that
information into "Integrations" and the new Linter and Formatter
sections.
- Breaking up "Configuration" into "Configuring Ruff" (for generic
configuration), and new Linter- and Formatter-specific sections.
- Updating all example configurations to use `[tool.ruff.lint]` and
`[tool.ruff.format]`.
My suggestion is to pull and build the docs locally, and review by
reading them in the browser rather than trying to parse all the code
changes.
Closes https://github.com/astral-sh/ruff/issues/7235.
Closes https://github.com/astral-sh/ruff/issues/7647.
<!--
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
Fixes https://github.com/astral-sh/ruff/issues/7448
Fixes https://github.com/astral-sh/ruff/issues/7892
I've removed automatic dangling comment formatting, we're doing manual
dangling comment formatting everywhere anyway (the
assert-all-comments-formatted ensures this) and dangling comments would
break the formatting there.
## Test Plan
New test file.
---------
Co-authored-by: Micha Reiser <micha@reiser.io>
Split out of #8044: In preview style, ellipsis are also collapsed in
non-stub files. This should only affect function/class contexts since
for other statements stub are generally not used. I've updated our tests
to use `pass` instead to reflect this, which makes tracking the preview
style changes much easier.
## Summary
Given an expression like `[x for (x) in y]`, we weren't skipping over
parentheses when searching for the `in` between `(x)` and `y`.
Closes https://github.com/astral-sh/ruff/issues/8053.
**Summary** Insert a newline after nested function and class
definitions, unless there is a trailing own line comment.
We need to e.g. format
```python
if platform.system() == "Linux":
if sys.version > (3, 10):
def f():
print("old")
else:
def f():
print("new")
f()
```
as
```python
if platform.system() == "Linux":
if sys.version > (3, 10):
def f():
print("old")
else:
def f():
print("new")
f()
```
even though `f()` is directly preceded by an if statement, not a
function or class definition. See the comments and fixtures for trailing
own line comment handling.
**Test Plan** I checked that the new content of `newlines.py` matches
black's formatting.
---------
Co-authored-by: Charlie Marsh <charlie.r.marsh@gmail.com>
**Summary** Handle comment before the default values of function
parameters correctly by inserting a line break instead of space after
the equals sign where required.
```python
def f(
a = # parameter trailing comment; needs line break
1,
b =
# default leading comment; needs line break
2,
c = ( # the default leading can only be end-of-line with parentheses; no line break
3
),
d = (
# own line leading comment with parentheses; no line break
4
)
)
```
Fixes#7603
**Test Plan** Added the different cases and one more complex case as
fixtures.
**Summary** Remove spaces from import statements such as
```python
import tqdm . tqdm
from tqdm . auto import tqdm
```
See also #7760 for a better solution.
**Test Plan** New fixtures
**Summary** Quoting of f-strings can change if they are triple quoted
and only contain single quotes inside.
Fixes#6841
**Test Plan** New fixtures
---------
Co-authored-by: Dhruv Manilawala <dhruvmanila@gmail.com>
## Summary
This PR fixes the bug where the formatter would panic if a class/function with
decorators had a suppression comment.
The fix is to use to correct start location to find the `async`/`def`/`class`
keyword when decorators are present which is the end of the last
decorator.
## Test Plan
Add test cases for the fix and update the snapshots.
## Summary
It turns out that _some_ identifiers can contain newlines --
specifically, dot-delimited import identifiers, like:
```python
import foo\
.bar
```
At present, we print all identifiers verbatim, which causes us to retain
the `\` in the formatted output. This also leads to violating some debug
assertions (see the linked issue, though that's a symptom of this
formatting failure).
This PR adds detection for import identifiers that contain newlines, and
formats them via `text` (slow) rather than `source_code_slice` (fast) in
those cases.
Closes https://github.com/astral-sh/ruff/issues/7734.
## Test Plan
`cargo test`
## Summary
At present, `quote-style` is used universally. However, [PEP
8](https://peps.python.org/pep-0008/) and [PEP
257](https://peps.python.org/pep-0257/) suggest that while either single
or double quotes are acceptable in general (as long as they're
consistent), docstrings and triple-quoted strings should always use
double quotes. In our research, the vast majority of Ruff users that
enable the `flake8-quotes` rules only enable them for inline strings
(i.e., non-triple-quoted strings).
Additionally, many Black forks (like Blue and Pyink) use double quotes
for docstrings and triple-quoted strings.
Our decision for now is to always prefer double quotes for triple-quoted
strings (which should include docstrings). Based on feedback, we may
consider adding additional options (e.g., a `"preserve"` mode, to avoid
changing quotes; or a `"multiline-quote-style"` to override this).
Closes https://github.com/astral-sh/ruff/issues/7615.
## Test Plan
`cargo test`
## Summary
Extends the pragma comment detection in the formatter to support
case-insensitive `noqa` (as supposed by Ruff), plus a variety of other
pragmas (`isort:`, `nosec`, etc.).
Also extracts the detection out into the trivia crate so that we can
reuse it in the linter (see:
https://github.com/astral-sh/ruff/issues/7471).
## Test Plan
`cargo test`
## Summary
No-op refactor, but we can evaluate early if the first part of
`preserve_parentheses || has_comments` is `true`, and thus avoid looking
up the node comments.
## Test Plan
`cargo test`
## Summary
The formatting for tuple patterns is now intended to match that of `for`
loops:
- Always parenthesize single-element tuples.
- Don't break on the trailing comma in single-element tuples.
- For other tuples, preserve the parentheses, and insert if-breaks.
Closes https://github.com/astral-sh/ruff/issues/7681.
## Test Plan
`cargo test`
I got confused and refactored a bit, now the naming should be more
consistent. This is the basis for the range formatting work.
Chages:
* `format_module` -> `format_module_source` (format a string)
* `format_node` -> `format_module_ast` (format a program parsed into an
AST)
* Added `parse_ok_tokens` that takes `Token` instead of `Result<Token>`
* Call the source code `source` consistently
* Added a `tokens_and_ranges` helper
* `python_ast` -> `module` (because that's the type)
## Summary
This is whitespace as per `is_python_whitespace`, and right now it tends
to lead to panics in the formatter. Seems reasonable to treat it as
whitespace in the `SimpleTokenizer` too.
Closes .https://github.com/astral-sh/ruff/issues/7624.
## Summary
Given:
```python
if True:
if True:
pass
else:
pass
# a
# b
# c
else:
pass
```
We want to preserve the newline after the `# c` (before the `else`).
However, the `last_node` ends at the `pass`, and the comments are
trailing comments on the `pass`, not trailing comments on the
`last_node` (the `if`). As such, when counting the trailing newlines on
the outer `if`, we abort as soon as we see the comment (`# a`).
This PR changes the logic to skip _all_ comments (even those with
newlines between them). This is safe as we know that there are no
"leading" comments on the `else`, so there's no risk of skipping those
accidentally.
Closes https://github.com/astral-sh/ruff/issues/7602.
## Test Plan
No change in compatibility.
Before:
| project | similarity index | total files | changed files |
|--------------|------------------:|------------------:|------------------:|
| cpython | 0.76083 | 1789 | 1631 |
| django | 0.99983 | 2760 | 36 |
| transformers | 0.99963 | 2587 | 319 |
| twine | 1.00000 | 33 | 0 |
| typeshed | 0.99979 | 3496 | 22 |
| warehouse | 0.99967 | 648 | 15 |
| zulip | 0.99972 | 1437 | 21 |
After:
| project | similarity index | total files | changed files |
|--------------|------------------:|------------------:|------------------:|
| cpython | 0.76083 | 1789 | 1631 |
| django | 0.99983 | 2760 | 36 |
| transformers | 0.99963 | 2587 | 319 |
| twine | 1.00000 | 33 | 0 |
| typeshed | 0.99983 | 3496 | 18 |
| warehouse | 0.99967 | 648 | 15 |
| zulip | 0.99972 | 1437 | 21 |
Similar to tuples, a generator _can_ be parenthesized or
unparenthesized. Only search for bracketed comments if it contains its
own parentheses.
Closes https://github.com/astral-sh/ruff/issues/7623.
## Summary
Given:
```python
if True:
if True:
if True:
pass
#a
#b
#c
else:
pass
```
When determining the placement of the various comments, we compute the
indentation depth of each comment, and then compare it to the depth of
the previous statement. It turns out this can lead to reordering
comments, e.g., above, `#b` is assigned as a trailing comment of `pass`,
and so gets reordered above `#a`.
This PR modifies the logic such that when we compute the indentation
depth of `#b`, we limit it to at most the indentation depth of `#a`. In
other words, when analyzing comments at the end of branches, we don't
let successive comments go any _deeper_ than their preceding comments.
Closes https://github.com/astral-sh/ruff/issues/7602.
## Test Plan
`cargo test`
No change in similarity.
Before:
| project | similarity index | total files | changed files |
|--------------|------------------:|------------------:|------------------:|
| cpython | 0.76083 | 1789 | 1631 |
| django | 0.99983 | 2760 | 36 |
| transformers | 0.99963 | 2587 | 319 |
| twine | 1.00000 | 33 | 0 |
| typeshed | 0.99979 | 3496 | 22 |
| warehouse | 0.99967 | 648 | 15 |
| zulip | 0.99972 | 1437 | 21 |
After:
| project | similarity index | total files | changed files |
|--------------|------------------:|------------------:|------------------:|
| cpython | 0.76083 | 1789 | 1631 |
| django | 0.99983 | 2760 | 36 |
| transformers | 0.99963 | 2587 | 319 |
| twine | 1.00000 | 33 | 0 |
| typeshed | 0.99979 | 3496 | 22 |
| warehouse | 0.99967 | 648 | 15 |
| zulip | 0.99972 | 1437 | 21 |
## Summary
Given:
```python
# -*- coding: utf-8 -*-
import random
# Defaults for arguments are defined here
# args.threshold = None;
logger = logging.getLogger("FastProject")
```
We want to count the number of newlines after `import random`, to ensure
that there's _at least one_, but up to two.
Previously, we used the end range of the statement (then skipped
trivia); instead, we need to use the end of the _last comment_. This is
similar to #7556.
Closes https://github.com/astral-sh/ruff/issues/7604.
Micha Reiser
Andrew Gallant
Charlie Marsh
Dhruv Manilawala
konsti
doolio
Zanie Blue
Chris Pryer
Cosmo