Format all attribute dot comments manually (#6825)

## Summary

This PR modifies our formatting of comments around the `.` in an
attribute. Specifically, the goal here is to avoid _reordering_
comments, and the net effect is that we generally leave comments
where-they-are when dealing with comments between around the dot (which
you can also think of as comments between attributes).

All comments around the dot are now treated as dangling and formatted
manually, with the exception of end-of-line or parenthesized comments on
the value, like those marked as trailing here, which remain trailing:

```python
(
    (
        a # trailing end-of-line
        # trailing own-line
    ) # dangling before dot end-of-line
    .b # trailing end-of-line
)
```

Closes https://github.com/astral-sh/ruff/issues/6823.

## Test Plan

`cargo test`

Before:

| project      | similarity index |
|--------------|------------------|
| cpython      | 0.76050          |
| django       | 0.99820          |
| transformers | 0.99800          |
| twine        | 0.99876          |
| typeshed     | 0.99953          |
| warehouse    | 0.99615          |
| zulip        | 0.99729          |

After:

| project      | similarity index |
|--------------|------------------|
| cpython      | 0.76050          |
| django       | 0.99820   |
| transformers | 0.99800          |
| twine        | 0.99876          |
| typeshed     | 0.99953          |
| warehouse    | 0.99615          |
| zulip        | 0.99729          |
This commit is contained in:
Charlie Marsh 2023-08-24 23:50:56 -04:00 committed by GitHub
parent 6f23469e00
commit 474e8fbcd4
No known key found for this signature in database
GPG key ID: 4AEE18F83AFDEB23
5 changed files with 216 additions and 128 deletions

View file

@ -1,8 +1,10 @@
use ruff_formatter::{write, FormatRuleWithOptions};
use ruff_python_ast::node::AnyNodeRef;
use ruff_python_ast::{Constant, Expr, ExprAttribute, ExprConstant};
use ruff_python_ast::{Constant, Expr, ExprAttribute, ExprConstant, Ranged};
use ruff_python_trivia::{find_only_token_in_range, SimpleTokenKind};
use ruff_text_size::TextRange;
use crate::comments::{leading_comments, trailing_comments, SourceComment};
use crate::comments::{dangling_comments, trailing_comments, SourceComment};
use crate::expression::parentheses::{
is_expression_parenthesized, NeedsParentheses, OptionalParentheses, Parentheses,
};
@ -44,13 +46,6 @@ impl FormatNodeRule<ExprAttribute> for FormatExprAttribute {
})
);
let comments = f.context().comments().clone();
let dangling_comments = comments.dangling(item);
let leading_attribute_comments_start = dangling_comments
.partition_point(|comment| comment.line_position().is_end_of_line());
let (trailing_dot_comments, leading_attribute_comments) =
dangling_comments.split_at(leading_attribute_comments_start);
if needs_parentheses {
value.format().with_options(Parentheses::Always).fmt(f)?;
} else if call_chain_layout == CallChainLayout::Fluent {
@ -88,55 +83,54 @@ impl FormatNodeRule<ExprAttribute> for FormatExprAttribute {
value.format().fmt(f)?;
}
if comments.has_trailing_own_line(value.as_ref()) {
hard_line_break().fmt(f)?;
}
if call_chain_layout == CallChainLayout::Fluent {
// Fluent style has line breaks before the dot
// ```python
// blogs3 = (
// Blog.objects.filter(
// entry__headline__contains="Lennon",
// )
// .filter(
// entry__pub_date__year=2008,
// )
// .filter(
// entry__pub_date__year=2008,
// )
// )
// ```
write!(
f,
[
(!leading_attribute_comments.is_empty()).then_some(hard_line_break()),
leading_comments(leading_attribute_comments),
text("."),
trailing_comments(trailing_dot_comments),
attr.format()
]
)
// Identify dangling comments before and after the dot:
// ```python
// (
// (
// a
// ) # `before_dot_end_of_line`
// # `before_dot_own_line`
// . # `after_dot_end_of_line`
// # `after_dot_own_line`
// b
// )
// ```
let comments = f.context().comments().clone();
let dangling = comments.dangling(item);
let (before_dot, after_dot) = if dangling.is_empty() {
(dangling, dangling)
} else {
// Regular style
// ```python
// blogs2 = Blog.objects.filter(
// entry__headline__contains="Lennon",
// ).filter(
// entry__pub_date__year=2008,
// )
// ```
write!(
f,
[
text("."),
trailing_comments(trailing_dot_comments),
(!leading_attribute_comments.is_empty()).then_some(hard_line_break()),
leading_comments(leading_attribute_comments),
attr.format()
]
let dot_token = find_only_token_in_range(
TextRange::new(item.value.end(), item.attr.start()),
SimpleTokenKind::Dot,
f.context().source(),
);
dangling.split_at(
dangling.partition_point(|comment| comment.start() < dot_token.start()),
)
}
};
let (before_dot_end_of_line, before_dot_own_line) = before_dot.split_at(
before_dot.partition_point(|comment| comment.line_position().is_end_of_line()),
);
let (after_dot_end_of_line, after_dot_own_line) = after_dot.split_at(
after_dot.partition_point(|comment| comment.line_position().is_end_of_line()),
);
write!(
f,
[
trailing_comments(before_dot_end_of_line),
(!before_dot.is_empty()).then_some(hard_line_break()),
dangling_comments(before_dot_own_line),
text("."),
trailing_comments(after_dot_end_of_line),
(!after_dot.is_empty()).then_some(hard_line_break()),
dangling_comments(after_dot_own_line),
attr.format()
]
)
});
let is_call_chain_root = self.call_chain_layout == CallChainLayout::Default
@ -169,13 +163,7 @@ impl NeedsParentheses for ExprAttribute {
== CallChainLayout::Fluent
{
OptionalParentheses::Multiline
} else if context
.comments()
.dangling(self)
.iter()
.any(|comment| comment.line_position().is_own_line())
|| context.comments().has_trailing_own_line(self)
{
} else if context.comments().has_dangling(self) {
OptionalParentheses::Always
} else {
self.value.needs_parentheses(self.into(), context)