Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Reuse BTreeSets in module resolver #13440

Merged
merged 1 commit into from
Sep 21, 2024
Merged

Reuse BTreeSets in module resolver #13440

merged 1 commit into from
Sep 21, 2024

Conversation

charliermarsh
Copy link
Member

Summary

For dependencies, there's no reason to re-allocate here, since we know the paths are unique.

@charliermarsh charliermarsh force-pushed the charlie/reuse branch 2 times, most recently from 2233fc1 to 66494eb Compare September 21, 2024 18:30
@charliermarsh charliermarsh added the performance Potential performance improvement label Sep 21, 2024
@@ -91,17 +91,24 @@ impl ModuleImports {

#[derive(Debug, Default)]
#[cfg_attr(feature = "serde", derive(Serialize, Deserialize))]
pub struct ImportMap(BTreeMap<SystemPathBuf, ModuleImports>);
pub struct ImportMap(BTreeMap<SystemPathBuf, BTreeSet<SystemPathBuf>>);
Copy link
Member

@MichaReiser MichaReiser Sep 21, 2024

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Would you mind naming the fields? The code is now kind of hard to understand where the types are generic data structures and the fields are not named

Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

You could also use File as values if you happen to have interned anyway.

A File is just a u32, so super fast to hash

Copy link
Member Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

What kinds of names would help here? Do you mean using named type aliases or named fields?

Copy link
Member Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Oh wow, smart. I do need it to be sorted / ordered in the output though 🤔

Copy link
Member Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I could try doing everything on u32 and then converting to sorted system paths at the very end

Copy link
Member Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

(I just retained the use of the struct here.)

Copy link
Contributor

github-actions bot commented Sep 21, 2024

ruff-ecosystem results

Linter (stable)

✅ ecosystem check detected no linter changes.

Linter (preview)

✅ ecosystem check detected no linter changes.

@charliermarsh charliermarsh enabled auto-merge (squash) September 21, 2024 20:10
@charliermarsh charliermarsh merged commit c2a5179 into main Sep 21, 2024
17 checks passed
@charliermarsh charliermarsh deleted the charlie/reuse branch September 21, 2024 20:14
netbsd-srcmastr pushed a commit to NetBSD/pkgsrc that referenced this pull request Oct 7, 2024
## 0.6.9

### Preview features

- Fix codeblock dynamic line length calculation for indented docstring examples ([#13523](astral-sh/ruff#13523))
- \[`refurb`\] Mark `FURB118` fix as unsafe ([#13613](astral-sh/ruff#13613))

### Rule changes

- \[`pydocstyle`\] Don't raise `D208` when last line is non-empty ([#13372](astral-sh/ruff#13372))
- \[`pylint`\] Preserve trivia (i.e. comments) in `PLR5501` autofix ([#13573](astral-sh/ruff#13573))

### Configuration

- \[`pyflakes`\] Add `allow-unused-imports` setting for `unused-import` rule (`F401`) ([#13601](astral-sh/ruff#13601))

### Bug fixes

- Support ruff discovery in pip build environments ([#13591](astral-sh/ruff#13591))
- \[`flake8-bugbear`\] Avoid short circuiting `B017` for multiple context managers ([#13609](astral-sh/ruff#13609))
- \[`pylint`\] Do not offer an invalid fix for `PLR1716` when the comparisons contain parenthesis ([#13527](astral-sh/ruff#13527))
- \[`pyupgrade`\] Fix `UP043` to apply to `collections.abc.Generator` and `collections.abc.AsyncGenerator` ([#13611](astral-sh/ruff#13611))
- \[`refurb`\] Fix handling of slices in tuples for `FURB118`, e.g., `x[:, 1]` ([#13518](astral-sh/ruff#13518))

### Documentation

- Update GitHub Action link to `astral-sh/ruff-action` ([#13551](astral-sh/ruff#13551))

## 0.6.8

### Preview features

- Remove unnecessary parentheses around `match case` clauses ([#13510](astral-sh/ruff#13510))
- Parenthesize overlong `if` guards in `match..case` clauses ([#13513](astral-sh/ruff#13513))
- Detect basic wildcard imports in `ruff analyze graph` ([#13486](astral-sh/ruff#13486))
- \[`pylint`\] Implement `boolean-chained-comparison` (`R1716`) ([#13435](astral-sh/ruff#13435))

### Rule changes

- \[`lake8-simplify`\] Detect `SIM910` when using variadic keyword arguments, i.e., `**kwargs` ([#13503](astral-sh/ruff#13503))
- \[`pyupgrade`\] Avoid false negatives with non-reference shadowed bindings of loop variables (`UP028`) ([#13504](astral-sh/ruff#13504))

### Bug fixes

- Detect tuples bound to variadic positional arguments i.e. `*args` ([#13512](astral-sh/ruff#13512))
- Exit gracefully on broken pipe errors ([#13485](astral-sh/ruff#13485))
- Avoid panic when analyze graph hits broken pipe ([#13484](astral-sh/ruff#13484))

### Performance

- Reuse `BTreeSets` in module resolver ([#13440](astral-sh/ruff#13440))
- Skip traversal for non-compound statements ([#13441](astral-sh/ruff#13441))

## 0.6.7

### Preview features

- Add Python version support to ruff analyze CLI ([#13426](astral-sh/ruff#13426))
- Add `exclude` support to `ruff analyze` ([#13425](astral-sh/ruff#13425))
- Fix parentheses around return type annotations ([#13381](astral-sh/ruff#13381))

### Rule changes

- \[`pycodestyle`\] Fix: Don't autofix if the first line ends in a question mark? (D400) ([#13399](astral-sh/ruff#13399))

### Bug fixes

- Respect `lint.exclude` in ruff check `--add-noqa` ([#13427](astral-sh/ruff#13427))

### Performance

- Avoid tracking module resolver files in Salsa ([#13437](astral-sh/ruff#13437))
- Use `forget` for module resolver database ([#13438](astral-sh/ruff#13438))

## 0.6.6

### Preview features

- \[`refurb`\] Skip `slice-to-remove-prefix-or-suffix` (`FURB188`) when non-trivial slice steps are present ([#13405](astral-sh/ruff#13405))
- Add a subcommand to generate dependency graphs ([#13402](astral-sh/ruff#13402))

### Formatter

- Fix placement of inline parameter comments ([#13379](astral-sh/ruff#13379))

### Server

- Fix off-by one error in the `LineIndex::offset` calculation ([#13407](astral-sh/ruff#13407))

### Bug fixes

- \[`fastapi`\] Respect FastAPI aliases in route definitions ([#13394](astral-sh/ruff#13394))
- \[`pydocstyle`\] Respect word boundaries when detecting function signature in docs ([#13388](astral-sh/ruff#13388))

### Documentation

- Add backlinks to rule overview linter ([#13368](astral-sh/ruff#13368))
- Fix documentation for editor vim plugin ALE ([#13348](astral-sh/ruff#13348))
- Fix rendering of `FURB188` docs ([#13406](astral-sh/ruff#13406))
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
performance Potential performance improvement
Projects
None yet
Development

Successfully merging this pull request may close these issues.

2 participants