Migrate to Swift concurrency. #675
Open
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This replaces `Dispatch.concurrentPerform` with a task group. One benefit of this approach is that we don't need to know the number of iterations that we want ahead of time. We can start tasks during the initial iteration over the `FileIterator` instead of eagerly expanding it first; this should potentially improve behavior for invocations that cover very large directories recursively, especially when they're on slower storage like a network drive.
Instead of having `StderrDiagnosticPrinter` internally manage its own synchronization (which currently uses a Dispatch queue), turn `DiagnosticsEngine` into an actor that processes diagnostics from an `AsyncStream`. When rules emit findings, they are enqueued in the stream, and a single task iterates over them to ensure that individual diagnostic handlers don't need to synchronize their own activity.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
I ran the old version and the new version in release mode for 15 iterations over about 400 source files and got the following mean wall times:
lint: 9.71 sec (Dispatch) vs. 9.83 sec (Concurrency)format: 20.61 sec (Dispatch) vs. 20.79 (Concurrency)Even though the means for Concurrency are slightly higher, Student's t-test yielded p = 0.416 (lint) and p = 0.439 (format) which aren't low enough to claim they're significant.