From 361d45f2b2f7e69d37d4e6d8270e448f72cae9a7 Mon Sep 17 00:00:00 2001
From: konstin <konstin@mailbox.org>
Date: Mon, 19 Jun 2023 11:40:09 +0200
Subject: [PATCH] Add `cargo dev repeat` for profiling (#5144)

## Summary

This adds a new subcommand that can be used as

```shell
cargo build --bin ruff_dev --profile=release-debug
perf record -g -F 999 target/release-debug/ruff_dev repeat --repeat 30 --exit-zero --no-cache path/to/cpython > /dev/null
flamegraph --perfdata perf.data
```

## Test Plan

This is a ruff internal script. I successfully used it to profile
cpython with the instructions above
---
 .gitignore                  |  5 +++++
 crates/ruff_cli/src/args.rs |  2 +-
 crates/ruff_cli/src/lib.rs  |  2 +-
 crates/ruff_dev/src/main.rs | 24 ++++++++++++++++++++++++
 4 files changed, 31 insertions(+), 2 deletions(-)

diff --git a/.gitignore b/.gitignore
index 55de3ba557e244..562d1b4c5a6dd0 100644
--- a/.gitignore
+++ b/.gitignore
@@ -6,6 +6,11 @@ github_search*.jsonl
 schemastore
 .venv*
 scratch.py
+perf.data
+perf.data.old
+flamegraph.svg
+# Additional target directories that don't invalidate the main compile cache when changing linker settings
+/target*
 
 ###
 # Rust.gitignore
diff --git a/crates/ruff_cli/src/args.rs b/crates/ruff_cli/src/args.rs
index 09662c827281ad..eb3efc09a5c111 100644
--- a/crates/ruff_cli/src/args.rs
+++ b/crates/ruff_cli/src/args.rs
@@ -68,7 +68,7 @@ pub enum Command {
     },
 }
 
-#[derive(Debug, clap::Args)]
+#[derive(Clone, Debug, clap::Args)]
 #[allow(clippy::struct_excessive_bools, clippy::module_name_repetitions)]
 pub struct CheckArgs {
     /// List of files or directories to check.
diff --git a/crates/ruff_cli/src/lib.rs b/crates/ruff_cli/src/lib.rs
index a9b53365fb7954..4307f742907e1e 100644
--- a/crates/ruff_cli/src/lib.rs
+++ b/crates/ruff_cli/src/lib.rs
@@ -159,7 +159,7 @@ fn format(files: &[PathBuf]) -> Result<ExitStatus> {
     Ok(ExitStatus::Success)
 }
 
-fn check(args: CheckArgs, log_level: LogLevel) -> Result<ExitStatus> {
+pub fn check(args: CheckArgs, log_level: LogLevel) -> Result<ExitStatus> {
     let (cli, overrides) = args.partition();
 
     // Construct the "default" settings. These are used when no `pyproject.toml`
diff --git a/crates/ruff_dev/src/main.rs b/crates/ruff_dev/src/main.rs
index c8beed5fb194c8..dbdfb4546693a7 100644
--- a/crates/ruff_dev/src/main.rs
+++ b/crates/ruff_dev/src/main.rs
@@ -4,6 +4,8 @@
 
 use anyhow::Result;
 use clap::{Parser, Subcommand};
+use ruff::logging::{set_up_logging, LogLevel};
+use ruff_cli::check;
 
 mod generate_all;
 mod generate_cli_help;
@@ -27,6 +29,7 @@ struct Args {
 }
 
 #[derive(Subcommand)]
+#[allow(clippy::large_enum_variant)]
 enum Command {
     /// Run all code and documentation generation steps.
     GenerateAll(generate_all::Args),
@@ -48,6 +51,16 @@ enum Command {
     PrintTokens(print_tokens::Args),
     /// Run round-trip source code generation on a given Python file.
     RoundTrip(round_trip::Args),
+    /// Run a ruff command n times for profiling/benchmarking
+    Repeat {
+        #[clap(flatten)]
+        args: ruff_cli::args::CheckArgs,
+        #[clap(flatten)]
+        log_level_args: ruff_cli::args::LogLevelArgs,
+        /// Run this many times
+        #[clap(long, short = 'n')]
+        repeat: usize,
+    },
 }
 
 fn main() -> Result<()> {
@@ -64,6 +77,17 @@ fn main() -> Result<()> {
         Command::PrintCST(args) => print_cst::main(args)?,
         Command::PrintTokens(args) => print_tokens::main(args)?,
         Command::RoundTrip(args) => round_trip::main(args)?,
+        Command::Repeat {
+            args,
+            repeat,
+            log_level_args,
+        } => {
+            let log_level = LogLevel::from(log_level_args);
+            set_up_logging(&log_level)?;
+            for _ in 0..*repeat {
+                check(args.clone(), log_level)?;
+            }
+        }
     }
     Ok(())
 }