feat: standalone file search CLI

codex-rs/Cargo.toml

@@ -8,6 +8,7 @@ members = [
     "core",
     "exec",
     "execpolicy",
+    "file-search",
     "linux-sandbox",
     "login",
     "mcp-client",

codex-rs/file-search/Cargo.toml

@@ -0,0 +1,20 @@
+[package]
+name = "codex-file-search"
+version = { workspace = true }
+edition = "2024"
+
+[[bin]]
+name = "codex-file-search"
+path = "src/main.rs"
+
+[lib]
+name = "codex_file_search"
+path = "src/lib.rs"
+
+[dependencies]
+anyhow = "1"
+clap = { version = "4", features = ["derive"] }
+ignore = "0.4.23"
+nucleo-matcher = "0.3.1"
+serde_json = "1.0.110"
+tokio = { version = "1", features = ["full"] }

codex-rs/file-search/README.md

@@ -0,0 +1,5 @@
+# codex_file_search
+
+Fast fuzzy file search tool for Codex.
+
+Uses <https://crates.io/crates/ignore> under the hood (which is what `ripgrep` uses) to traverse a directory (while honoring `.gitignore`, etc.) to produce the list of files to search and then uses <https://crates.io/crates/nucleo-matcher> to fuzzy-match the user supplied `PATTERN` against the corpus.

codex-rs/file-search/src/cli.rs

@@ -0,0 +1,38 @@
+use std::num::NonZero;
+use std::path::PathBuf;
+
+use clap::ArgAction;
+use clap::Parser;
+
+/// Fuzzy matches filenames under a directory.
+#[derive(Parser)]
+#[command(version)]
+pub struct Cli {
+    /// Whether to output results in JSON format.
+    #[clap(long, default_value = "false")]
+    pub json: bool,
+
+    /// Maximum number of results to return.
+    #[clap(long, short = 'l', default_value = "64")]
+    pub limit: NonZero<usize>,
+
+    /// Directory to search.
+    #[clap(long, short = 'C')]
+    pub cwd: Option<PathBuf>,
+
+    // While it is common to default to the number of logical CPUs when creating
+    // a thread pool, empirically, the I/O of the filetree traversal offers
+    // limited parallelism and is the bottleneck, so using a smaller number of
+    // threads is more efficient. (Empirically, using more than 2 threads doesn't seem to provide much benefit.)
+    //
+    /// Number of worker threads to use.
+    #[clap(long, default_value = "2")]
+    pub threads: NonZero<usize>,
+
+    /// Exclude patterns
+    #[arg(short, long, action = ArgAction::Append)]
+    pub exclude: Vec<String>,
+
+    /// Search pattern.
+    pub pattern: Option<String>,
+}