[ty] File inclusion and exclusion (astral-sh#18498)

Author: MichaReiser

Cargo.lock

@@ -126,6 +126,7 @@ quote = { version = "1.0.23" }
 rand = { version = "0.9.0" }
 rayon = { version = "1.10.0" }
 regex = { version = "1.10.2" }
+regex-automata = { version = "0.4.9" }
 rustc-hash = { version = "2.0.0" }
 rustc-stable-hash = { version = "0.1.2" }
 # When updating salsa, make sure to also update the revision in `fuzz/Cargo.toml`
@@ -165,7 +166,7 @@ tracing-subscriber = { version = "0.3.18", default-features = false, features =
     "env-filter",
     "fmt",
     "ansi",
-    "smallvec",
+    "smallvec"
 ] }
 tryfn = { version = "0.2.1" }
 typed-arena = { version = "2.0.2" }
@@ -175,7 +176,11 @@ unicode-width = { version = "0.2.0" }
 unicode_names2 = { version = "1.2.2" }
 unicode-normalization = { version = "0.1.23" }
 url = { version = "2.5.0" }
-uuid = { version = "1.6.1", features = ["v4", "fast-rng", "macro-diagnostics"] }
+uuid = { version = "1.6.1", features = [
+    "v4",
+    "fast-rng",
+    "macro-diagnostics",
+] }
 walkdir = { version = "2.3.2" }
 wasm-bindgen = { version = "0.2.92" }
 wasm-bindgen-test = { version = "0.3.42" }
@@ -210,8 +215,8 @@ must_use_candidate = "allow"
 similar_names = "allow"
 single_match_else = "allow"
 too_many_lines = "allow"
-needless_continue = "allow"            # An explicit continue can be more readable, especially if the alternative is an empty block.
-unnecessary_debug_formatting = "allow" # too many instances, the display also doesn't quote the path which is often desired in logs where we use them the most often.
+needless_continue = "allow"       # An explicit continue can be more readable, especially if the alternative is an empty block.
+unnecessary_debug_formatting = "allow"  # too many instances, the display also doesn't quote the path which is often desired in logs where we use them the most often.
 # Without the hashes we run into a `rustfmt` bug in some snapshot tests, see #13250
 needless_raw_string_hashes = "allow"
 # Disallowed restriction lints

Cargo.toml

@@ -665,6 +665,9 @@ pub enum DiagnosticId {
 
     /// No rule with the given name exists.
     UnknownRule,
+
+    /// A glob pattern doesn't follow the expected syntax.
+    InvalidGlob,
 }
 
 impl DiagnosticId {
@@ -699,6 +702,7 @@ impl DiagnosticId {
             DiagnosticId::Lint(name) => name.as_str(),
             DiagnosticId::RevealedType => "revealed-type",
             DiagnosticId::UnknownRule => "unknown-rule",
+            DiagnosticId::InvalidGlob => "invalid-glob",
         }
     }
 

crates/ruff_db/src/diagnostic/mod.rs

@@ -45,6 +45,30 @@ impl SystemPath {
         SystemPath::from_std_path(dunce::simplified(self.as_std_path())).unwrap()
     }
 
+    /// Returns `true` if the `SystemPath` is absolute, i.e., if it is independent of
+    /// the current directory.
+    ///
+    /// * On Unix, a path is absolute if it starts with the root, so
+    ///   `is_absolute` and [`has_root`] are equivalent.
+    ///
+    /// * On Windows, a path is absolute if it has a prefix and starts with the
+    ///   root: `c:\windows` is absolute, while `c:temp` and `\temp` are not.
+    ///
+    /// # Examples
+    ///
+    /// ```
+    /// use ruff_db::system::SystemPath;
+    ///
+    /// assert!(!SystemPath::new("foo.txt").is_absolute());
+    /// ```
+    ///
+    /// [`has_root`]: Utf8Path::has_root
+    #[inline]
+    #[must_use]
+    pub fn is_absolute(&self) -> bool {
+        self.0.is_absolute()
+    }
+
     /// Extracts the file extension, if possible.
     ///
     /// The extension is:
@@ -538,6 +562,10 @@ impl SystemPathBuf {
         self.0.into_std_path_buf()
     }
 
+    pub fn into_string(self) -> String {
+        self.0.into_string()
+    }
+
     #[inline]
     pub fn as_path(&self) -> &SystemPath {
         SystemPath::new(&self.0)

crates/ruff_db/src/system/path.rs

@@ -5,7 +5,9 @@ use clap::{ArgAction, ArgMatches, Error, Parser};
 use ruff_db::system::SystemPathBuf;
 use ty_project::combine::Combine;
 use ty_project::metadata::options::{EnvironmentOptions, Options, SrcOptions, TerminalOptions};
-use ty_project::metadata::value::{RangedValue, RelativePathBuf, ValueSource};
+use ty_project::metadata::value::{
+    RangedValue, RelativeExcludePattern, RelativePathBuf, ValueSource,
+};
 use ty_python_semantic::lint;
 
 #[derive(Debug, Parser)]
@@ -148,6 +150,13 @@ pub(crate) struct CheckCommand {
     respect_ignore_files: Option<bool>,
     #[clap(long, overrides_with("respect_ignore_files"), hide = true)]
     no_respect_ignore_files: bool,
+
+    /// Glob patterns for files to exclude from type checking.
+    ///
+    /// Uses gitignore-style syntax to exclude files and directories from type checking.
+    /// Supports patterns like `tests/`, `*.tmp`, `**/__pycache__/**`.
+    #[arg(long, help_heading = "File selection")]
+    exclude: Option<Vec<String>>,
 }
 
 impl CheckCommand {
@@ -195,6 +204,12 @@ impl CheckCommand {
             }),
             src: Some(SrcOptions {
                 respect_ignore_files,
+                exclude: self.exclude.map(|excludes| {
+                    excludes
+                        .iter()
+                        .map(|exclude| RelativeExcludePattern::cli(exclude))
+                        .collect()
+                }),
                 ..SrcOptions::default()
             }),
             rules,