Move glob implementation to new crate.

It is isolated and complex enough that it deserves attention all on its
own. It's also eminently reusable.

.gitignore

@@ -2,3 +2,4 @@
 tags
 target
 /grep/Cargo.lock
+/globset/Cargo.lock

Cargo.lock

@@ -5,8 +5,8 @@ dependencies = [
  "deque 0.3.1 (registry+https://github.com/rust-lang/crates.io-index)",
  "docopt 0.6.86 (registry+https://github.com/rust-lang/crates.io-index)",
  "env_logger 0.3.5 (registry+https://github.com/rust-lang/crates.io-index)",
- "fnv 1.0.5 (registry+https://github.com/rust-lang/crates.io-index)",
  "glob 0.2.11 (registry+https://github.com/rust-lang/crates.io-index)",
+ "globset 0.1.0",
  "grep 0.1.3",
  "kernel32-sys 0.2.2 (registry+https://github.com/rust-lang/crates.io-index)",
  "lazy_static 0.2.1 (registry+https://github.com/rust-lang/crates.io-index)",
@@ -78,6 +78,16 @@ name = "glob"
 version = "0.2.11"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 
+[[package]]
+name = "globset"
+version = "0.1.0"
+dependencies = [
+ "fnv 1.0.5 (registry+https://github.com/rust-lang/crates.io-index)",
+ "lazy_static 0.2.1 (registry+https://github.com/rust-lang/crates.io-index)",
+ "memchr 0.1.11 (registry+https://github.com/rust-lang/crates.io-index)",
+ "regex 0.1.77 (registry+https://github.com/rust-lang/crates.io-index)",
+]
+
 [[package]]
 name = "grep"
 version = "0.1.3"

Cargo.toml

@@ -26,7 +26,7 @@ path = "tests/tests.rs"
 deque = "0.3"
 docopt = "0.6"
 env_logger = "0.3"
-fnv = "1.0"
+globset = { version = "0.1.0", path = "globset" }
 grep = { version = "0.1.3", path = "grep" }
 lazy_static = "0.2"
 libc = "0.2"

globset/Cargo.toml

@@ -0,0 +1,10 @@
+[package]
+name = "globset"
+version = "0.1.0"
+authors = ["Andrew Gallant <jamslam@gmail.com>"]
+
+[dependencies]
+fnv = "1.0"
+lazy_static = "0.2"
+memchr = "0.1"
+regex = "0.1.77"

globset/src/lib.rs

@@ -26,6 +26,12 @@ to make its way into `glob` proper.
 // at the .gitignore for the chromium repo---just about every pattern satisfies
 // that assumption.)
 
+extern crate fnv;
+#[macro_use]
+extern crate lazy_static;
+extern crate memchr;
+extern crate regex;
+
 use std::borrow::Cow;
 use std::collections::HashMap;
 use std::error::Error as StdError;
@@ -36,12 +42,12 @@ use std::iter;
 use std::path::Path;
 use std::str;
 
-use fnv;
-use regex;
 use regex::bytes::Regex;
 
 use pathutil::file_name;
 
+mod pathutil;
+
 lazy_static! {
     static ref FILE_SEPARATORS: String = regex::quote(r"/\");
 }

globset/src/pathutil.rs

@@ -0,0 +1,38 @@
+use std::ffi::OsStr;
+use std::path::Path;
+
+/// The final component of the path, if it is a normal file.
+///
+/// If the path terminates in ., .., or consists solely of a root of prefix,
+/// file_name will return None.
+#[cfg(unix)]
+pub fn file_name<'a, P: AsRef<Path> + ?Sized>(
+    path: &'a P,
+) -> Option<&'a OsStr> {
+    use std::os::unix::ffi::OsStrExt;
+    use memchr::memrchr;
+
+    let path = path.as_ref().as_os_str().as_bytes();
+    if path.is_empty() {
+        return None;
+    } else if path.len() == 1 && path[0] == b'.' {
+        return None;
+    } else if path.last() == Some(&b'.') {
+        return None;
+    } else if path.len() >= 2 && &path[path.len() - 2..] == &b".."[..] {
+        return None;
+    }
+    let last_slash = memrchr(b'/', path).map(|i| i + 1).unwrap_or(0);
+    Some(OsStr::from_bytes(&path[last_slash..]))
+}
+
+/// The final component of the path, if it is a normal file.
+///
+/// If the path terminates in ., .., or consists solely of a root of prefix,
+/// file_name will return None.
+#[cfg(not(unix))]
+pub fn file_name<'a, P: AsRef<Path> + ?Sized>(
+    path: &'a P,
+) -> Option<&'a OsStr> {
+    path.as_ref().file_name()
+}

src/gitignore.rs

@@ -28,15 +28,15 @@ use std::fs::File;
 use std::io::{self, BufRead};
 use std::path::{Path, PathBuf};
 
+use globset;
 use regex;
 
-use glob;
 use pathutil::{is_file_name, strip_prefix};
 
 /// Represents an error that can occur when parsing a gitignore file.
 #[derive(Debug)]
 pub enum Error {
-    Glob(glob::Error),
+    Glob(globset::Error),
     Regex(regex::Error),
     Io(io::Error),
 }
@@ -61,8 +61,8 @@ impl fmt::Display for Error {
     }
 }
 
-impl From<glob::Error> for Error {
+impl From<globset::Error> for Error {
-    fn from(err: glob::Error) -> Error {
+    fn from(err: globset::Error) -> Error {
         Error::Glob(err)
     }
 }
@@ -82,7 +82,7 @@ impl From<io::Error> for Error {
 /// Gitignore is a matcher for the glob patterns in a single gitignore file.
 #[derive(Clone, Debug)]
 pub struct Gitignore {
-    set: glob::Set,
+    set: globset::Set,
     root: PathBuf,
     patterns: Vec<Pattern>,
     num_ignores: u64,
@@ -207,7 +207,7 @@ impl<'a> Match<'a> {
 /// GitignoreBuilder constructs a matcher for a single set of globs from a
 /// .gitignore file.
 pub struct GitignoreBuilder {
-    builder: glob::SetBuilder,
+    builder: globset::SetBuilder,
     root: PathBuf,
     patterns: Vec<Pattern>,
 }
@@ -237,7 +237,7 @@ impl GitignoreBuilder {
     pub fn new<P: AsRef<Path>>(root: P) -> GitignoreBuilder {
         let root = strip_prefix("./", root.as_ref()).unwrap_or(root.as_ref());
         GitignoreBuilder {
-            builder: glob::SetBuilder::new(),
+            builder: globset::SetBuilder::new(),
             root: root.to_path_buf(),
             patterns: vec![],
         }
@@ -299,7 +299,7 @@ impl GitignoreBuilder {
             whitelist: false,
             only_dir: false,
         };
-        let mut opts = glob::MatchOptions::default();
+        let mut opts = globset::MatchOptions::default();
         let has_slash = line.chars().any(|c| c == '/');
         let is_absolute = line.chars().nth(0).unwrap() == '/';
         if line.starts_with("\\!") || line.starts_with("\\#") {

src/main.rs

@@ -1,7 +1,7 @@
 extern crate deque;
 extern crate docopt;
 extern crate env_logger;
-extern crate fnv;
+extern crate globset;
 extern crate grep;
 #[cfg(windows)]
 extern crate kernel32;
@@ -61,7 +61,6 @@ macro_rules! eprintln {
 mod args;
 mod atty;
 mod gitignore;
-mod glob;
 mod ignore;
 mod out;
 mod pathutil;

src/types.rs

@@ -11,7 +11,7 @@ use std::path::Path;
 use regex;
 
 use gitignore::{Match, Pattern};
-use glob::{self, MatchOptions};
+use globset::{self, MatchOptions};
 
 const TYPE_EXTENSIONS: &'static [(&'static str, &'static [&'static str])] = &[
     ("asm", &["*.asm", "*.s", "*.S"]),
@@ -93,7 +93,7 @@ pub enum Error {
     /// A user specified file type definition could not be parsed.
     InvalidDefinition,
     /// There was an error building the matcher (probably a bad glob).
-    Glob(glob::Error),
+    Glob(globset::Error),
     /// There was an error compiling a glob as a regex.
     Regex(regex::Error),
 }
@@ -125,8 +125,8 @@ impl fmt::Display for Error {
     }
 }
 
-impl From<glob::Error> for Error {
+impl From<globset::Error> for Error {
-    fn from(err: glob::Error) -> Error {
+    fn from(err: globset::Error) -> Error {
         Error::Glob(err)
     }
 }
@@ -160,8 +160,8 @@ impl FileTypeDef {
 #[derive(Clone, Debug)]
 pub struct Types {
     defs: Vec<FileTypeDef>,
-    selected: Option<glob::SetYesNo>,
+    selected: Option<globset::SetYesNo>,
-    negated: Option<glob::SetYesNo>,
+    negated: Option<globset::SetYesNo>,
     has_selected: bool,
     unmatched_pat: Pattern,
 }
@@ -174,8 +174,8 @@ impl Types {
     /// If has_selected is true, then at least one file type was selected.
     /// Therefore, any non-matches should be ignored.
     fn new(
-        selected: Option<glob::SetYesNo>,
+        selected: Option<globset::SetYesNo>,
-        negated: Option<glob::SetYesNo>,
+        negated: Option<globset::SetYesNo>,
         has_selected: bool,
         defs: Vec<FileTypeDef>,
     ) -> Types {
@@ -271,7 +271,7 @@ impl TypesBuilder {
             if self.selected.is_empty() {
                 None
             } else {
-                let mut bset = glob::SetBuilder::new();
+                let mut bset = globset::SetBuilder::new();
                 for name in &self.selected {
                     let globs = match self.types.get(name) {
                         Some(globs) => globs,
@@ -290,7 +290,7 @@ impl TypesBuilder {
             if self.negated.is_empty() {
                 None
             } else {
-                let mut bset = glob::SetBuilder::new();
+                let mut bset = globset::SetBuilder::new();
                 for name in &self.negated {
                     let globs = match self.types.get(name) {
                         Some(globs) => globs,