twin/ripgrep
1
0
Fork 0
mirror of https://github.com/BurntSushi/ripgrep.git synced 2025-07-31 20:21:59 -07:00
Code Issues Packages Projects Releases Wiki Activity

repo: move all source code in crates directory

Browse Source 
The top-level listing was just getting a bit too long for my taste. So
put all of the code in one directory and shrink the large top-level mess
to a small top-level mess.

NOTE: This commit only contains renames. The subsequent commit will
actually make ripgrep build again. We do it this way with the naive hope
that this will make it easier for git history to track the renames.
Sigh.
This commit is contained in:
Andrew Gallant
2020-02-17 18:19:19 -05:00
parent 0bc4f0447b
commit fdd8510fdd
113 changed files with 0 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files

3
crates/ignore/COPYING Normal file
View File

@@ -0,0 +1,3 @@
This project is dual-licensed under the Unlicense and MIT licenses.
You may use this code under the terms of either license.

36
crates/ignore/Cargo.toml Normal file
View File

@@ -0,0 +1,36 @@
[package]
name = "ignore"
version = "0.4.11" #:version
authors = ["Andrew Gallant <jamslam@gmail.com>"]
description = """
A fast library for efficiently matching ignore files such as `.gitignore`
against file paths.
"""
documentation = "https://docs.rs/ignore"
homepage = "https://github.com/BurntSushi/ripgrep/tree/master/ignore"
repository = "https://github.com/BurntSushi/ripgrep/tree/master/ignore"
readme = "README.md"
keywords = ["glob", "ignore", "gitignore", "pattern", "file"]
license = "Unlicense/MIT"
[lib]
name = "ignore"
bench = false
[dependencies]
crossbeam-channel = "0.4.0"
crossbeam-utils = "0.7.0"
globset = { version = "0.4.3", path = "../globset" }
lazy_static = "1.1"
log = "0.4.5"
memchr = "2.1"
regex = "1.1"
same-file = "1.0.4"
thread_local = "1"
walkdir = "2.2.7"
[target.'cfg(windows)'.dependencies.winapi-util]
version = "0.1.2"
[features]
simd-accel = ["globset/simd-accel"]

21
crates/ignore/LICENSE-MIT Normal file
View File

@@ -0,0 +1,21 @@
The MIT License (MIT)
Copyright (c) 2015 Andrew Gallant
Permission is hereby granted, free of charge, to any person obtaining a copy
of this software and associated documentation files (the "Software"), to deal
in the Software without restriction, including without limitation the rights
to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
copies of the Software, and to permit persons to whom the Software is
furnished to do so, subject to the following conditions:
The above copyright notice and this permission notice shall be included in
all copies or substantial portions of the Software.
THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
THE SOFTWARE.

66
crates/ignore/README.md Normal file
View File

@@ -0,0 +1,66 @@
ignore
======
The ignore crate provides a fast recursive directory iterator that respects
various filters such as globs, file types and `.gitignore` files. This crate
also provides lower level direct access to gitignore and file type matchers.
[![Linux build status](https://api.travis-ci.org/BurntSushi/ripgrep.svg)](https://travis-ci.org/BurntSushi/ripgrep)
[![Windows build status](https://ci.appveyor.com/api/projects/status/github/BurntSushi/ripgrep?svg=true)](https://ci.appveyor.com/project/BurntSushi/ripgrep)
[![](https://img.shields.io/crates/v/ignore.svg)](https://crates.io/crates/ignore)
Dual-licensed under MIT or the [UNLICENSE](http://unlicense.org).
### Documentation
[https://docs.rs/ignore](https://docs.rs/ignore)
### Usage
Add this to your `Cargo.toml`:
```toml
[dependencies]
ignore = "0.4"
```
and this to your crate root:
```rust
extern crate ignore;
```
### Example
This example shows the most basic usage of this crate. This code will
recursively traverse the current directory while automatically filtering out
files and directories according to ignore globs found in files like
`.ignore` and `.gitignore`:
```rust,no_run
use ignore::Walk;
for result in Walk::new("./") {
// Each item yielded by the iterator is either a directory entry or an
// error, so either print the path or the error.
match result {
Ok(entry) => println!("{}", entry.path().display()),
Err(err) => println!("ERROR: {}", err),
}
}
```
### Example: advanced
By default, the recursive directory iterator will ignore hidden files and
directories. This can be disabled by building the iterator with `WalkBuilder`:
```rust,no_run
use ignore::WalkBuilder;
for result in WalkBuilder::new("./").hidden(false).build() {
println!("{:?}", result);
}
```
See the documentation for `WalkBuilder` for many other options.

24
crates/ignore/UNLICENSE Normal file
View File

@@ -0,0 +1,24 @@
This is free and unencumbered software released into the public domain.
Anyone is free to copy, modify, publish, use, compile, sell, or
distribute this software, either in source code form or as a compiled
binary, for any purpose, commercial or non-commercial, and by any
means.
In jurisdictions that recognize copyright laws, the author or authors
of this software dedicate any and all copyright interest in the
software to the public domain. We make this dedication for the benefit
of the public at large and to the detriment of our heirs and
successors. We intend this dedication to be an overt act of
relinquishment in perpetuity of all present and future rights to this
software under copyright law.
THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.
IN NO EVENT SHALL THE AUTHORS BE LIABLE FOR ANY CLAIM, DAMAGES OR
OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
OTHER DEALINGS IN THE SOFTWARE.
For more information, please refer to <http://unlicense.org/>

84
crates/ignore/examples/walk.rs Normal file
View File

@@ -0,0 +1,84 @@
extern crate crossbeam_channel as channel;
extern crate ignore;
extern crate walkdir;
use std::env;
use std::io::{self, Write};
use std::path::Path;
use std::thread;
use ignore::WalkBuilder;
use walkdir::WalkDir;
fn main() {
let mut path = env::args().nth(1).unwrap();
let mut parallel = false;
let mut simple = false;
let (tx, rx) = channel::bounded::<DirEntry>(100);
if path == "parallel" {
path = env::args().nth(2).unwrap();
parallel = true;
} else if path == "walkdir" {
path = env::args().nth(2).unwrap();
simple = true;
}
let stdout_thread = thread::spawn(move || {
let mut stdout = io::BufWriter::new(io::stdout());
for dent in rx {
write_path(&mut stdout, dent.path());
}
});
if parallel {
let walker = WalkBuilder::new(path).threads(6).build_parallel();
walker.run(|| {
let tx = tx.clone();
Box::new(move |result| {
use ignore::WalkState::*;
tx.send(DirEntry::Y(result.unwrap())).unwrap();
Continue
})
});
} else if simple {
let walker = WalkDir::new(path);
for result in walker {
tx.send(DirEntry::X(result.unwrap())).unwrap();
}
} else {
let walker = WalkBuilder::new(path).build();
for result in walker {
tx.send(DirEntry::Y(result.unwrap())).unwrap();
}
}
drop(tx);
stdout_thread.join().unwrap();
}
enum DirEntry {
X(walkdir::DirEntry),
Y(ignore::DirEntry),
}
impl DirEntry {
fn path(&self) -> &Path {
match *self {
DirEntry::X(ref x) => x.path(),
DirEntry::Y(ref y) => y.path(),
}
}
}
#[cfg(unix)]
fn write_path<W: Write>(mut wtr: W, path: &Path) {
use std::os::unix::ffi::OsStrExt;
wtr.write(path.as_os_str().as_bytes()).unwrap();
wtr.write(b"\n").unwrap();
}
#[cfg(not(unix))]
fn write_path<W: Write>(mut wtr: W, path: &Path) {
wtr.write(path.to_string_lossy().as_bytes()).unwrap();
wtr.write(b"\n").unwrap();
}

244
crates/ignore/src/default_types.rs Normal file
View File

@@ -0,0 +1,244 @@
/// This list represents the default file types that ripgrep ships with. In
/// general, any file format is fair game, although it should generally be
/// limited to reasonably popular open formats. For other cases, you can add
/// types to each invocation of ripgrep with the '--type-add' flag.
///
/// If you would like to add or improve this list, please file a PR:
/// https://github.com/BurntSushi/ripgrep
///
/// Please try to keep this list sorted lexicographically and wrapped to 79
/// columns (inclusive).
#[rustfmt::skip]
pub const DEFAULT_TYPES: &[(&str, &[&str])] = &[
("agda", &["*.agda", "*.lagda"]),
("aidl", &["*.aidl"]),
("amake", &["*.mk", "*.bp"]),
("asciidoc", &["*.adoc", "*.asc", "*.asciidoc"]),
("asm", &["*.asm", "*.s", "*.S"]),
("asp", &[
"*.aspx", "*.aspx.cs", "*.aspx.cs", "*.ascx", "*.ascx.cs", "*.ascx.vb",
]),
("ats", &["*.ats", "*.dats", "*.sats", "*.hats"]),
("avro", &["*.avdl", "*.avpr", "*.avsc"]),
("awk", &["*.awk"]),
("bazel", &["*.bzl", "WORKSPACE", "BUILD", "BUILD.bazel"]),
("bitbake", &["*.bb", "*.bbappend", "*.bbclass", "*.conf", "*.inc"]),
("brotli", &["*.br"]),
("buildstream", &["*.bst"]),
("bzip2", &["*.bz2", "*.tbz2"]),
("c", &["*.[chH]", "*.[chH].in", "*.cats"]),
("cabal", &["*.cabal"]),
("cbor", &["*.cbor"]),
("ceylon", &["*.ceylon"]),
("clojure", &["*.clj", "*.cljc", "*.cljs", "*.cljx"]),
("cmake", &["*.cmake", "CMakeLists.txt"]),
("coffeescript", &["*.coffee"]),
("config", &["*.cfg", "*.conf", "*.config", "*.ini"]),
("cpp", &[
"*.[ChH]", "*.cc", "*.[ch]pp", "*.[ch]xx", "*.hh", "*.inl",
"*.[ChH].in", "*.cc.in", "*.[ch]pp.in", "*.[ch]xx.in", "*.hh.in",
]),
("creole", &["*.creole"]),
("crystal", &["Projectfile", "*.cr"]),
("cs", &["*.cs"]),
("csharp", &["*.cs"]),
("cshtml", &["*.cshtml"]),
("css", &["*.css", "*.scss"]),
("csv", &["*.csv"]),
("cython", &["*.pyx", "*.pxi", "*.pxd"]),
("d", &["*.d"]),
("dart", &["*.dart"]),
("dhall", &["*.dhall"]),
("diff", &["*.patch", "*.diff"]),
("docker", &["*Dockerfile*"]),
("edn", &["*.edn"]),
("elisp", &["*.el"]),
("elixir", &["*.ex", "*.eex", "*.exs"]),
("elm", &["*.elm"]),
("erb", &["*.erb"]),
("erlang", &["*.erl", "*.hrl"]),
("fidl", &["*.fidl"]),
("fish", &["*.fish"]),
("fortran", &[
"*.f", "*.F", "*.f77", "*.F77", "*.pfo",
"*.f90", "*.F90", "*.f95", "*.F95",
]),
("fsharp", &["*.fs", "*.fsx", "*.fsi"]),
("gap", &["*.g", "*.gap", "*.gi", "*.gd", "*.tst"]),
("gn", &["*.gn", "*.gni"]),
("go", &["*.go"]),
("gradle", &["*.gradle"]),
("groovy", &["*.groovy", "*.gradle"]),
("gzip", &["*.gz", "*.tgz"]),
("h", &["*.h", "*.hpp"]),
("haml", &["*.haml"]),
("haskell", &["*.hs", "*.lhs", "*.cpphs", "*.c2hs", "*.hsc"]),
("hbs", &["*.hbs"]),
("hs", &["*.hs", "*.lhs"]),
("html", &["*.htm", "*.html", "*.ejs"]),
("idris", &["*.idr", "*.lidr"]),
("java", &["*.java", "*.jsp", "*.jspx", "*.properties"]),
("jinja", &["*.j2", "*.jinja", "*.jinja2"]),
("jl", &["*.jl"]),
("js", &["*.js", "*.jsx", "*.vue"]),
("json", &["*.json", "composer.lock"]),
("jsonl", &["*.jsonl"]),
("julia", &["*.jl"]),
("jupyter", &["*.ipynb", "*.jpynb"]),
("kotlin", &["*.kt", "*.kts"]),
("less", &["*.less"]),
("license", &[
// General
"COPYING", "COPYING[.-]*",
"COPYRIGHT", "COPYRIGHT[.-]*",
"EULA", "EULA[.-]*",
"licen[cs]e", "licen[cs]e.*",
"LICEN[CS]E", "LICEN[CS]E[.-]*", "*[.-]LICEN[CS]E*",
"NOTICE", "NOTICE[.-]*",
"PATENTS", "PATENTS[.-]*",
"UNLICEN[CS]E", "UNLICEN[CS]E[.-]*",
// GPL (gpl.txt, etc.)
"agpl[.-]*",
"gpl[.-]*",
"lgpl[.-]*",
// Other license-specific (APACHE-2.0.txt, etc.)
"AGPL-*[0-9]*",
"APACHE-*[0-9]*",
"BSD-*[0-9]*",
"CC-BY-*",
"GFDL-*[0-9]*",
"GNU-*[0-9]*",
"GPL-*[0-9]*",
"LGPL-*[0-9]*",
"MIT-*[0-9]*",
"MPL-*[0-9]*",
"OFL-*[0-9]*",
]),
("lisp", &["*.el", "*.jl", "*.lisp", "*.lsp", "*.sc", "*.scm"]),
("lock", &["*.lock", "package-lock.json"]),
("log", &["*.log"]),
("lua", &["*.lua"]),
("lz4", &["*.lz4"]),
("lzma", &["*.lzma"]),
("m4", &["*.ac", "*.m4"]),
("make", &[
"[Gg][Nn][Uu]makefile", "[Mm]akefile",
"[Gg][Nn][Uu]makefile.am", "[Mm]akefile.am",
"[Gg][Nn][Uu]makefile.in", "[Mm]akefile.in",
"*.mk", "*.mak"
]),
("mako", &["*.mako", "*.mao"]),
("man", &["*.[0-9lnpx]", "*.[0-9][cEFMmpSx]"]),
("markdown", &["*.markdown", "*.md", "*.mdown", "*.mkdn"]),
("matlab", &["*.m"]),
("md", &["*.markdown", "*.md", "*.mdown", "*.mkdn"]),
("mk", &["mkfile"]),
("ml", &["*.ml"]),
("msbuild", &[
"*.csproj", "*.fsproj", "*.vcxproj", "*.proj", "*.props", "*.targets",
]),
("nim", &["*.nim", "*.nimf", "*.nimble", "*.nims"]),
("nix", &["*.nix"]),
("objc", &["*.h", "*.m"]),
("objcpp", &["*.h", "*.mm"]),
("ocaml", &["*.ml", "*.mli", "*.mll", "*.mly"]),
("org", &["*.org", "*.org_archive"]),
("pascal", &["*.pas", "*.dpr", "*.lpr", "*.pp", "*.inc"]),
("pdf", &["*.pdf"]),
("perl", &["*.perl", "*.pl", "*.PL", "*.plh", "*.plx", "*.pm", "*.t"]),
("php", &["*.php", "*.php3", "*.php4", "*.php5", "*.phtml"]),
("pod", &["*.pod"]),
("postscript", &["*.eps", "*.ps"]),
("protobuf", &["*.proto"]),
("ps", &["*.cdxml", "*.ps1", "*.ps1xml", "*.psd1", "*.psm1"]),
("puppet", &["*.erb", "*.pp", "*.rb"]),
("purs", &["*.purs"]),
("py", &["*.py"]),
("qmake", &["*.pro", "*.pri", "*.prf"]),
("qml", &["*.qml"]),
("r", &["*.R", "*.r", "*.Rmd", "*.Rnw"]),
("rdoc", &["*.rdoc"]),
("readme", &["README*", "*README"]),
("robot", &["*.robot"]),
("rst", &["*.rst"]),
("ruby", &["Gemfile", "*.gemspec", ".irbrc", "Rakefile", "*.rb"]),
("rust", &["*.rs"]),
("sass", &["*.sass", "*.scss"]),
("scala", &["*.scala", "*.sbt"]),
("sh", &[
// Portable/misc. init files
".login", ".logout", ".profile", "profile",
// bash-specific init files
".bash_login", "bash_login",
".bash_logout", "bash_logout",
".bash_profile", "bash_profile",
".bashrc", "bashrc", "*.bashrc",
// csh-specific init files
".cshrc", "*.cshrc",
// ksh-specific init files
".kshrc", "*.kshrc",
// tcsh-specific init files
".tcshrc",
// zsh-specific init files
".zshenv", "zshenv",
".zlogin", "zlogin",
".zlogout", "zlogout",
".zprofile", "zprofile",
".zshrc", "zshrc",
// Extensions
"*.bash", "*.csh", "*.ksh", "*.sh", "*.tcsh", "*.zsh",
]),
("slim", &["*.skim", "*.slim", "*.slime"]),
("smarty", &["*.tpl"]),
("sml", &["*.sml", "*.sig"]),
("soy", &["*.soy"]),
("spark", &["*.spark"]),
("spec", &["*.spec"]),
("sql", &["*.sql", "*.psql"]),
("stylus", &["*.styl"]),
("sv", &["*.v", "*.vg", "*.sv", "*.svh", "*.h"]),
("svg", &["*.svg"]),
("swift", &["*.swift"]),
("swig", &["*.def", "*.i"]),
("systemd", &[
"*.automount", "*.conf", "*.device", "*.link", "*.mount", "*.path",
"*.scope", "*.service", "*.slice", "*.socket", "*.swap", "*.target",
"*.timer",
]),
("taskpaper", &["*.taskpaper"]),
("tcl", &["*.tcl"]),
("tex", &["*.tex", "*.ltx", "*.cls", "*.sty", "*.bib", "*.dtx", "*.ins"]),
("textile", &["*.textile"]),
("tf", &["*.tf"]),
("thrift", &["*.thrift"]),
("toml", &["*.toml", "Cargo.lock"]),
("ts", &["*.ts", "*.tsx"]),
("twig", &["*.twig"]),
("txt", &["*.txt"]),
("typoscript", &["*.typoscript", "*.ts"]),
("vala", &["*.vala"]),
("vb", &["*.vb"]),
("verilog", &["*.v", "*.vh", "*.sv", "*.svh"]),
("vhdl", &["*.vhd", "*.vhdl"]),
("vim", &["*.vim"]),
("vimscript", &["*.vim"]),
("webidl", &["*.idl", "*.webidl", "*.widl"]),
("wiki", &["*.mediawiki", "*.wiki"]),
("xml", &[
"*.xml", "*.xml.dist", "*.dtd", "*.xsl", "*.xslt", "*.xsd", "*.xjb",
"*.rng", "*.sch", "*.xhtml",
]),
("xz", &["*.xz", "*.txz"]),
("yacc", &["*.y"]),
("yaml", &["*.yaml", "*.yml"]),
("zig", &["*.zig"]),
("zsh", &[
".zshenv", "zshenv",
".zlogin", "zlogin",
".zlogout", "zlogout",
".zprofile", "zprofile",
".zshrc", "zshrc",
"*.zsh",
]),
("zstd", &["*.zst", "*.zstd"]),
];

1200
crates/ignore/src/dir.rs Normal file
View File

File diff suppressed because it is too large Load Diff

786
crates/ignore/src/gitignore.rs Normal file
View File

@@ -0,0 +1,786 @@
/*!
The gitignore module provides a way to match globs from a gitignore file
against file paths.
Note that this module implements the specification as described in the
`gitignore` man page from scratch. That is, this module does *not* shell out to
the `git` command line tool.
*/
use std::cell::RefCell;
use std::env;
use std::fs::File;
use std::io::{self, BufRead, Read};
use std::path::{Path, PathBuf};
use std::str;
use std::sync::Arc;
use globset::{Candidate, GlobBuilder, GlobSet, GlobSetBuilder};
use regex::bytes::Regex;
use thread_local::ThreadLocal;
use pathutil::{is_file_name, strip_prefix};
use {Error, Match, PartialErrorBuilder};
/// Glob represents a single glob in a gitignore file.
///
/// This is used to report information about the highest precedent glob that
/// matched in one or more gitignore files.
#[derive(Clone, Debug)]
pub struct Glob {
/// The file path that this glob was extracted from.
from: Option<PathBuf>,
/// The original glob string.
original: String,
/// The actual glob string used to convert to a regex.
actual: String,
/// Whether this is a whitelisted glob or not.
is_whitelist: bool,
/// Whether this glob should only match directories or not.
is_only_dir: bool,
}
impl Glob {
/// Returns the file path that defined this glob.
pub fn from(&self) -> Option<&Path> {
self.from.as_ref().map(|p| &**p)
}
/// The original glob as it was defined in a gitignore file.
pub fn original(&self) -> &str {
&self.original
}
/// The actual glob that was compiled to respect gitignore
/// semantics.
pub fn actual(&self) -> &str {
&self.actual
}
/// Whether this was a whitelisted glob or not.
pub fn is_whitelist(&self) -> bool {
self.is_whitelist
}
/// Whether this glob must match a directory or not.
pub fn is_only_dir(&self) -> bool {
self.is_only_dir
}
/// Returns true if and only if this glob has a `**/` prefix.
fn has_doublestar_prefix(&self) -> bool {
self.actual.starts_with("**/") || self.actual == "**"
}
}
/// Gitignore is a matcher for the globs in one or more gitignore files
/// in the same directory.
#[derive(Clone, Debug)]
pub struct Gitignore {
set: GlobSet,
root: PathBuf,
globs: Vec<Glob>,
num_ignores: u64,
num_whitelists: u64,
matches: Option<Arc<ThreadLocal<RefCell<Vec<usize>>>>>,
}
impl Gitignore {
/// Creates a new gitignore matcher from the gitignore file path given.
///
/// If it's desirable to include multiple gitignore files in a single
/// matcher, or read gitignore globs from a different source, then
/// use `GitignoreBuilder`.
///
/// This always returns a valid matcher, even if it's empty. In particular,
/// a Gitignore file can be partially valid, e.g., when one glob is invalid
/// but the rest aren't.
///
/// Note that I/O errors are ignored. For more granular control over
/// errors, use `GitignoreBuilder`.
pub fn new<P: AsRef<Path>>(
gitignore_path: P,
) -> (Gitignore, Option<Error>) {
let path = gitignore_path.as_ref();
let parent = path.parent().unwrap_or(Path::new("/"));
let mut builder = GitignoreBuilder::new(parent);
let mut errs = PartialErrorBuilder::default();
errs.maybe_push_ignore_io(builder.add(path));
match builder.build() {
Ok(gi) => (gi, errs.into_error_option()),
Err(err) => {
errs.push(err);
(Gitignore::empty(), errs.into_error_option())
}
}
}
/// Creates a new gitignore matcher from the global ignore file, if one
/// exists.
///
/// The global config file path is specified by git's `core.excludesFile`
/// config option.
///
/// Git's config file location is `$HOME/.gitconfig`. If `$HOME/.gitconfig`
/// does not exist or does not specify `core.excludesFile`, then
/// `$XDG_CONFIG_HOME/git/ignore` is read. If `$XDG_CONFIG_HOME` is not
/// set or is empty, then `$HOME/.config/git/ignore` is used instead.
pub fn global() -> (Gitignore, Option<Error>) {
GitignoreBuilder::new("").build_global()
}
/// Creates a new empty gitignore matcher that never matches anything.
///
/// Its path is empty.
pub fn empty() -> Gitignore {
Gitignore {
set: GlobSet::empty(),
root: PathBuf::from(""),
globs: vec![],
num_ignores: 0,
num_whitelists: 0,
matches: None,
}
}
/// Returns the directory containing this gitignore matcher.
///
/// All matches are done relative to this path.
pub fn path(&self) -> &Path {
&*self.root
}
/// Returns true if and only if this gitignore has zero globs, and
/// therefore never matches any file path.
pub fn is_empty(&self) -> bool {
self.set.is_empty()
}
/// Returns the total number of globs, which should be equivalent to
/// `num_ignores + num_whitelists`.
pub fn len(&self) -> usize {
self.set.len()
}
/// Returns the total number of ignore globs.
pub fn num_ignores(&self) -> u64 {
self.num_ignores
}
/// Returns the total number of whitelisted globs.
pub fn num_whitelists(&self) -> u64 {
self.num_whitelists
}
/// Returns whether the given path (file or directory) matched a pattern in
/// this gitignore matcher.
///
/// `is_dir` should be true if the path refers to a directory and false
/// otherwise.
///
/// The given path is matched relative to the path given when building
/// the matcher. Specifically, before matching `path`, its prefix (as
/// determined by a common suffix of the directory containing this
/// gitignore) is stripped. If there is no common suffix/prefix overlap,
/// then `path` is assumed to be relative to this matcher.
pub fn matched<P: AsRef<Path>>(
&self,
path: P,
is_dir: bool,
) -> Match<&Glob> {
if self.is_empty() {
return Match::None;
}
self.matched_stripped(self.strip(path.as_ref()), is_dir)
}
/// Returns whether the given path (file or directory, and expected to be
/// under the root) or any of its parent directories (up to the root)
/// matched a pattern in this gitignore matcher.
///
/// NOTE: This method is more expensive than walking the directory hierarchy
/// top-to-bottom and matching the entries. But, is easier to use in cases
/// when a list of paths are available without a hierarchy.
///
/// `is_dir` should be true if the path refers to a directory and false
/// otherwise.
///
/// The given path is matched relative to the path given when building
/// the matcher. Specifically, before matching `path`, its prefix (as
/// determined by a common suffix of the directory containing this
/// gitignore) is stripped. If there is no common suffix/prefix overlap,
/// then `path` is assumed to be relative to this matcher.
///
/// # Panics
///
/// This method panics if the given file path is not under the root path
/// of this matcher.
pub fn matched_path_or_any_parents<P: AsRef<Path>>(
&self,
path: P,
is_dir: bool,
) -> Match<&Glob> {
if self.is_empty() {
return Match::None;
}
let mut path = self.strip(path.as_ref());
assert!(!path.has_root(), "path is expected to be under the root");
match self.matched_stripped(path, is_dir) {
Match::None => (), // walk up
a_match => return a_match,
}
while let Some(parent) = path.parent() {
match self.matched_stripped(parent, /* is_dir */ true) {
Match::None => path = parent, // walk up
a_match => return a_match,
}
}
Match::None
}
/// Like matched, but takes a path that has already been stripped.
fn matched_stripped<P: AsRef<Path>>(
&self,
path: P,
is_dir: bool,
) -> Match<&Glob> {
if self.is_empty() {
return Match::None;
}
let path = path.as_ref();
let _matches = self.matches.as_ref().unwrap().get_or_default();
let mut matches = _matches.borrow_mut();
let candidate = Candidate::new(path);
self.set.matches_candidate_into(&candidate, &mut *matches);
for &i in matches.iter().rev() {
let glob = &self.globs[i];
if !glob.is_only_dir() || is_dir {
return if glob.is_whitelist() {
Match::Whitelist(glob)
} else {
Match::Ignore(glob)
};
}
}
Match::None
}
/// Strips the given path such that it's suitable for matching with this
/// gitignore matcher.
fn strip<'a, P: 'a + AsRef<Path> + ?Sized>(
&'a self,
path: &'a P,
) -> &'a Path {
let mut path = path.as_ref();
// A leading ./ is completely superfluous. We also strip it from
// our gitignore root path, so we need to strip it from our candidate
// path too.
if let Some(p) = strip_prefix("./", path) {
path = p;
}
// Strip any common prefix between the candidate path and the root
// of the gitignore, to make sure we get relative matching right.
// BUT, a file name might not have any directory components to it,
// in which case, we don't want to accidentally strip any part of the
// file name.
//
// As an additional special case, if the root is just `.`, then we
// shouldn't try to strip anything, e.g., when path begins with a `.`.
if self.root != Path::new(".") && !is_file_name(path) {
if let Some(p) = strip_prefix(&self.root, path) {
path = p;
// If we're left with a leading slash, get rid of it.
if let Some(p) = strip_prefix("/", path) {
path = p;
}
}
}
path
}
}
/// Builds a matcher for a single set of globs from a .gitignore file.
#[derive(Clone, Debug)]
pub struct GitignoreBuilder {
builder: GlobSetBuilder,
root: PathBuf,
globs: Vec<Glob>,
case_insensitive: bool,
}
impl GitignoreBuilder {
/// Create a new builder for a gitignore file.
///
/// The path given should be the path at which the globs for this gitignore
/// file should be matched. Note that paths are always matched relative
/// to the root path given here. Generally, the root path should correspond
/// to the *directory* containing a `.gitignore` file.
pub fn new<P: AsRef<Path>>(root: P) -> GitignoreBuilder {
let root = root.as_ref();
GitignoreBuilder {
builder: GlobSetBuilder::new(),
root: strip_prefix("./", root).unwrap_or(root).to_path_buf(),
globs: vec![],
case_insensitive: false,
}
}
/// Builds a new matcher from the globs added so far.
///
/// Once a matcher is built, no new globs can be added to it.
pub fn build(&self) -> Result<Gitignore, Error> {
let nignore = self.globs.iter().filter(|g| !g.is_whitelist()).count();
let nwhite = self.globs.iter().filter(|g| g.is_whitelist()).count();
let set = self
.builder
.build()
.map_err(|err| Error::Glob { glob: None, err: err.to_string() })?;
Ok(Gitignore {
set: set,
root: self.root.clone(),
globs: self.globs.clone(),
num_ignores: nignore as u64,
num_whitelists: nwhite as u64,
matches: Some(Arc::new(ThreadLocal::default())),
})
}
/// Build a global gitignore matcher using the configuration in this
/// builder.
///
/// This consumes ownership of the builder unlike `build` because it
/// must mutate the builder to add the global gitignore globs.
///
/// Note that this ignores the path given to this builder's constructor
/// and instead derives the path automatically from git's global
/// configuration.
pub fn build_global(mut self) -> (Gitignore, Option<Error>) {
match gitconfig_excludes_path() {
None => (Gitignore::empty(), None),
Some(path) => {
if !path.is_file() {
(Gitignore::empty(), None)
} else {
let mut errs = PartialErrorBuilder::default();
errs.maybe_push_ignore_io(self.add(path));
match self.build() {
Ok(gi) => (gi, errs.into_error_option()),
Err(err) => {
errs.push(err);
(Gitignore::empty(), errs.into_error_option())
}
}
}
}
}
}
/// Add each glob from the file path given.
///
/// The file given should be formatted as a `gitignore` file.
///
/// Note that partial errors can be returned. For example, if there was
/// a problem adding one glob, an error for that will be returned, but
/// all other valid globs will still be added.
pub fn add<P: AsRef<Path>>(&mut self, path: P) -> Option<Error> {
let path = path.as_ref();
let file = match File::open(path) {
Err(err) => return Some(Error::Io(err).with_path(path)),
Ok(file) => file,
};
let rdr = io::BufReader::new(file);
let mut errs = PartialErrorBuilder::default();
for (i, line) in rdr.lines().enumerate() {
let lineno = (i + 1) as u64;
let line = match line {
Ok(line) => line,
Err(err) => {
errs.push(Error::Io(err).tagged(path, lineno));
break;
}
};
if let Err(err) = self.add_line(Some(path.to_path_buf()), &line) {
errs.push(err.tagged(path, lineno));
}
}
errs.into_error_option()
}
/// Add each glob line from the string given.
///
/// If this string came from a particular `gitignore` file, then its path
/// should be provided here.
///
/// The string given should be formatted as a `gitignore` file.
#[cfg(test)]
fn add_str(
&mut self,
from: Option<PathBuf>,
gitignore: &str,
) -> Result<&mut GitignoreBuilder, Error> {
for line in gitignore.lines() {
self.add_line(from.clone(), line)?;
}
Ok(self)
}
/// Add a line from a gitignore file to this builder.
///
/// If this line came from a particular `gitignore` file, then its path
/// should be provided here.
///
/// If the line could not be parsed as a glob, then an error is returned.
pub fn add_line(
&mut self,
from: Option<PathBuf>,
mut line: &str,
) -> Result<&mut GitignoreBuilder, Error> {
#![allow(deprecated)]
if line.starts_with("#") {
return Ok(self);
}
if !line.ends_with("\\ ") {
line = line.trim_right();
}
if line.is_empty() {
return Ok(self);
}
let mut glob = Glob {
from: from,
original: line.to_string(),
actual: String::new(),
is_whitelist: false,
is_only_dir: false,
};
let mut is_absolute = false;
if line.starts_with("\\!") || line.starts_with("\\#") {
line = &line[1..];
is_absolute = line.chars().nth(0) == Some('/');
} else {
if line.starts_with("!") {
glob.is_whitelist = true;
line = &line[1..];
}
if line.starts_with("/") {
// `man gitignore` says that if a glob starts with a slash,
// then the glob can only match the beginning of a path
// (relative to the location of gitignore). We achieve this by
// simply banning wildcards from matching /.
line = &line[1..];
is_absolute = true;
}
}
// If it ends with a slash, then this should only match directories,
// but the slash should otherwise not be used while globbing.
if let Some((i, c)) = line.char_indices().rev().nth(0) {
if c == '/' {
glob.is_only_dir = true;
line = &line[..i];
}
}
glob.actual = line.to_string();
// If there is a literal slash, then this is a glob that must match the
// entire path name. Otherwise, we should let it match anywhere, so use
// a **/ prefix.
if !is_absolute && !line.chars().any(|c| c == '/') {
// ... but only if we don't already have a **/ prefix.
if !glob.has_doublestar_prefix() {
glob.actual = format!("**/{}", glob.actual);
}
}
// If the glob ends with `/**`, then we should only match everything
// inside a directory, but not the directory itself. Standard globs
// will match the directory. So we add `/*` to force the issue.
if glob.actual.ends_with("/**") {
glob.actual = format!("{}/*", glob.actual);
}
let parsed = GlobBuilder::new(&glob.actual)
.literal_separator(true)
.case_insensitive(self.case_insensitive)
.backslash_escape(true)
.build()
.map_err(|err| Error::Glob {
glob: Some(glob.original.clone()),
err: err.kind().to_string(),
})?;
self.builder.add(parsed);
self.globs.push(glob);
Ok(self)
}
/// Toggle whether the globs should be matched case insensitively or not.
///
/// When this option is changed, only globs added after the change will be
/// affected.
///
/// This is disabled by default.
pub fn case_insensitive(
&mut self,
yes: bool,
) -> Result<&mut GitignoreBuilder, Error> {
// TODO: This should not return a `Result`. Fix this in the next semver
// release.
self.case_insensitive = yes;
Ok(self)
}
}
/// Return the file path of the current environment's global gitignore file.
///
/// Note that the file path returned may not exist.
fn gitconfig_excludes_path() -> Option<PathBuf> {
// git supports $HOME/.gitconfig and $XDG_CONFIG_HOME/git/config. Notably,
// both can be active at the same time, where $HOME/.gitconfig takes
// precedent. So if $HOME/.gitconfig defines a `core.excludesFile`, then
// we're done.
match gitconfig_home_contents().and_then(|x| parse_excludes_file(&x)) {
Some(path) => return Some(path),
None => {}
}
match gitconfig_xdg_contents().and_then(|x| parse_excludes_file(&x)) {
Some(path) => return Some(path),
None => {}
}
excludes_file_default()
}
/// Returns the file contents of git's global config file, if one exists, in
/// the user's home directory.
fn gitconfig_home_contents() -> Option<Vec<u8>> {
let home = match home_dir() {
None => return None,
Some(home) => home,
};
let mut file = match File::open(home.join(".gitconfig")) {
Err(_) => return None,
Ok(file) => io::BufReader::new(file),
};
let mut contents = vec![];
file.read_to_end(&mut contents).ok().map(|_| contents)
}
/// Returns the file contents of git's global config file, if one exists, in
/// the user's XDG_CONFIG_HOME directory.
fn gitconfig_xdg_contents() -> Option<Vec<u8>> {
let path = env::var_os("XDG_CONFIG_HOME")
.and_then(|x| if x.is_empty() { None } else { Some(PathBuf::from(x)) })
.or_else(|| home_dir().map(|p| p.join(".config")))
.map(|x| x.join("git/config"));
let mut file = match path.and_then(|p| File::open(p).ok()) {
None => return None,
Some(file) => io::BufReader::new(file),
};
let mut contents = vec![];
file.read_to_end(&mut contents).ok().map(|_| contents)
}
/// Returns the default file path for a global .gitignore file.
///
/// Specifically, this respects XDG_CONFIG_HOME.
fn excludes_file_default() -> Option<PathBuf> {
env::var_os("XDG_CONFIG_HOME")
.and_then(|x| if x.is_empty() { None } else { Some(PathBuf::from(x)) })
.or_else(|| home_dir().map(|p| p.join(".config")))
.map(|x| x.join("git/ignore"))
}
/// Extract git's `core.excludesfile` config setting from the raw file contents
/// given.
fn parse_excludes_file(data: &[u8]) -> Option<PathBuf> {
// N.B. This is the lazy approach, and isn't technically correct, but
// probably works in more circumstances. I guess we would ideally have
// a full INI parser. Yuck.
lazy_static! {
static ref RE: Regex =
Regex::new(r"(?im)^\s*excludesfile\s*=\s*(.+)\s*$").unwrap();
};
let caps = match RE.captures(data) {
None => return None,
Some(caps) => caps,
};
str::from_utf8(&caps[1]).ok().map(|s| PathBuf::from(expand_tilde(s)))
}
/// Expands ~ in file paths to the value of $HOME.
fn expand_tilde(path: &str) -> String {
let home = match home_dir() {
None => return path.to_string(),
Some(home) => home.to_string_lossy().into_owned(),
};
path.replace("~", &home)
}
/// Returns the location of the user's home directory.
fn home_dir() -> Option<PathBuf> {
// We're fine with using env::home_dir for now. Its bugs are, IMO, pretty
// minor corner cases. We should still probably eventually migrate to
// the `dirs` crate to get a proper implementation.
#![allow(deprecated)]
env::home_dir()
}
#[cfg(test)]
mod tests {
use super::{Gitignore, GitignoreBuilder};
use std::path::Path;
fn gi_from_str<P: AsRef<Path>>(root: P, s: &str) -> Gitignore {
let mut builder = GitignoreBuilder::new(root);
builder.add_str(None, s).unwrap();
builder.build().unwrap()
}
macro_rules! ignored {
($name:ident, $root:expr, $gi:expr, $path:expr) => {
ignored!($name, $root, $gi, $path, false);
};
($name:ident, $root:expr, $gi:expr, $path:expr, $is_dir:expr) => {
#[test]
fn $name() {
let gi = gi_from_str($root, $gi);
assert!(gi.matched($path, $is_dir).is_ignore());
}
};
}
macro_rules! not_ignored {
($name:ident, $root:expr, $gi:expr, $path:expr) => {
not_ignored!($name, $root, $gi, $path, false);
};
($name:ident, $root:expr, $gi:expr, $path:expr, $is_dir:expr) => {
#[test]
fn $name() {
let gi = gi_from_str($root, $gi);
assert!(!gi.matched($path, $is_dir).is_ignore());
}
};
}
const ROOT: &'static str = "/home/foobar/rust/rg";
ignored!(ig1, ROOT, "months", "months");
ignored!(ig2, ROOT, "*.lock", "Cargo.lock");
ignored!(ig3, ROOT, "*.rs", "src/main.rs");
ignored!(ig4, ROOT, "src/*.rs", "src/main.rs");
ignored!(ig5, ROOT, "/*.c", "cat-file.c");
ignored!(ig6, ROOT, "/src/*.rs", "src/main.rs");
ignored!(ig7, ROOT, "!src/main.rs\n*.rs", "src/main.rs");
ignored!(ig8, ROOT, "foo/", "foo", true);
ignored!(ig9, ROOT, "**/foo", "foo");
ignored!(ig10, ROOT, "**/foo", "src/foo");
ignored!(ig11, ROOT, "**/foo/**", "src/foo/bar");
ignored!(ig12, ROOT, "**/foo/**", "wat/src/foo/bar/baz");
ignored!(ig13, ROOT, "**/foo/bar", "foo/bar");
ignored!(ig14, ROOT, "**/foo/bar", "src/foo/bar");
ignored!(ig15, ROOT, "abc/**", "abc/x");
ignored!(ig16, ROOT, "abc/**", "abc/x/y");
ignored!(ig17, ROOT, "abc/**", "abc/x/y/z");
ignored!(ig18, ROOT, "a/**/b", "a/b");
ignored!(ig19, ROOT, "a/**/b", "a/x/b");
ignored!(ig20, ROOT, "a/**/b", "a/x/y/b");
ignored!(ig21, ROOT, r"\!xy", "!xy");
ignored!(ig22, ROOT, r"\#foo", "#foo");
ignored!(ig23, ROOT, "foo", "./foo");
ignored!(ig24, ROOT, "target", "grep/target");
ignored!(ig25, ROOT, "Cargo.lock", "./tabwriter-bin/Cargo.lock");
ignored!(ig26, ROOT, "/foo/bar/baz", "./foo/bar/baz");
ignored!(ig27, ROOT, "foo/", "xyz/foo", true);
ignored!(ig28, "./src", "/llvm/", "./src/llvm", true);
ignored!(ig29, ROOT, "node_modules/ ", "node_modules", true);
ignored!(ig30, ROOT, "**/", "foo/bar", true);
ignored!(ig31, ROOT, "path1/*", "path1/foo");
ignored!(ig32, ROOT, ".a/b", ".a/b");
ignored!(ig33, "./", ".a/b", ".a/b");
ignored!(ig34, ".", ".a/b", ".a/b");
ignored!(ig35, "./.", ".a/b", ".a/b");
ignored!(ig36, "././", ".a/b", ".a/b");
ignored!(ig37, "././.", ".a/b", ".a/b");
ignored!(ig38, ROOT, "\\[", "[");
ignored!(ig39, ROOT, "\\?", "?");
ignored!(ig40, ROOT, "\\*", "*");
ignored!(ig41, ROOT, "\\a", "a");
ignored!(ig42, ROOT, "s*.rs", "sfoo.rs");
ignored!(ig43, ROOT, "**", "foo.rs");
ignored!(ig44, ROOT, "**/**/*", "a/foo.rs");
not_ignored!(ignot1, ROOT, "amonths", "months");
not_ignored!(ignot2, ROOT, "monthsa", "months");
not_ignored!(ignot3, ROOT, "/src/*.rs", "src/grep/src/main.rs");
not_ignored!(ignot4, ROOT, "/*.c", "mozilla-sha1/sha1.c");
not_ignored!(ignot5, ROOT, "/src/*.rs", "src/grep/src/main.rs");
not_ignored!(ignot6, ROOT, "*.rs\n!src/main.rs", "src/main.rs");
not_ignored!(ignot7, ROOT, "foo/", "foo", false);
not_ignored!(ignot8, ROOT, "**/foo/**", "wat/src/afoo/bar/baz");
not_ignored!(ignot9, ROOT, "**/foo/**", "wat/src/fooa/bar/baz");
not_ignored!(ignot10, ROOT, "**/foo/bar", "foo/src/bar");
not_ignored!(ignot11, ROOT, "#foo", "#foo");
not_ignored!(ignot12, ROOT, "\n\n\n", "foo");
not_ignored!(ignot13, ROOT, "foo/**", "foo", true);
not_ignored!(
ignot14,
"./third_party/protobuf",
"m4/ltoptions.m4",
"./third_party/protobuf/csharp/src/packages/repositories.config"
);
not_ignored!(ignot15, ROOT, "!/bar", "foo/bar");
not_ignored!(ignot16, ROOT, "*\n!**/", "foo", true);
not_ignored!(ignot17, ROOT, "src/*.rs", "src/grep/src/main.rs");
not_ignored!(ignot18, ROOT, "path1/*", "path2/path1/foo");
not_ignored!(ignot19, ROOT, "s*.rs", "src/foo.rs");
fn bytes(s: &str) -> Vec<u8> {
s.to_string().into_bytes()
}
fn path_string<P: AsRef<Path>>(path: P) -> String {
path.as_ref().to_str().unwrap().to_string()
}
#[test]
fn parse_excludes_file1() {
let data = bytes("[core]\nexcludesFile = /foo/bar");
let got = super::parse_excludes_file(&data).unwrap();
assert_eq!(path_string(got), "/foo/bar");
}
#[test]
fn parse_excludes_file2() {
let data = bytes("[core]\nexcludesFile = ~/foo/bar");
let got = super::parse_excludes_file(&data).unwrap();
assert_eq!(path_string(got), super::expand_tilde("~/foo/bar"));
}
#[test]
fn parse_excludes_file3() {
let data = bytes("[core]\nexcludeFile = /foo/bar");
assert!(super::parse_excludes_file(&data).is_none());
}
// See: https://github.com/BurntSushi/ripgrep/issues/106
#[test]
fn regression_106() {
gi_from_str("/", " ");
}
#[test]
fn case_insensitive() {
let gi = GitignoreBuilder::new(ROOT)
.case_insensitive(true)
.unwrap()
.add_str(None, "*.html")
.unwrap()
.build()
.unwrap();
assert!(gi.matched("foo.html", false).is_ignore());
assert!(gi.matched("foo.HTML", false).is_ignore());
assert!(!gi.matched("foo.htm", false).is_ignore());
assert!(!gi.matched("foo.HTM", false).is_ignore());
}
ignored!(cs1, ROOT, "*.html", "foo.html");
not_ignored!(cs2, ROOT, "*.html", "foo.HTML");
not_ignored!(cs3, ROOT, "*.html", "foo.htm");
not_ignored!(cs4, ROOT, "*.html", "foo.HTM");
}

499
crates/ignore/src/lib.rs Normal file
View File

@@ -0,0 +1,499 @@
/*!
The ignore crate provides a fast recursive directory iterator that respects
various filters such as globs, file types and `.gitignore` files. The precise
matching rules and precedence is explained in the documentation for
`WalkBuilder`.
Secondarily, this crate exposes gitignore and file type matchers for use cases
that demand more fine-grained control.
# Example
This example shows the most basic usage of this crate. This code will
recursively traverse the current directory while automatically filtering out
files and directories according to ignore globs found in files like
`.ignore` and `.gitignore`:
```rust,no_run
use ignore::Walk;
for result in Walk::new("./") {
// Each item yielded by the iterator is either a directory entry or an
// error, so either print the path or the error.
match result {
Ok(entry) => println!("{}", entry.path().display()),
Err(err) => println!("ERROR: {}", err),
}
}
```
# Example: advanced
By default, the recursive directory iterator will ignore hidden files and
directories. This can be disabled by building the iterator with `WalkBuilder`:
```rust,no_run
use ignore::WalkBuilder;
for result in WalkBuilder::new("./").hidden(false).build() {
println!("{:?}", result);
}
```
See the documentation for `WalkBuilder` for many other options.
*/
#![deny(missing_docs)]
extern crate crossbeam_channel as channel;
extern crate globset;
#[macro_use]
extern crate lazy_static;
#[macro_use]
extern crate log;
extern crate memchr;
extern crate regex;
extern crate same_file;
extern crate thread_local;
extern crate walkdir;
#[cfg(windows)]
extern crate winapi_util;
use std::error;
use std::fmt;
use std::io;
use std::path::{Path, PathBuf};
pub use walk::{
DirEntry, ParallelVisitor, ParallelVisitorBuilder, Walk, WalkBuilder,
WalkParallel, WalkState,
};
mod default_types;
mod dir;
pub mod gitignore;
pub mod overrides;
mod pathutil;
pub mod types;
mod walk;
/// Represents an error that can occur when parsing a gitignore file.
#[derive(Debug)]
pub enum Error {
/// A collection of "soft" errors. These occur when adding an ignore
/// file partially succeeded.
Partial(Vec<Error>),
/// An error associated with a specific line number.
WithLineNumber {
/// The line number.
line: u64,
/// The underlying error.
err: Box<Error>,
},
/// An error associated with a particular file path.
WithPath {
/// The file path.
path: PathBuf,
/// The underlying error.
err: Box<Error>,
},
/// An error associated with a particular directory depth when recursively
/// walking a directory.
WithDepth {
/// The directory depth.
depth: usize,
/// The underlying error.
err: Box<Error>,
},
/// An error that occurs when a file loop is detected when traversing
/// symbolic links.
Loop {
/// The ancestor file path in the loop.
ancestor: PathBuf,
/// The child file path in the loop.
child: PathBuf,
},
/// An error that occurs when doing I/O, such as reading an ignore file.
Io(io::Error),
/// An error that occurs when trying to parse a glob.
Glob {
/// The original glob that caused this error. This glob, when
/// available, always corresponds to the glob provided by an end user.
/// e.g., It is the glob as written in a `.gitignore` file.
///
/// (This glob may be distinct from the glob that is actually
/// compiled, after accounting for `gitignore` semantics.)
glob: Option<String>,
/// The underlying glob error as a string.
err: String,
},
/// A type selection for a file type that is not defined.
UnrecognizedFileType(String),
/// A user specified file type definition could not be parsed.
InvalidDefinition,
}
impl Clone for Error {
fn clone(&self) -> Error {
match *self {
Error::Partial(ref errs) => Error::Partial(errs.clone()),
Error::WithLineNumber { line, ref err } => {
Error::WithLineNumber { line: line, err: err.clone() }
}
Error::WithPath { ref path, ref err } => {
Error::WithPath { path: path.clone(), err: err.clone() }
}
Error::WithDepth { depth, ref err } => {
Error::WithDepth { depth: depth, err: err.clone() }
}
Error::Loop { ref ancestor, ref child } => Error::Loop {
ancestor: ancestor.clone(),
child: child.clone(),
},
Error::Io(ref err) => match err.raw_os_error() {
Some(e) => Error::Io(io::Error::from_raw_os_error(e)),
None => Error::Io(io::Error::new(err.kind(), err.to_string())),
},
Error::Glob { ref glob, ref err } => {
Error::Glob { glob: glob.clone(), err: err.clone() }
}
Error::UnrecognizedFileType(ref err) => {
Error::UnrecognizedFileType(err.clone())
}
Error::InvalidDefinition => Error::InvalidDefinition,
}
}
}
impl Error {
/// Returns true if this is a partial error.
///
/// A partial error occurs when only some operations failed while others
/// may have succeeded. For example, an ignore file may contain an invalid
/// glob among otherwise valid globs.
pub fn is_partial(&self) -> bool {
match *self {
Error::Partial(_) => true,
Error::WithLineNumber { ref err, .. } => err.is_partial(),
Error::WithPath { ref err, .. } => err.is_partial(),
Error::WithDepth { ref err, .. } => err.is_partial(),
_ => false,
}
}
/// Returns true if this error is exclusively an I/O error.
pub fn is_io(&self) -> bool {
match *self {
Error::Partial(ref errs) => errs.len() == 1 && errs[0].is_io(),
Error::WithLineNumber { ref err, .. } => err.is_io(),
Error::WithPath { ref err, .. } => err.is_io(),
Error::WithDepth { ref err, .. } => err.is_io(),
Error::Loop { .. } => false,
Error::Io(_) => true,
Error::Glob { .. } => false,
Error::UnrecognizedFileType(_) => false,
Error::InvalidDefinition => false,
}
}
/// Returns a depth associated with recursively walking a directory (if
/// this error was generated from a recursive directory iterator).
pub fn depth(&self) -> Option<usize> {
match *self {
Error::WithPath { ref err, .. } => err.depth(),
Error::WithDepth { depth, .. } => Some(depth),
_ => None,
}
}
/// Turn an error into a tagged error with the given file path.
fn with_path<P: AsRef<Path>>(self, path: P) -> Error {
Error::WithPath {
path: path.as_ref().to_path_buf(),
err: Box::new(self),
}
}
/// Turn an error into a tagged error with the given depth.
fn with_depth(self, depth: usize) -> Error {
Error::WithDepth { depth: depth, err: Box::new(self) }
}
/// Turn an error into a tagged error with the given file path and line
/// number. If path is empty, then it is omitted from the error.
fn tagged<P: AsRef<Path>>(self, path: P, lineno: u64) -> Error {
let errline =
Error::WithLineNumber { line: lineno, err: Box::new(self) };
if path.as_ref().as_os_str().is_empty() {
return errline;
}
errline.with_path(path)
}
/// Build an error from a walkdir error.
fn from_walkdir(err: walkdir::Error) -> Error {
let depth = err.depth();
if let (Some(anc), Some(child)) = (err.loop_ancestor(), err.path()) {
return Error::WithDepth {
depth: depth,
err: Box::new(Error::Loop {
ancestor: anc.to_path_buf(),
child: child.to_path_buf(),
}),
};
}
let path = err.path().map(|p| p.to_path_buf());
let mut ig_err = Error::Io(io::Error::from(err));
if let Some(path) = path {
ig_err = Error::WithPath { path: path, err: Box::new(ig_err) };
}
ig_err
}
}
impl error::Error for Error {
#[allow(deprecated)]
fn description(&self) -> &str {
match *self {
Error::Partial(_) => "partial error",
Error::WithLineNumber { ref err, .. } => err.description(),
Error::WithPath { ref err, .. } => err.description(),
Error::WithDepth { ref err, .. } => err.description(),
Error::Loop { .. } => "file system loop found",
Error::Io(ref err) => err.description(),
Error::Glob { ref err, .. } => err,
Error::UnrecognizedFileType(_) => "unrecognized file type",
Error::InvalidDefinition => "invalid definition",
}
}
}
impl fmt::Display for Error {
fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result {
match *self {
Error::Partial(ref errs) => {
let msgs: Vec<String> =
errs.iter().map(|err| err.to_string()).collect();
write!(f, "{}", msgs.join("\n"))
}
Error::WithLineNumber { line, ref err } => {
write!(f, "line {}: {}", line, err)
}
Error::WithPath { ref path, ref err } => {
write!(f, "{}: {}", path.display(), err)
}
Error::WithDepth { ref err, .. } => err.fmt(f),
Error::Loop { ref ancestor, ref child } => write!(
f,
"File system loop found: \
{} points to an ancestor {}",
child.display(),
ancestor.display()
),
Error::Io(ref err) => err.fmt(f),
Error::Glob { glob: None, ref err } => write!(f, "{}", err),
Error::Glob { glob: Some(ref glob), ref err } => {
write!(f, "error parsing glob '{}': {}", glob, err)
}
Error::UnrecognizedFileType(ref ty) => {
write!(f, "unrecognized file type: {}", ty)
}
Error::InvalidDefinition => write!(
f,
"invalid definition (format is type:glob, e.g., \
html:*.html)"
),
}
}
}
impl From<io::Error> for Error {
fn from(err: io::Error) -> Error {
Error::Io(err)
}
}
#[derive(Debug, Default)]
struct PartialErrorBuilder(Vec<Error>);
impl PartialErrorBuilder {
fn push(&mut self, err: Error) {
self.0.push(err);
}
fn push_ignore_io(&mut self, err: Error) {
if !err.is_io() {
self.push(err);
}
}
fn maybe_push(&mut self, err: Option<Error>) {
if let Some(err) = err {
self.push(err);
}
}
fn maybe_push_ignore_io(&mut self, err: Option<Error>) {
if let Some(err) = err {
self.push_ignore_io(err);
}
}
fn into_error_option(mut self) -> Option<Error> {
if self.0.is_empty() {
None
} else if self.0.len() == 1 {
Some(self.0.pop().unwrap())
} else {
Some(Error::Partial(self.0))
}
}
}
/// The result of a glob match.
///
/// The type parameter `T` typically refers to a type that provides more
/// information about a particular match. For example, it might identify
/// the specific gitignore file and the specific glob pattern that caused
/// the match.
#[derive(Clone, Debug)]
pub enum Match<T> {
/// The path didn't match any glob.
None,
/// The highest precedent glob matched indicates the path should be
/// ignored.
Ignore(T),
/// The highest precedent glob matched indicates the path should be
/// whitelisted.
Whitelist(T),
}
impl<T> Match<T> {
/// Returns true if the match result didn't match any globs.
pub fn is_none(&self) -> bool {
match *self {
Match::None => true,
Match::Ignore(_) | Match::Whitelist(_) => false,
}
}
/// Returns true if the match result implies the path should be ignored.
pub fn is_ignore(&self) -> bool {
match *self {
Match::Ignore(_) => true,
Match::None | Match::Whitelist(_) => false,
}
}
/// Returns true if the match result implies the path should be
/// whitelisted.
pub fn is_whitelist(&self) -> bool {
match *self {
Match::Whitelist(_) => true,
Match::None | Match::Ignore(_) => false,
}
}
/// Inverts the match so that `Ignore` becomes `Whitelist` and
/// `Whitelist` becomes `Ignore`. A non-match remains the same.
pub fn invert(self) -> Match<T> {
match self {
Match::None => Match::None,
Match::Ignore(t) => Match::Whitelist(t),
Match::Whitelist(t) => Match::Ignore(t),
}
}
/// Return the value inside this match if it exists.
pub fn inner(&self) -> Option<&T> {
match *self {
Match::None => None,
Match::Ignore(ref t) => Some(t),
Match::Whitelist(ref t) => Some(t),
}
}
/// Apply the given function to the value inside this match.
///
/// If the match has no value, then return the match unchanged.
pub fn map<U, F: FnOnce(T) -> U>(self, f: F) -> Match<U> {
match self {
Match::None => Match::None,
Match::Ignore(t) => Match::Ignore(f(t)),
Match::Whitelist(t) => Match::Whitelist(f(t)),
}
}
/// Return the match if it is not none. Otherwise, return other.
pub fn or(self, other: Self) -> Self {
if self.is_none() {
other
} else {
self
}
}
}
#[cfg(test)]
mod tests {
use std::env;
use std::error;
use std::fs;
use std::path::{Path, PathBuf};
use std::result;
/// A convenient result type alias.
pub type Result<T> =
result::Result<T, Box<dyn error::Error + Send + Sync>>;
macro_rules! err {
($($tt:tt)*) => {
Box::<dyn error::Error + Send + Sync>::from(format!($($tt)*))
}
}
/// A simple wrapper for creating a temporary directory that is
/// automatically deleted when it's dropped.
///
/// We use this in lieu of tempfile because tempfile brings in too many
/// dependencies.
#[derive(Debug)]
pub struct TempDir(PathBuf);
impl Drop for TempDir {
fn drop(&mut self) {
fs::remove_dir_all(&self.0).unwrap();
}
}
impl TempDir {
/// Create a new empty temporary directory under the system's configured
/// temporary directory.
pub fn new() -> Result<TempDir> {
use std::sync::atomic::{AtomicUsize, Ordering};
static TRIES: usize = 100;
static COUNTER: AtomicUsize = AtomicUsize::new(0);
let tmpdir = env::temp_dir();
for _ in 0..TRIES {
let count = COUNTER.fetch_add(1, Ordering::SeqCst);
let path = tmpdir.join("rust-ignore").join(count.to_string());
if path.is_dir() {
continue;
}
fs::create_dir_all(&path).map_err(|e| {
err!("failed to create {}: {}", path.display(), e)
})?;
return Ok(TempDir(path));
}
Err(err!("failed to create temp dir after {} tries", TRIES))
}
/// Return the underlying path to this temporary directory.
pub fn path(&self) -> &Path {
&self.0
}
}
}

262
crates/ignore/src/overrides.rs Normal file
View File

@@ -0,0 +1,262 @@
/*!
The overrides module provides a way to specify a set of override globs.
This provides functionality similar to `--include` or `--exclude` in command
line tools.
*/
use std::path::Path;
use gitignore::{self, Gitignore, GitignoreBuilder};
use {Error, Match};
/// Glob represents a single glob in an override matcher.
///
/// This is used to report information about the highest precedent glob
/// that matched.
///
/// Note that not all matches necessarily correspond to a specific glob. For
/// example, if there are one or more whitelist globs and a file path doesn't
/// match any glob in the set, then the file path is considered to be ignored.
///
/// The lifetime `'a` refers to the lifetime of the matcher that produced
/// this glob.
#[derive(Clone, Debug)]
pub struct Glob<'a>(GlobInner<'a>);
#[derive(Clone, Debug)]
enum GlobInner<'a> {
/// No glob matched, but the file path should still be ignored.
UnmatchedIgnore,
/// A glob matched.
Matched(&'a gitignore::Glob),
}
impl<'a> Glob<'a> {
fn unmatched() -> Glob<'a> {
Glob(GlobInner::UnmatchedIgnore)
}
}
/// Manages a set of overrides provided explicitly by the end user.
#[derive(Clone, Debug)]
pub struct Override(Gitignore);
impl Override {
/// Returns an empty matcher that never matches any file path.
pub fn empty() -> Override {
Override(Gitignore::empty())
}
/// Returns the directory of this override set.
///
/// All matches are done relative to this path.
pub fn path(&self) -> &Path {
self.0.path()
}
/// Returns true if and only if this matcher is empty.
///
/// When a matcher is empty, it will never match any file path.
pub fn is_empty(&self) -> bool {
self.0.is_empty()
}
/// Returns the total number of ignore globs.
pub fn num_ignores(&self) -> u64 {
self.0.num_whitelists()
}
/// Returns the total number of whitelisted globs.
pub fn num_whitelists(&self) -> u64 {
self.0.num_ignores()
}
/// Returns whether the given file path matched a pattern in this override
/// matcher.
///
/// `is_dir` should be true if the path refers to a directory and false
/// otherwise.
///
/// If there are no overrides, then this always returns `Match::None`.
///
/// If there is at least one whitelist override and `is_dir` is false, then
/// this never returns `Match::None`, since non-matches are interpreted as
/// ignored.
///
/// The given path is matched to the globs relative to the path given
/// when building the override matcher. Specifically, before matching
/// `path`, its prefix (as determined by a common suffix of the directory
/// given) is stripped. If there is no common suffix/prefix overlap, then
/// `path` is assumed to reside in the same directory as the root path for
/// this set of overrides.
pub fn matched<'a, P: AsRef<Path>>(
&'a self,
path: P,
is_dir: bool,
) -> Match<Glob<'a>> {
if self.is_empty() {
return Match::None;
}
let mat = self.0.matched(path, is_dir).invert();
if mat.is_none() && self.num_whitelists() > 0 && !is_dir {
return Match::Ignore(Glob::unmatched());
}
mat.map(move |giglob| Glob(GlobInner::Matched(giglob)))
}
}
/// Builds a matcher for a set of glob overrides.
pub struct OverrideBuilder {
builder: GitignoreBuilder,
}
impl OverrideBuilder {
/// Create a new override builder.
///
/// Matching is done relative to the directory path provided.
pub fn new<P: AsRef<Path>>(path: P) -> OverrideBuilder {
OverrideBuilder { builder: GitignoreBuilder::new(path) }
}
/// Builds a new override matcher from the globs added so far.
///
/// Once a matcher is built, no new globs can be added to it.
pub fn build(&self) -> Result<Override, Error> {
Ok(Override(self.builder.build()?))
}
/// Add a glob to the set of overrides.
///
/// Globs provided here have precisely the same semantics as a single
/// line in a `gitignore` file, where the meaning of `!` is inverted:
/// namely, `!` at the beginning of a glob will ignore a file. Without `!`,
/// all matches of the glob provided are treated as whitelist matches.
pub fn add(&mut self, glob: &str) -> Result<&mut OverrideBuilder, Error> {
self.builder.add_line(None, glob)?;
Ok(self)
}
/// Toggle whether the globs should be matched case insensitively or not.
///
/// When this option is changed, only globs added after the change will be affected.
///
/// This is disabled by default.
pub fn case_insensitive(
&mut self,
yes: bool,
) -> Result<&mut OverrideBuilder, Error> {
// TODO: This should not return a `Result`. Fix this in the next semver
// release.
self.builder.case_insensitive(yes)?;
Ok(self)
}
}
#[cfg(test)]
mod tests {
use super::{Override, OverrideBuilder};
const ROOT: &'static str = "/home/andrew/foo";
fn ov(globs: &[&str]) -> Override {
let mut builder = OverrideBuilder::new(ROOT);
for glob in globs {
builder.add(glob).unwrap();
}
builder.build().unwrap()
}
#[test]
fn empty() {
let ov = ov(&[]);
assert!(ov.matched("a.foo", false).is_none());
assert!(ov.matched("a", false).is_none());
assert!(ov.matched("", false).is_none());
}
#[test]
fn simple() {
let ov = ov(&["*.foo", "!*.bar"]);
assert!(ov.matched("a.foo", false).is_whitelist());
assert!(ov.matched("a.foo", true).is_whitelist());
assert!(ov.matched("a.rs", false).is_ignore());
assert!(ov.matched("a.rs", true).is_none());
assert!(ov.matched("a.bar", false).is_ignore());
assert!(ov.matched("a.bar", true).is_ignore());
}
#[test]
fn only_ignores() {
let ov = ov(&["!*.bar"]);
assert!(ov.matched("a.rs", false).is_none());
assert!(ov.matched("a.rs", true).is_none());
assert!(ov.matched("a.bar", false).is_ignore());
assert!(ov.matched("a.bar", true).is_ignore());
}
#[test]
fn precedence() {
let ov = ov(&["*.foo", "!*.bar.foo"]);
assert!(ov.matched("a.foo", false).is_whitelist());
assert!(ov.matched("a.baz", false).is_ignore());
assert!(ov.matched("a.bar.foo", false).is_ignore());
}
#[test]
fn gitignore() {
let ov = ov(&["/foo", "bar/*.rs", "baz/**"]);
assert!(ov.matched("bar/lib.rs", false).is_whitelist());
assert!(ov.matched("bar/wat/lib.rs", false).is_ignore());
assert!(ov.matched("wat/bar/lib.rs", false).is_ignore());
assert!(ov.matched("foo", false).is_whitelist());
assert!(ov.matched("wat/foo", false).is_ignore());
assert!(ov.matched("baz", false).is_ignore());
assert!(ov.matched("baz/a", false).is_whitelist());
assert!(ov.matched("baz/a/b", false).is_whitelist());
}
#[test]
fn allow_directories() {
// This tests that directories are NOT ignored when they are unmatched.
let ov = ov(&["*.rs"]);
assert!(ov.matched("foo.rs", false).is_whitelist());
assert!(ov.matched("foo.c", false).is_ignore());
assert!(ov.matched("foo", false).is_ignore());
assert!(ov.matched("foo", true).is_none());
assert!(ov.matched("src/foo.rs", false).is_whitelist());
assert!(ov.matched("src/foo.c", false).is_ignore());
assert!(ov.matched("src/foo", false).is_ignore());
assert!(ov.matched("src/foo", true).is_none());
}
#[test]
fn absolute_path() {
let ov = ov(&["!/bar"]);
assert!(ov.matched("./foo/bar", false).is_none());
}
#[test]
fn case_insensitive() {
let ov = OverrideBuilder::new(ROOT)
.case_insensitive(true)
.unwrap()
.add("*.html")
.unwrap()
.build()
.unwrap();
assert!(ov.matched("foo.html", false).is_whitelist());
assert!(ov.matched("foo.HTML", false).is_whitelist());
assert!(ov.matched("foo.htm", false).is_ignore());
assert!(ov.matched("foo.HTM", false).is_ignore());
}
#[test]
fn default_case_sensitive() {
let ov =
OverrideBuilder::new(ROOT).add("*.html").unwrap().build().unwrap();
assert!(ov.matched("foo.html", false).is_whitelist());
assert!(ov.matched("foo.HTML", false).is_ignore());
assert!(ov.matched("foo.htm", false).is_ignore());
assert!(ov.matched("foo.HTM", false).is_ignore());
}
}

142
crates/ignore/src/pathutil.rs Normal file
View File

@@ -0,0 +1,142 @@
use std::ffi::OsStr;
use std::path::Path;
use walk::DirEntry;
/// Returns true if and only if this entry is considered to be hidden.
///
/// This only returns true if the base name of the path starts with a `.`.
///
/// On Unix, this implements a more optimized check.
#[cfg(unix)]
pub fn is_hidden(dent: &DirEntry) -> bool {
use std::os::unix::ffi::OsStrExt;
if let Some(name) = file_name(dent.path()) {
name.as_bytes().get(0) == Some(&b'.')
} else {
false
}
}
/// Returns true if and only if this entry is considered to be hidden.
///
/// On Windows, this returns true if one of the following is true:
///
/// * The base name of the path starts with a `.`.
/// * The file attributes have the `HIDDEN` property set.
#[cfg(windows)]
pub fn is_hidden(dent: &DirEntry) -> bool {
use std::os::windows::fs::MetadataExt;
use winapi_util::file;
// This looks like we're doing an extra stat call, but on Windows, the
// directory traverser reuses the metadata retrieved from each directory
// entry and stores it on the DirEntry itself. So this is "free."
if let Ok(md) = dent.metadata() {
if file::is_hidden(md.file_attributes() as u64) {
return true;
}
}
if let Some(name) = file_name(dent.path()) {
name.to_str().map(|s| s.starts_with(".")).unwrap_or(false)
} else {
false
}
}
/// Returns true if and only if this entry is considered to be hidden.
///
/// This only returns true if the base name of the path starts with a `.`.
#[cfg(not(any(unix, windows)))]
pub fn is_hidden(dent: &DirEntry) -> bool {
if let Some(name) = file_name(dent.path()) {
name.to_str().map(|s| s.starts_with(".")).unwrap_or(false)
} else {
false
}
}
/// Strip `prefix` from the `path` and return the remainder.
///
/// If `path` doesn't have a prefix `prefix`, then return `None`.
#[cfg(unix)]
pub fn strip_prefix<'a, P: AsRef<Path> + ?Sized>(
prefix: &'a P,
path: &'a Path,
) -> Option<&'a Path> {
use std::os::unix::ffi::OsStrExt;
let prefix = prefix.as_ref().as_os_str().as_bytes();
let path = path.as_os_str().as_bytes();
if prefix.len() > path.len() || prefix != &path[0..prefix.len()] {
None
} else {
Some(&Path::new(OsStr::from_bytes(&path[prefix.len()..])))
}
}
/// Strip `prefix` from the `path` and return the remainder.
///
/// If `path` doesn't have a prefix `prefix`, then return `None`.
#[cfg(not(unix))]
pub fn strip_prefix<'a, P: AsRef<Path> + ?Sized>(
prefix: &'a P,
path: &'a Path,
) -> Option<&'a Path> {
path.strip_prefix(prefix).ok()
}
/// Returns true if this file path is just a file name. i.e., Its parent is
/// the empty string.
#[cfg(unix)]
pub fn is_file_name<P: AsRef<Path>>(path: P) -> bool {
use memchr::memchr;
use std::os::unix::ffi::OsStrExt;
let path = path.as_ref().as_os_str().as_bytes();
memchr(b'/', path).is_none()
}
/// Returns true if this file path is just a file name. i.e., Its parent is
/// the empty string.
#[cfg(not(unix))]
pub fn is_file_name<P: AsRef<Path>>(path: P) -> bool {
path.as_ref().parent().map(|p| p.as_os_str().is_empty()).unwrap_or(false)
}
/// 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 memchr::memrchr;
use std::os::unix::ffi::OsStrExt;
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()
}

591
crates/ignore/src/types.rs Normal file
View File

@@ -0,0 +1,591 @@
/*!
The types module provides a way of associating globs on file names to file
types.
This can be used to match specific types of files. For example, among
the default file types provided, the Rust file type is defined to be `*.rs`
with name `rust`. Similarly, the C file type is defined to be `*.{c,h}` with
name `c`.
Note that the set of default types may change over time.
# Example
This shows how to create and use a simple file type matcher using the default
file types defined in this crate.
```
use ignore::types::TypesBuilder;
let mut builder = TypesBuilder::new();
builder.add_defaults();
builder.select("rust");
let matcher = builder.build().unwrap();
assert!(matcher.matched("foo.rs", false).is_whitelist());
assert!(matcher.matched("foo.c", false).is_ignore());
```
# Example: negation
This is like the previous example, but shows how negating a file type works.
That is, this will let us match file paths that *don't* correspond to a
particular file type.
```
use ignore::types::TypesBuilder;
let mut builder = TypesBuilder::new();
builder.add_defaults();
builder.negate("c");
let matcher = builder.build().unwrap();
assert!(matcher.matched("foo.rs", false).is_none());
assert!(matcher.matched("foo.c", false).is_ignore());
```
# Example: custom file type definitions
This shows how to extend this library default file type definitions with
your own.
```
use ignore::types::TypesBuilder;
let mut builder = TypesBuilder::new();
builder.add_defaults();
builder.add("foo", "*.foo");
// Another way of adding a file type definition.
// This is useful when accepting input from an end user.
builder.add_def("bar:*.bar");
// Note: we only select `foo`, not `bar`.
builder.select("foo");
let matcher = builder.build().unwrap();
assert!(matcher.matched("x.foo", false).is_whitelist());
// This is ignored because we only selected the `foo` file type.
assert!(matcher.matched("x.bar", false).is_ignore());
```
We can also add file type definitions based on other definitions.
```
use ignore::types::TypesBuilder;
let mut builder = TypesBuilder::new();
builder.add_defaults();
builder.add("foo", "*.foo");
builder.add_def("bar:include:foo,cpp");
builder.select("bar");
let matcher = builder.build().unwrap();
assert!(matcher.matched("x.foo", false).is_whitelist());
assert!(matcher.matched("y.cpp", false).is_whitelist());
```
*/
use std::cell::RefCell;
use std::collections::HashMap;
use std::path::Path;
use std::sync::Arc;
use globset::{GlobBuilder, GlobSet, GlobSetBuilder};
use regex::Regex;
use thread_local::ThreadLocal;
use default_types::DEFAULT_TYPES;
use pathutil::file_name;
use {Error, Match};
/// Glob represents a single glob in a set of file type definitions.
///
/// There may be more than one glob for a particular file type.
///
/// This is used to report information about the highest precedent glob
/// that matched.
///
/// Note that not all matches necessarily correspond to a specific glob.
/// For example, if there are one or more selections and a file path doesn't
/// match any of those selections, then the file path is considered to be
/// ignored.
///
/// The lifetime `'a` refers to the lifetime of the underlying file type
/// definition, which corresponds to the lifetime of the file type matcher.
#[derive(Clone, Debug)]
pub struct Glob<'a>(GlobInner<'a>);
#[derive(Clone, Debug)]
enum GlobInner<'a> {
/// No glob matched, but the file path should still be ignored.
UnmatchedIgnore,
/// A glob matched.
Matched {
/// The file type definition which provided the glob.
def: &'a FileTypeDef,
/// The index of the glob that matched inside the file type definition.
which: usize,
/// Whether the selection was negated or not.
negated: bool,
},
}
impl<'a> Glob<'a> {
fn unmatched() -> Glob<'a> {
Glob(GlobInner::UnmatchedIgnore)
}
/// Return the file type defintion that matched, if one exists. A file type
/// definition always exists when a specific definition matches a file
/// path.
pub fn file_type_def(&self) -> Option<&FileTypeDef> {
match self {
Glob(GlobInner::UnmatchedIgnore) => None,
Glob(GlobInner::Matched { def, .. }) => Some(def),
}
}
}
/// A single file type definition.
///
/// File type definitions can be retrieved in aggregate from a file type
/// matcher. File type definitions are also reported when its responsible
/// for a match.
#[derive(Clone, Debug, Eq, PartialEq)]
pub struct FileTypeDef {
name: String,
globs: Vec<String>,
}
impl FileTypeDef {
/// Return the name of this file type.
pub fn name(&self) -> &str {
&self.name
}
/// Return the globs used to recognize this file type.
pub fn globs(&self) -> &[String] {
&self.globs
}
}
/// Types is a file type matcher.
#[derive(Clone, Debug)]
pub struct Types {
/// All of the file type definitions, sorted lexicographically by name.
defs: Vec<FileTypeDef>,
/// All of the selections made by the user.
selections: Vec<Selection<FileTypeDef>>,
/// Whether there is at least one Selection::Select in our selections.
/// When this is true, a Match::None is converted to Match::Ignore.
has_selected: bool,
/// A mapping from glob index in the set to two indices. The first is an
/// index into `selections` and the second is an index into the
/// corresponding file type definition's list of globs.
glob_to_selection: Vec<(usize, usize)>,
/// The set of all glob selections, used for actual matching.
set: GlobSet,
/// Temporary storage for globs that match.
matches: Arc<ThreadLocal<RefCell<Vec<usize>>>>,
}
/// Indicates the type of a selection for a particular file type.
#[derive(Clone, Debug)]
enum Selection<T> {
Select(String, T),
Negate(String, T),
}
impl<T> Selection<T> {
fn is_negated(&self) -> bool {
match *self {
Selection::Select(..) => false,
Selection::Negate(..) => true,
}
}
fn name(&self) -> &str {
match *self {
Selection::Select(ref name, _) => name,
Selection::Negate(ref name, _) => name,
}
}
fn map<U, F: FnOnce(T) -> U>(self, f: F) -> Selection<U> {
match self {
Selection::Select(name, inner) => {
Selection::Select(name, f(inner))
}
Selection::Negate(name, inner) => {
Selection::Negate(name, f(inner))
}
}
}
fn inner(&self) -> &T {
match *self {
Selection::Select(_, ref inner) => inner,
Selection::Negate(_, ref inner) => inner,
}
}
}
impl Types {
/// Creates a new file type matcher that never matches any path and
/// contains no file type definitions.
pub fn empty() -> Types {
Types {
defs: vec![],
selections: vec![],
has_selected: false,
glob_to_selection: vec![],
set: GlobSetBuilder::new().build().unwrap(),
matches: Arc::new(ThreadLocal::default()),
}
}
/// Returns true if and only if this matcher has zero selections.
pub fn is_empty(&self) -> bool {
self.selections.is_empty()
}
/// Returns the number of selections used in this matcher.
pub fn len(&self) -> usize {
self.selections.len()
}
/// Return the set of current file type definitions.
///
/// Definitions and globs are sorted.
pub fn definitions(&self) -> &[FileTypeDef] {
&self.defs
}
/// Returns a match for the given path against this file type matcher.
///
/// The path is considered whitelisted if it matches a selected file type.
/// The path is considered ignored if it matches a negated file type.
/// If at least one file type is selected and `path` doesn't match, then
/// the path is also considered ignored.
pub fn matched<'a, P: AsRef<Path>>(
&'a self,
path: P,
is_dir: bool,
) -> Match<Glob<'a>> {
// File types don't apply to directories, and we can't do anything
// if our glob set is empty.
if is_dir || self.set.is_empty() {
return Match::None;
}
// We only want to match against the file name, so extract it.
// If one doesn't exist, then we can't match it.
let name = match file_name(path.as_ref()) {
Some(name) => name,
None if self.has_selected => {
return Match::Ignore(Glob::unmatched());
}
None => {
return Match::None;
}
};
let mut matches = self.matches.get_or_default().borrow_mut();
self.set.matches_into(name, &mut *matches);
// The highest precedent match is the last one.
if let Some(&i) = matches.last() {
let (isel, iglob) = self.glob_to_selection[i];
let sel = &self.selections[isel];
let glob = Glob(GlobInner::Matched {
def: sel.inner(),
which: iglob,
negated: sel.is_negated(),
});
return if sel.is_negated() {
Match::Ignore(glob)
} else {
Match::Whitelist(glob)
};
}
if self.has_selected {
Match::Ignore(Glob::unmatched())
} else {
Match::None
}
}
}
/// TypesBuilder builds a type matcher from a set of file type definitions and
/// a set of file type selections.
pub struct TypesBuilder {
types: HashMap<String, FileTypeDef>,
selections: Vec<Selection<()>>,
}
impl TypesBuilder {
/// Create a new builder for a file type matcher.
///
/// The builder contains *no* type definitions to start with. A set
/// of default type definitions can be added with `add_defaults`, and
/// additional type definitions can be added with `select` and `negate`.
pub fn new() -> TypesBuilder {
TypesBuilder { types: HashMap::new(), selections: vec![] }
}
/// Build the current set of file type definitions *and* selections into
/// a file type matcher.
pub fn build(&self) -> Result<Types, Error> {
let defs = self.definitions();
let has_selected = self.selections.iter().any(|s| !s.is_negated());
let mut selections = vec![];
let mut glob_to_selection = vec![];
let mut build_set = GlobSetBuilder::new();
for (isel, selection) in self.selections.iter().enumerate() {
let def = match self.types.get(selection.name()) {
Some(def) => def.clone(),
None => {
let name = selection.name().to_string();
return Err(Error::UnrecognizedFileType(name));
}
};
for (iglob, glob) in def.globs.iter().enumerate() {
build_set.add(
GlobBuilder::new(glob)
.literal_separator(true)
.build()
.map_err(|err| Error::Glob {
glob: Some(glob.to_string()),
err: err.kind().to_string(),
})?,
);
glob_to_selection.push((isel, iglob));
}
selections.push(selection.clone().map(move |_| def));
}
let set = build_set
.build()
.map_err(|err| Error::Glob { glob: None, err: err.to_string() })?;
Ok(Types {
defs: defs,
selections: selections,
has_selected: has_selected,
glob_to_selection: glob_to_selection,
set: set,
matches: Arc::new(ThreadLocal::default()),
})
}
/// Return the set of current file type definitions.
///
/// Definitions and globs are sorted.
pub fn definitions(&self) -> Vec<FileTypeDef> {
let mut defs = vec![];
for def in self.types.values() {
let mut def = def.clone();
def.globs.sort();
defs.push(def);
}
defs.sort_by(|def1, def2| def1.name().cmp(def2.name()));
defs
}
/// Select the file type given by `name`.
///
/// If `name` is `all`, then all file types currently defined are selected.
pub fn select(&mut self, name: &str) -> &mut TypesBuilder {
if name == "all" {
for name in self.types.keys() {
self.selections.push(Selection::Select(name.to_string(), ()));
}
} else {
self.selections.push(Selection::Select(name.to_string(), ()));
}
self
}
/// Ignore the file type given by `name`.
///
/// If `name` is `all`, then all file types currently defined are negated.
pub fn negate(&mut self, name: &str) -> &mut TypesBuilder {
if name == "all" {
for name in self.types.keys() {
self.selections.push(Selection::Negate(name.to_string(), ()));
}
} else {
self.selections.push(Selection::Negate(name.to_string(), ()));
}
self
}
/// Clear any file type definitions for the type name given.
pub fn clear(&mut self, name: &str) -> &mut TypesBuilder {
self.types.remove(name);
self
}
/// Add a new file type definition. `name` can be arbitrary and `pat`
/// should be a glob recognizing file paths belonging to the `name` type.
///
/// If `name` is `all` or otherwise contains any character that is not a
/// Unicode letter or number, then an error is returned.
pub fn add(&mut self, name: &str, glob: &str) -> Result<(), Error> {
lazy_static! {
static ref RE: Regex = Regex::new(r"^[\pL\pN]+$").unwrap();
};
if name == "all" || !RE.is_match(name) {
return Err(Error::InvalidDefinition);
}
let (key, glob) = (name.to_string(), glob.to_string());
self.types
.entry(key)
.or_insert_with(|| FileTypeDef {
name: name.to_string(),
globs: vec![],
})
.globs
.push(glob);
Ok(())
}
/// Add a new file type definition specified in string form. There are two
/// valid formats:
/// 1. `{name}:{glob}`. This defines a 'root' definition that associates the
/// given name with the given glob.
/// 2. `{name}:include:{comma-separated list of already defined names}.
/// This defines an 'include' definition that associates the given name
/// with the definitions of the given existing types.
/// Names may not include any characters that are not
/// Unicode letters or numbers.
pub fn add_def(&mut self, def: &str) -> Result<(), Error> {
let parts: Vec<&str> = def.split(':').collect();
match parts.len() {
2 => {
let name = parts[0];
let glob = parts[1];
if name.is_empty() || glob.is_empty() {
return Err(Error::InvalidDefinition);
}
self.add(name, glob)
}
3 => {
let name = parts[0];
let types_string = parts[2];
if name.is_empty()
|| parts[1] != "include"
|| types_string.is_empty()
{
return Err(Error::InvalidDefinition);
}
let types = types_string.split(',');
// Check ahead of time to ensure that all types specified are
// present and fail fast if not.
if types.clone().any(|t| !self.types.contains_key(t)) {
return Err(Error::InvalidDefinition);
}
for type_name in types {
let globs =
self.types.get(type_name).unwrap().globs.clone();
for glob in globs {
self.add(name, &glob)?;
}
}
Ok(())
}
_ => Err(Error::InvalidDefinition),
}
}
/// Add a set of default file type definitions.
pub fn add_defaults(&mut self) -> &mut TypesBuilder {
static MSG: &'static str = "adding a default type should never fail";
for &(name, exts) in DEFAULT_TYPES {
for ext in exts {
self.add(name, ext).expect(MSG);
}
}
self
}
}
#[cfg(test)]
mod tests {
use super::TypesBuilder;
macro_rules! matched {
($name:ident, $types:expr, $sel:expr, $selnot:expr,
$path:expr) => {
matched!($name, $types, $sel, $selnot, $path, true);
};
(not, $name:ident, $types:expr, $sel:expr, $selnot:expr,
$path:expr) => {
matched!($name, $types, $sel, $selnot, $path, false);
};
($name:ident, $types:expr, $sel:expr, $selnot:expr,
$path:expr, $matched:expr) => {
#[test]
fn $name() {
let mut btypes = TypesBuilder::new();
for tydef in $types {
btypes.add_def(tydef).unwrap();
}
for sel in $sel {
btypes.select(sel);
}
for selnot in $selnot {
btypes.negate(selnot);
}
let types = btypes.build().unwrap();
let mat = types.matched($path, false);
assert_eq!($matched, !mat.is_ignore());
}
};
}
fn types() -> Vec<&'static str> {
vec![
"html:*.html",
"html:*.htm",
"rust:*.rs",
"js:*.js",
"foo:*.{rs,foo}",
"combo:include:html,rust",
]
}
matched!(match1, types(), vec!["rust"], vec![], "lib.rs");
matched!(match2, types(), vec!["html"], vec![], "index.html");
matched!(match3, types(), vec!["html"], vec![], "index.htm");
matched!(match4, types(), vec!["html", "rust"], vec![], "main.rs");
matched!(match5, types(), vec![], vec![], "index.html");
matched!(match6, types(), vec![], vec!["rust"], "index.html");
matched!(match7, types(), vec!["foo"], vec!["rust"], "main.foo");
matched!(match8, types(), vec!["combo"], vec![], "index.html");
matched!(match9, types(), vec!["combo"], vec![], "lib.rs");
matched!(not, matchnot1, types(), vec!["rust"], vec![], "index.html");
matched!(not, matchnot2, types(), vec![], vec!["rust"], "main.rs");
matched!(not, matchnot3, types(), vec!["foo"], vec!["rust"], "main.rs");
matched!(not, matchnot4, types(), vec!["rust"], vec!["foo"], "main.rs");
matched!(not, matchnot5, types(), vec!["rust"], vec!["foo"], "main.foo");
matched!(not, matchnot6, types(), vec!["combo"], vec![], "leftpad.js");
#[test]
fn test_invalid_defs() {
let mut btypes = TypesBuilder::new();
for tydef in types() {
btypes.add_def(tydef).unwrap();
}
// Preserve the original definitions for later comparison.
let original_defs = btypes.definitions();
let bad_defs = vec![
// Reference to type that does not exist
"combo:include:html,python",
// Bad format
"combo:foobar:html,rust",
"",
];
for def in bad_defs {
assert!(btypes.add_def(def).is_err());
// Ensure that nothing changed, even if some of the includes were valid.
assert_eq!(btypes.definitions(), original_defs);
}
}
}

2162
crates/ignore/src/walk.rs Normal file
View File

File diff suppressed because it is too large Load Diff

216
crates/ignore/tests/gitignore_matched_path_or_any_parents_tests.gitignore Normal file
View File

@@ -0,0 +1,216 @@
# Based on https://github.com/behnam/gitignore-test/blob/master/.gitignore
### file in root
# MATCH /file_root_1
file_root_00
# NO_MATCH
file_root_01/
# NO_MATCH
file_root_02/*
# NO_MATCH
file_root_03/**
# MATCH /file_root_10
/file_root_10
# NO_MATCH
/file_root_11/
# NO_MATCH
/file_root_12/*
# NO_MATCH
/file_root_13/**
# NO_MATCH
*/file_root_20
# NO_MATCH
*/file_root_21/
# NO_MATCH
*/file_root_22/*
# NO_MATCH
*/file_root_23/**
# MATCH /file_root_30
**/file_root_30
# NO_MATCH
**/file_root_31/
# NO_MATCH
**/file_root_32/*
# NO_MATCH
**/file_root_33/**
### file in sub-dir
# MATCH /parent_dir/file_deep_1
file_deep_00
# NO_MATCH
file_deep_01/
# NO_MATCH
file_deep_02/*
# NO_MATCH
file_deep_03/**
# NO_MATCH
/file_deep_10
# NO_MATCH
/file_deep_11/
# NO_MATCH
/file_deep_12/*
# NO_MATCH
/file_deep_13/**
# MATCH /parent_dir/file_deep_20
*/file_deep_20
# NO_MATCH
*/file_deep_21/
# NO_MATCH
*/file_deep_22/*
# NO_MATCH
*/file_deep_23/**
# MATCH /parent_dir/file_deep_30
**/file_deep_30
# NO_MATCH
**/file_deep_31/
# NO_MATCH
**/file_deep_32/*
# NO_MATCH
**/file_deep_33/**
### dir in root
# MATCH /dir_root_00
dir_root_00
# MATCH /dir_root_01
dir_root_01/
# MATCH /dir_root_02
dir_root_02/*
# MATCH /dir_root_03
dir_root_03/**
# MATCH /dir_root_10
/dir_root_10
# MATCH /dir_root_11
/dir_root_11/
# MATCH /dir_root_12
/dir_root_12/*
# MATCH /dir_root_13
/dir_root_13/**
# NO_MATCH
*/dir_root_20
# NO_MATCH
*/dir_root_21/
# NO_MATCH
*/dir_root_22/*
# NO_MATCH
*/dir_root_23/**
# MATCH /dir_root_30
**/dir_root_30
# MATCH /dir_root_31
**/dir_root_31/
# MATCH /dir_root_32
**/dir_root_32/*
# MATCH /dir_root_33
**/dir_root_33/**
### dir in sub-dir
# MATCH /parent_dir/dir_deep_00
dir_deep_00
# MATCH /parent_dir/dir_deep_01
dir_deep_01/
# NO_MATCH
dir_deep_02/*
# NO_MATCH
dir_deep_03/**
# NO_MATCH
/dir_deep_10
# NO_MATCH
/dir_deep_11/
# NO_MATCH
/dir_deep_12/*
# NO_MATCH
/dir_deep_13/**
# MATCH /parent_dir/dir_deep_20
*/dir_deep_20
# MATCH /parent_dir/dir_deep_21
*/dir_deep_21/
# MATCH /parent_dir/dir_deep_22
*/dir_deep_22/*
# MATCH /parent_dir/dir_deep_23
*/dir_deep_23/**
# MATCH /parent_dir/dir_deep_30
**/dir_deep_30
# MATCH /parent_dir/dir_deep_31
**/dir_deep_31/
# MATCH /parent_dir/dir_deep_32
**/dir_deep_32/*
# MATCH /parent_dir/dir_deep_33
**/dir_deep_33/**

300
crates/ignore/tests/gitignore_matched_path_or_any_parents_tests.rs Normal file
View File

@@ -0,0 +1,300 @@
extern crate ignore;
use std::path::Path;
use ignore::gitignore::{Gitignore, GitignoreBuilder};
const IGNORE_FILE: &'static str =
"tests/gitignore_matched_path_or_any_parents_tests.gitignore";
fn get_gitignore() -> Gitignore {
let mut builder = GitignoreBuilder::new("ROOT");
let error = builder.add(IGNORE_FILE);
assert!(error.is_none(), "failed to open gitignore file");
builder.build().unwrap()
}
#[test]
#[should_panic(expected = "path is expected to be under the root")]
fn test_path_should_be_under_root() {
let gitignore = get_gitignore();
let path = "/tmp/some_file";
gitignore.matched_path_or_any_parents(Path::new(path), false);
assert!(false);
}
#[test]
fn test_files_in_root() {
let gitignore = get_gitignore();
let m = |path: &str| {
gitignore.matched_path_or_any_parents(Path::new(path), false)
};
// 0x
assert!(m("ROOT/file_root_00").is_ignore());
assert!(m("ROOT/file_root_01").is_none());
assert!(m("ROOT/file_root_02").is_none());
assert!(m("ROOT/file_root_03").is_none());
// 1x
assert!(m("ROOT/file_root_10").is_ignore());
assert!(m("ROOT/file_root_11").is_none());
assert!(m("ROOT/file_root_12").is_none());
assert!(m("ROOT/file_root_13").is_none());
// 2x
assert!(m("ROOT/file_root_20").is_none());
assert!(m("ROOT/file_root_21").is_none());
assert!(m("ROOT/file_root_22").is_none());
assert!(m("ROOT/file_root_23").is_none());
// 3x
assert!(m("ROOT/file_root_30").is_ignore());
assert!(m("ROOT/file_root_31").is_none());
assert!(m("ROOT/file_root_32").is_none());
assert!(m("ROOT/file_root_33").is_none());
}
#[test]
fn test_files_in_deep() {
let gitignore = get_gitignore();
let m = |path: &str| {
gitignore.matched_path_or_any_parents(Path::new(path), false)
};
// 0x
assert!(m("ROOT/parent_dir/file_deep_00").is_ignore());
assert!(m("ROOT/parent_dir/file_deep_01").is_none());
assert!(m("ROOT/parent_dir/file_deep_02").is_none());
assert!(m("ROOT/parent_dir/file_deep_03").is_none());
// 1x
assert!(m("ROOT/parent_dir/file_deep_10").is_none());
assert!(m("ROOT/parent_dir/file_deep_11").is_none());
assert!(m("ROOT/parent_dir/file_deep_12").is_none());
assert!(m("ROOT/parent_dir/file_deep_13").is_none());
// 2x
assert!(m("ROOT/parent_dir/file_deep_20").is_ignore());
assert!(m("ROOT/parent_dir/file_deep_21").is_none());
assert!(m("ROOT/parent_dir/file_deep_22").is_none());
assert!(m("ROOT/parent_dir/file_deep_23").is_none());
// 3x
assert!(m("ROOT/parent_dir/file_deep_30").is_ignore());
assert!(m("ROOT/parent_dir/file_deep_31").is_none());
assert!(m("ROOT/parent_dir/file_deep_32").is_none());
assert!(m("ROOT/parent_dir/file_deep_33").is_none());
}
#[test]
fn test_dirs_in_root() {
let gitignore = get_gitignore();
let m = |path: &str, is_dir: bool| {
gitignore.matched_path_or_any_parents(Path::new(path), is_dir)
};
// 00
assert!(m("ROOT/dir_root_00", true).is_ignore());
assert!(m("ROOT/dir_root_00/file", false).is_ignore());
assert!(m("ROOT/dir_root_00/child_dir", true).is_ignore());
assert!(m("ROOT/dir_root_00/child_dir/file", false).is_ignore());
// 01
assert!(m("ROOT/dir_root_01", true).is_ignore());
assert!(m("ROOT/dir_root_01/file", false).is_ignore());
assert!(m("ROOT/dir_root_01/child_dir", true).is_ignore());
assert!(m("ROOT/dir_root_01/child_dir/file", false).is_ignore());
// 02
assert!(m("ROOT/dir_root_02", true).is_none()); // dir itself doesn't match
assert!(m("ROOT/dir_root_02/file", false).is_ignore());
assert!(m("ROOT/dir_root_02/child_dir", true).is_ignore());
assert!(m("ROOT/dir_root_02/child_dir/file", false).is_ignore());
// 03
assert!(m("ROOT/dir_root_03", true).is_none()); // dir itself doesn't match
assert!(m("ROOT/dir_root_03/file", false).is_ignore());
assert!(m("ROOT/dir_root_03/child_dir", true).is_ignore());
assert!(m("ROOT/dir_root_03/child_dir/file", false).is_ignore());
// 10
assert!(m("ROOT/dir_root_10", true).is_ignore());
assert!(m("ROOT/dir_root_10/file", false).is_ignore());
assert!(m("ROOT/dir_root_10/child_dir", true).is_ignore());
assert!(m("ROOT/dir_root_10/child_dir/file", false).is_ignore());
// 11
assert!(m("ROOT/dir_root_11", true).is_ignore());
assert!(m("ROOT/dir_root_11/file", false).is_ignore());
assert!(m("ROOT/dir_root_11/child_dir", true).is_ignore());
assert!(m("ROOT/dir_root_11/child_dir/file", false).is_ignore());
// 12
assert!(m("ROOT/dir_root_12", true).is_none()); // dir itself doesn't match
assert!(m("ROOT/dir_root_12/file", false).is_ignore());
assert!(m("ROOT/dir_root_12/child_dir", true).is_ignore());
assert!(m("ROOT/dir_root_12/child_dir/file", false).is_ignore());
// 13
assert!(m("ROOT/dir_root_13", true).is_none());
assert!(m("ROOT/dir_root_13/file", false).is_ignore());
assert!(m("ROOT/dir_root_13/child_dir", true).is_ignore());
assert!(m("ROOT/dir_root_13/child_dir/file", false).is_ignore());
// 20
assert!(m("ROOT/dir_root_20", true).is_none());
assert!(m("ROOT/dir_root_20/file", false).is_none());
assert!(m("ROOT/dir_root_20/child_dir", true).is_none());
assert!(m("ROOT/dir_root_20/child_dir/file", false).is_none());
// 21
assert!(m("ROOT/dir_root_21", true).is_none());
assert!(m("ROOT/dir_root_21/file", false).is_none());
assert!(m("ROOT/dir_root_21/child_dir", true).is_none());
assert!(m("ROOT/dir_root_21/child_dir/file", false).is_none());
// 22
assert!(m("ROOT/dir_root_22", true).is_none());
assert!(m("ROOT/dir_root_22/file", false).is_none());
assert!(m("ROOT/dir_root_22/child_dir", true).is_none());
assert!(m("ROOT/dir_root_22/child_dir/file", false).is_none());
// 23
assert!(m("ROOT/dir_root_23", true).is_none());
assert!(m("ROOT/dir_root_23/file", false).is_none());
assert!(m("ROOT/dir_root_23/child_dir", true).is_none());
assert!(m("ROOT/dir_root_23/child_dir/file", false).is_none());
// 30
assert!(m("ROOT/dir_root_30", true).is_ignore());
assert!(m("ROOT/dir_root_30/file", false).is_ignore());
assert!(m("ROOT/dir_root_30/child_dir", true).is_ignore());
assert!(m("ROOT/dir_root_30/child_dir/file", false).is_ignore());
// 31
assert!(m("ROOT/dir_root_31", true).is_ignore());
assert!(m("ROOT/dir_root_31/file", false).is_ignore());
assert!(m("ROOT/dir_root_31/child_dir", true).is_ignore());
assert!(m("ROOT/dir_root_31/child_dir/file", false).is_ignore());
// 32
assert!(m("ROOT/dir_root_32", true).is_none()); // dir itself doesn't match
assert!(m("ROOT/dir_root_32/file", false).is_ignore());
assert!(m("ROOT/dir_root_32/child_dir", true).is_ignore());
assert!(m("ROOT/dir_root_32/child_dir/file", false).is_ignore());
// 33
assert!(m("ROOT/dir_root_33", true).is_none()); // dir itself doesn't match
assert!(m("ROOT/dir_root_33/file", false).is_ignore());
assert!(m("ROOT/dir_root_33/child_dir", true).is_ignore());
assert!(m("ROOT/dir_root_33/child_dir/file", false).is_ignore());
}
#[test]
fn test_dirs_in_deep() {
let gitignore = get_gitignore();
let m = |path: &str, is_dir: bool| {
gitignore.matched_path_or_any_parents(Path::new(path), is_dir)
};
// 00
assert!(m("ROOT/parent_dir/dir_deep_00", true).is_ignore());
assert!(m("ROOT/parent_dir/dir_deep_00/file", false).is_ignore());
assert!(m("ROOT/parent_dir/dir_deep_00/child_dir", true).is_ignore());
assert!(m("ROOT/parent_dir/dir_deep_00/child_dir/file", false).is_ignore());
// 01
assert!(m("ROOT/parent_dir/dir_deep_01", true).is_ignore());
assert!(m("ROOT/parent_dir/dir_deep_01/file", false).is_ignore());
assert!(m("ROOT/parent_dir/dir_deep_01/child_dir", true).is_ignore());
assert!(m("ROOT/parent_dir/dir_deep_01/child_dir/file", false).is_ignore());
// 02
assert!(m("ROOT/parent_dir/dir_deep_02", true).is_none());
assert!(m("ROOT/parent_dir/dir_deep_02/file", false).is_none());
assert!(m("ROOT/parent_dir/dir_deep_02/child_dir", true).is_none());
assert!(m("ROOT/parent_dir/dir_deep_02/child_dir/file", false).is_none());
// 03
assert!(m("ROOT/parent_dir/dir_deep_03", true).is_none());
assert!(m("ROOT/parent_dir/dir_deep_03/file", false).is_none());
assert!(m("ROOT/parent_dir/dir_deep_03/child_dir", true).is_none());
assert!(m("ROOT/parent_dir/dir_deep_03/child_dir/file", false).is_none());
// 10
assert!(m("ROOT/parent_dir/dir_deep_10", true).is_none());
assert!(m("ROOT/parent_dir/dir_deep_10/file", false).is_none());
assert!(m("ROOT/parent_dir/dir_deep_10/child_dir", true).is_none());
assert!(m("ROOT/parent_dir/dir_deep_10/child_dir/file", false).is_none());
// 11
assert!(m("ROOT/parent_dir/dir_deep_11", true).is_none());
assert!(m("ROOT/parent_dir/dir_deep_11/file", false).is_none());
assert!(m("ROOT/parent_dir/dir_deep_11/child_dir", true).is_none());
assert!(m("ROOT/parent_dir/dir_deep_11/child_dir/file", false).is_none());
// 12
assert!(m("ROOT/parent_dir/dir_deep_12", true).is_none());
assert!(m("ROOT/parent_dir/dir_deep_12/file", false).is_none());
assert!(m("ROOT/parent_dir/dir_deep_12/child_dir", true).is_none());
assert!(m("ROOT/parent_dir/dir_deep_12/child_dir/file", false).is_none());
// 13
assert!(m("ROOT/parent_dir/dir_deep_13", true).is_none());
assert!(m("ROOT/parent_dir/dir_deep_13/file", false).is_none());
assert!(m("ROOT/parent_dir/dir_deep_13/child_dir", true).is_none());
assert!(m("ROOT/parent_dir/dir_deep_13/child_dir/file", false).is_none());
// 20
assert!(m("ROOT/parent_dir/dir_deep_20", true).is_ignore());
assert!(m("ROOT/parent_dir/dir_deep_20/file", false).is_ignore());
assert!(m("ROOT/parent_dir/dir_deep_20/child_dir", true).is_ignore());
assert!(m("ROOT/parent_dir/dir_deep_20/child_dir/file", false).is_ignore());
// 21
assert!(m("ROOT/parent_dir/dir_deep_21", true).is_ignore());
assert!(m("ROOT/parent_dir/dir_deep_21/file", false).is_ignore());
assert!(m("ROOT/parent_dir/dir_deep_21/child_dir", true).is_ignore());
assert!(m("ROOT/parent_dir/dir_deep_21/child_dir/file", false).is_ignore());
// 22
// dir itself doesn't match
assert!(m("ROOT/parent_dir/dir_deep_22", true).is_none());
assert!(m("ROOT/parent_dir/dir_deep_22/file", false).is_ignore());
assert!(m("ROOT/parent_dir/dir_deep_22/child_dir", true).is_ignore());
assert!(m("ROOT/parent_dir/dir_deep_22/child_dir/file", false).is_ignore());
// 23
// dir itself doesn't match
assert!(m("ROOT/parent_dir/dir_deep_23", true).is_none());
assert!(m("ROOT/parent_dir/dir_deep_23/file", false).is_ignore());
assert!(m("ROOT/parent_dir/dir_deep_23/child_dir", true).is_ignore());
assert!(m("ROOT/parent_dir/dir_deep_23/child_dir/file", false).is_ignore());
// 30
assert!(m("ROOT/parent_dir/dir_deep_30", true).is_ignore());
assert!(m("ROOT/parent_dir/dir_deep_30/file", false).is_ignore());
assert!(m("ROOT/parent_dir/dir_deep_30/child_dir", true).is_ignore());
assert!(m("ROOT/parent_dir/dir_deep_30/child_dir/file", false).is_ignore());
// 31
assert!(m("ROOT/parent_dir/dir_deep_31", true).is_ignore());
assert!(m("ROOT/parent_dir/dir_deep_31/file", false).is_ignore());
assert!(m("ROOT/parent_dir/dir_deep_31/child_dir", true).is_ignore());
assert!(m("ROOT/parent_dir/dir_deep_31/child_dir/file", false).is_ignore());
// 32
// dir itself doesn't match
assert!(m("ROOT/parent_dir/dir_deep_32", true).is_none());
assert!(m("ROOT/parent_dir/dir_deep_32/file", false).is_ignore());
assert!(m("ROOT/parent_dir/dir_deep_32/child_dir", true).is_ignore());
assert!(m("ROOT/parent_dir/dir_deep_32/child_dir/file", false).is_ignore());
// 33
// dir itself doesn't match
assert!(m("ROOT/parent_dir/dir_deep_33", true).is_none());
assert!(m("ROOT/parent_dir/dir_deep_33/file", false).is_ignore());
assert!(m("ROOT/parent_dir/dir_deep_33/child_dir", true).is_ignore());
assert!(m("ROOT/parent_dir/dir_deep_33/child_dir/file", false).is_ignore());
}