micromatch
Glob matching for javascript/node.js. A drop-in replacement and faster alternative to minimatch and multimatch. Just use micromatch.isMatch()
instead of minimatch()
, or use micromatch()
instead of multimatch()
.
Install
Install with npm
$ npm i micromatch --save
Table of contents
Features
Micromatch is 10-55x faster than minimatch, resulting from a combination of caching, tokenization, parsing, runtime compilation and regex optimization strategies.
- Drop-in replacement for minimatch and multimatch
- Built-in support for multiple glob patterns, like
['foo/*.js', '!bar.js']
- Better support for the Bash 4.3 specification, and less buggy
- Extensive unit tests (approx. 1,300 tests). Minimatch fails many of the tests.
Mainstream glob features:
- Brace Expansion (
foo/bar-{1..5}.md
, one/{two,three}/four.md
) - Typical glob patterns, like
**/*
, a/b/*.js
, or ['foo/*.js', '!bar.js']
Extended globbing features:
- Logical
OR
(foo/bar/(abc|xyz).js
) - Regex character classes (
foo/bar/baz-[1-5].js
) - POSIX bracket expressions (
**/[[:alpha:][:digit:]]/
) - extglobs (
**/+(x|y)
, !(a|b)
, etc)
You can combine these to create whatever matching patterns you need.
Usage
var mm = require('micromatch');
mm(array, patterns);
Examples
mm(['a.js', 'b.md', 'c.txt'], '*.{js,txt}');
Multiple patterns
Multiple patterns can also be passed:
mm(['a.md', 'b.js', 'c.txt', 'd.json'], ['*.md', '*.txt']);
Negation patterns:
Behavior;
- when the pattern is a string, minimatch behavior is used, so patterns are inclusive by default.
- when an array of patterns is passed, multimatch behavior is used, so patterns are exclusive by default
mm(['a.js', 'b.md', 'c.txt'], '!*.{js,txt}');
mm(['a.md', 'b.js', 'c.txt', 'd.json'], ['*.*', '!*.{js,txt}']);
Switch from minimatch
Use micromatch.isMatch()
instead of minimatch()
Minimatch
The main minimatch()
function returns true/false for a single file path and pattern:
var minimatch = require('minimatch');
minimatch('foo.js', '*.js');
Micromatch
With micromatch, .isMatch()
to get the same result:
var mm = require('micromatch');
mm.isMatch('foo.js', '*.js');
This implementation difference is necessary since the main micromatch()
method supports matching on multiple globs, with behavior similar to multimatch.
Methods
var mm = require('micromatch');
.isMatch
mm.isMatch(filepath, globPattern);
Returns true if a file path matches the given glob pattern.
Example
mm.isMatch('.verb.md', '*.md');
mm.isMatch('.verb.md', '*.md', {dot: true});
.contains
Returns true if any part of a file path matches the given glob pattern. Think of this is "has path" versus "is path".
Example
.isMatch()
would return false for both of the following:
mm.contains('a/b/c', 'a/b');
mm.contains('a/b/c', 'a/*');
.matcher
Returns a function for matching using the supplied pattern. e.g. create your own "matcher". The advantage of this method is that the pattern can be compiled outside of a loop.
Pattern
Can be any of the following:
Example
var isMatch = mm.matcher('*.md');
var files = [];
['a.md', 'b.txt', 'c.md'].forEach(function(fp) {
if (isMatch(fp)) {
files.push(fp);
}
});
.filter
Returns a function that can be passed to Array#filter()
.
Params
Examples
Single glob:
var fn = mm.filter('*.md');
['a.js', 'b.txt', 'c.md'].filter(fn);
var fn = mm.filter('[a-c]');
['a', 'b', 'c', 'd', 'e'].filter(fn);
Array of glob patterns:
var arr = [1,2,3,4,5,6,7,8,9,10,11,12,13,14,15];
var fn = mm.filter(['{1..10}', '![7-9]', '!{3..4}']);
arr.filter(fn);
(Internally this function generates the matching function by using the matcher method. You can use the matcher method directly to create your own filter function)
.any
Returns true if a file path matches any of the given patterns.
mm.any(filepath, patterns, options);
Params
- filepath
{String}
: The file path to test. - patterns
{String|Array}
: One or more glob patterns - options:
{Object}
: options to pass to the .matcher()
method.
Example
mm.any('abc', ['!*z']);
mm.any('abc', ['a*', 'z*']);
mm.any('abc', 'a*');
mm.any('abc', ['z*']);
.expand
Returns an object with a regex-compatible string and tokens.
mm.expand('*.js');
{ options: { track: false, dot: undefined, makeRe: true, negated: false },
pattern: '(.*\\/|^)bar\\/(?:(?!(?:^|\\/)\\.).)*?',
history: [],
tokens:
{ path:
{ whole: '**/bar/**',
dirname: '**/bar/',
filename: '**',
basename: '**',
extname: '',
ext: '' },
is:
{ glob: true,
negated: false,
globstar: true,
dotfile: false,
dotdir: false },
match: {},
original: '**/bar/**',
pattern: '**/bar/**',
base: '' } }
.makeRe
Create a regular expression for matching file paths based on the given pattern:
mm.makeRe('*.js');
Options
options.unixify
Normalize slashes in file paths and glob patterns to forward slashes.
Type: {Boolean}
Default: undefined
on non-windows, true
on windows.
options.dot
Match dotfiles. Same behavior as minimatch.
Type: {Boolean}
Default: false
options.unescape
Unescape slashes in glob patterns. Use cautiously, especially on windows.
Type: {Boolean}
Default: undefined
Example
mm.isMatch('abc', '\\a\\b\\c', {unescape: true});
options.nodupes
Remove duplicate elements from the result array.
Type: {Boolean}
Default: undefined
Example
Example of using the unescape
and nodupes
options together:
mm.match(['abc', '\\a\\b\\c'], '\\a\\b\\c', {unescape: true});
mm.match(['abc', '\\a\\b\\c'], '\\a\\b\\c', {unescape: true, nodupes: true});
options.matchBase
Allow glob patterns without slashes to match a file path based on its basename. . Same behavior as minimatch.
Type: {Boolean}
Default: false
Example
mm(['a/b.js', 'a/c.md'], '*.js');
mm(['a/b.js', 'a/c.md'], '*.js', {matchBase: true});
options.nobraces
Don't expand braces in glob patterns. Same behavior as minimatch nobrace
.
Type: {Boolean}
Default: undefined
See braces for more information about extended brace expansion.
options.nobrackets
Don't expand POSIX bracket expressions.
Type: {Boolean}
Default: undefined
See expand-brackets for more information about extended bracket expressions.
options.noextglob
Don't expand extended globs.
Type: {Boolean}
Default: undefined
See extglob for more information about extended globs.
options.nocase
Use a case-insensitive regex for matching files. Same behavior as minimatch.
Type: {Boolean}
Default: false
options.nonull
If true
, when no matches are found the actual (array-ified) glob pattern is returned instead of an empty array. Same behavior as minimatch.
Type: {Boolean}
Default: false
options.cache
Cache the platform (e.g. win32
) to prevent this from being looked up for every filepath.
Type: {Boolean}
Default: true
Other features
Micromatch also supports the following.
Extended globbing
extglobs
Extended globbing, as described by the bash man page:
pattern | regex equivalent | description |
---|
?(pattern-list) | `(... | ...)?` |
*(pattern-list) | `(... | ...)*` |
+(pattern-list) | `(... | ...)+` |
@(pattern-list) | `(... | ...)` * |
!(pattern-list) | N/A | Matches anything except one of the given patterns |
* @
isn't a RegEx character.
Powered by extglob. Visit that library for the full range of options or to report extglob related issues.
See extglob for more information about extended globs.
brace expansion
In simple cases, brace expansion appears to work the same way as the logical OR
operator. For example, (a|b)
will achieve the same result as {a,b}
.
Here are some powerful features unique to brace expansion (versus character classes):
- range expansion:
a{1..3}b/*.js
expands to: ['a1b/*.js', 'a2b/*.js', 'a3b/*.js']
- nesting:
a{c,{d,e}}b/*.js
expands to: ['acb/*.js', 'adb/*.js', 'aeb/*.js']
Visit braces to ask questions and create an issue related to brace-expansion, or to see the full range of features and options related to brace expansion.
regex character classes
With the exception of brace expansion ({a,b}
, {1..5}
, etc), most of the special characters convert directly to regex, so you can expect them to follow the same rules and produce the same results as regex.
For example, given the list: ['a.js', 'b.js', 'c.js', 'd.js', 'E.js']
:
[ac].js
: matches both a
and c
, returning ['a.js', 'c.js']
[b-d].js
: matches from b
to d
, returning ['b.js', 'c.js', 'd.js']
[b-d].js
: matches from b
to d
, returning ['b.js', 'c.js', 'd.js']
a/[A-Z].js
: matches and uppercase letter, returning ['a/E.md']
Learn about regex character classes.
regex groups
Given ['a.js', 'b.js', 'c.js', 'd.js', 'E.js']
:
(a|c).js
: would match either a
or c
, returning ['a.js', 'c.js']
(b|d).js
: would match either b
or d
, returning ['b.js', 'd.js']
(b|[A-Z]).js
: would match either b
or an uppercase letter, returning ['b.js', 'E.js']
As with regex, parenthese can be nested, so patterns like ((a|b)|c)/b
will work. But it might be easier to achieve your goal using brace expansion.
POSIX bracket expressions
Example
mm.isMatch('a1', '[[:alpha:][:digit:]]');
See expand-brackets for more information about extended bracket expressions.
Notes
Whenever possible parsing behavior for patterns is based on globbing specifications in Bash 4.3. Patterns that aren't described by Bash follow wildmatch spec (used by git).
Benchmarks
Run the benchmarks:
node benchmark
As of November 25, 2015:
micromatch x 26,420 ops/sec ±0.89% (91 runs sampled)
minimatch x 3,507 ops/sec ±0.64% (97 runs sampled)
micromatch x 25,315 ops/sec ±0.82% (93 runs sampled)
minimatch x 4,398 ops/sec ±0.86% (94 runs sampled)
micromatch x 341,254 ops/sec ±0.78% (93 runs sampled)
minimatch x 30,197 ops/sec ±1.12% (91 runs sampled)
micromatch x 54,649 ops/sec ±0.74% (94 runs sampled)
minimatch x 3,095 ops/sec ±0.82% (95 runs sampled)
micromatch x 16,719 ops/sec ±0.79% (95 runs sampled)
minimatch x 4,348 ops/sec ±0.86% (96 runs sampled)
micromatch x 721 ops/sec ±0.77% (94 runs sampled)
minimatch x 17.73 ops/sec ±1.08% (50 runs sampled)
micromatch x 5,051 ops/sec ±0.87% (97 runs sampled)
minimatch x 628 ops/sec ±0.83% (94 runs sampled)
micromatch x 51,280 ops/sec ±0.80% (95 runs sampled)
minimatch x 1,923 ops/sec ±0.84% (95 runs sampled)
micromatch x 22,440 ops/sec ±0.97% (94 runs sampled)
minimatch x 2,481 ops/sec ±1.10% (94 runs sampled)
micromatch x 722,823 ops/sec ±1.30% (87 runs sampled)
minimatch x 52,967 ops/sec ±1.09% (94 runs sampled)
micromatch x 243,471 ops/sec ±0.79% (94 runs sampled)
minimatch x 11,736 ops/sec ±0.82% (96 runs sampled)
micromatch x 190,874 ops/sec ±0.98% (95 runs sampled)
minimatch x 21,699 ops/sec ±0.81% (97 runs sampled)
micromatch x 496,393 ops/sec ±3.86% (90 runs sampled)
minimatch x 53,765 ops/sec ±0.75% (95 runs sampled)
Run tests
Install dev dependencies:
$ npm i -d && npm test
Contributing
Pull requests and stars are always welcome. For bugs and feature requests, please create an issue.
Please be sure to run the benchmarks before/after any code changes to judge the impact before you do a PR. thanks!
Related
Author
Jon Schlinkert
License
Copyright © 2014-2015 Jon Schlinkert
Released under the MIT license.
This file was generated by verb-cli on November 25, 2015.