Table of Contents
Prettier is an opinionated code formatter inspired by refmt with advanced support for language features from:
It removes all original styling* and ensures that all outputted code conforms to a consistent style. (See this blog post)
If you are interested in the details, you can watch those two conference talks:
A few of the many projects using Prettier**:
In the case of JavaScript, this goes way beyond ESLint and other projects built on it. Unlike ESLint, there aren't a million configuration options and rules. But more importantly: everything is fixable. This works because Prettier never "checks" anything; it takes JavaScript as input and delivers the formatted JavaScript as output.
In technical terms: Prettier parses your JavaScript into an AST (Abstract Syntax Tree) and pretty-prints the AST, completely ignoring any of the original formatting*. Say hello to completely consistent syntax!
There's an extremely important piece missing from existing styling tools: the maximum line length. Sure, you can tell ESLint to warn you when you have a line that's too long, but that's an after-thought (ESLint never knows how to fix it). The maximum line length is a critical piece the formatter needs for laying out and wrapping code.
For example, take the following code:
foo(arg1, arg2, arg3, arg4);
That looks like the right way to format it. However, we've all run into this situation:
foo(reallyLongArg(), omgSoManyParameters(), IShouldRefactorThis(), isThereSeriouslyAnotherOne());
Suddenly our previous format for calling function breaks down because this is too long. What you would probably do is this instead:
foo(
reallyLongArg(),
omgSoManyParameters(),
IShouldRefactorThis(),
isThereSeriouslyAnotherOne()
);
This clearly shows that the maximum line length has a direct impact on the style of code we desire. The fact that current style tools ignore this means they can't really help with the situations that are actually the most troublesome. Individuals on teams will all format these differently according to their own rules and we lose the consistency we sought after.
Even if we disregard line lengths, it's too easy to sneak in various styles of code in all other linters. The most strict linter I know happily lets all these styles happen:
foo({ num: 3 },
1, 2)
foo(
{ num: 3 },
1, 2)
foo(
{ num: 3 },
1,
2
)
Prettier bans all custom styling* by parsing it away and re-printing the parsed AST with its own rules that take the maximum line length into account, wrapping code when necessary.
*Well actually, some original styling is preserved when practical—see empty lines and multi-line objects.
**See Issue #1351 for discussion about how these projects using Prettier were chosen.
Install:
yarn add prettier --dev
You can install it globally if you like:
yarn global add prettier
We're defaulting to yarn
but you can use npm
if you like:
npm install [-g] prettier
Run Prettier through the CLI with this script. Run it without any arguments to see the options.
To format a file in-place, use --write
. You may want to consider
committing your code before doing that, just in case.
prettier [opts] [filename ...]
In practice, this may look something like:
prettier --single-quote --trailing-comma es5 --write "{app,__{tests,mocks}__}/**/*.js"
Don't forget the quotes around the globs! The quotes make sure that Prettier expands the globs rather than your shell, for cross-platform usage. The glob syntax from the glob module is used.
Prettier CLI will ignore files located in node_modules
directory. To opt-out from this behavior use --with-node-modules
flag.
If you're worried that Prettier will change the correctness of your code, add --debug-check
to the command.
This will cause Prettier to print an error message if it detects that code correctness might have changed.
Note that --write
cannot be used with --debug-check
.
Another useful flag is --list-different
(or -l
) which prints the filenames of files that are different from Prettier formatting. If there are differences the script errors out, which is useful in a CI scenario.
prettier --single-quote --list-different "src/**/*.js"
You can use this with a pre-commit tool. This can re-format your files that are marked as "staged" via git add
before you commit.
1. lint-staged
Install it along with husky:
yarn add lint-staged husky --dev
and add this config to your package.json
:
{
"scripts": {
"precommit": "lint-staged"
},
"lint-staged": {
"*.js": [
"prettier --write",
"git add"
]
}
}
See https://github.com/okonet/lint-staged#configuration for more details about how you can configure lint-staged.
2. pre-commit
Copy the following config in your pre-commit config yaml file:
- repo: https://github.com/awebdeveloper/pre-commit-prettier
sha: '' # Use the sha or tag you want to point at
hooks:
- id: prettier
additional_dependencies: ['prettier@1.4.2']
Find more info from here.
Alternately you can save this script as .git/hooks/pre-commit
and give it execute permission:
#!/bin/sh
jsfiles=$(git diff --cached --name-only --diff-filter=ACM | grep '\.jsx\?$' | tr '\n' ' ')
[ -z "$jsfiles" ] && exit 0
diffs=$(node_modules/.bin/prettier -l $jsfiles)
[ -z "$diffs" ] && exit 0
echo "here"
echo >&2 "Javascript files must be formatted with prettier. Please run:"
echo >&2 "node_modules/.bin/prettier --write "$diffs""
exit 1
Prettier ships with a handful of customizable format options, usable in both the CLI and API.
Option | Default | Override |
---|---|---|
Print Width - Specify the length of line that the printer will wrap on. We strongly recommend against using more than 80 columns. Prettier works by cramming as much content as possible until it reaches the limit, which happens to work well for 80 columns but makes lines that are very crowded. When a bigger column count is used in styleguides, it usually means that code is allowed to go beyond 80 columns, but not to make every single line go there, like prettier would do. |
80 |
CLI: --print-width <int> API: printWidth: <int> |
Tab Width - Specify the number of spaces per indentation-level. | 2 |
CLI: --tab-width <int> API: tabWidth: <int> |
Tabs - Indent lines with tabs instead of spaces. | false |
CLI: --use-tabs API: useTabs: <bool> |
Semicolons - Print semicolons at the ends of statements. Valid options:
|
true |
CLI: --no-semi API: semi: <bool> |
Quotes - Use single quotes instead of double quotes. Notes:
|
false |
CLI: --single-quote API: singleQuote: <bool> |
Trailing Commas - Print trailing commas wherever possible. Valid options:
|
"none" |
CLI: --trailing-comma <none|es5|all> API: trailingComma: "<none|es5|all>" |
Bracket Spacing - Print spaces between brackets in object literals. Valid options:
|
true |
CLI: --no-bracket-spacing API: bracketSpacing: <bool> |
JSX Brackets on Same Line - Put the > of a multi-line JSX element at the end of the last line instead of being alone on the next line |
false |
CLI: --jsx-bracket-same-line API: jsxBracketSameLine: <bool> |
Cursor Offset - Specify where the cursor is. This option only works with prettier.formatWithCursor , and cannot be used with rangeStart and rangeEnd . |
-1 |
CLI: --cursor-offset <int> API: cursorOffset: <int> |
Range Start - Format code starting at a given character offset. The range will extend backwards to the start of the first line containing the selected statement. This option cannot be used with cursorOffset . |
0 |
CLI: --range-start <int> API: rangeStart: <int> |
Range End - Format code ending at a given character offset (exclusive). The range will extend forwards to the end of the selected statement. This option cannot be used with cursorOffset . |
Infinity |
CLI: --range-end <int> API: rangeEnd: <int> |
Parser - Specify which parser to use. Both the babylon and flow parsers support the same set of JavaScript features (including Flow). Prettier automatically infers the parser from the input file path, so you shouldn't have to change this setting. Built-in parsers:
|
babylon |
CLI: --parser <string> --parser ./path/to/my-parser API: parser: "<string>" parser: require("./my-parser") |
Filepath - Specify the input filepath this will be used to do parser inference. Example: cat foo | prettier --stdin-filepath foo.css will default to use postcss parser |
CLI: --stdin-filepath API: filepath: "<string>" |
The API has three functions, exported as format
, check
, and formatWithCursor
. format
usage is as follows:
const prettier = require("prettier");
const options = {} // optional
prettier.format(source, options);
check
checks to see if the file has been formatted with Prettier given those options and returns a Boolean.
This is similar to the --list-different
parameter in the CLI and is useful for running Prettier in CI scenarios.
formatWithCursor
both formats the code, and translates a cursor position from unformatted code to formatted code.
This is useful for editor integrations, to prevent the cursor from moving when code is formatted. For example:
const prettier = require("prettier");
prettier.formatWithCursor(" 1", { cursorOffset: 2 });
// -> { formatted: '1;\n', cursorOffset: 1 }
If you need to make modifications to the AST (such as codemods), or you want to provide an alternate parser, you can do so by setting the parser
option to a function. The function signature of the parser function is:
(text: string, parsers: object, options: object) => AST;
Prettier's built-in parsers are exposed as properties on the parsers
argument.
prettier.format("lodash ( )", {
parser(text, { babylon }) {
const ast = babylon(text);
ast.program.body[0].expression.callee.name = "_";
return ast;
}
}); // ==> "_();\n"
The --parser
CLI option may be a path to a node.js module exporting a parse function.
A JavaScript comment of // prettier-ignore
will exclude the next node in the abstract syntax tree from formatting.
For example:
matrix(
1, 0, 0,
0, 1, 0,
0, 0, 1
)
// prettier-ignore
matrix(
1, 0, 0,
0, 1, 0,
0, 0, 1
)
will be transformed to:
matrix(1, 0, 0, 0, 1, 0, 0, 0, 1);
// prettier-ignore
matrix(
1, 0, 0,
0, 1, 0,
0, 0, 1
)
Atom users can simply install the prettier-atom package and use
Ctrl+Alt+F
to format a file (or format on save if enabled).
Emacs users should see this repository for on-demand formatting.
Vim users can simply install either sbdchd/neoformat or mitermayer/vim-prettier, for more details see this directory
Can be installed using the extension sidebar. Search for Prettier - JavaScript formatter
.
Can also be installed using ext install prettier-vscode
.
Check its repository for configuration and shortcuts
Install the JavaScript Prettier extension.
Sublime Text support is available through Package Control and the JsPrettier plug-in.
See the WebStorm guide.
Prettier attempts to support all JavaScript language features,
including non-standardized ones. By default it uses the
Babylon parser with all language
features enabled, but you can also use the
Flow parser with the
parser
API or --parser
CLI option.
All of JSX and Flow syntax is supported. In fact, the test suite in
tests
is the entire Flow test suite and they all pass.
Prettier also supports TypeScript, CSS, LESS, and SCSS.
The minimum version of TypeScript supported is 2.1.3 as it introduces the ability to have leading |
for type definitions which prettier outputs.
eslint-plugin-prettier
plugs Prettier into your ESLint workfloweslint-config-prettier
turns off all ESLint rules that are unnecessary or might conflict with Prettierprettier-eslint
passesprettier
output toeslint --fix
prettier-standard
usesprettier
andprettier-eslint
to format code with standard rulesprettier-standard-formatter
passesprettier
output tostandard --fix
prettier-miscellaneous
prettier
with a few minor extra optionsneutrino-preset-prettier
allows you to use Prettier as a Neutrino presetprettier_d
runs Prettier as a server to avoid Node.js startup delayPrettier Bookmarklet
provides a bookmarklet and exposes a REST API for Prettier that allows to format CodeMirror editor in your browserprettier-github
formats code in GitHub comments
This printer is a fork of recast's printer with its algorithm replaced by the one described by Wadler in "A prettier printer". There still may be leftover code from recast that needs to be cleaned up.
The basic idea is that the printer takes an AST and returns an intermediate representation of the output, and the printer uses that to generate a string. The advantage is that the printer can "measure" the IR and see if the output is going to fit on a line, and break if not.
This means that most of the logic of printing an AST involves
generating an abstract representation of the output involving certain
commands. For example, concat(["(", line, arg, line ")"])
would
represent a concatenation of opening parens, an argument, and closing
parens. But if that doesn't fit on one line, the printer can break
where line
is specified.
More (rough) details can be found in commands.md.
Show the world you're using Prettier →
[![styled with prettier](https://img.shields.io/badge/styled_with-prettier-ff69b4.svg)](https://github.com/prettier/prettier)
See CONTRIBUTING.md.