@babel/parser
The Babel parser (previously Babylon) is a JavaScript parser used in Babel.
- The latest ECMAScript version enabled by default (ES2020).
- Comment attachment.
- Support for JSX, Flow, Typescript.
- Support for experimental language proposals (accepting PRs for anything at least stage-0).
Credits
Heavily based on acorn and acorn-jsx, thanks to the awesome work of @RReverser and @marijnh.
API
babelParser.parse(code, [options])
babelParser.parseExpression(code, [options])
parse() parses the provided code as an entire ECMAScript program, while
parseExpression() tries to parse a single Expression with performance in
mind. When in doubt, use .parse().
Options
History
| Version | Changes |
|---|---|
v7.16.0 | Added startColumn |
v7.15.0 | Added attachComment |
v7.7.0 | Added errorRecovery |
v7.5.0 | Added allowUndeclaredExports |
v7.2.0 | Added createParenthesizedExpressions |
allowImportExportEverywhere: By default,
importandexportdeclarations can only appear at a program's top level. Setting this option totrueallows them anywhere where a statement is allowed.allowAwaitOutsideFunction: By default,
awaituse is only allowed inside of an async function or, when thetopLevelAwaitplugin is enabled, in the top-level scope of modules. Set this totrueto also accept it in the top-level scope of scripts. This option is discouraged in favor oftopLevelAwaitplugin.allowReturnOutsideFunction: By default, a return statement at the top level raises an error. Set this to
trueto accept such code.allowSuperOutsideMethod: By default,
superuse is not allowed outside of class and object methods. Set this totrueto accept such code.allowUndeclaredExports: By default, exporting an identifier that was not declared in the current module scope will raise an error. While this behavior is required by the ECMAScript modules specification, Babel's parser cannot anticipate transforms later in the plugin pipeline that might insert the appropriate declarations, so it is sometimes important to set this option to
trueto prevent the parser from prematurely complaining about undeclared exports that will be added later.attachComment: By default, Babel attaches comments to adjacent AST nodes. When this option is set to
false, comments are not attached. It can provide up to 30% performance improvement when the input code has many comments.@babel/eslint-parserwill set it for you. It is not recommended to useattachComment: falsewith Babel transform, as doing so removes all the comments in output code, and renders annotations such as/* istanbul ignore next */nonfunctional.createParenthesizedExpressions: By default, the parser sets
extra.parenthesizedon the expression nodes. When this option is set totrue,ParenthesizedExpressionAST nodes are created instead.errorRecovery: By default, Babel always throws an error when it finds some invalid code. When this option is set to
true, it will store the parsing error and try to continue parsing the invalid input file. The resulting AST will have anerrorsproperty representing an array of all the parsing errors. Note that even when this option is enabled,@babel/parsercould throw for unrecoverable errors.plugins: Array containing the plugins that you want to enable.
sourceType: Indicate the mode the code should be parsed in. Can be one of
"script","module", or"unambiguous". Defaults to"script"."unambiguous"will make @babel/parser attempt to guess, based on the presence of ES6importorexportstatements. Files with ES6imports andexports are considered"module"and are otherwise"script".sourceFilename: Correlate output AST nodes with their source filename. Useful when generating code and source maps from the ASTs of multiple input files.
startColumn: By default, the parsed code is treated as if it starts from line 1, column 0. You can provide a column number to alternatively start with. Useful for integration with other source tools.
startLine: By default, the parsed code is treated as if it starts from line 1, column 0. You can provide a line number to alternatively start with. Useful for integration with other source tools.
strictMode: By default, ECMAScript code is parsed as strict only if a
"use strict";directive is present or if the parsed file is an ECMAScript module. Set this option totrueto always parse files in strict mode.ranges: Adds a
rangeproperty to each node:[node.start, node.end]tokens: Adds all parsed tokens to a
tokensproperty on theFilenode
Output
The Babel parser generates AST according to Babel AST format. It is based on ESTree spec with the following deviations:
- Literal token is replaced with StringLiteral, NumericLiteral, BigIntLiteral, BooleanLiteral, NullLiteral, RegExpLiteral
- Property token is replaced with ObjectProperty and ObjectMethod
- MethodDefinition is replaced with ClassMethod
- PropertyDefinition is replaced with ClassProperty
- Program and BlockStatement contain additional
directivesfield with Directive and DirectiveLiteral - ClassMethod, ObjectProperty, and ObjectMethod value property's properties in FunctionExpression is coerced/brought into the main method node.
- ChainExpression is replaced with OptionalMemberExpression and OptionalCallExpression
- ImportExpression is replaced with a CallExpression whose
calleeis an Import node.
There is now an
estreeplugin which reverts these deviations
AST for JSX code is based on Facebook JSX AST.
Semver
The Babel Parser follows semver in most situations. The only thing to note is that some spec-compliancy bug fixes may be released under patch versions.
For example: We push a fix to early error on something like #107 - multiple default exports per file. That would be considered a bug fix even though it would cause a build to fail.
Example
require("@babel/parser").parse("code", {
// parse in strict mode and allow module declarations
sourceType: "module",
plugins: [
// enable jsx and flow syntax
"jsx",
"flow",
],
});
Plugins
Miscellaneous
| Name | Code Example |
|---|---|
estree (repo) | n/a |
Language extensions
History
| Version | Changes |
|---|---|
v7.6.0 | Added v8intrinsic |
| Name | Code Example |
|---|---|
flow (repo) | var a: string = ""; |
flowComments (docs) | /*:: type Foo = {...}; */ |
jsx (repo) | <a attr="b">{s}</a> |
typescript (repo) | var a: string = ""; |
v8intrinsic | %DebugPrint(foo); |
ECMAScript proposals
History
| Version | Changes |
|---|---|
v7.17.0 | Added regexpUnicodeSets, destructuringPrivate, decoratorAutoAccessors |
v7.15.0 | Added hack to the proposal option of pipelineOperator. Moved topLevelAwait, privateIn to Latest ECMAScript features |
v7.14.0 | Added asyncDoExpressions. Moved classProperties, classPrivateProperties, classPrivateMethods, moduleStringNames to Latest ECMAScript features |
v7.13.0 | Added moduleBlocks |
v7.12.0 | Added classStaticBlock, moduleStringNames |
v7.11.0 | Added decimal |
v7.10.0 | Added privateIn |
v7.9.0 | Added recordAndTuple |
v7.7.0 | Added topLevelAwait |
v7.4.0 | Added partialApplication |
v7.2.0 | Added classPrivateMethods |
| Name | Code Example |
|---|---|
asyncDoExpressions (proposal) | async do { await requestAPI().json() } |
decimal (proposal) | 0.3m |
decorators (proposal) decorators-legacy | @a class A {} |
decoratorAutoAccessors (proposal) | class Example { @reactive accessor myBool = false; } |
destructuringPrivate (proposal) | class Example { #x = 1; method() { const { #x: x } = this; } } |
doExpressions (proposal) | var a = do { if (true) { 'hi'; } }; |
exportDefaultFrom (proposal) | export v from "mod" |
functionBind (proposal) | a::b, ::console.log |
importAssertions (proposal) | import json from "./foo.json" assert { type: "json" }; |
moduleBlocks (proposal) | let m = module { export let y = 1; }; |
partialApplication (proposal) | f(?, a) |
pipelineOperator (proposal) | a |> b |
recordAndTuple (proposal) | #{x: 1}, #[1, 2] |
regexpUnicodeSets (proposal) | /[\p{Decimal_Number}--[0-9]]/v; |
throwExpressions (proposal) | () => throw new Error("") |
Latest ECMAScript features
The following features are already enabled on the latest version of @babel/parser, and cannot be disabled because they are part of the language.
You should enable these features only if you are using an older version.
| Name | Code Example |
|---|---|
asyncGenerators (proposal) | async function*() {}, for await (let a of b) {} |
bigInt (proposal) | 100n |
classProperties (proposal) | class A { b = 1; } |
classPrivateProperties (proposal) | class A { #b = 1; } |
classPrivateMethods (proposal) | class A { #c() {} } |
classStaticBlock (proposal) | class A { static {} } |
dynamicImport (proposal) | import('./guy').then(a) |
exportNamespaceFrom (proposal) | export * as ns from "mod" |
functionSent (proposal) | function.sent |
logicalAssignment (proposal) | a &&= b |
moduleStringNames (proposal) | import { "😄" as smile } from "emoji"; |
nullishCoalescingOperator (proposal) | a ?? b |
numericSeparator (proposal) | 1_000_000 |
objectRestSpread (proposal) | var a = { b, ...c }; |
optionalCatchBinding (proposal) | try {throw 0;} catch{do();} |
optionalChaining (proposal) | a?.b |
privateIn (proposal) | #p in obj |
topLevelAwait (proposal) | await promise in modules |
Plugins options
History
| Version | Changes |
|---|---|
7.19.0 | The syntaxType option of the recordAndTuple plugin defaults to hash; added allowCallParenthesized option for the decorators plugin. |
7.17.0 | Added @@ and ^^ to the topicToken option of the hack pipeline operator |
7.16.0 | Added disallowAmbiguousJSXLike for typescript plugin. Added ^ to the topicToken option of the hack pipeline operators |
7.14.0 | Added dts for typescript plugin |
NOTE: When a plugin is specified multiple times, only the first options are considered.
decorators:decoratorsBeforeExport(boolean)// decoratorsBeforeExport: true @dec export class C {} // decoratorsBeforeExport: false export @dec class C {}allowCallParenthesized(boolean, defaults totrue)When
false, disallow decorators in the@(...)()form in favor of@(...()). The stage 3 decorators proposal usesallowCallParenthesized: false.
pipelineOperator:proposal(required, accepted values:minimal,fsharp,hack,(deprecated)) There are several different proposals for the pipeline operator. This option chooses which proposal to use. See plugin-proposal-pipeline-operator for more information, including a table comparing their behavior.smarttopicToken(required whenproposalishack, accepted values:%,#,^,@@,^^) Thehackproposal uses a “topic” placeholder in its pipe. There are two different choices for this topic placeholder. This option chooses what token to use to refer to the topic.topicToken: "#"is incompatible withrecordAndTuplewithsyntaxType: "hash". See plugin-proposal-pipeline-operator for more information.
recordAndtuple:syntaxType(hashorbar, defaults tohash) There are two syntax variants forrecordAndTuple. They share exactly same runtime semantics.
See Ergonomics ofSyntaxType Record Example Tuple Example "hash"#{ a: 1 }#[1, 2]"bar"{| a: 1 |}[|1, 2|]#{}/#[]for more information.
flow:all(boolean, default:false) Some code has different meaning in Flow and in vanilla JavaScript. For example,foo<T>(x)is parsed as a call expression with a type argument in Flow, but as a comparison (foo < T > x) accordingly to the ECMAScript specification. By default,babel-parserparses those ambiguous constructs as Flow types only if the file starts with a// @flowpragma. Set this option totrueto always parse files as if// @flowwas specified.
typescriptdts(boolean, defaultfalse) This option will enable parsing within a TypeScript ambient context, where certain syntax have different rules (like.d.tsfiles and insidedeclare moduleblocks). Please see https://www.typescriptlang.org/docs/handbook/declaration-files/introduction.html and https://basarat.gitbook.io/typescript/type-system/intro for more information about ambient contexts.disallowAmbiguousJSXLike(boolean, defaultfalse) Even when thejsxplugin is not enabled, this option disallows using syntax that would be ambiguous with JSX (<X> ytype assertions and<X>() => {}type arguments). It matches thetscbehavior when parsing.mtsand.mjsfiles.
Error codes
History
| Version | Changes |
|---|---|
v7.14.0 | Added error codes |
Error codes are useful for handling the errors thrown by @babel/parser.
There are two error codes, code and reasonCode.
code- Rough classification of errors (e.g.
BABEL_PARSER_SYNTAX_ERROR,BABEL_PARSER_SOURCETYPE_MODULE_REQUIRED).
- Rough classification of errors (e.g.
reasonCode- Detailed classification of errors (e.g.
MissingSemicolon,VarRedeclaration).
- Detailed classification of errors (e.g.
Example of using error codes with errorRecovery:
const { parse } = require("@babel/parser");
const ast = parse(`a b`, { errorRecovery: true });
console.log(ast.errors[0].code); // BABEL_PARSER_SYNTAX_ERROR
console.log(ast.errors[0].reasonCode); // MissingSemicolon
FAQ
Will the Babel parser support a plugin system?
Previous issues: #1351, #6694.
We currently aren't willing to commit to supporting the API for plugins or the resulting ecosystem (there is already enough work maintaining Babel's own plugin system). It's not clear how to make that API effective, and it would limit our ability to refactor and optimize the codebase.
Our current recommendation for those that want to create their own custom syntax is for users to fork the parser.
To consume your custom parser, you can add a plugin to your options to call the parser via its npm package name or require it if using JavaScript,
const parse = require("custom-fork-of-babel-parser-on-npm-here");
module.exports = {
plugins: [
{
parserOverride(code, opts) {
return parse(code, opts);
},
},
],
};