Learn about our RFC process, Open RFC meetings & more.Join in the discussion! »


0.7.6 • Public • Published


Generic JSDoc-like comment parser. This library is not intended to be documentation generator, but rather composite unit for it.

npm install comment-parser

Module provides parse(s: string[, opts: object]): object[] function which takes /** ... */ comment string and returns array of objects with parsed data.

It is not trying to detect relations between tags or somehow recognize their meaning. Any tag can be used, as long as it satisfies the format.

 * Singleline or multiline description text. Line breaks are preserved.
 * @some-tag {Type} name Singleline or multiline description text
 * @some-tag {Type} name.subname Singleline or multiline description text
 * @some-tag {Type} name.subname.subsubname Singleline or
 * multiline description text
 * @some-tag {Type} [optionalName=someDefault]
 * @another-tag

this would be parsed into following

  "tags": [{
    "tag": "some-tag",
    "type": "Type",
    "name": "name",
    "optional": false,
    "description": "Singleline or multiline description text",
    "line": 3,
    "source": "@some-tag {Type} name Singleline or multiline description text"
  }, {
    "tag": "some-tag",
    "type": "Type",
    "name": "name.subname",
    "optional": false,
    "description": "Singleline or multiline description text",
    "line": 4,
    "source": "@some-tag {Type} name.subname Singleline or multiline description text"
  }, {
    "tag": "some-tag",
    "type": "Type",
    "name": "name.subname.subsubname",
    "optional": false,
    "description": "Singleline or\nmultiline description text",
    "line": 5,
    "source": "@some-tag {Type} name.subname.subsubname Singleline or\nmultiline description text"
  }, {
    "tag": "some-tag",
    "type": "Type",
    "name": "optionalName",
    "optional": true,
    "description": "",
    "line": 7,
    "default": "someDefault",
    "source": "@some-tag {Type} [optionalName=someDefault]"
  }, {
    "tag": "another-tag",
    "name": "",
    "optional": false,
    "type": "",
    "description": "",
    "line": 8,
    "source": "@another-tag"
  "line": 0,
  "description": "Singleline or multiline description text. Line breaks are preserved.",
  "source": "Singleline or multiline description text. Line breaks are preserved.\n\n@some-tag {Type} name Singleline or multiline description text\n@some-tag {Type} name.subname Singleline or multiline description text\n@some-tag {Type} name.subname.subsubname Singleline or\nmultiline description text\n@another-tag"

Below are examples of acceptable comment formats

/** online comment */
/** first line
 * second line */
   No *s on middle lines is acceptable too
   which might be convenient for writing big
   chunks of text.
 * keeping *s on some lines
 * would work either
   left bound is determined by opening marker position
   and white space will be trimmed as there was '* '

Comments starting with /*** and /* are ignored.

Also you can parse entire file with parse.file('path/to/file', callback) or acquire an instance of Transform stream with parse.stream().


dotted_names: boolean

By default dotted names like name.subname.subsubname will be expanded into nested sections, this can be prevented by passing opts.dotted_names = false.

trim: boolean

Set this to false to avoid the default of trimming whitespace at the start and end of each line.

join: string | number | boolean

If the following lines of a multiline comment do not start with a star, join will have the following effect on tag source (and description) when joining the lines together:

  1. If a string, use that string in place of the leading whitespace (and avoid newlines).
  2. If a non-zero number (e.g., 1), do no trimming and avoid newlines.
  3. If undefined, false, or 0, use the default behavior of not trimming but adding a newline.
  4. Otherwise (e.g., if join is true), replace any leading whitespace with a single space and avoid newlines.

Note that if a multi-line comment has lines that start with a star, these will be appended with initial whitespace as is and with newlines regardless of the join setting.

Note also that the comment source will not be changed by this setting.

fence: string | RegExp | ((source: string) => boolean)

Set to a string or regular expression to toggle state upon finding an odd number of matches within a line. Defaults to ```.

If set to a function, it should return true to toggle fenced state; upon returning true the first time, this will prevent subsequent lines from being interpreted as starting a new jsdoc tag until such time as the function returns true again to indicate that the state has toggled back.

parsers: Parser[] (Custom parsers)

In case you need to parse tags in different way you can pass opts.parsers = [parser1, ..., parserN], where each parser is function name(str:String, data:Object):{source:String, data:Object}.

Each parser function takes string left after previous parsers applied and data produced by them. And returns null or {source: '', data:{}} where source is consumed substring and data is a payload with tag node fields.

Tag node data is build by merging result bits from all parsers. Here is some example that is not doing actual parsing but is demonstrating the flow:

 * Source to be parsed below
 * @tag {type} name Description
parse(source, {parsers: [
    // takes entire string
    function parse_tag(str, data) {
        return {source: ' @tag', data: {tag: 'tag'}};
    // parser throwing exception
    function check_tag(str, data) {
        if (allowed_tags.indexOf(data.tag) === -1) {
            throw new Error('Unrecognized tag "' + data.tag + '"');
    // takes the rest of the string after ' @tag''
    function parse_name1(str, data) {
        return {source: ' name', data: {name: 'name1'}};
    // alternative name parser
    function parse_name2(str, data) {
        return {source: ' name', data: {name: 'name2'}};

This would produce following:

  "tags": [{
    "tag": "tag",
    "errors": [
      "check_tag: Unrecognized tag \"tag\""
    "name": "name2",
    "optional": false,
    "type": "",
    "description": "",
    "line": 2,
    "source": "@tag {type} name Description"
  "line": 0,
  "description": "Source to be parsed below",
  "source": "Source to be parsed below\n@tag {type} name Description"


One may also convert comment-parser JSON structures back into strings using the stringify method (stringify(o: (object|Array) [, opts: object]): string).

This method accepts the JSON as its first argument and an optional options object with an indent property set to either a string or a number that will be used to determine the number of spaces of indent. The indent of the start of the doc block will be one space less than the indent of each line of asterisks for the sake of alignment as per usual practice.

The stringify export delegates to the specialized methods stringifyBlocks, stringifyBlock, and stringifyTag, which are available on the stringify function object.


comment-parser is CommonJS module and was primarely designed to be used with Node. Module index.js includes stream and file functionality. Use prser-only module in browser comment-parser/parse.js


> npm info --registry https://registry.npmjs.org comment-parser contributors


npm i comment-parser

DownloadsWeekly Downloads






Unpacked Size

35.2 kB

Total Files


Last publish


  • avatar