Comment Serializer parses a source string for documentation comment blocks and returns a serialized object. It is language and comment syntax style agnostic. It configured to support most documentation comment block styles.
Try out an example on RunKit.
source-code.txt:
//!
// This is the general description.
//
// ^title Button
// ^markup <button class="btn">My Button</button>
///
Blah, blah, blah... Some code...
my-serializer.js
const { readFileSync } = require('fs');
const serializer = require('comment-serializer');
const src = readFileSync('source-code.txt', { encoding: 'utf8' });
const mySerializer = serializer({
tokens: {
commentBegin: '//!',
commentLinePrefix: '//',
tagPrefix: '^',
commentEnd: '///',
},
});
const result = mySerializer(src);
- Required: No
- Type:
object
Custom tag parsers. An object of functions, keyed by the name of the tag to be parsed. The function receives the tag's content and must return the parsed value as a string
.
When no parsers are provided, tags are serialized into tag
and value
pairs, but their values are not parsed in any way.
You can find some examples of custom parsers in examples/custom-parsers/parsers.js
Source to parse:
/**
* @mySpecialTag This value is special!
*/
The custom parser:
const mySerializer = serializer({
parsers: {
mySpecialTag: (value) => value.toUpperCase(),
},
});
Result:
[
{
line: 1,
source: '/**\n * @mySpecialTag This value is special!\n */',
context: '',
content: '@mySpecialTag This value is special!',
preface: '',
tags: [
{
line: 2,
tag: 'mySpecialTag',
value: 'This value is special!',
valueParsed: 'THIS VALUE IS SPECIAL!',
source: '@mySpecialTag This value is special!',
},
],
},
];
A more advanced example:
/**
* @mySpecialTag
* - item-1
* - item-2
* - item-3
*/
const mySerializer = serializer({
parsers: {
mySpecialTag: (value) => {
const match = value.match(/\s*[-*]\s+[\w-_]*/g);
return [
{
type: 'items',
value: match.map((item) => item.trim().replace(/[-*]\s/, '')),
},
];
},
},
});
Result:
[
{
line: 1,
source:
'/**\n * @mySpecialTag\n * - item-1\n * - item-2\n * - item-3\n */',
context: '',
content: '@mySpecialTag\n - item-1\n - item-2\n - item-3',
preface: '',
tags: [
{
line: 2,
tag: 'mySpecialTag',
value: '\n - item-1\n - item-2\n - item-3',
valueParsed: [
{
type: 'items',
value: ['item-1', 'item-2', 'item-3'],
},
],
source: '@mySpecialTag\n - item-1\n - item-2\n - item-3',
},
],
},
];
Errors that occur while parsing a tag's value are caught. When this happens, tag's the valueParsed
property will be an empty array
, and the error object is added to the error
property.
Example:
{
line: 2,
tag: 'mySpecialTag',
value: '\n - item-1\n - item-2\n - item-3',
valueParsed: [],
source: '@mySpecialTag\n - item-1\n - item-2\n - item-3',
error: { Error }
}
- Required: No
- Type:
object
Customize the comment delimiters. The default tokens use JSDoc comment block syntax.
- Required: No
- Type:
string
- Default:
'/**'
The delimiter marking the beginning of a comment block.
- Required: No
- Type:
string
- Default:
'*'
The delimiter marking a new line in the body of a comment block.
- Required: No
- Type:
string
- Default:
'@'
The delimiter marking the start of a tag in the comment body.
- Required: No
- Type:
string
- Default:
'*/'
The delimiter marking the end of a comment block.
Comment delimiters:
/** <- commentBegin
* <- commentLinePrefix
* @tag <- tagPrefix (the "@" symbol)
*/ <- commentEnd
While most documentation comment styles should be supported, there are a few rules around the syntax:
-
The
commentBegin
,commentLinePrefix
,commentEnd
andtagPrefix
delimiter tokens must be distinct from each other. -
Delimiters should not rely on a whitespace character as a delimiter. For example, the following style would not be supported:
/**
An unsupported style.
@tag
*/