Extract text from HTML. Excludes content from metadata tags by default.
For example, script and style. Reduces multiple spaces to a single space
and trims whitespace from the start and end by default. Set preserveWhitespace
to true to disable this behavior. Optionally, replace tags with text.
Offers a much nicer out-of-the-box experience compared to striptags. See comparison here.
Single dependency on htmlparser2
export interface Replacement {
/** Tag name to match (without brackets) */
matchTag: string
/** Text to replace the tag with */
text: string
/** Is the tag self-closing? */
isSelfClosing?: boolean
}
export interface Options {
/** Exclude content from the set of tags. Defaults to all HTML metadata tags. */
excludeContentFromTags?: string[]
/** Whitespace is trimmed by default. Set this to true to preserve whitespace. */
preserveWhitespace?: boolean
/** Replace a tag with some text. Flag self-closing tags with isSelfClosing: true. */
replacements?: Replacement[]
}
// Content from the following tags are excluded by default
export const defaultExcludeContentFromTags = [
'head',
'base',
'link',
'meta',
'noscript',
'script',
'style',
'title',
]import { extractText } from 'extract-text-html'
const html = `
<!doctype html>
<html lang="en">
<head>
<meta charset="utf-8">
<link rel="stylesheet" href="https://static-production.npmjs.com/styles.74f9073cf68d3c5f4990.css" />
<title data-react-helmet="true">extracttext - npm search</title>
</head>
<body>
<h1>Some Title</h1>
<div style="font-weight: bold">Some text</div>
<script crossorigin="anonymous" src="https://static-production.npmjs.com/minicssextractbug.536095f4b1a94d2b149c.js"></script>
<script crossorigin="anonymous" src="https://static-production.npmjs.com/search/search.9fbe393f02970084bce5.js"></script>
<script>
const FOO = 'bar'
</script>
<br />
<br />
</body>
</html>
`
const extracted = extractText(html)
// Some Title Some textReplacements example usage
const html = `<b>bold <span>text</span></b>
<div>some text</div><br /><br><p>more text</p>`
const extracted = extractText(html, {
preserveWhitespace: true,
replacements: [
{ matchTag: 'br', text: '\n', isSelfClosing: true },
{ matchTag: 'b', text: '__' },
],
})
/*
__bold text__
some text
more text
*/