@iarna/rtf-to-html

Convert RTF to HTML in pure JavaScript.

const rtfToHTML = require('@iarna/rtf-to-html')

fs.createReadStream('example.rtf').pipe(rtfToHTML((err, html) => {
  // …
})
rtfToHTML.fromStream(fs.createReadStream('example.rtf'), (err, html) => {
  // …
})
rtfToHTML.fromString('{\\rtf1\\ansi\\b hi there\\b0}', (err, html) => {
  console.log(html)
  // prints a document containing:
  // <p><strong>hi there</strong></p>
})

This is built on rtf-parser and shares its limitations.

This generates complete HTML documents from RTF documents. It does not currently have the facility to work on snippets of either.

Supported features:

Paragraph detection (results in  tags) plus empty paragraph trimming.
Font (as font-family: Font Name, Font Family). Font families in RTF don't map perfectly to HTML. This is the mapping we currently use:
- roman: serif
- swiss: sans-serif
- script: cursive
- decor: fantasy
- modern: sans-serif
- tech: monospace
- bidi: serif
Font size (as font-size: #pt)
Bold (as )
Italic (as )
Underline (as )
Strikethrough (as <s>)
Superscript (as )
Subscript (as )
Foreground color (as color: rgb(#,#,#))
Background color (as color: rgb(#,#,#))
Paragraph first-line indents (as text-indent: #pt)
Idented regions (as padding-left: #pt)
Text alignment: left, right, center, justify (as text-align:)

rtfToHTML([opts], cb) → WritableStream

opts - Optional options to pass to the HTML generator. See the section on Options for details.
cb - A callback accepting (err, html), see the section on the Callback for details.

Returns a WritableStream that you can pipe into.

rtfToHTML.fromStream(stream[, opts], cb)

stream - A readable stream that should contain RTF.
opts - Optional options to pass to the HTML generator. See the section on Options for details.
cb - A callback accepting (err, html), see the section on the Callback for details.

rtfToHTML.fromString(string[, opts], cb)

string - A string containing RTF.
opts - Optional options to pass to the HTML generator. See the section on Options for details.
cb - A callback accepting (err, html), see the section on the Callback for details.

Callback

rtfToHTML returns HTML produced from RTF using a standard Node.js style callback, which should accept the following arguments: `(err, html)`.

If we encounter an error in parsing then it will be set in err. Otherwise the resulting HTML will be in html.

Options

Options are always optional. You can configure how HTML is generated with the following:

paraBreaks - (Defaults to \n\n) Inserted between resulting paragraphs.
paraTag - (Defaults to p) The tagname to use for paragraphs.

template - A function that is used to generate the final HTML document, defaults to:

function outputTemplate (doc, defaults, content) {
  return `<!DOCTYPE html>
<html>
  <head>
    <meta charset="UTF-8">
    <style>
    body {
      margin-left: ${doc.marginLeft / 20}pt;
      margin-right: ${doc.marginRight / 20}pt;
      margin-top: ${doc.marginTop / 20}pt;
      margin-bottom: ${doc.marginBottom / 20}pt;
      font-size: ${defaults.fontSize / 2}pt;
      text-indent: ${defaults.firstLineIndent / 20}pt;
    }
    </style>
  </head>
  <body>
    ${content.replace(/\n/, '\n    ')}
  </body>
</html>`
}

You can also configure some of the starting (default) state of the output formatting with:

disableFonts - Defaults to true. If you set this to false then we'll output font change information when we encounter it. This is a bit broken due to our not supporting styles.
fontSize - Defaults to the document-wide declared font size, or if that's missing, 24.
bold - Defaults to false
italic - Defaults to false
underline - Defaults to false
strikethrough - Defaults to false
foreground - Defaults to {red: 0, blue: 0, green: 0}
background - Defaults to {red: 255, blue: 255, green: 255}
firstLineIndent - Defaults to the document-wide value, or if that's missing, 0. This is how far to indent the first line of new paragraphs.
indent: Defaults to 0
align: Defaults to left.
valign: Defaults to normal

const rtfToHTML = require('rtf-to-html/rtf-to-html.js')

rtfToHTML(doc[, opts]) → HTML

This is internally how the other interfaces are implemented. Unlike the other interfaces, this one is synchronous.

doc - A parsed RTF document as produced by the rtf-parser library.
opts - Optional options, see the section on Options for details.

robert843/rtf-to-html