Disallow unrecognized character references (unrecognized-char-ref)

HTML5 defines a set of named character references (sometimes called HTML entities) which can be used as &name; where name is the entity name, e.g. ' (') or & (&). Only entities from the list can be used. Each entity is case sensitive but some entities is defined in multiple casing variants, e.g. both © and © are acceptable.

This rule ignores numerical entities such as — or —.

Rule details

Examples of incorrect code for this rule:

<p>&foobar;</p>
error: Unrecognized character reference "&foobar;" (unrecognized-char-ref) at inline:1:4:
> 1 | <p>&foobar;</p>
    |    ^^^^^^^^


1 error found.

Examples of correct code for this rule:

<p>&amp;</p>

Options

This rule takes an optional object:

{
  "ignoreCase": false,
  "requireSemicolon": true
}

ignoreCase

If set to true this rule ignores the casing of the entity.

With this option disabled the following is incorrect:

<p>&Amp;</p>
error: Unrecognized character reference "&Amp;" (unrecognized-char-ref) at inline:1:4:
> 1 | <p>&Amp;</p>
    |    ^^^^^


1 error found.

With this option enabled the following is correct:

<p>&Amp;</p>

requireSemicolon

By default named character references are terminated by a semicolon (;) but for legacy compatibility some are listed without.

If set to false legacy variants without semicolon are allowed.

With this option enabled the following is incorrect:

<p>&copy</p>
error: Character reference "&copy" must be terminated by a semicolon (unrecognized-char-ref) at inline:1:4:
> 1 | <p>&copy</p>
    |    ^^^^^


1 error found.

With this option disabled the following is correct:

<p>&copy</p>

Attribute values with a ? is treated as a querystring and unless terminated with a ; is considered to be a parameter and not a character reference, e.g. the following is always valid even if &bar looks like a reference without semicolon to terminate it:

<a href="foo.php?foo=1&bar=2">...</a>

Version history