Web Foundations

Attribute Selector (CSS Selector)

By Adam Roberts


An attribute selector will match elements on the basis of either the presence of an attribute, or the exact or partial match of an attribute value. Attribute selectors were introduced in CSS2, and CSS3 added a few more.

Attribute selectors are delimited by square brackets; the simplest form of an attribute selector consists of an attribute name surrounded by square brackets:

[href] {
  ⋮ declarations

This example selector matches any element that has an href attribute. It also contains an implied universal selector, and is equivalent to *[href].

Here’s another example:

a[href] {
  ⋮ declarations

This selector matches any a element that has an href attribute, so it matches a hypertext link, but not a named anchor.

Attribute selectors can also specify a value, or a partial value, to match. The values must be strings, in which case they’re surrounded by single or double quotes, or identifiers, without quotes. All the examples below use strings.

Case Sensitivity

The value specified in an attribute selector is case sensitive if the attribute value in the markup language is case sensitive. Thus, values for id and class attributes in HTML are case sensitive, while values for lang and type attributes are not.

XHTML, when served as XML, is always case sensitive; see Differences Between HTML and XHTML for more on this.

It’s not always easy to remember which HTML attributes are case sensitive and which aren’t. It’s usually best to assume that everything is case sensitive, but don’t rely on it!

We can use the = operator to have an attribute selector match elements that have specific values:

input[type="submit"] {
  ⋮ declarations

This selector matches any input element that has a type attribute with a value equal to “submit”.

Default Attributes

Attribute selectors can only match attributes and values that exist in the document tree, and there’s no guarantee that a default value specified in a DTD (or similar) can be matched. For instance, in HTML, the default value for a form element’s method attribute is "get", but you can’t rely on a selector like form[method="get"] to select a form element with the start tag <form action="comment.php">.

Attribute Selectors for IDs

Note that [id="foo"] isn’t equivalent to #foo. Although both selectors would match the same element in HTML, there’s a significant difference between their levels of specificity.

We can use the |= operator to cause an attribute selector to match elements which have an attribute containing a list of hyphenated words that begin with a specific value:

[hreflang|="en"] {
  ⋮ declarations

This example selector matches any element that has an hreflang attribute containing a value of “en“, whether or not that value is followed by a hyphen and more characters. In other words, it matches hreflang attribute values of “en“, “en-US“, “en-GB“, and so on. This selector syntax was intended to allow language subcode matches.

Hyphen or No Hyphen?

All supporting browsers allow a hyphen to appear in the value in a selector like [hreflang|="en"]. It’s unclear whether or not this is illegal, because the CSS specification doesn’t contain any guidelines to help us deal with this situation.

We can use the ~= operator to make an attribute selector match elements that have an attribute that contains a list of space-separated words, one of which is the specified value:

[class~="warning"] {
  ⋮ declarations

This selector matches any HTML element with a class attribute that contains a space-separated list of words, one of which is "warning". So it matches <p class="warning"> and <strong class="important warning"> and <div class="warning highlight">, but not <p class="my-warning"> or <ul class="warnings">.


This selector matches all input elements with a type attribute that’s equal to "submit" (in other words, submit buttons):

input[type="submit"] {
  ⋮ declarations

No Reader comments