Element: querySelectorAll() method
The Element
method querySelectorAll()
returns a static (not live) NodeList
representing a list of elements
matching the specified group of selectors which are descendants of the element on which
the method was called.
Syntax
querySelectorAll(selectors)
Parameters
selectors
-
A string containing one or more selectors to match. This string must be a valid CSS selector string; if it isn't, a
SyntaxError
exception is thrown.Note that the HTML specification does not require attribute values to be valid CSS identifiers. If a
class
orid
attribute value is not a valid CSS identifier, then you must escape it before using it in a selector, either by callingCSS.escape()
on the value, or using one of the techniques described in Escaping characters. See Escaping attribute values for an example.The selectors are applied to the entire document, not just the particular element on which
querySelectorAll()
is called. To restrict the selector to the element on whichquerySelectorAll()
is called, include the:scope
pseudo-class at the start of the selector. See the selector scope example.
Return value
A non-live NodeList
containing one Element
object for
each descendant node that matches at least one of the specified selectors.
Note: If the specified selectors
include a CSS pseudo-element, the returned list
is always empty.
Exceptions
SyntaxError
DOMException
-
Thrown if the syntax of the specified
selectors
string is not valid.
Examples
Getting all elements with a custom data value
This example uses the attribute selector to select multiple elements with a data-name
data attribute that contains "funnel-chart-percent".
<section class="box" id="sect1">
<div data-name="funnel-chart-percent1">10.900%</div>
<div data-name="funnel-chart-percent2">3700.00%</div>
<div data-name="funnel-chart-percent3">0.00%</div>
</section>
const refs = [
...document.querySelectorAll(`[data-name*="funnel-chart-percent"]`),
];
Obtaining a list of matches
To obtain a NodeList
of all of the <p>
elements
contained within the element "myBox"
:
const matches = myBox.querySelectorAll("p");
This example returns a list of all <div>
elements within
"myBox"
with a class of either "note
" or "alert
":
const matches = myBox.querySelectorAll("div.note, div.alert");
Here, we get a list of the document's <p>
elements whose immediate
parent element is a <div>
with the class "highlighted"
and
which are located inside a container whose ID is "test"
.
const container = document.querySelector("#test");
const matches = container.querySelectorAll("div.highlighted > p");
This example uses an attribute selector to return a list of the <iframe>
elements in the document
that contain an attribute named "data-src"
:
const matches = document.querySelectorAll("iframe[data-src]");
Here, an attribute selector is used to return a list of the list items contained within
a list whose ID is "userlist"
which have a "data-active"
attribute whose value is "1"
:
const container = document.querySelector("#userlist");
const matches = container.querySelectorAll("li[data-active='1']");
Accessing the matches
Once the NodeList
of matching elements is returned, you can examine it
just like any array. If the array is empty (that is, its length
property is
0
), then no matches were found.
Otherwise, you can use standard array notation to access the contents of the list. You can use any common looping statement, such as:
const highlightedItems = userList.querySelectorAll(".highlighted");
highlightedItems.forEach((userItem) => {
deleteUser(userItem);
});
Note: NodeList
is not a genuine array, that is to say it doesn't
have array methods like slice
, some
, map
, etc. To convert it into an array, try
Array.from(nodeList)
.
Selector scope
The querySelectorAll()
method applies its selectors to the whole document: they are not scoped to the element on which the method is called. To scope the selectors, include the :scope
pseudo-class at the start of the selector string.
HTML
In this example the HTML contains:
- two buttons:
#select
and#select-scope
- three nested
<div>
elements:#outer
,#subject
, and#inner
- a
<pre>
element which the example uses for output.
<button id="select">Select</button>
<button id="select-scope">Select with :scope</button>
<div id="outer">
#outer
<div id="subject">
#subject
<div id="inner">#inner</div>
</div>
</div>
<pre id="output"></pre>
JavaScript
In the JavaScript, we first select the #subject
element.
When the #select
button is pressed, we call querySelectorAll()
on #subject
, passing "#outer #inner"
as the selector string.
When the #select-scope
button is pressed, we again call querySelectorAll()
on #subject
, but this time we pass ":scope #outer #inner"
as the selector string.
const subject = document.querySelector("#subject");
const select = document.querySelector("#select");
select.addEventListener("click", () => {
const selected = subject.querySelectorAll("#outer #inner");
output.textContent = `Selection count: ${selected.length}`;
});
const selectScope = document.querySelector("#select-scope");
selectScope.addEventListener("click", () => {
const selected = subject.querySelectorAll(":scope #outer #inner");
output.textContent = `Selection count: ${selected.length}`;
});
Result
When we press "Select", the selector selects all elements with an ID of inner
that also have an ancestor with an ID of outer
. Note that even though #outer
is outside the #subject
element, it is still used in selection, so our #inner
element is found.
When we press "Select with :scope", the :scope
pseudo-class restricts the selector scope to #subject
, so #outer
is not used in selector matching, and we don't find the #inner
element.
Escaping attribute values
This example shows that if an HTML document contains an id
which is not a valid CSS identifier, then we must escape the attribute value before using it in querySelectorAll()
.
HTML
In the following code, a <div>
element has an id
of "this?element"
, which is not a valid CSS identifier, because the "?"
character is not allowed in CSS identifiers.
We also have three buttons, and a <pre>
element for logging errors.
<div id="container">
<div id="this?element"></div>
</div>
<button id="no-escape">No escape</button>
<button id="css-escape">CSS.escape()</button>
<button id="manual-escape">Manual escape</button>
<pre id="log"></pre>
CSS
div {
background-color: blue;
margin: 1rem 0;
height: 100px;
width: 200px;
}
JavaScript
All three buttons, when clicked, try to select the <div>
, and then set its background color to a random value.
- The first button uses the
"this?element"
value directly. - The second button escapes the value using
CSS.escape()
. - The third button explicitly escapes the
"?"
character using a backslash. Note that we must also escape the backslash itself, using another backslash, like:"\\?"
.
const container = document.querySelector("#container");
const log = document.querySelector("#log");
function random(number) {
return Math.floor(Math.random() * number);
}
function setBackgroundColor(id) {
log.textContent = "";
try {
const elements = container.querySelectorAll(`#${id}`);
const randomColor = `rgb(${random(255)} ${random(255)} ${random(255)})`;
elements[0].style.backgroundColor = randomColor;
} catch (e) {
log.textContent = e;
}
}
document.querySelector("#no-escape").addEventListener("click", () => {
setBackgroundColor("this?element");
});
document.querySelector("#css-escape").addEventListener("click", () => {
setBackgroundColor(CSS.escape("this?element"));
});
document.querySelector("#manual-escape").addEventListener("click", () => {
setBackgroundColor("this\\?element");
});
Result
Clicking the first button gives an error, while the second and third buttons work properly.
Specifications
Specification |
---|
DOM Standard # ref-for-dom-parentnode-queryselectorall① |
Browser compatibility
BCD tables only load in the browser
See also
- Locating DOM elements using selectors
- Attribute selectors in the CSS Guide
- Attribute selectors in the MDN Learning Area
Element.querySelector()
-
Document.querySelector()
andDocument.querySelectorAll()
-
DocumentFragment.querySelector()
andDocumentFragment.querySelectorAll()