score.dom¶
This module provides a convenient wrapper for DOM nodes. Its API resembles that of jQuery, since this has become the best known DOM manipulation API over time.
Quickstart¶
<script src="score.init.js"></script>
<script src="score.dom.js"></script>
<script>
(function() {
var romans = score.dom('.roman');
score.dom('.home').append(romans);
})();
</script>
Details¶
Constructor¶
The “constructor” [1] will return an object deriving from score.dom.proto
.
It accepts either of the following:
- Nothing: Creates an empty node list.
score.dom()
- Selector: Selects all nodes matching the selector globally
score.dom('body')
- NodeList object:
score.dom(document.getElementsByTagName('a'))
- A
score.dom
object:
var bodyNode = score.dom('body');
score.dom(bodyNode).addClass('foo');
[1] | The term constructor is actually incorrect, as score.dom is just a
normal function. |
Creating new nodes¶
The constructor above is for wrapping already existing DOM nodes. It is also
possible to create score.dom objects with brand new nodes by using the
create()
function:
var node = score.dom.create('div');
score.dom Object¶
Whenever you call score.dom()
, you will receive an array-like object
containing a list of nodes. This object actually inherits the Array prototype,
allowing you to use all array features:
var nodes = score.dom('.spam');
nodes.length; // 3
nodes[0]; // A native HTMLDivElement: <div class="spam">...</div>
There are a few exceptions: we are overriding the forEach()
, filter()
,
map()
and reduce()
functions to provide score.dom objects in the
callback:
nodes.forEach(function(node) {
// node is a score.dom object
console.log(node.attr('class'));
});
For convenience filter()
also accepts a selector instead of a callback
function:
nodes.filter(':not(.spam)');
Since score.dom objects are actually arrays, all operations are always performed on all nodes in your array. This might come as a surprise in certain cases, where jQuery is a bit inconsistent:
// Remove all nodes, that hava a "spam" class, from the document:
score.dom('.spam').detach();
// Select all child nodes of all ".knight" nodes, i.e. the nodes
// ".foo" and ".bar" in the following document:
// <div class="knight"><span class="foo"></span></div>
// <div class="knight"><span class="bar"></span></div>
score.dom('.knight').children();
// Select all parent nodes of all ".cheese" nodes; the resulting object
// will contain the "#cheese-shop" *twice*:
// <div id="cheese-shop">
// <div class="customer"></div>
// <div class="customer"></div>
// </div>
score.dom('.customer').parent().length; // 2
score.dom('.customer').parent().uniq().length; // 1
Safety Measures¶
Apart from simplifying the complex DOM API, this module also tries to aid development by throwing errors, whenever it assumes that something might have gone wrong. Currently, we have two distinct checks for some of the operations available. Note, that the precautions are tied to operations, not whole functions: Some usage of the function might have a different set of constraints than another.
Minimum Length 1¶
Some operations intend to change a node. These operations are marked (min1) and will fail if the score.dom object is empty:
// create a new node:
var banana = score.dom.create('span').addClass('banana');
// select an existing node from the document
var fruits = score.dom('#fruits')
// What we're not realizing at this point is that our colleague has changed
// the ID of the "fruits"-node to "weapons", and that our fruits-variable
// is nothing but an empty array. This is why the next function will cause
// an Error to be thrown:
fruits.append(banana); // Error: "Empty list"
Single Node Operations¶
Some operations are marked as Single Node Operations (sno for short). These operations will fail if the score.dom object contains more than one element:
try {
var nodes = score.dom('.spam');
if (nodes.length > 1) {
// this will throw an error since we
// have more than one node:
nodes.text()
}
} catch (e) {
console.log(e); // "Attempting Single-Node-Operation on multiple nodes"
}
Assertion¶
Sometimes it can be handy to check if your query will return exactly one result.
Therefor assertOne()
will throw errors on empty or multiple results in a query:
try {
score.dom('#fish-tank').find('.fish').assertOne();
} catch (e) {
console.log(e); // "Multiple results found"
}
try {
score.dom('#elusive-fish').assertOne();
} catch (e) {
console.log(e); // "No result found"
}
// Since `assertOne()` doesn't actually do anything,
// it will simply pass the object:
var oneMoreMint = score.dom('#mint');
oneMoreMint === oneMoreMint.assertOne(); // true
Filtering¶
If you have a score.dom
object, you can reduce its list of nodes using the
following methods:
eq(index)
will return a newscore.dom
object containing a single node, the one at the given index. Will throw an Error, if the index it out of range.The dynamic value
x.first
returns the same asx.eq(0)
,x.last
returns the same asx.eq(x.length - 1)
:score.dom('.knight').first; // The first knight score.dom('.samurai').last; // The last samurai score.dom('#cheese-shop').find('.cheese').first; // throws an Error
The function
uniq()
will remove duplicates from your node list:score.dom('.customer').parent().uniq();
Cloning¶
The represented Nodes can be duplicated using clone()
:
var spams = score.dom('.spam');
spams.first.parent().append(spams.clone());
The function makes deep clones of all elements unless called with false
:
var spams = score.dom('.spam');
var shallowCopies = spams.clone(false);
Querying¶
You can query, if all nodes in your list match a given selector using
matches()
(min1):
var spams = score.dom('.spam');
spams.matches('.spam');
Node Operations¶
There are two operations you can perform on individual nodes:
text()
will return the textContent of the node (sno) or set the textContent of all nodes to a given value:// setting the text content: score.dom('body').text('hello world'); // retrieving the text content (sno): score.dom('body').text();
attr()
does the same for the value of an attribute:// setting an attribute: score.dom('#parrot').attr('data-state', 'deceased'); // getting the value of an attribute (sno) // (returns null if attribute does not exist): score.dom('.customer').attr('data-state'); // removing an attribute: score.dom('.customer').attr('data-state', null);
Restructuring¶
You can remove nodes from the document using detach()
, removing them from
the DOM. You can then attach them beneath another given node using
prepend()
or append()
, depending on whether they should be inserted at the
beginning, or the end of the children list. Both functions are single node
operations and will throw an error, if the score.dom object
they were called on does not contain exactly one node:
score.dom('.parrot').detach();
score.dom('.fruits').append(score.dom.fromString('<li>Banana</li>'));
score.dom('.fruits').prepend(score.dom.fromString('<li>Carrot</li>'));
score.dom('.fruits').children().first.text() // 'Carrot'
Both functions also accept a second node that can serve as an anchor, if the insert operation should be performed at a specific position in the children list:
// insert spam *after* eggs
score.dom('.meal').append(spam, eggs);
// insert spam *before* eggs
score.dom('.meal').prepend(spam, eggs);
Traversal¶
The function parent()
returns a new score.dom
containing each node’s
parent. children()
returns a new score.dom
containing all child nodes
of every node.
find()
will find all nodes beneath the original nodes matching given
selector.
closest()
queries the document upward until the given selector matches.
This is done for each node in the original list.
// Assuming the following document:
// <div id="top>
// <div class="lvl2">
// <span class="bottom">
// <span class="bottom">
// </div>
// <div class="lvl2">
// <span class="bottom">
// <span class="bottom">
// </div>
// </div>
var bottoms = score.dom('.bottom');
bottoms.length === 4;
var bottomParents = bottoms.parent();
bottomParents.length === 4;
bottomParents.hasClass('lvl2');
bottomParents[0] === bottomParents[1];
bottomParents[2] === bottomParents[3];
bottomParents[0] !== bottomParents[2];
var tops = bottomParents.closest('#top');
tops.length == 4;
tops[0] === tops[1];
tops[0] === tops[2];
tops[0] === tops[3];
var secondLevels = score.dom('#top').find('.lvl2');
secondLevels.length === 2;
CSS Class Manipulation¶
The module allows adding/removing css classes using the usual method names:
if (score.dom('body').hasClass('spam')) {
score.dom('.knight').addClass('ni');
score.dom('#cheese-shop').removeClass('cheese');
score.dom('.self-defense').toggleClass('fruit');
}
Note that hasClass()
will only return true
, if all nodes have the
given css class.
Acknowledgments¶
Many thanks to BrowserStack and Travis CI for providing automated tests for our open source projects! We wouldn’t be able to maintain our high quality standards without them!