Ggtcf

Is there any way to get the collection of all textNode objects within a document?

getElementsByTagName() works great for Elements, but textNodes are not Elements.

Update: I realize this can be accomplished by walking the DOM - as many below suggest. I know how to write a DOM-walker function that looks at every node in the document. I was hoping there was some browser-native way to do it. After all it's a little strange that I can get all the <input>s with a single built-in call, but not all textNodes.

Update:

I have outlined some basic performance tests for each of these 6 methods over 1000 runs. getElementsByTagName is the fastest but it does a half-assed job, as it does not select all elements, but only one particular type of tag ( i think p) and blindly assumes that its firstChild is a text element. It might be little flawed but its there for demonstration purpose and comparing its performance to TreeWalker. Run the tests yourselves on jsfiddle to see the results.

1. Using a TreeWalker


2. Custom Iterative Traversal


3. Custom Recursive Traversal


4. Xpath query


5. querySelectorAll


6. getElementsByTagName

Let's assume for a moment that there is a method that allows you to get all Text nodes natively. You would still have to traverse each resulting text node and call node.nodeValue to get the actual text as you would do with any DOM Node. So the issue of performance is not with iterating through text nodes, but iterating through all nodes that are not text and checking their type. I would argue (based on the results) that TreeWalker performs just as fast as getElementsByTagName, if not faster (even with getElementsByTagName playing handicapped).

Ran each test 1000 times.

Method Total ms Average ms
--------------------------------------------------
document.TreeWalker 301 0.301
Iterative Traverser 769 0.769
Recursive Traverser 7352 7.352
XPath query 1849 1.849
querySelectorAll 1725 1.725
getElementsByTagName 212 0.212

Source for each method:

TreeWalker

function nativeTreeWalker() 
 var walker = document.createTreeWalker(
 document.body, 
 NodeFilter.SHOW_TEXT, 
 null, 
 false
 );

 var node;
 var textNodes = [];

 while(node = walker.nextNode()) 
 textNodes.push(node.nodeValue);
 

Recursive Tree Traversal

function customRecursiveTreeWalker() 
 var result = [];

 (function findTextNodes(current) 
 for(var i = 0; i < current.childNodes.length; i++) 
 var child = current.childNodes[i];
 if(child.nodeType == 3) 
 result.push(child.nodeValue);
 
 else 
 findTextNodes(child);
 
 
 )(document.body);

Iterative Tree Traversal

function customIterativeTreeWalker() 
 var result = [];
 var root = document.body;

 var node = root.childNodes[0];
 while(node != null) 
 if(node.nodeType == 3) /* Fixed a bug here. Thanks @theazureshadow */
 result.push(node.nodeValue);
 

 if(node.hasChildNodes()) 
 node = node.firstChild;
 
 else 
 while(node.nextSibling == null && node != root) 
 node = node.parentNode;
 
 node = node.nextSibling;
 
 

querySelectorAll

function nativeSelector() 
 var elements = document.querySelectorAll("body, body *"); /* Fixed a bug here. Thanks @theazureshadow */
 var results = [];
 var child;
 for(var i = 0; i < elements.length; i++) 
 child = elements[i].childNodes[0];
 if(elements[i].hasChildNodes() && child.nodeType == 3) 
 results.push(child.nodeValue);
 
 

getElementsByTagName (handicap)

function getElementsByTagName() 
 var elements = document.getElementsByTagName("p");
 var results = [];
 for(var i = 0; i < elements.length; i++) 
 results.push(elements[i].childNodes[0].nodeValue);
 

XPath

function xpathSelector() 
 var xpathResult = document.evaluate(
 "//*/text()", 
 document, 
 null, 
 XPathResult.ORDERED_NODE_ITERATOR_TYPE, 
 null
 );

 var results = [], res;
 while(res = xpathResult.iterateNext()) 
 results.push(res.nodeValue); /* Fixed a bug here. Thanks @theazureshadow */
 

Also, you might find this discussion helpful - http://bytes.com/topic/javascript/answers/153239-how-do-i-get-elements-text-node

I know you specifically asked for a collection, but if you just meant that informally and didn't care if they were all joined together into one big string, you can use:

var allTextAsString = document.documentElement.textContent || document.documentElement.innerText;

...with the first item being the DOM3 standard approach. Note however that innerText appears to exclude script or style tag contents in implementations that support it (at least IE and Chrome) while textContent includes them (in Firefox and Chrome).

Here's a modern Iterator version of the fastest TreeWalker method:

function getTextNodesIterator(el) // Returns an iterable TreeWalker
 const walker = document.createTreeWalker(el, NodeFilter.SHOW_TEXT);
 const next = () => 
 const value = walker.nextNode();
 return 
 value,
 done: !value
 ;
 ;
 walker[Symbol.iterator] = () => (next);
 return walker;

Usage:

for (const textNode of getTextNodesIterator(document.body)) 
 console.log(textNode)

But the loop might get stuck if you move the nodes around while looping.

This is safer:

function getTextNodes(el) // Returns an array of Text nodes
 const walker = document.createTreeWalker(el, NodeFilter.SHOW_TEXT);
 const nodes = [];
 while (walker.nextNode()) 
 nodes.push(walker.currentNode);
 
 return nodes;

 document.deepText= function(hoo, fun)
 var A= [], tem;
 if(hoo)
 hoo= hoo.firstChild;
 while(hoo!= null)
 if(hoo.nodeType== 3)
 if(typeof fun== 'function')
 tem= fun(hoo);
 if(tem!= undefined) A[A.length]= tem;
 
 else A[A.length]= hoo;
 
 else A= A.concat(document.deepText(hoo, fun));
 hoo= hoo.nextSibling;
 
 
 return A;
 

/*
You can return an array of all the descendant text nodes of some parent element,
or you can pass it some function and do something (find or replace or whatever)
to the text in place.

This example returns the text of the non-whitespace textnodes in the body:

var A= document.deepText(document.body, function(t)
 var tem= t.data;
 return /S/.test(tem)? tem: undefined;
);
alert(A.join('n'))

*/

Handy for search and replace, highlighting and so on

var el1 = document.childNodes[0]
function get(node,ob)

 ob = ob 



var o = get(el1)
console.log(o)

Here's an alternative that's a bit more idiomatic and (hopefully) easier to understand.

function getText(node) 
 // recurse into each child node
 if (node.hasChildNodes()) 
 node.childNodes.forEach(getText);
 
 // get content of each non-empty text node
 else if (node.nodeType === Node.TEXT_NODE) 
 const text = node.textContent.trim();
 if (text) 
 console.log(text); // do something