I've tried this
<html>
<head>
<title>None</title>
</head>
<body>
<p id="text">just some random text. random text</p>
<button type="button" onclick="strReplace();">Replace</button>
<script>
function strReplace(){
var myStr = document.getElementById("text");
var mySte = myStr.textContent;
console.log(mySte);
</script>
</body>
and I want this following outcome
just some random text
random text
You can use this regex to find and replace string without breaking html: /(?!<[^>]+)\.(?![^<]+>)/g
[myattr]
{
background-color: pink;
}
[myattr]:after
{
content: "this element's attribute is: " attr(myattr);
background-color: lightgreen;
margin: 1em;
}
p > span
{
background-color: lightblue;
}<html>
<head>
<title>None</title>
</head>
<body>
<p id="text">just some. random text. <span myattr="text.with.dots">nested.html</span> end.
<span>i < 40. But i is also > 30. What are valid values of i?</span>
</p>
<button type="button" onclick="strReplace();">Replace</button>
<script>
function strReplace(){
var myStr = document.getElementById("text");
myStr.innerHTML = myStr.innerHTML.replace(/(?!<[^>]+)\.(?![^<]+>)/g, "<br>");
console.log(myStr.innerHTML);
}
</script>
</body>
To directly answer the question:
var outputHtml = inputText.replace(/\./g, '<br />');
The Javascript replace method will by default replace the first instance of something (the first . in this case), so the g regex modifier tells it to replace them all. The \. in the regex is because . is a special character in regexes. This will replace every dot in the text.
This technique won't work well on text that contains literal ellipsis, like this:
Hello world...
If your text is likely to contain this, and you want that to be ignored, then the following regex is more appropriate:
/[^\.](\.)[^\.]/g
This'll match any . which is not surrounded by other .
var outputHtml = inputText.replace(/[^\.](\.)[^\.]/g, '<br />');
In the question here, the input is actually coming from the DOM. We can't replace . on its innerHTML as it would replace content inside HTML tags as well. So, if your input text is coming from the DOM like this (and not, say, a textarea), then the safest route is to apply the replacement only to text nodes. That is done like this:
document.getElementById("start").addEventListener("click", () => {
var inputNode = document.getElementById("text");
replace(inputNode);
});
function replace(node) {
if(!node) {
return;
}
if (node.nodeName == '#text') {
// We've found a text node. Apply the regex to it now.
// Note that you can use either of the above regexes here.
var inputText = node.textContent;
var lines = inputText.split(/\./g);
if(lines.length > 1) {
// It had at least one dot in it.
// Swap in this new set, each with a <br /> between them.
var parent = node.parentNode;
var nextSibling = node.nextSibling;
parent.removeChild(node);
lines.forEach((line, i) => {
if(i != 0){
// insert the <br>
var br = document.createElement('br');
parent.insertBefore(br, nextSibling);
}
var textNode = document.createTextNode(line);
parent.insertBefore(textNode, nextSibling);
});
}
} else {
// Loop through each child node.
// We go backwards such that completed replacements don't affect the loop.
for(var i=node.childNodes.length - 1; i>=0; i--) {
replace(node.childNodes[i]);
}
}
}<html>
<head>
<title>None</title>
</head>
<body>
<p id="text">just some random text. random text</p>
<button type="button" id="start">Replace</button>
</body>
</html>If you're starting from a HTML string inside a browser, you can then use the above replace method and the browsers internal parsing to safely only affect the actual text:
function replaceHtmlString(html) {
var div = document.createElement('div');
div.innerHTML = html;
replace(div);
return div.innerHTML;
}
')` – TJBlackman Jul 24 '21 at 14:25