I have a basic chat that reads from a database. Each chat message is read from the database by this.
<span>{{formatChat text}}</span>
text being the message read. And then I use the formatChat registerHelper to detect URLs.
Template.registerHelper('formatChat', function(text) {
var urlRegex = /https?:\/\/([a-zA-Z0-9\-\.]+)(\.[a-zA-Z0-9]+)((([a-zA-Z0-9\?\=\/])+)?((\#|\?)(.+)?)?)?$/
var urlRegexMini = /(www(\d{0,3})\.)?([a-zA-Z0-9\-\.]+)(\.(com|net|org|gov|co\.uk|edu|io\b)+)([a-zA-Z0-9\?\=\/]+((([a-zA-Z0-9\?\=\/])+)?((\#|\?)(.+)?)?)?)?$/
finalString = "";
//Parse every word individually
split = text.split(' ');
for (i = 0; i < split.length; i++) {
finalString += " ";
if (urlRegex.test(split[i])) {
finalString += "<a href='" + split[i] + "'>" + split[i] + "</a>";
}
else if (urlRegexMini.test(split[i])) {
finalString += "<a href='http://" + split[i] + "'>" + split[i] + "</a>";
}else{
finalString += split[i];
}
}
return finalString.substring(1,finalString.length);
});
The problem is that meteor doesn't allow injection, so it will literally show the anchor tag as plain text.
One solution that I thought of was to have a registerHelper for each individual word, but that seems rather foolish.
How can I efficiently get around this rule?
I believe this should do it:
{{{formatChat text}}}