Manipulating Javascript RegExp results as non-strings

Question

It's much easier to show the Javascript code than offer a complex explanation:

var html = '<input tabindex="1" type="text">'; html.replace(/tabindex="([0-9]+)"/g, 'tabindex="' + ("$1" * i) + '"');

In this code, "$1" will always be a string, meaning this RegExp will always result in 'tabindex="NaN"' ... Using parseInt doesn't work.

Is it possible to cast the response from the RegExp as an integer so I can perform maths on the replacement?

CMS and outis have given you your answer. Just out of curiosity, what are you attempting to do with this code? — jessegavin
– jessegavin, Commented Feb 1, 2010 at 7:46
I have a block of HTML being cloned within the DOM, and want to ensure the attributes are correctly updated once cloned. Simple stuff, and the above is a further simplification, as a better example. — beseku
– beseku, Commented Feb 1, 2010 at 7:52
As outis said, then, you should definitely consider cloneNode and not HTML string hacking. Then you can use simple assignments like newelement.tabindex= oldelement.tabindex*i, and you don't have to worry about exact HTML formats (eg. IE's innerHTML returning you tabindex=2 without any quotes). The only gotcha is that in IE (oh, IE, again, how we love you), assigning the name attribute has some wrinkles. — bobince
– bobince, Commented Feb 1, 2010 at 15:26
This is a very specific case where a larger chunk of HTML, (lots of elements), is available as a string in the JS, rather than in the DOM. It's way quicker to use a simple search and replace or RegExp rather than to create the element and then trawl through the various child nodes and update them - I wrote the DOM manipulation first but clocked it about 10 x as slow within Firebug, hence looking to improve the speed. — beseku
– beseku, Commented Feb 3, 2010 at 2:55

outis · Accepted Answer · 2010-02-01 08:15:25Z

4

The replacement argument can be a function:

html.replace(/tabindex="([0-9]+)"/g, function (all, tabindex) {
        return 'tabindex="' + (tabindex * i) + '"'
    });

However, you should only be using regexps to manipulate HTML if you don't have a parser. If this code is to run in a browser, then you have a parser. If the node is coming from another document, use document.importNode. If the node is in the same document, use Node.cloneNode. Both support deep copying.

edited Feb 1, 2010 at 8:15

answered Feb 1, 2010 at 7:42

outis

77.7k23 gold badges155 silver badges228 bronze badges

Sign up to request clarification or add additional context in comments.

Collectives™ on Stack Overflow

Manipulating Javascript RegExp results as non-strings

1 Answer 1

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

Comments

Your Answer

Sign up or log in

Post as a guest

Related