Caching from URLs with a query string

问题:

The web application I'm working on serves up images with URLs like /image?name=a.gif. Pages are loading slowly partly because browsers are not caching the images.

Is there any combination of http headers that will persuade IE to cache the images even though the URL has a query string in it? I'm trying to avoid the browser making any unnecessary requests (including if-modified-since type requests)

I'm happy with a solution that works with IE (6+) only. Also, I know that I can avoid the problem by URL rewriting - I'm just interested in understanding browser caching better.

Thanks

回答1:

Set the expires header in your web server to some far away date.

回答2:

It's commonly thought that a lot of older browsers (including IE6) and some older versions of caches like Squid won't cache any file that has a querystring attached, regardless of the expires date.

The more robust alternative to querystring versioning is to put versioning information into the path, such as:

site/static/23212903/css/bundle.css

Or:

site/images/logo.r23212903.png

For more, here's a great article on setting cache headers correctly by Sergey Chernyshev.