通过XMLHttpRequest（）下载时，如何避免缓存数据

Question

I am writing an extension for the Chrome browser (and later hope to port to Firefox). The extension downloads a configuration file from my server - an xml file via XMLHttpRequest. What I am find is that it downloads the file once and every subsequent call simply seems to use the cached original version of the file. It doesn't matter whether or not I change the file on the server.

I read that you could try

xmlhttp.setRequestHeader( 'Pragma', 'Cache-Control: no-cache');

and so I've done this but it doesn't seem to make any difference. The only way I can get the new file seems to be to delete the browser cache - which obviously is not a solution for my growing users.

This seems like a problem that I wouldn't be the first person to experience - so given thatcacheing rules seem to uphold this as a policy that can't be easily avoided, my question is, what's the better design? is there a best practice I don't know about? Should I be pushing rather than pulling somehow?

Answer 1

An easy way is to add a useless parameter containing the time to the request. Since time tends to go forwards and never backwards, you can be reasonably sure that your query is unique and therefore won't be cached.

For instance (assuming the URL is in a string url ):

url += '?_time=' + (new Date()).getTime();

or, if your URL already has query parameters,

url += '&_time=' + (new Date()).getTime();