Ok, this is driving me batty. I'm trying to remove any instance of index.html or index.php on any URL, preserving the entire path before the matched string and also preserving any arguments that may be passed after.

The goal here is that whatever index file may be passed on the URL, the default/existing index file for that path will always be used, since the server already searches for various index file extensions automatically. This has the benefit of allowing me to change from html to php for any index file without affecting existing links elsewhere on the web and also cleans up the various base/sub-base URLs in browsers.

The other reason is to overcome a limitation with Google analytics where they only allow a single "default" page, such as "index.html" to be defined, even though my site uses both static and dynamic pages and therefor has both .html and .php scattered around.

Just testing with index.php for a moment, I tried this:



However it fails, causing nothing on the site to work at all.



Any ideas?

What's in the Apache error log?

mod_rewrite isn't usually enabled by most default Apache configs -- are you sure it's loaded?

Mod_rewrite is definitely loaded because I have another rewrite that works properly to rewrite the URL without www, among others.

As far as the apache log goes, I'm not certain where to find it on this shared host, nor if I can find it. Took a quick look through my logs folder already.

Try nuking the "HTTP" in the RewriteCond -- the "HTTP" predicate is used for rewriting header values (e.g. HTTTP:User-Agent) but I don't knwow if just plain putting "HTTP" in a rewrite condition makes sense.

But you should really try to find the error log to see what the error is -- grep for ErrorLog in your httpd.conf.

Thanks Tony.

I think I've got it now. The HTTP as you mentioned had to go - damn other people's examples.

I also wanted the rewrite to be reflected in the browser, so I had to force a redirect. This seems to work:



I also escape the periods for good measure.