Posted: 13 January 2005 at 8:13pm | IP Logged

I've seen this rule posted here, which redirects requests for directories that do not end with a slash.

RewriteCond Host: (.*)
RewriteRule ([^.?]+[^.?/]) http\://$1$2/ [I,R]

I don't understand how the RewriteRule pattern works. Yaroslav, Lexey, or anybody else would you please walk me through it?

([^.?]+[^.?/])

My attempt at interpretation:

[^.?] means "Anything except (any single character (zero or one time only))."

Sorry not making sense. I understand you can match either a single character one time or a single character zero times. But what matches anything except that (signified by the caret)?

+ means "The preceding expression repeated one or more times."
[^.?/] means "Followed by anything except (any single character (zero or one time only) or a slash)."

Also I was glad to see the pattern doesn't match requests ending in file names, so for example you don't get redirected to /index.html/ when you request /index.html. However I would like to understand how this works.

Thanks,
Kevin

Posted: 14 January 2005 at 4:27am | IP Logged

>My attempt at interpretation:

>[^.?] means "Anything except (any single character (zero or one time only))."

:) Inside a set '?' and '.' are normal characters. So that means any character except '?' or '.'.

>+ means "The preceding expression repeated one or more times."

>[^.?/] means "Followed by anything except (any single character (zero or one time only) or a slash)."

That means "followed by anything except '?', '.' or '/'.

Posted: 15 January 2005 at 9:35pm | IP Logged

>Inside a set '?' and '.' are normal characters.
Thanks very much for clearing that up!

Is it possible for this to work when just the hostname is requested? For example, redirect http://www.example.com to http://www.example.com/
I'm not sure it's possible because even when you don't type the trailing slash in the browser it appears in the actual HTTP request.

Thanks