Forum Moderators: phranque
We dont need any directory specific redirects. Will rewrite rules work if placed in httpd.conf?
Are there any other ways to tune rewrites?
Never forget that Pattern is applied to a complete URL in per-server configuration files. However, in per-directory configuration files, the per-directory prefix (which always is the same for a specific directory) is automatically removed for the pattern matching and automatically added after the substitution has been done. This feature is essential for many sorts of rewriting - without this, you would always have to match the parent directory which is not always possible.
In /.htaccess:
RewriteRule ^index\.html$ /index.php [L]
In httpd.conf:
RewriteRule ^/index\.html$ /index.php [L]
In other words, when coding in .htaccess, the URL-path matched by the RewriteRule pattern will not include the path to the current directory. So, in the case above, the leading slash is stripped before the RewriteRule in .htaccess 'sees' it. To make it even clearer, take the case where the .htaccess file and the rule are in a subdirectory:
In /foo/.htaccess:
RewriteRule ^index\.html$ /foo/index.php [L]
In httpd.conf:
RewriteRule ^/foo/index\.html$ /foo/index.php [L]
Also, look at your regular-expressions patterns carefully: Are they as specific as possible, or do you have a bunch of patterns with multiple instances of ".*" in them? That is a sure sign of inefficient pattern-coding. If it is not possible to determine what characters you want to match, then how about determining what characters you don't want to match? Example:
# Replace underscore with hyphen
RewriteRule ^(.*)_(.*)$ http://www.example.com/$1-$2 [R=301,L]
can be much more efficiently coded as:
RewriteRule ^([^_]*)_(.*)$ http://www.example.com/$1-$2 [R=301,L]
In the first case, we tell the parser to match any and all characters into the first pattern, because the first pattern accepts any number of any characters. So, it does that, and the match fails because the rest of the pattern is 'starved' by having no characters left to match. So the parser retries -- this time taking one character out of the first match, and trying it against the remainder of the pattern. This can repeat N-1 times, where N is the number of characters in the requested URL. If three or more ".*" patterns are used, the number of retries increases geometrically, and can reach ridiculous numbers.
In the second case, in the first sub-pattern, we tell the parser, "match any number of characters NOT equal to an underscore." Therefore, we tell the parser exactly what character should terminate a matching attempt for the first sub-pattern. This is much more efficient.
Jim
