Works pretty well in capturing the full URL when using this in a search (like in Sublime Text 2).
(https?|ftps?)\:\/\/[a-zA-Z0-9\-\.]+\.[a-zA-Z]{2,3}(\/\S*)?/?
The following will capture the URL in a SQL dump including escaped quotation marks.
- URL pattern:
http://www.yourwebsite.org/calculator/degrees/sociology - RegEx:
(https?|ftps?)\:\/\/[a-zA-Z0-9\-\.]+\.[a-zA-Z]{2,3}(\/\S*)?/calculator/degrees/([^\\"]+) - Replacement:
http://www.yourwebsite.org/degrees/$3/ - Result:
http://www.yourwebsite.org/degrees/sociology/
This will find all hrefs that do not contain a trailing slash. NOTE: it will detect links in your <head> and ones that end in .html that purposfully do not have a trailing slash, so becareful performing a find/replace.
- URL pattern:
href="http://www.yourwebsite.org/calculator/degrees/financial" - RegEx:
href="(\S)+[^/]"
Convert from http://old.yourwebsite.org/whatever/ to http://new.yourwebsite.com/whatever/
EXPLAINED:
- since the pattern will contain literal forward slashes for the url (eg "schema://domain/path"), we're delimiting the path with pipe chars to avoid having to backslash-escape each forward slash
- just in case we have a mix of http & https urls, we'll match both with "https?" which means match "http" followed by one or zero "s" chars
- we're backslash-escaping the dots in the domain, since in regex syntax an unescaped dot normally means "any single character other than a newline"
- we're capturing everything before and after the "request" in "requestinfo" with parentheses in the pattern, then joining them together in the replacement using backreferences
- we're making the entire pattern match as case insensitive by adding an "i" flag after the closing pattern delimiter
preg_replace('|(https?://)old(new\.yourwebsite\.)org|i', '$1$2com', $content);
#RewriteRule ^calculator/degrees(?:/([\w-]+?)(?:-in.+)?)?/?$ /degrees/$1/ [L,R=301]