Wrestling with .htaccess syntax is a pleasure that I engage in every six months or so! I grasp the basic principles, but have never had cause to become fully fluent. These two rules are the culmination of a requirement to redirect a specific subset of pages from a site that used different URI structures; with and without query strings.
There were two 301 permanent redirects that I wanted to implement:
Sounds simple, but I was unhappy with my initial attempt because it was retaining the query string after the rewrite. Even worse, the legacy code on the old site was already rewriting URIs such as /blog/firstpost into /index.php?sec=blog&post=firstpost and I didn't want that rearing its ugly head again.
After a degree of head-scratching, the following two instructions achieve what I wanted. (I'm not an Apache expert, and there may be a better way.)
Example 1:
# legacy site 301 redirects
RewriteRule ^blog/firstpost/?$ https://newsite.com/information/first-post-on-this-blog [NC,L,R=301]I included this rule near the top of the .htaccess file (obviously after the RewriteEngine On, etc) but before the various legacy rules that were manipulating query strings.
Example 2:
# legacy site 301 query-string redirects
RewriteCond %{QUERY_STRING} post=secpost
RewriteRule ^index\.php$ https://newsite.com/information/second-post-here/? [NC,L,R=301]Apache 2.4 introduced the QSD flag, Query String Discard. The final ? in the redirect above predates that, but it works and I didn't feel concerned by it.
It turned out to be fairly straightforward in the end. Lots of examples that I found are using regex expressions to redirect whole sites, which is fine, but in this case a limited subset of pages required the redirect. Also the destination URI was very different. An alternative is to implement a route alias on the new page, but that's displayed in the address bar, and I didn't like it.
Good luck! ;-)