Forum Moderators: phranque
# mod_rewrite in use
RewriteEngine On
RewriteRule (.*)$ http://www.example.com/
I am taking my first baby steps in this and am using the above code to redirect any links which point to a particular domain, to another place.
I just wanted to make sure this is the correct way to do this, and that I'm not in danger of creating any infinite loops or other chaos which might incur the wrath of the server guy.
# mod_rewrite in use
RewriteEngine On
RewriteRule .*$ http://www.example.com/
I've done a bit more reading... I guess I don't need the brackets in there? But I still do need the $ right?
Finally found a tutorial on regular expressions which isn't complete greek to me, so I'm getting there!
To do what?
That code will certainly loop if it is placed and executed on the www.example.com server. It will happily redirect requests for all pages and objects on www.example.com to the home page of www.example.com and then redirect that home page request to itself repeatedly.
If www.example.com is on another server, then it won't loop. It will simply redirect all page and object requests made to the server on which it resides to the home page on the www.example.com server.
So the answer depends on what you are trying to accomplish and where this code is located.
Jim
[edited by: jdMorgan at 8:10 pm (utc) on Jan. 15, 2008]
Sorry if I didn't give enough info...
I generally want to use it for places where documents have been moved or removed in the past. I'm getting a few errors from external sites in the Google webmaster tools, which say people are trying to go to pages like
www.example.com/mydirectory/someoldfile.pdf
www.example.com/mydirectory/someotherfile.pdf
So I'm putting my code in an htaccess file within www.example.com/mydirectory/ to send all such requests straight to the home page or another main page as appropriate.
Just wanted to make sure I'm not doing anything st00pid :)
Then use a custom 410 error document to explain the situation, and provide alternatives to help the visitor find what they were looking for.
Be aware that a client requesting a .pdf document may not be able to display an HTML error page. Nonetheless, the server should still signal that the requested document is gone, since that is the factually-correct response.
301-redirecting *all* requests for *any* document in /mydirectory creates a massive-duplicate-content problem for you home page, and opens you up to willful abuse by competitors or malicious ne'er-do-wells. A 410 is the proper response to avoid this.
404 and 410 error documents should contain an explanation --apologetic in tone-- that the document is missing for unknown reasons (404) or has been intentionally removed (401), and that no direct replacement is available. They should also provide text links to as many as the following resources as possible:
Optionally, an on-page HTML meta-refresh (with a timer long enough for all of the above text and links to be read, understood, and considered) can be used to forward the visitor to the home page if no choice is made. If a meta-refresh is to be used, a notice of this should also be included ("Please choose one of the links above. If you do not choose a link, you will be automatically forwarded to our home page in 30 seconds").
The syntax for a 410-Gone RewriteRule and ErrorDocument looks like this:
ErrorDocument 410 /path-to-410-error-document.html
RewriteRule .* - [G]
The above code (yours and mine) assumes that all documents in /mydirectory have been removed -- All requests to that directory will result in a 410-Gone response.
If possible, it would be better to specify all of the documents which have been removed, returning a 410-Gone only for those; Any other requests should get a 404-Not Found. In that case, you'll likely need several RewriteRules, using specific old URL-paths or patterns which only match valid-but-removed URLs.
RewriteRule ^old-doc1\.pdf$ - [G]
RewriteRule ^older-doc3\.pdf$ - [G]
Jim
I'll get onto it...
