robots.txt disallow searching reoccuring folders

[jchristi]

I'm searching my heart out, but must not be using the correct terms to get to the following answer.

We have a large site that has plenty of subdirectories of similar construction.

Within each subdir, there are specific folders that we want to disallow searches of.

I'm trying to ascertain how little I can get by with in the robots.txt file to achieve the desired results.

If the server has the four following structures:

www.server.com/Images/picture.gif
www.server.com/subdirectory1/Images/picture.gif
www.server.com/subdirectory2/Images/picture.gif
www.server.com/subdirectory3/andevenmorefolderstructure/Images/picture.gif

what disallow language would tell the search engines not to search any "Images" folder?

Would */Images/ do it?

or /Images/
*/Images/
to catch the root, and then all deeper buried folders?


or, must I specify entire subdir paths to the folders I want blocked?
/Images/
/subdirectory1/Images/
/subdirectory2/Images/
/subdirectory3/andevenmorefolderstructure/Images/

[g1smd]

The pattern matching is "from the left".

So /images disallows any URL path "beginning" /images

And /*images disallows any URL path "containing" images

[jchristi]

thanks...

and i need to include the trailing / ( /*images/ ) if I want to be sure i match only "images" folders and not "imageofsomething" folders?

[g1smd]

Yes.

[phranque]

also note that the wildcarding/file globbing patterns are not specified in the robots exclusion protocol but rather are extensions supported by most of the big players including google.
in other words, don't expect EVERY "well-behaved" bot to necessarily understand and respect your exclusions.

[jchristi]

I understand it's not honored by everything.

Basically, we have a Sharepoint deployment, and for every site listed below the domain, they have similar structures. I do not want to spell out every disallow for every site.

if this gets me through the big dogs... that will be enough for me.