Robots.txt quick question

[speedshopping]

Hi,

We have a bunch of 404 errors from Webmaster Tools that sit in a directory that has thousands of other valid URLs.

We want to disallow these 404 URLs without affecting the others in the directory like so:

404 error URL: /xyz/keyword/abc/
Valid URL: /xyz/keyword 1/abc/

If I disallow using:

Disallow: /xyz/keyword/abc/

will it disallow the URL above and continue to crawl all other URLs in the /xyz/ or will it try to disallow all of them?

Thanks in advance for your help.

[tangor]

This sounds like something better handled via .htaccess. Robots.txt is not going to help against those SEs which already found the URIs as they will keep hitting their already collected info to see it the link still exists. Feed it a 410 (gone), 404 (not found, ie, do nothing) or 301 to the page you want the SE to find.

This assumes that the bad URIs are NOT THERE. If they are, then WHY?