Shell Programming and Scripting

First time poster (so please excuse me in advance)

I have a webserver running linux, apache, etc. I have a list of HTML webpages that I want to delete because I think they are old. While I could delete them then check for broken links, I'd like to be more pro-active.

I want to write a shell script that will search all the pages in my site for links to the pages in my list.

Let's say I have a potential file to delete at www.fake-url.com/foo/bar/index.html

I can't just grep for it because the page can be written within pages in a number of way:
1) It could be a full or root relative (that's easy enough to search for)
2) It could be a relative link!

I can't grep for "index.html" because there are multiple index pages. I've written some shell scripts, but searching for relative links like this seems overwhelming.

Hopefully my question makes some sense and I'm posting it in an appropriate place. I was thinking of writing my own script, but you know of an existing script or program that does this it would certainly be appreciated!

I would approach this in another way. Configure Apache for logging so that you can see the referring page (%{Referer}i) AND the file being served (%f). Now, crawl the site using a standard crawler (wget -r -l inf -nd --delete-after ). Next, analyze the logfile against your deletion-candidate list. You can do something like:

Code:

cat candidate-deletion-files.txt |
while read old_file;   do 
   grep -- "$old_file" access.log >/dev/null || echo "Safe to delete: $old_file"
done

To be more particular, I'd make sure I'm looking at the correct field in the logfile. You can do this with awk, or by first extracting the file names from the logfile and searching that output using the grep above.

That's a different approach. I think it makes sense. Thanks!