checker for broken links on html pages

[bigearsbilly]

Can anyone recommend an open source checker to find
stale, broken links on html pages?

I've tried:

linkcheck-1.4: Checks a web site for bad links
linkchecker-5.0.1: Check HTML documents for broken links
linklint-2.3.6.d: Perl script that checks links on web sites


and none of them work.

 

[bb]

Depends on how intelligent it should be. You can find absolute links very easily, and check them with curl (print all failed links):

grep -Eo -e 'https?://[^"[:space:]]*' input.html | sort -u |\
while read u; do curl -sfI "$u" > /dev/null || echo "$u"; done

Or with csh:

foreach u (`grep -Eo -e 'https?://[^"[:space:]]*' input.html | sort -u`)
curl -sfI "$u" > /dev/null || echo "$u"
end

If you want to check relative links as well, you'll need a more sophisticated tool that you can point to a pages URL, so that it can resolve the relative links like a browser.

 

[bigearsbilly]

yes well.
I downloaded a web site using wget so the links are
rewritten as relative.
but, strangely, wget missed some links, dunno why it looks simple enough.
So I do actually need more of a relative link checker.

I'm writing one myself now.