Question

My company would like to run a program that searches for broken links on our several TWiki webs. I'm trying Flinch and TagReader. It's not easygoing, considering my lack of linux knowledge, but added to this is the fact that the TWiki pages are not html until the perl mods render them so (I could be using any number of words incorrectly here). Is there an existing plugin for checking links? Should we use the GenHTML Add on, then check them? Anything?

• TWiki version:01 DEC 2001
• Web server:apache
• Server OS:linux
• Web browser: Mozilla
• Client OS: RH 7.3

-- HowardHolden - 24 Sep 2002

Answer

Existing link checkers should run fine on TWiki - it would be best if they respect robots.txt, then you can use Google:robots.txt+twiki to ensure that they don't generate spurious Edit and other actions. It's much easier to just check the HTML than try to check the raw text IMO, unless you want to write a server side tool that strips the URLs out of the raw text.

-- RichardDonkin - 25 Sep 2002

Not sure if a link checker can help here because links to non existing topics do not return a 404 error. Rather, a real HTML page is returned when following a question mark link.

The answer could be Codev.FindReferencedButNotDefinedWikiWords (not tested); the opposite problem is HowToFindOrphanedTopics.

-- PeterThoeny - 26 Sep 2002