Re: lynx-dev -traversal -crawl gets into infinite loop (fwd)

[Klaus Weide]

This message never made it to lynx-dev, so I am resending it.

---------- Forwarded message ----------
Date: Wed, 26 Apr 2000 09:38:42 -0500 (CDT)
From: Klaus Weide <address@hidden>
Reply-To: address@hidden
To: address@hidden
cc: address@hidden
Subject: Re: lynx-dev -traversal -crawl gets into infinite loop
In-Reply-To: <address@hidden>
Message-ID: <address@hidden>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII

On Wed, 26 Apr 2000, Bob Izenberg wrote:

> [ Forwarded message from unsubscribed address
>   address@hidden ]
> 
> Hello,
> 
> I was trying to be smart and use lynx to convert my javadoc
> generated .html files to text.
> I think it's a bug that when using -crawl -traversal lynx
> doesn't remember it has already visited a link.
> This causes lynx to get into an infinite loop, pingponging
> between two html files that reference each other.
> No, I can't change the .html files for they are generated
> from java source code and not man-made.

Someone has reported something like this before.  But nobody
has been able to provide a reproducible case.
> 
> Try to fix this if you can, thanks.

Please try to help us figure out what the problem is, if you want to
see it fixed.  Is there a *small* set of files that shows the same
problem?  Is there anything unusual about the link (or near it) that
does not get remembered - like length, unusual characters, or anything
else?  Does this only occur after a certain (high) number of documents?
Does lynx produce any unusal messages (maybe only in a -trace logfile)?

Also make sure you have enough disk space for all generated files
(and some more to spare).

   Klaus