On Mon, 1 Apr 2002, Ronnie Koch wrote:

> 
> Problem resolved. I removed the "Disallow: /index" from robots.txt.
> 
> FWIW
> 
> ht://Dig 3.2.0b3

The problem with zillion refreshes is/was caused by this version of htdig 
calling itself "htdig" in the headers and not "htdig/.....". The test was 
for the first 6 digits "htdig/" - local mod to wwwoffle fixed problem for 
now.

> wwwoffled version 2.7a
> 
> Thanks - once I realised that */search/index had hidden links the rest was 
> easy :-)
> 
> 
> 
> Ronnie
> 
> On 31 Mar 2002, Andrew M. Bishop wrote:
> 
> > Ronnie Koch <[EMAIL PROTECTED]> writes:
> > 
> > > I have been using wwwoffle for a couple of years and only recently decided
> > > to get the htdig search operational - with minimal success. I upgraded to
> > > 2.7a, from source, this week from 2.6d on RedHat 7.2 (Intel). The original
> > > cache was xfered from another machine that ran 2.5x and was converted to
> > > work with 2.6d (months ago).
> > > 
> > > 0. I can only search the WWWOFFLE FAQ & Welcome pages with htdig.
> > 
> > I use htdig with WWWOFFLE (I don't use the other search options
> > although I have tested them).  I don't have any problems, so I believe
> > that it works in the default installation (my installation is not very
> > different from the default).
> > 
> > What are the error messages that you get from htdig?
> 
> No errors - just nothing indexed.
> 
> > > 1. I changed htdig-full.conf to point to a existing page in the cache and
> > > got a lot of good stuff pluss a zillion refresh requests that I did not
> > > need. It kind of confirmed that the basic search is working.
> > 
> > The htdig configuration file is set up to search to a depth of 4. This
> > is why the first search index page is start4.html so that the depth 4
> > pages are the cached ones and no further links are followed.
> > 
> > There should not be any refresh requests since WWWOFFLE should detect
> > that it is htdig that is requesting the page and not store the
> > request.
> 
> Noticed in the source - to rusty to realistically follow the logic though.
> 
> > > 2. The localhost:8080/search/index/ page is empty (link from  
> > > localhost:8080/search/start4.html)
> > 
> > It may look empty, but if you examine the source code you should find
> > that it is not really empty.
> 
> Lynx cleared that one up.
> 
> 
> --------------------------------------------------------------------
> Ronnie Koch                   | Snailmail: POBox 60124
> Bsc Eng (Elec)(Pret) MSAIEE   |            Pierre van Ryneveld 0045
> Tel: +27-(0)83-2270548        |            South Africa
>                Email: [EMAIL PROTECTED]
>                http://www.infinityio.co.za
> --------------------------------------------------------------------
> 
> 
> 
> 

Regards




Ronnie
--------------------------------------------------------------------
Ronnie Koch                   | Snailmail: POBox 60124
Bsc Eng (Elec)(Pret) MSAIEE   |            Pierre van Ryneveld 0045
Tel: +27-(0)83-2270548        |            South Africa
               Email: [EMAIL PROTECTED]
               http://www.infinityio.co.za
--------------------------------------------------------------------


Reply via email to