Tobias Freitag wrote:
> 
> Hi there,
> 
> I am trying to index about 1000 HTML-Articles on our website via local
> filesystem. I use the "Alias" command to map the links to the real
> location.
> 
> The problem is that when I search for any pattern, most results show
> directories not files. It seems that the indexer is parsing the
> directory listings for matching filenames. Nice Feature, but nothing
> what I really need.
> 
> For Example a search result for "linux" looks like this:
> 
> Displaying documents 1-20 of total 55 found.
> 
> 1. http://www.linux-magazin.de/ausgabe/1996/08/ [1]
> 
> 2.  [1]
> 
> [...]
> 
> 13. http://www.linux-magazin.de/ausgabe/1996/06/News/biodata.html [1]
> 
> The Allow/Disallow settings are kept simple:
> 
> Allow \.html$ \.htm$ \.txt$ \/$
> Disallow .*
> 
> I tried it without the \/$ to prevent directory indexing, but then the
> program indexed nothing except the rootdir
> (http://www.linux-magazin.de/ausgabe/). I have also tried "CheckOnly
> \/$", but it doesnt seem to change anything.
> 
> Ideas anyone?
> 
> # This is sample indexer config file
> # To start using it please edit and rename to indexer.conf
> # You may want to keep the original indexer.conf-dist for future references.
> # Use '#' to comment out lines.
> # All command names are case insensitive (DBHost=DBHOST=dbhost).
> # You may use '\' character to prolong current command to next line
> # when it is required.
>  ...
> UrlPathWeight -1
> UrlFileWeight -1
>  ... 

This is because of -1 in UrlFileWeight and UrlPathWeight.
______________
If you want to unsubscribe send "unsubscribe udmsearch"
to [EMAIL PROTECTED]

Reply via email to