I use xav search engine II in my hp. I host my hp at hypermart. It work extremely slow!!!
How can I modify it to make it faster??
Also, I want the script update the url I entered only, during maintaining realm. I don't want it follow the url in the html files and index them automatically. How can I do it??
Thanks a lot!!!

I host a page on hypermart, and I find that their service of web pages themselves is slow as well, its doubtful that any changes you could make to the script would help at all. Most likely, the best bet is to get your own server or find someone faster.

Just an idea.


------------------
Fred Hirsch
Web Consultant & Programmer

I think I can explain why Hypermart is so slow. I remember a month or so ago, we had this conversation on another thread in one of the forums. Linux (one of the various flavors of Unix), can be run easily on systems that would grind to a halt if you tried to run Win9x them. I wouldn't be a bit surprised if they are using seriously overloaded, underpowered servers to host their sites. I would be willing to bet that their systems are probably Pentium 133's if you're lucky. One of the great things about Linux (at least for the ISP's) is that you can put together a system from spare parts and it will run Linux. You can't do this with HP-UX, Solaris, or the higher end versions of Unix.

Please... don't get me wrong, I think Hypermart is great if you want to host a free site, but if you decide that you need more disk space, I'd go to a provider that doesn't offer free services. I'm sure that some of the regulars on the forum can easily point you to a good provider based on what you're doing with your site and what you're looking for in a provider.

Excluding Mass Directories in the Xav Search II - NT Server

I was wondering if there is a way to exclude common hidden directories on a NT server. I am using Xav Indexed Search (http://www.xav.com). The problem is that the index file always includes hidden directories, including /vti_bin/, /_private/, etc. I would like to ignore these types of directories without specifying them on an individual basis in the $DMZ code.

Anyone know how to do this??

TIA

------------------
Eliot Lee
Founder and Editor
Anthro TECH, L.L.C
http://www.anthrotech.com/
info@anthrotech.com
==========================
Coconino Community College
http://www.coco.cc.az.us/
Web Technology
Coordinator
elee@coco.cc.az.us

[This message has been edited by Eliot (edited June 07, 1999).]

Hi

I looked at this NT problem a while ago but couldn't work out how to fix it and I have sinced decided to move the site to a unix box to avoid all the front page stuff (to name but one NT gripe...)

Chris

I figured how to avoid the hidden directories in the build.pl file:

Insert Codes:
(After @directories = ($basedir)
--------------------------------------------
@gDontSearchTheseDirs = qw(_vti_bin _derived _themes _vti_cnf _borders);
grep($gDontSearchTheseDirs{$_}++, @gDontSearchTheseDirs);
--------------------------------------------

Replace the following Codes:

next if (($entry eq ".") | | ($entry eq "..")

WITH

next if (($entry eq ".") | | ($entry eq "..") | | $gDontSearchTheseDirs{$entry});

It works great now!

------------------
Eliot Lee
Founder and Editor
Anthro TECH, L.L.C
http://www.anthrotech.com/
info@anthrotech.com
==========================
Coconino Community College
http://www.coco.cc.az.us/
Web Technology
Coordinator
elee@coco.cc.az.us

Let me get this straight. It seems that you all know what this index script does, but I don't. Does it index remote html files? In other words, it can crawl pages and store them into a databae which then can be recalled through a search? Can it support any file type and not just html? Thanks for any help. I was looking for some sort of script like this. The closest I could find was a script called: Chatologica Global Search CGI Script.

------------------
XanthisHP
http://boom.home.dhs.org
BOOM! Flash 3 Resource - When there's a Flash, there's a BOOM!

Yes...You execute the "build.pl" file via a web browser and it builds a flat file from a specified web site. You can specify many different file types, including image files, .pdf, .doc, etc.

The search engine searches through the flat file. It is much faster than spidering through regular static web pages.

For more information, check out:

www.xav.com

Regards,

------------------
Eliot Lee
Founder and Editor
Anthro TECH, L.L.C
http://www.anthrotech.com/
info@anthrotech.com
==========================
Coconino Community College
http://www.coco.cc.az.us/
Web Technology
Coordinator
elee@coco.cc.az.us

[This message has been edited by Eliot (edited June 21, 1999).]