While I have other aspects of my website, I want to be able to utilize one search engine to search both areas, is Ksearch able to be modified to do this.

I apologize up front as I am not the greatest of coders, actually not sure if I should consider myself one LOL, but your tool is very easy to use and I was hoping for this to be able to handle both aspects.

Thanks for all and any help.
...Mike

I have been over and over the configurations.pl, and everything seems correct.

cgi and perl work fine, including for other applications.
Permissions are set.
This is a Windows 2008 server using IIS 7.

We have used previous versions of ksearch before on other servers, so I am familiar with the requirements for the configurations, but this is the first time using this version.  (Previous versions used on Windows 2003 server and Linux, both worked marvelously.

Is there something specific in this version or Windows 2008 that would be causing it to not index?

Thank you.

I have followed your clear instructions and when i attempt to run the indexing script I am left with a blank page.
I understand there should be some notification that the indexing process has begun, but i am left with a completely blank page.

I have spent more time than I am prepared to admit to trying to install KSearch on a nginx server and solve what I am sure must be an elementary fault on my part.

Anyway:

Perl is running and i have confirmed the path is correct.

I double checked all permissions.

I always use Filezilla.

This my first time using nginx, but i did have experience of using Apache servers many years ago. I think the problem possibly lies in the naming of the absolute paths for the config file. I have asked my hosting company for this information, but will take sometime.

I tried running the search out of curiosity and receive this error message:

Software error:

Can't "next" outside a loop block at /home/heyhouses.com/public_html/search/ksearch.cgi line 1184.

I'd be grateful for any pointers or help.

Regards

Michael

was able to troble shoot the path errors ...

but it is still not indexing any pages

I point it to the only page on the site /TEMP.html as the start path. when I run the cgi I get:

Please wait until this page is finished loading before closing this window
or browsing the internet. If the script prematurely terminates, your server
may have timed out.

When the indexer completes, the last line should read: Total run time: ....

Deleting existing db files:

Loading files to ignore:
/home/www/Your-Website.com/folder_name

Using stop words file: //home/user/www/search/configuration/stop_terms.txt

Starting indexer at /home/darkone/www/TEMP.html

No common terms were present in over 80 percent of all files

Finished: Indexed files (KB) with total terms.
Total run time: 0 seconds

and the database.txt has no data in it... not enterprising any search renters no results. How do fix this so it will index the page?

The hosting server is a shared Linux type running Apache which I do not control. It has a private cgi-bin directory for running Perl scripts and a public httpdocs directory for running http. I feel that this is the problem because search_form must live in the cgi-bin directory which cannot run external http requests.

The failure message is 500 http scripts prematurely terminated. Anyone know a solution without changing my hosting provider?  I'm not convinced that increasing http timeouts as suggested in the FAQ will help.

Thanks in advance.

I need a search engine for a very large website with very high traffic.  There are thousands of pages and thousands of PDFs.

Our site has a non-member side with restricted access to the PDFs and member side that is virtually the same but with full access.

Can Ksearch be set to search only one side or the other?  I don't want member only pages returned in a search to a person that is not logged in.  I also don't want the members getting results that are the non-logged in watermarked copies.

Can this be done without listing each and every page as non-indexed?

I planned to put up two instances of Ksearch, one for the non member side that blocks all member pages, and one for the member side that blocks all non-member pages...if it can be done semi-easily.

Thanks in advance for any responses.

I have made simple site that contains product documentation, the site consists from three separate parts and I would like to allow user to choose from which part of the site he wants to see search result.

Is it possible? If yes - how can I do that?

Thanks in advance

The indexer on our site seems to hand at a certain point, and the error log reveals the following:-

Use of uninitialized value in concatenation (.) or string at indexer.cgi line 364.

Does anybody know how to resolve this problem?

Is there and error handler for this?

Many thanks for your help.

Best regards,

Alex.

Cannot open log file http://www.gpsl.at/cgi-bin/ksearch1.6/search/log.txt: No such file or directory at indexer.cgi line 131., referer: http://www.gpsl.at/cgi-bin/ksearch1.6/search/indexer.cgi

Cannot open http://www.gpsl.at/cgi-bin/ksearch1.6/search/database/files: No such file or directory at indexer.cgi line 149., referer: http://www.gpsl.at/cgi-bin/ksearch1.6/search/indexer.cgi

here the config:

# Where you want the indexer to start (full path; end with /)
$INDEXER_START = '/var/www/vhosts/gpsl.at/'; # re-index

# The base url of your site being indexed
# Be sure it corresponds to the $INDEXER_START directory (end with /)
$BASE_URL = 'http://www.gpsl.at/';

# The url where ksearch.cgi is located
$SEARCH_URL =

I started using Ksearch 1.6 and it works great !

But I have a problem. My Web site has an English and French version. In the French version my html pages have the charset UTF-8.  So all the words like é è ê are modified. I tried to find where to change the charset, but I can't 

1. when I set $USE_DBM = 1, if the site has about 10 web pages, the indexer.cgi could run successfully. However, if the site has a little bit more files, when I run the indexer.cgi, sometime, it goes to an error page saying"Internet Explorer cannot display the webpage", and sometime the indexer.cgi will never ends nor show any error message.
   When I set $USE_DBM = 0, running indexer.cgi can completed but some time gets message saying "Use of uninitialized value $meta_desc_entry in concatenation (.) or string at C:\Inetpub\wwwroot\kkte\ksearch\indexer.cgi line 397"

2. After I run the indexer.cgi, I tried the search function. It is very annoying. If the word I searched has less total matches, I can get the result, such as "Results 1-8 of 8 (0KB) for the word "ksearchfold" with 8 total matches". Then if I search for a word having many total matches (such like 66 total matches), when I click the "search" button, it goes to an error page "Internet Explorer cannot display the webpage"
3. Regarding the pdf search, I placed a very simple pdf file on the site. When I run the indexer.cgi, it always says "Ignoring PDF file 'C:/Inetpub/wwwroot/kkte/kt/tbd.pdf': filename has illegal characters"

Does anyone of you have any suggestion or advice about it?
Thank you in advance and happy holiday:)

For kScripts, I'm using a small, customized search box at the upper-right corner of each page. The search results page is also customized, with the CGI stuffed into one of my own page templates. Note that in the score, I removed the file size and date information.

Also, when changes are made to the site, do I need to re-index manually? I suppose I can set up a chron job, but I'm not sure how to handle the password.