************************************************************************
* Orca Search v2.0b *
* A robust auto-spidering search engine for single/multiple sites *
* Copyright (C) 2005 GreyWyvern *
* *
* This program is free software; you can redistribute it and/or modify *
* it under the terms of the GNU General Public License as published by *
* the Free Software Foundation; either version 2 of the License, or *
* (at your option) any later version. *
* *
* This program is distributed in the hope that it will be useful, *
* but WITHOUT ANY WARRANTY; without even the implied warranty of *
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the *
* GNU General Public License for more details. *
* *
* You should have received a copy of the GNU General Public License *
* along with this program; if not, write to the Free Software *
* Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 *
* USA *
************************************************************************
*** Changelog
- See changelog.txt for this script's complete change history
*** Contents
1. Script Requirements
2. Quick Start
3. Introduction
4. Upgrading
5. Detailed Installation
6. Spider Configuration
7. Crontab Spidering
8. Entry List Panel
9. Search Options
10. Searching
11. Sitemap
12. JWriter
13. FAQ (see faq.txt)
************************************************************************
1. Script Requirements:
PHP 4.2.0+ (4.3.0+ recommended)
MySQL 3.23+
2. Quick Start
If you are an advanced user and wish to skip the hand-holding guide
later on in this readme file, you can use the Quick Start guide below.
Good Luck!
i. Extract all the files from the zipped package to a temporary
local directory.
ii. Open the config.php file and edit the seven user variables to your
preference. The fifth variable is the prefix the script will use
to create three tables in your database. The sixth and seventh
will be the username and password you'll use to log in to the
Control Panel.
iii. In your www or public HTML folder, create a folder named "os2".
iv. Upload the following files into the "os2" directory:
body.xhtml.php
body.xhtml.css
config.php
control.css
control.php
head.php
spider.php
lang.txt
v. Upload the following file into "os2"s parent folder:
_search.php
vi. If you plan on using the offline javascript search (JWriter),
upload "jwriter.php" and "egg.js" into the "os2" directory. If
you plan on using the Google sitemap generator, upload
"sitemap.xml" into "os2"s parent folder. CHMOD "egg.js" and
"sitemap.xml" to 766 or 777. If you are using a Windows server,
set appropriate permissions so PHP can read and write to these
files.
vii. Visit "/os2/control.php" with your web browser. On your first
visit to this page, it will create the tables required for
running the script. If there are no errors, installation was a
success. You can now log into the Control Panel and configure
the script as you wish.
3. Introduction
Welcome to the Orca Search script. This script will index the pages
at a single domain, or group of specified domains by spidering the
contents at an interval you specify. You may even set up the spider to
be run at a specific time via *NIX cron tab.
Once the script is up and running, you'll need to teach your spider
what to eat and what to leave alone. This will take some tweaking for
days or weeks to come, but eventually, what you'll end up with is an
automatic self-updating search system you never have to think about
again! Well, maybe not "never", but pretty close :)
BEFORE YOU BEGIN: Please follow the installation steps as close as
you are able. Doing things out of order may require you to start over
from scratch. You have been warned.
IF YOU ARE UPGRADING FROM A BETA VERSION: The script does not have an
automatic upgrade system from beta versions. Installing over a previous
version will NOT work as expected, unless you open phpMyAdmin and
manually make all the visible MySQL table changes. It is recommended
that you instead install the new version into a fresh directory using a
new MySQL table. Then manually transfer the settings of the old version
to the new via each GUI. After this is done, change the name of the new
directory to the name of the old one.
************************************************************************
* Please report any bugs to: orcas <at> greywyvern <dot> com *
************************************************************************
4. Upgrading
The Orca Search is designed to make upgrading easy! Upgrade code is
included which will silently upgrade any version 2.0 Final or later, to
the version of the release you have just downloaded. If, rather, you
are installing this script fresh, you can skip to the installation
section #5 later on in this manual.
To upgrade your Orca Search, just follow these simple steps:
i. From your *existing* installation, open the "config.php" file in
any text editor.
ii. Copy the seven setup variables, five for MySQL access and two for
Control Panel username and password. I usually copy the entire
beginning of this file down to the "Functions" comment lines.
iii. Extract all the files from this package into a temporary
directory and open the *new* "config.php".
iv. Paste the old config variables (the ones you copied) into the new
"config.php", overwriting the ones that are already there. Save
this file.
v. Overwrite all files in the "os2" directory (or whatever you named
this subdirectory) on your webserver, with the corresponding files
from the temporary directory (the new version files).
vi. On the next visit to the Control Panel or search interface with
any web browser, the Orca Search will be automatically upgraded!
5. Detailed Installation
i. Ensure you have all files
The Orca Search v2.0 attempts to be completely modular, unlike the
other scripts in the Orca series. Most pieces of the script are 100%
swappable with newer or modified versions, should they become available.
Each file has a specific function which can be built upon by future
modders. As such, there are three types of files for this script:
Core, Output and Tools.
So let's run through all the files you got in your Orca Search package:
There are six (6) Core files. These files MUST be installed in order
for the search script to work:
config.php - Global configuration file
control.css - CSS for Control Panel
control.php - The Control Panel
head.php - Accepts search requests and builds an array of results
lang.txt - Language cartridge for spider and control panel
spider.php - Crawls your site and indexes pages
The default language file is English. To use a different language file,
if available, name the file "lang.txt" and overwrite the existing text
file. You may need to adjust the Control Panel Display Charset as per
instructions in the language file you are now using (explained later).
The control panel script will always look for a file named "lang.txt" in
the same directory in which the control.php file resides.
There are three (3) Output files. Two of them, "body.xhtml.php" and
"body.xhtml.css" go together, while "body.rss20.php" stands alone. Only
one set of these needs to be installed depending on what type of output
you want your search script to generate:
body.xhtml.php + body.xhtml.css
- Generates XHTML output from a result array
body.rss20.php - Generates an RSS 2.0 feed of search results
There are four (4) Tools files. The first three files are for a tool
called the JWriter. This tool will take the data from your search
database and compress it into a javascript file which can be used to
search a site which has been mirrored for offline use by a program such
as HTTrack. I've designed the JWriter mainly using output from this
mirroring program so I highly recommend it: <http://www.httrack.com/>
The JWriter tool uses these files:
egg.js - Offline Javascript target file
jwriter.php - JWriter workhorse
_search.html - Sample offline search page
The third file is the target file for Sitemap writing:
sitemap.xml - Empty Sitemap file
Google is currently beta testing a sitemap service, where you can submit
an XML or gzipped XML list of pages at the site you want indexed. The
script will output this XML data into the sitemap.xml file. There is
also the option to gzip this data to save space and bandwidth. If you
select this option, you will need to rename this file with an .xml.gz
extension.
Among all of these files there is a small file named "_search.php".
This file shows a sample XHTML search result page (using the
"body.xhtml.php" output file) and where each piece of the search engine
is included. Following the same system of PHP includes, you should be
able to embed the search engine into your already existing webpage
layout.
The .zip file also includes an .htaccess file which turns off your
server's zlib output compression within the os2/ directory. The
spider's progressive output requires delivery of the page to the browser
in plain HTML. You don't need to upload this file if you are having no
problems viewing the spider output, when triggered from the Control
Panel. Servers usually have zlib compression turned off by default.
ii. Edit config.php
Before uploading, open the "config.php" file. There is a short list
of variables you need to assign manually in order for the script to work
on your server.
First, there are five variables under the MySQL header. These variables
allow the script to access the MySQL database system on your server to
manage script data and store search indices. If you don't know these
variables offhand, ask your host.
The first four variables will be specific to your server and MySQL
installation. However, the fifth variable will be used as a prefix for
creating three tables in your database. You can give this variable any
name you want, as long as it is only letters and numbers and does not
begin with a number.
Next there are two variables under the Admin header. These are the
login name and password for the Control Panel. Change them to something
hard to guess!
Once everything is the way you like it, you can save and close the
"config.php" file.
iii. Upload the files
Create a directory in your public HTTP area to contain the search
script files. The default is "os2" but you can specify any directory
you want, provided you change the include statements in any search page
you create, and in appropriate places in the Control Panel, to point to
the new directory.
**********************************************************************
* For the purpose of convenience, the remainder of this manual will *
* assume you are using the default "os2" directory and the XHTML *
* Output files! *
**********************************************************************
Upload the following Core files into the "os2" directory:
config.php
control.css
control.php
head.php
lang.txt
spider.php
Then upload the following Output files into the "os2" directory:
body.xhtml.php
body.xhtml.css
Finally, upload the following file into the parent directory of "os2":
_search.php
After you have uploaded all the files, your directory structure should
look like this: (* if desired)
/_search.php
/os2/body.xhtml.php
/os2/body.xhtml.css
/os2/config.php
/os2/control.css
/os2/control.php
/os2/head.php
/os2/lang.txt
/os2/spider.php
iv. Activate the Control Panel
When the above files have been uploaded, visit "os2/control.php" via
HTTP with your web browser. If you are prompted with a login screen,
script setup was a success! The Control Panel is now installed and
ready to be configured.
6. Spider Configuration
i. Logging in and setting up
Log in using the username and password you entered in the "config.php"
file. Once you log in, you'll be confronted by the Spider setup area.
Scroll down and set all your spidering options, they should be well
explained by the help text included in the form. Some of the more
obscure form elements will be explained below. Make sure you "Submit"
to save your changes!
If you have trouble entering or viewing special characters in any of the
fields, the problem may be that the Control Panel is not being served in
the character set of your input. If this is the case, click the Tools
button in the menu and change the Display Charset to your preference.
If you load a different language file than the default (English), make
sure to check which character set(s) it is compatible with. The Control
Panel Display Charset *must* be changed to match the language file or
else dialogues may not display properly.
ii. Automatic Categorisation
There is a special textarea in the spider options labelled "Automatic
Categorisation". You can use this field to make the spider
automatically assign certain categories to newly found pages. The field
uses this special syntax:
CategoryName:::URIMatch
or
CategoryName;;;TitleMatch
Type the name of the category you want automatically assigned first,
then choose whether a URI or Title string match will work best. If URI,
add three colons, if Title, add three semi-colons. After the colons/
semi-colons, type a plain text matching string which will trigger the
assignment of this category. These matches will be compared to the
entire title or entire URI (including the "http://"). Here are some
examples:
a) Products:::products/
This rule will match these example URIs:
http://www.example.com/products/
http://www.example.com/products/item1.php
http://www.example.com/donotindex/products/item1.php
If any of these URIs are found, they will automatically be assigned to
the "Products" category.
b) My Blog;;;My Blog
This rule will match these example Titles:
My Blog
My Blog - A Day At The Farm
Add a Comment to My Blog
If any of these Titles are found, they will automatically be assigned
to the "My Blog" category.
Because of the three colons and semi-colons system being used, you
cannot assign category names which contain these character sequences.
However, the match string may contain them, if needed.
The results of the spider can also be emailed to you. The Email Results
field accepts email addresses in the same format as PHP's mail()
function. Read more about it here:
http://php.net/manual/en/function.mail.php
iii. Starting the engine
After the spider has been found and you have set all your desired Spider
options, hit the "Go" button in the form up top to begin the spider.
Then watch it crawl! Because crawling a site requires a lot of error
tolerance, if anything goes wrong with this search script, it will
probably happen now. If an error does happen, the spider will stop and
a message will be displayed. As I aim to make this script work with as
many different PHP installations and URI formats as possible, if you
could email me any error messages I would be very grateful :) As we
move on, I will assume the spider completed its crawl sucessfully.
Unless you were really meticulous, you'll probably notice that the
spider ate a lot more or a lot less than you were expecting it to. This
is normal. Just look through the list of files the spider ate, adjust
your rules on the Spider page, and try again. You don't have to get it
perfect just yet though, since you can make manual edits using the Entry
List section.
Remember, that if the spider crawled and indexed some pages it wasn't
supposed to, those pages will not be purged if you add a blocking rule
and spider again. They will only be marked as "Blocked" or "Unread".
You will need to manually delete them from the Entry List panel.
7. Crontab Spidering
This section assumes you are hosting the spider on a *NIX server which
has a crontab daemon installed.
To trigger a spider by crontab, you'll need to check the "Enable Cron"
checkbox in the Spider panel. This only tells the script to expect a
crontab trigger, you'll have to set up the crontab yourself using the
server tools you have available. Usually your host will have provided a
control panel which allows you to set up a crontab.
Note that enabling the crontab trigger will disable PHP-based automatic
spidering, spider result emailing and spider HTML output. The spider
will depend entirely on you to provide the correct trigger, otherwise it
will do nothing but ignore requests to spider.
The rest of this guide assumes you know how the basic options of a
crontab work. See the Wikipedia entry on the crontab command for more
information: <http://en.wikipedia.org/wiki/Crontab>
As for the value of the crontab itself, first you will need to know the
path to the PHP binary on your server. It is usually "usr/bin/php" but
depending on who set up the server it may be somewhere else. If so,
you'll need to find the correct path by asking your host.
Next, you need to know the full server path to the spider script. It
needs to be on the same server as the crontab daemon and the PHP binary.
You can find this by examining a phpinfo() file on your server. It
should list your website's server path several times.
Once you know where the spider file is located, the value of the crontab
should look like this:
/usr/bin/php /path/to/the/spider.php
... where the first path is your path to the PHP binary and the second
is the path to your spider script. This crontab will call the PHP
binary and tell it to execute your spider. So set your crontab times
however you like and save this.
Your spider should now run by crontab, although make sure you receive an
email notification at first (usually automatic) so you know it is
running properly.
8. Entry List Panel
Click the "Entry List" button in the menu to go to the Entry List.
Here you'll find a big list of every page your spider has indexed. You
can go through and make any edits you want, like adding custom keywords,
titles and descriptions to entries, changing their category and even
manually unlisting and/or deleting them.
It's important to note right away, that even if you have many thousands
of pages listed, you can use the various filters in the Filters row to
narrow down your list. Make it a habit to experiment with the different
filters and using them together for very powerful matching.
The Entry List has four main columns: Title/URI; Category; Status; and
Edit/Sitemap. You can sort the list based on Title, URI and Category;
just use the links along the top. Initially, you will see each entry's
URI, Category and Status listed.
Each entry falls into one of Five status types:
OK - Page is indexed and ready for searching
Orphan - Page is indexed, but is one of:
- not linked to from any other page
- not within your specified Allowed Domains
Unlisted - Page is indexed but will not be searched
Unread - Page is not indexed because it was at least one of:
- unreadable MIME-type
- recently added manually (not yet spidered)
- URI was HTTP redirected elsewhere
Blocked - Page is not indexed because it was at least one of:
- blocked via robots.txt
- blocked by a user-defined Ignore URI rule
- contains a <meta> tag redirect
Not Found - Page is not indexed
- This page used to be indexed but can no longer be found
You can filter entries by status type by selecting a status from the
dropdown menu and pressing the "Set" button to set this filter. When
an entry's status is displayed in boldface, it means it has been
manually set using the Edit button.
If you only ran the spider once before coming to the List page, you
will notice that all of the entry URI's down the left-hand of the page
are listed in boldface as well. After each spider, new pages, or
existing pages with updated content are marked in bold. You can filter
the list to view only pages updated since the last spider by checking
the "New" checkbox next to the "Filters" title and hitting the "Set"
button. If there are no "New" pages in your list, this checkbox will be
disabled.
By default, the Entry List lists 100 entries per page. To change this,
use the text field above the column containing the "Edit" buttons. Your
selection will be remembered until you change it, even if you log out.
Get used to using this interface, as it will become your main interface
for managing the pages crawled by your spider! If you have any
suggestions for making it easier to use, or for new features, be sure to
let me know :)
9. Search Options
After massaging your newly indexed database, you can set some search
options in the "Search" panel. These options will affect what results
ultimately get sent to the user after a query.
You don't need to change anything here for the script to work properly.
These are mainly "tweak" type settings, so you'll probably want to come
back to it after you've run the search script for a few days. If so,
feel free to skip to the next section for now.
At the top of the page you'll find some simple cache settings. If you
are hard up for MySQL database space, you can limit the size of the
cache and/or enable GZip compression if your server supports it.
Below that are the main search options. The first large text box is
probably the element you'll end up editing most. Entries with URIs
matching lines in this textarea are prevented from appearing in any
search results. This is a useful complement to the "Ignore URIs" in the
Spider panel since the pages you block here will still be spidered for
links. For example, if one of your pages is just a few links to other
pages on your site, it would be a good idea to list it here, since it
would be practically useless as a search result, but you still want the
spider to crawl those links.
The script also allows you to change the weighting applied to any match
as you see fit. The first group of values are arithmetic bonuses; they
will be added for each match found in a specific section of an entry, to
a maximum of three to prevent spamming. To disable searching of any
location within an entry, just set the corresponding Match Weight to
zero (0). URI is set to zero, and thus disabled, by default.
The second group are geometric bonuses; the total score for each entry
will be multiplied by these bonuses for each instance they apply. For
example, if someone searched for a query with three terms and on single
entry had all three, the final score for that entry will be multiplied
by the Multi-word Bonus twice; once for each additional term found.
A useful option here is Latin Accent Matching. Latin Accent Matching
enables search queries such as "starkste" to match "stärkste". Because
it allows for many more possible matches, enabling Latin Accent Matching
will slow down search speeds significantly, especially if you are
indexing many pages, so think about whether you actually need it before
enabling it. This option should only be enabled if the pages you are
indexing are in either UTF-8 (with UTF-8 Indexing enabled) or ISO-8859-1
encoding.
The Maximum Returned Results field determines the maximum number of
search results any single query can return. If you set this to a value
other than zero (0), the script will use that value as a hard limit.
If you set it to zero, the script will try to calculate an appropriate
number of search results to return.
When an entry is displayed as a search result, some text from the page
will be displayed along with it with the search term highlighted. You
can limit the amount of this text by adjusting the Maximum Matched Text
Displayed value.
Finally, by default, the script will not display pages which have the
"Orphan" status in any search results. You can display them anyway by
enabling the Show Orphans option.
The other items in the form should be adequately explained in the inline
help text.
10. Searching
Visit the "_search.php" page with your web browser. You should find
a search input form waiting to be used. If you added more categories
than just "Main" using the Entry List page, the ability to filter
results by category will appear via a handy dropdown box.
A feature of the Orca Search is that the body.xhtml.php file, which
displays the output, can be modified however you wish to fit your own
website style. It takes all of the output from the "head.php" file,
interprets it, and displays it in a logical form. If you want to make
your own output file(s), examine "head.php" for a description of the PHP
variables that file creates.
It is possible to create practically any output format from the
"head.php" output, even archiving it straight into another database!
All that's needed is an appropriate body file to crunch the output.
If you want to add a searchbox elsewhere on your site, just use one of
these sample bits of HTML code:
i. Search box with submit button:
<form action="_search.php" method="get">
<input type="text" name="q">
<input type="submit" value="Search">
</form>
... replace _search.php with the name of your search page.
ii. Without submit button (press enter to submit):
<form action="_search.php" method="get">
<label>Search: <input type="text" name="q"></label>
</form>
To preselect a category for any search box, just add the following
<input> element to the form, replacing categoryname with the name of
your desired category:
<input type="hidden" name="c" value="categoryname">
_OR_ you can add the category drop down menu from the search output
page. The category drop-down menu appears there only if you have more
than one category, and is dynamically generated depending on what
categories you have. If you'd like to include this drop down in a
search box on another page, just copy the <select> element from the HTML
on the output page and paste it, as-is into your form. It's that
simple! Make sure you recopy the HTML if you ever add more categories.
11. Sitemap
If you're happy with your new search engine, let's look at the other
tools of the Orca Search script. First, click on the Tools button in
the menu to go to the Tools panel. Initially there won't be much to see
on this page. Both the Sitemap and JWriter dialogues will be collapsed,
so check the "Enable Sitemap" box and hit Submit to make the Sitemap
options appear.
If you haven't uploaded the "sitemap.xml" file yet, you'll notice that
the script will be searching for it now and will tell you if it cannot
be found. Since sitemaps are only applicable to pages beneath them in
the directory tree, it is recommended to upload the sitemap file into
the root HTML directory.
The default location for the sitemap might be incorrect, so make sure
it's pointing to the correct location. The rest of the items in this
dialogue should be relatively self-explanatory. I recommend enabling
gzipping and the Automatic Changefreq option. Make sure to rename the
sitemap file from an .xml extension to an .xml.gz extension if you
enable gzipping. See Google's sitemap FAQ for more information about
sitemaps and how search engines deal with them:
<http://www.google.com/webmasters/sitemaps/docs/en/about.html>
Once the options are set up to your liking (remember to hit "Submit"!),
return to the Spider panel and initiate another spider. Just before the
spider finishes, it will output a fresh sitemap, ready for submission to
search engines!
Enabling the sitemap will also add editing options to the Entry List
section which you may want to check out. The last column will display a
small bar and a letter, indicating the sitemap priority and change
frequency respectively. You can hover over any sitemap element here to
see more information. Just hit the Edit button to manually adjust an
individual page's sitemap settings.
12. JWriter
The Orca Search, in all its glory, only works with a PHP/MySQL backend
to support it. However, if you need it to work, in a limited fashion,
with an offline copy, the script can compress and output a copy of the
database in javascript. This is great for having your search engine
still function if you mirror your site to CD, for example.
As a mirroring program, I recommend HTTrack, especially since the
JWriter has been designed using this program's output. Visit their
website: <http://www.httrack.com/> for more information. Despite
favouring HTTrack, the Javascript output should work well enough with
any decent offline mirroring program, as long as it does not rename
downloaded HTML-content files (other than their extensions).
******************************************
****************** NOTE ******************
The JWriter works best with sites that do not use query strings to
reference content pages. Many popular Content Management Systems these
days, such as PostNuke, use a query string system for page access:
- http://www.example.com/modules.php?op=modload&name=gallery
Such a link would be renamed to a static HTML file by HTTrack like so:
- /modules0y5sj.html
However, the URI with the query string would remain in the JWriter
output with just the extension altered, requiring you to manually change
the URIs in the egg.js file.
- /modules.html?op=modload&name=gallery
******************************************
******************************************
First you will need to upload the JWriter tools files. These are the
"egg.js" and "jwriter.php" files and the "_search.html" sample page.
Upload "egg.js" into the "os2" directory and place "_search.html" into
the parent directory, like so:
/_search.html
/os2/egg.js
/os2/jwriter.php
PHP will need permission to write to "egg.js" so apply the appropriate
properties to this file, like CHMOD'ing to 766 or 777.
Next, head back to the Control Panel and check the Enable JWriter
checkbox to make the JWriter interface appear. There are two parts to
the interface: an options form above, and a "Write to File" button
below. Some of these items may not appear initially if the paths to the
jwriter.php and egg.js files are not correct. Make sure these are
pointed to the correct locations.
Although the options are relatively well-explained via the inline help,
you may find that you'll understand them better if you actually mirror
your site and examine the output. File extensions get renamed, full
URI's become relative, and "/" URIs are given an actual filename,
usually index.html. All of these actions can be accounted for using the
JWriter, just set them up to match.
Since the JWriter wraps up searching and output all in one, you'll need
to specify how many search results you want per page, and the HTML
template to use for search results; just like is required in the
"body.xhtml.php" file. Like the XHTML output file, the HTML template
here uses a system of replace-codes to insert result-data in the correct
HTML tags. These codes in both systems are identical and are as
follows:
{R_NUMBER} - Number of the result in the result list
{R_RELEVANCE} - Relevance score to one decimal place
{R_URI} - Result URI
{R_CATEGORY} - Result Category
{R_TITLE} - Result Title
{R_DESCRIPTION} - Result Description
- If there is no page description, the first 200 characters of body
text will be used instead
{R_MATCH} - A selection of body text with highlighted matches
- If there is no match-text, the contents of the description will be
used instead
These codes will be globally replaced in the HTML result template, thus
any single code can be used more than once and all instances will be
replaced.
When you are ready, hit the "Write to File" button to watch the script
go to work. A percentage complete reading will be listed, along with an
estimate of the script's memory usage. If the script times out and gets
stuck at a certain percentage for a few minutes, the problem may be that
PHP does not have enough memory to complete the compression. The script
will try to increase the memory limit before this happens, but presently
that is all it will try. The JWriter is not yet robust enough to
overcome this, but hopefully it will be in the future.
If your "egg laying" was sucessful, you will be taken back to the
control panel where you can continue on with other modifications.
Download the egg.js file and include it in a <script> tag where you want
your search results to appear like so:
<script type="text/javascript" src="egg.js"></script>
The script will capture the same query strings that the online version
uses and output search results using the exact same format as the
default body.xhtml.php file. These search scripts should be very close
in functionality, but due to the compression process, results won't be
identical, and there are no phrase matches.
************************************************************************
* Please send all bug reports and questions to orcas <at> greywyvern <dot> com *
* *
* Thanks for choosing the Orca Search script! *
************************************************************************