|
 |
The
visual|catalogues
Search Engine - Technical
Issues
Searches
on a catalogue distributed on CD-Rom, electronic business card,
local network or local hard drive are carried out using the visual|catalogues
viewer for Windows (95,98,2000,NT
4.0). This interface will also soon allow searches to be carried out
optionally on an on-line resource. Searches on an Internet web-page
based catalogue are carried out use the visual|catalogues
Internet viewer.
Both these viewers are supplied as part of the distributable
software included with this system.
Suitable
ISP
If
you want to use this software to create a searchable catalogue to
run from your Internet site server, you will need to check
with your ISP what type of computer (hardware) (PC, HP, Sun, Mac
etc.), and operating system (Windows, Mac, Unix, Linux) they are
using to hold their cgi-bin,
and
if there is a visual|catalogues search
engine version compiled for that
hardware/software combination. See below for further information, and
visit the downloads
page for the most up-to-date list of compilations of the search
engine. Note that if this is all too complex we offer catalogue
hosting services. and your
catalogue can be linked to your current web pages seemlessly. Contact
us for more details.
The
following is a technical explanation of how the visual|catalogue
search engine works, along with
issues relating to its installation
and use.
Technical
Searches
are generated from keywords you input against each of the items you
catalogue. You add relevant keywords to your catalogue and when
you build the catalogue for publication the visual|catalogues Professional
and Enterprise
versions automatically generate a flat-file
database tables (if
you have licensed the software!).
A
flat-file
database is the simplest format that a
database can be created in. If properly structured it is also
arguably the fastest to access, since all data is held in long binary
format, accessed directly at run time, and special techniques such as
binary chop can be used to speed access to records. A flat file
database is not dependent on intermediate program code for
interpretation. For example reading a flat file database does not
rely on a proprietary database system such as Microsoft Access, which
also requires the Microsoft Windows operating system to be installed
(excellent though these programs are !). A flat file database can
happily be stored on and accessed from any suitable computer
hardware or operating system, whether Internet server or not, without
regard to the hardware or operating system software that is being run.
The
down side of this is that to read the data in a flat file database
on any specific computer hardware and operating system combination
requires a search engine program (executable file) to be compiled
for that combination of computer hardware and operating system.
The visual|catalogues
search engine is written in the C
programing language for maximum portability.
For
CD-Rom based catalogues the search engine is already compiled
to run on Windows (95,98,2000,and NT 4) based systems. So If you
click the "Publish to CD"
button in the v|c
data entr modules (Lite, Pro, Enterprise), all required files,
including the search engine and flat file database tables, are
automatically copied into the drive/directory you are using to hold
your CD data while you cut your CD.
For
Internet publishing these files need to be copied to your web site
server. So If you click the "Publish to Web Site",
the software automates this copying process if you have added the http://
address to copy to. Refer to the relevant help topic in the
documentation that came with your software for the steps to go
through to achieve this.
Unless
you have your own Internet server, you will be copying your
catalogue and all the files it needs onto your ISP's (Interent
Service Provider) site. Many ISP's maintain a seperate computer on
their server network to run cgi scripts and other executable files,
often held in a directory called \cgi-bin. You may well have
your own cgi-bin. For searches to work from your web site your ISP
must allow you to run cgi executable programs. Your ISP will
understand all this and advise. It is to the cgi-server that holds
the cgi-bin you have access to, that the search engine and database
files for your catalogue need to be copied. Again if you give the
software the address of your \cgi-bin directory the ftp function will
copy all the relevant files over for you.
However
before you do this you must first establish what computer hardware
and operating system your ISP is using for the server you are copying
to, and if we have a search engine version compiled for that
hardware/software combination. See the downloads
page for the most up-to-date list of compilations of the search
engine for different platforms, and to request a compilation for a
specific computer hardware/operating system combination. We undertake
that, wherever possible, we will be adding to the platforms and OS's
that the search engine is compiled for, as required by users.
If
this is a problem, and because it may be more convenient for some
customers, we can also offer catalogue
hosting services. This service will allow you to download
all files (or just your search files) to your own secure space on our
server system. You will have to be prepared to pay the bandwidth
costs (volume of data transferred to and from the site as a result of
you uploading data to the site, and users accessing your catalogue).
See bandwith prices.
If
we host your catalogue you can of course link to this from your own
or any other web site you wish to provide a link from. |
