Old Version History (V3.1 and earlier)
Click here for the latest version history
Note: We are no longer working on
Version 3.1, as development have moved on to Version
7.0, but you can download the final
4.2 release (Build 1013) here and get the Users
Guide for this version here.
The final V3.1 release (build 1022)
is available here. This is purely for supporting users of the previous version. We
recommend upgrading to Version 7.0 where
possible, as we have improved many aspects of Zoom and have addressed
many issues based on user feedback.
Version 3.1.1022 (5
- Fixed bug with missing last character in plugin
supported binary filenames that appear as the search results title
eg. "mydocument.pd" instead of "mydocument.pdf"
Version 3.1.1021 (1
- Increased max title length for a search result
to accomodate long filenames.
- Fixed bug with long filenames for PDF (and other
plugin-supported files), where the filename appearing in the title
of the search result is corrupted and/or joined to another page
- Fixed bug with converting HTML entities within
a URL causing only the first entity to be converted.
- Fixed bug with parsing registration usernames.
- Prevents start/pause/resume button from being
pressed whilst in the process of cleaning up after indexing (which
caused a crash).
- Fixed rare crash bug when using the "browse
folder" window on some computers.
Version 3.1.1020 (14
- Fixed bug when browsing local folders, causing
the indexer to crash on some computers.
- Fixed bug with downloading from https:// (secure
socket layer) URLs.
Version 3.1.1019 (13
- Fixed bug with running the PDF, DOC (and other)
plugins when Zoom is installed under a long file path, on some
Version 3.1.1018 (8
- Browse for folder window now opens to currently
specified directory instead of always opening in the root directory.
- Fixed browse for folder window, cancel no longer
resets existing directory entered.
- Fixed "Start points scanned" in the
Indexing Status tab (total number of start points incorrect)
- Fixed "File extensions scanned" in
the Indexing Status tab (bug with incorrect count)
Version 3.1.1017 (30
with windows-1251 (Cyrillic/Russian) so that search words do not
get corrupted in an incorrect encoding by the browser.
- Fixed logging bug in CGI version with the IP
address of the user not being recorded.
- Fixed logging bug in CGI and PHP versions with
additional carriage return characters on some platforms.
- Fixed path bug in Linux build of CGI search application
causing search forms to point to the absolute location of the
CGI rather than the relative location.
Version 3.1.1016 (12
- Fixed bug with redirected URLs which have not
been scanned, being skipped with message "Redirected URL
- Added support for indexing PDF files in UTF-8
- Added support for indexing words broken-up by
<span> tags as one word.
Version 3.1.1015 (10
- Fixed bug with retrieving context descriptions
for large sites, when performing an exact phrase search which
occurs in the title of a page (applies to PHP, ASP and CGI).
- Fixed bug with searching and indexing some UTF-8
- "Disable lowercase conversion..." option
now applies to the Indexer as well as the script, when indexing
Version 3.1.1014 (6
- Fixed bug with indexing HTML numeric entities
which were specified in hexadecimal (eg. ]). This resulted
in pages being partially indexed.
- Added indexing support for HTML "quote"
entities (eg. ", &ldquot, &lsquot;, etc.)
- Fixed bug in Statistics graphs: some accented
characters were missing in the graphs, and the graph image was
Version 3.1.1013 (27
- Fixed bug with ASP script, when exact phrase
matching some words on very large sites.
- Fixed bug with PHP script, when exact phrase
matching words containing word join characters.
- Trims trailing and leading spaces when importing
start points from text file.
- Fixed possible crash bug after "Not enough
Version 3.1.1012 (14
- Fixed "access violation" crash bug
when indexing meta data on computers with older versions of Internet
Explorer installed (versions prior to 6.0) .
- Fixed rare start up crash issue on some Windows
- Fixed missing "Result pages:" number
links in the CGI version.
- Search scripts now removes '/' and '\' from a
query so that searches containing slashes will not be invalid.
Version 3.1.1011 (9
- Fixed bug with indexing case variant words (this
caused some words appearing in the context description to not
return a hit for the page when searched for)
- Minor changes to search scripts (PHP and ASP).
Version 3.1.1010 (1
- Fixed bug with loading in an existing config
file (.zcfg) with the "Linkback URL" set. This may have
caused other configuration settings to be incorrect after loading,
such as the Results Format.
- Fixed crash bug with CGI version when searching
for exact phrase and logging is enabled.
Version 3.1.1009 (30
- Fixed bug with garbled context description on
large dictionary files - this bug also caused crashes when searching
for certain words using the CGI script.
Version 3.1.1008 (29
- Fixed crash bug when words exceed the max word
length with accent marks and "accent insensitivity"
- Fixed reported crash and incorrect meta description
and keywords retrieved when an invalid meta syntax is provided.
Version 3.1.1007 (28
- Fixed incorrect error message "Failed to
close pagetext file" whenever results format is set to "Simple
(no descriptions)" or "Meta description only"
- More changes made to the appearance of punctuation
marks in context descriptions.
Version 3.1.1006 (18
- Fixed crash bug when using command line parameter
for repot mode generator.
for iso-8859-2 code page (Polish).
- Added VBScript engine version check in ASP search
Version 3.1.1005 (15
links to result pages.
Version 3.1.1004 (10
- Fixed missing characters in the context description
- single occurances of characters enabled for "word joining"
(eg. "this - that" appeared as "this that").
- Fixed search script bug when searching with the
"match all word" option and query includes a skipped
- Fixed indexer "Edit HTML and CSS template
page (source)" menu option when CGI is selected - previously
opened the PHP/ASP template.
- Fixed highlighting bug when a search term (besides
the first) is a skipped word.
- Fixed highlighting bug in the CGI version when
"always substring match" is enabled.
- Fixed bug with CGI version printing out "These
words have been skipped..." when "always substring match"
is enabled (results were correct).
Version 3.1.1003 (7
- Fixed CGI problem with working in virtual paths
or virtual homes (Windows/IIS only)
- Improved context description for sites with many
unique words (over 100,000) and exact phrase searches
- Fixed crash bug with loading .zcfg files with
long additional start points and additional base URLs
- Fixed bug with handling plugin converted file
which exceeds max file size limit - all subsequent plugin files
were not scanned.
- Fixed bug with not scanning entire page when
multiple HTTP requests were required
- Removed incorrect "Queued URL" messages
that were printing out in verbose mode
- Added more error messages to CGI builds
- Updated default HTML template to include new
result_title CSS tag.
- Minor bug fixes: CGI template rendered incorrectly.
- Fixed offline mode base URL from being overwritten
by spider mode base URL.
- Fixed rare case for non-wordjoin characters in
Version 3.1 introduces the option of high-performance
searching with the CGI platform.
- New CGI platform
option: provides high performance searching
to support enterprise-scale sites (for example, over 60,000 pages
depending on the number of unique words). This options is currently
available for web servers running Windows, Linux, and BSD. More
information on our CGI platform page.
- Super fast indexing: Highly optimised
indexing process allows sites that previously took hours to index
to be indexed in a matter of minutes
- Additional spider URL features: override
base URL, index page and follow external links, spider options
for main start point and more.
- New plugin support for
WPD (Word Perfect documents) files.
- Improved search word highlighting method, faster,
and no limit necessary.
- and many other new features, and bug fixes.
Click here to jump to the other
big features introduced in Version 3.0
- Fixed display issues with DPI settings greater
than 96 DPI
- Fixed bug with indexing URLs in spider mode,
which have been redirected to a site outside of the base URL.
- Added log window messages when Statistics Reports
Generator is run from the command-line
- Fixed indexing of skipped words based on min.
- Fixed ASP script issue with creating invalid
log format when locale setting uses commas as decimal places
versions of IE not supporting the push function.
- Fixed indexer crash bug caused by accent insensitive
option and very long words
- Fixed bug with skipwords being indexed with the
- Fixed "mode" text for current mode
loaded from a .zcfg file
- Minor script changes
- Support for HTTP redirect to relative URLs
- Fixed crash bug in indexer caused by accent insensitive
option and very long words
- Fixed minor ASP script bugs (rounding off page
numbers, and skipped word consideration)
- Scripts reports plural form of "result(s)
- Fixed score calculation issue with non-weighted
- Stats generator no longer require log files to
end in ".log"
quote entries in index
- Fixed "Substring match for all searches"
option in PHP and ASP.
- Added "result" css tag for changing
the appearance of the results listing from the search template.
- Added section in Users Guide on "How to
customize the look of your search page".
- Add "What to do after indexing" section
to Help file.
- New "Browse Directory" windows with
"Create new folder" option
- Changed accent insensitive mappings to characters
more appropriate for most international languages. ("ä"
to "ae", etc.)
- Context description now shows matched word(s)
closer to the middle of the context extracted, rather than the
- Accent insensitivity option restricted to certain
- Removed the type ligature "oe" character
from accent insensitivity option, to prevent "o" from
being mapped to "oe" for character sets that omitt the
character (eg. utf-8)
- Category patterns are now case-insensitive
- Fixed "jumbled" context description
in PHP for index files where the longest entry is a non-indexed
- Fixed bug with accent insensitive feature for
- Fixed bug with PPT plugin ("can not open
- Fixed crash at end of indexing (bug re: charset
output on HTML template)
- Fixed "zoom_pagetext.tmp.zdat" not
being renamed to "zoom_pagetext.zdat" on Windows 98
- Fixed context description bug for sites with
unique words over 60,000
- Fixed commandline parameters in 98/ME
- Other minor changes
Version 3.0 features an overhaul of the existing design
and introduces a vast number of new features and improvements over
- Improved support for large websites
New algorithms and compression techniques optimize the indexing
process to use less memory and resources than before.
- Context descriptions
Search results can now show the surrounding words which appear
on the page where the matched word was found, providing a context-based
description for each search (PHP/ASP only)
- Exact phrase
Search for phrases (words in a specific order) when specified
in quotes. For example, you can search for "zoom search engine"
without matching "zoom engine search" (PHP/ASP only)
- Search statistics reports
Create HTML and GIF based reports with pie-charts and graphs
such as "Top 10 searches" and more. Click
here for more information.
- Multiple start spider URLs (Spider mode)
Index multiple sites on different domains or sub-domains by adding
them to a list of spider URLs. You no longer need "additional
base URLs" and a page to link your various sites and domains
together. You can simply add these URLs directly into the Indexer,
or use the Import/Export spider URL facility.
- Customizable HTML templates with CSS support
Search results are now rendered with CSS tags,
so you can completely change the appearance of the results without
making any changes to the script.
- Improved International language support
- Indexer now supports Unicode
- Specify encoding or charset and Zoom will
automatically configure your search templates and script for
- Improved UTF-8 support
- Accent insensitivity option. With
this option enabled, users searching for the word "cliché"
will find "cliche", and vice versa.
- Easier to use scripts and templates
All search script settings can now be configured from the
Indexer. This means no script hacking is required (but you are
still free to do so).
- Option to have result links open in a new window
- Pause and resume indexing
- User definable index word rules allow
for a greater variety of words to be selected for indexing. Searches
for monetary values ("$99"), dot-appended words or acronyms
("a.b.c."), version numbers ("V1.95.XP"),
and underscored words ("search_template") ... and more,
are now possible.
- Detailed indexing status gives your more
information about the indexing process. Including "number
of words indexed", "number of unique words per page",
"URLs crawled by spider", number of files scanned for
each file extension, and much more.
- <!--ZOOMSTOPFOLLOW--> and <!--ZOOMRESTARTFOLLOW-->
allows you to specify sections of a page which contains links
that the spider indexer should NOT follow.
- Many bug fixes and minor features
Zoom 2.1 Build 1004:
- Bug fixed for crashes due to invalid HTML links
- Added scanning keywords in .desc files for plugin supported
Zoom 2.1 Build 1003:
- Bug fixes for log length buffer
- Added support for HTML entities in four digit numeral
form (eg. greek and mathematical symbols)
Zoom 2.1 Build 1002:
- Fixed bug with scanning files with no extensions
- Improved ability to determine the absolute/resolved URL
of a redirected link in spider mode
- Bug fixes for PHP, ASP and JS script files re: total
result count, and substring matching
Zoom 2.1 Build 1001:
- Minor bug fixes for configuration window and memory estimation
Zoom 2.1 Build 1000:
- Plug-in support for .XLS and .PPT
files (Std and Pro editions only)
- Authentication support for indexing secure websites in spider
- Optimized all search scripts (search.php, search.asp, search.js)
producing much faster search results, of 2x, up to 10x the performance
of previous versions.
- Improved support for searching large sites of over a thousand
- New tabbed interface for the configuration window
- Introduced new mandatory index data file (zoom_keywords_idx.dat)
for the server-side search scripts to help improve search performance
Zoom 2.0 Build 1019:
- Now supporting and indexing decimal numbers, product/serial
numbers and abbreviations. Eg. "2.3" "4.1-102A",
- Fix for indexing and searching hyphenated words
- Fix for indexing words in keywords, description and title
on 'normal' weighting level
- Changed default settings in Indexer and search scripts
and templates to use "windows-1252" encoding (English,
French, German, etc.) instead of UTF-8
- Autorun now ensures it runs in the mode specified by
overriding the mode setting (offline or spider) saved in the .zcfg
Zoom 2.0 Build 1018:
- New features: Add categories support to
your website. Allow your users to restrict searches to certain
categories on your website from a drop-down box. Manage the categories
via the Indexer's configuration.
- Configurable word weighting for indexed words
found within the title, meta description, or
meta keywords/zoomwords (Indexer configuration)
- Option to enable or disable UTF-8 encoding for HTML character
entities (Indexer configuration)
Zoom 2.0 Build 1017:
- Additional verbose mode message when retrieving files
over HTTP in spider mode. To help with determining a slow or non-responsive
- Other minor code changes.
Zoom 2.0 Build 1016:
- Forces plugin temporary files to be write enabled (to
prevent a read-only temporary file due to files originating from
a CD-ROM source, or various network folder related issues)
- Extra check for previously existing temporary files (and
removes them accordingly)
Zoom 2.0 Build 1015:
- Added support for indexing UTF-8 encoded websites.
- HTML character entities are now converted to UTF-8 encoded
characters for indexing.
- Additional error messages for skipping files due to plugins
failing to write temporary files as well as files skipped due
to a beginning underscore character.
Zoom 2.0 Build 1014:
- All search script "result pages" links will
now wrap accordingly.
- Search scripts now only display up to 10 other pages
of search results either before or after the current page.
Zoom 2.0 Build 1013:
- Fixed support for Lotus Notes websites with .nsf directory
structure and no extension filenames.
Zoom 2.0 Build 1012:
1.5 (such as IE 5.0 or older). Note that some foreign language
support with extended charsets will require IE 5.5 or later to
- Temporary plugin buffer files are now being deleted automatically.
Zoom 2.0 Build 1011:
- Fixed issue with reading in .zcfg files that have been
modified in an external text editor with DOS style carriage returns.
- Modified attributes required for indexing: now indexes
all files except hidden and system files.
Zoom 2.0 Build 1010:
- Option to use meta descriptions when they are available
for page results - or to always use page content instead.
- Fix for search words beginning with an accented character
- Improved support for foreign (non-English) single-byte
Zoom 2.0 Build 1009:
engine from a CD-ROM or DVD or any offline distribution where
a PHP or ASP is not available, and where network traffic and latency
is not an issue.
- Option to scan files with out any extension (eg: "home",
- Bug fixes for long URLs (over 200 characters)
Zoom 2.0 Build 1008:
- Support for custom meta description files (.desc)
for plug-in supported extensions (Standard and Pro editions only)
- Support for <meta name="keywords"> tag
in addition to <meta name="ZOOMWORDS"> tag.
- Bug fixes for "Load Configuration" and "Save
Configuration As" problems with Win98
- Fixed issues with DPI display settings of greater than
- Improved performance of verbose mode
- Step-by-step guide, and other updates added to the Help
Zoom 2.0 Build 1007:
- Search script timing function
- Speed optimisations to the ASP script
- Bug fixes in ASP script to handle Unix style carriage
returns in the search_template.html
- Fixed trailing blanks problem in PHP script
- Indexer will now quit automatically when it is run in
auto-run mode (to allow for remote, web-based operation)
- Misc. bug fixes
Zoom 2.0 Build 1006:
- Option to accept multiple base URLs, allowing indexes
that span across multiple domains or sub-domains. This allows
you to index multiple domains and sub-domains together as one
site (eg: http://www.mywebsite.com, http://news.mywebsite.com)
- Search script: keywords highlighting feature
- Better support for older PHP servers (PHP 3.0 to PHP
Click here for the latest version history