Old version history (V4.2 and earlier)
Click here for the latest version history
Note: We are no longer working on
Version 4.2, as development have moved on to Version
6.0, but you can download the final 4.2 release (Build 1013)
here and get the Users
Guide for this version here.
This is purely for supporting users of the previous version. We
recommend upgrading to Version 6.0 where
possible, as we have improved many aspects of Zoom and have addressed
many issues based on user feedback.
Version 4.2.1013
(15 May 2006)
- Fixed rare stalling bug when indexing from an
unreliable internet connection or bad web server causing internet
read faults. Previously, this may have caused Zoom to be stuck
at a start point and not move on to the next one.
Version 4.2.1012
(26 April 2006)
- Fixed crash bug caused by buffer overflow when
indexing a large number of pages without meta descriptions or
titles.
- Updated configuration window to indicate that
the skip underscore option applies for Offline Mode only.
Version 4.2.1011
(10 April 2006)
- Fixed bug in the PHP and Javascript search scripts
that caused some characters to be dropped from the search query
when followed by a space.
- Fixed bug with remembering last window size and
position when a user closes the ZoomIndexer application while
it is minimized or maximized. This caused Zoom to start up next
time outside of the visible desktop (if it was minimized), or
sized to the full size of the desktop (if it was maximized).
- Fixed issue with the latest PDF plugin, where
complicated layouts may cause words in the content to be indexed
out of order.
- Fixed bug with the "Unique words found"
tally occasionally reporting inaccurately at the end of indexing
in the "Index Summary".
Version 4.2.1010
(14 Feb 2006)
- Fixed crash bug when indexing fails (due to out
of memory, download error, etc.), and Zoom is attempting to clean
up memory.
- Fixed issues with pausing/resuming threads to
avoid a race condition which may cause Zoom to stop after indexing
and not finish.
- Fixed bug when reporting file size for a file
which the web server fails to report content-length.
- Fixed bug with download progress bars in the
Indexer window.
- Fixed bug with Indexing Status sometimes reporting
"URLs visited by spider" as "N/A" after indexing.
- Fixed bug when attempting to load in too many
skip words and/or scan extensions beyond the maximum allowed in
the Configuration window.
- Fixed bugs in search scripts (PHP, ASP, JS) dealing
with the handling of punctuation characters in search queries.
- Fixed bug in ASP script causing exact phrase
searches to fail when the phrase occurs near the end of the web
page.
- Fixed bug in Javascript version when indexing
words with "comma" enabled as a wordjoin character.
- Fixed bug in Javascript version causing the indexed
page score to be less than 255.
Version 4.2.1009
(30 Jan 2006)
- Fixed crash bug introduced in build 4.2.1008
which caused "Access violation" errors when downloading
files larger than the "Max file size limit" specified.
Version 4.2.1008
(27 Jan 2006)
- Fixed ASP search script bug causing exclusion
searches to be handled incorrectly.
- Fixed bug with using custom titles from .DESC
files for an Excel document (ie: titles in ".xls.desc"
files).
- Added error message and error handling for invalid
security certificate when spider indexing a secure website.
- Fixed Indexer issues when dealing with very short
meta descriptions and titles containing HTML entities.
- Fixed minor spelling suggestion issues for words
containing the letter "X".
- Fixed issues with Indexer when spider crawling
multibyte encoded files on a slow server which caused some multibyte
words to be corrupted.
Version 4.2.1007
(11 Jan 2006)
- Corrected a translation mistake in the Dutch
language file (Dutch.zlang).
- Fixed bug in Indexer (Spider mode) when indexing
multiple start points, and using the "Max. pages per start
point" limit causing error messages such as "406 HTTP
error" and "Server redirect error".
- Fixed "unable to open file" error when
attempting to open a log file in the same folder as the statistics
report.
- Fixed bug causing the Indexer to crash when attempting
to index a file (in Offline Mode) which has an invalid path name
that is over the limit of the Windows operating system.
- Increased the length of the acceptable login
name that can be entered in the "Authentication" configuration
window to 40 (to allow for "DomainName\UserName" style
logins).
- Added error messages when a user attempts to
enter a login or password that is too long.
- Fixed possible problem with PHP script when magic
quotes are enabled on the server.
- Fixed issues in the PHP script with performing
case insensitive searches for words in a foreign language (with
UTF-8 encoding) such as Russian.
- Fixed bug in PHP script which caused a warning
message ("Warning: strpos() [function.strpos]: Empty delimiter")
to appear occasionally, when spelling suggestions is enabled.
- Fixed bug causing the Javascript version to not
correctly match words which begin with a number, but contains
an alphabetic character in a different uppercase/lowercase form
than the search query.
- Fixed "option explicit" compatibility
in ASP script.
Version 4.2.1006
(30 Nov 2005)
- Fixed crash bug causing the Indexer to crash
when encountering a Content-Disposition header with an empty filename
attribute.
- Improved XSS patch in the Javascript version.
Version 4.2.1005 (24
Nov 2005)
- Added support for Thai charset (windows-874)
as well as iso-8859-9 and utf-85.
- Fixed bug with not indexing titles from PPT files
when "Use meta info from plugins" is enabled.
- Fixed bug with incorrect skipped messages displayed
in Offline Mode.
- Fixed bugs with the statistic report generator
(pie chart legend was occasionally cropped, and some log files
were being incorrectly identified as being invalid.
- Fixed bug with reporting the wrong line number
in error messages when parsing searchwords log file.
- Updated PHP version so that it can now be included
from other PHP pages located in different directories.
- Fixed minor bugs in the ASP version with the
spelling suggestion feature and some other errors.
- Patched XSS vulnerability in the Javascript version.
Version 4.2.1004 (27
Oct 2005)
- Fixed bug with not correctly indexing plugin
supported files linked via "file://" style links in
Spider Mode.
- Fixed bug with indexing incorrect title, description
or datetime from "file://" linked documents.
- Fixed bug with spelling suggestions in PHP search
script.
- Added verbose indexing messages to Offline Mode.
- Added "application/x-shockwave-flash"
content-type support
- Added precautions to saved window positions so
as to avoid the Indexer window starting outside of the visible
desktop area.
- Added error message when user attempts to add
a category with a name which already exists.
- Added support for <strong> and <em>
tags, to avoid splitting words which contained these tags inside
them.
Version 4.2.1003
(6 Oct 2005)
- Fixed crash bug when a cylical synonym entry
is entered (eg. "cat"->"feline" and "feline"->"cat").
Note that this is NOT necessary as the first synonym entry already
implies equivalency both ways. It will now correctly ignore the
redundant entry if it is entered.
- Fixed a rare stalling bug when indexing a site
which returns an unexpected HTTP redirect status code when no
redirection is actually required.
- Fixed search script bug which caused the "Jump
to highlighting" parameter ("zoom_highlight=")
to be omitted from the result links when "Title of page"
is disabled from the "Results Layout" tab.
- Added highlighting for page titles in the search
results of the Javascript version.
- Fixed Indexer bug with converting HTML entities
before removing HTML tags which may have potentially caused some
page content to be excluded from indexing.
- Fixed Indexer bug which was listing more file
extensions than specified in the currently loaded ZCFG file, at
the very start of indexing (ie. "Will scan files with extensions:
...")
Version 4.2.1002
(21 Sept 2005)
- Fixed bugs causing indexing to stall indefinitely
until the user clicks "Stop".
- Fixed bugs with drawing the index log when paused
occasionally.
- Added Javascript workaround for issue with Arabic
(windows-1256) in the JS version.
- Fixed bug with CGI not using the translated text
for the message, "The following phrase contains very common
words..."
- Fixed minor bug with PHP spelling suggestions.
- ASP script is now "Option Explicit"
compatible.
- Updated Portuguese language file with some translation
corrections.
Version 4.2.1001
(23 Aug 2005)
- Improved support for detecting file types of
documents served via a download script. Zoom will now look at
the Content-Disposition header even when the Content-Type is unsupported.
This means it will now index files served with a content-type
of "octet-stream" or "x-msdownload", etc.
- Fixed "Help" button for Synonyms tab
in the Configuration window.
- Fixed crash bug in CGI search when using spelling
suggestions.
- Updated ASP script to avoid variable names conflicting
with other scripts if you are embedding search.asp in other ASP
scripts.
- Removed charset enforcement in the Javascript
version. This allows the Russian/Cyrillic workaround with UTF-8
parameter decoding to work correctly.
Version 4.2.1000 (17
Aug 2005)
- Synonyms: This feature allows
you to specify synonyms, variations of words, common misspellings,
etc. and allow you to map them to an equivalent word in the index.
This can be very helpful if many users on your site are searching
for names which are different to the ones you used on your website,
but have the same meaning. For example, the word "question"
could be made equivalent to "inquiry,enquiry,query,questions",
so that when a user searches for any of the aforementioned words,
they would get the same results.
- Exclusion/negative searches:
This allows your users to perform search queries which exclude
a certain keyword by preceding the search term with a hyphen (eg.
"cats -dog" would return all pages containing the word
"cat" but not the word "dog").
- Improved spelling suggestions: We have improved
the spelling suggestions algorithm to provide more accurate and
useful suggestions when a user's search returns few results. It
will also return multiple suggestions for single word searches.
- Includes new language files: Japanese, Russian, and Portuguese.
- Validated XHTML 1.0 Transitional output.
- Added HTML entities for yacute, thorn, times, oelig, scaron,
yuml, and more.
- Fixed various issues with "Jump to highlighting" script
(highlight.js).
- Fixed issues with searching for an exact phrase in the meta
description or title but the individual words of the phrase can
be found in the content of the document.
- Fixed issues causing the warning message "Template file
has different charset" to be shown incorrectly (eg. when
both the template and Indexer is set to use UTF-8).
- Prevents Zoom from remembering Window positions which are outside
the visible area of the desktop.
- Fixed several bugs in the Javascript version, including broken
"Next" links when using UTF-8.
- Fixed bug causing "aaaaa"s to appear in the context
description.
- Fixed bug with some HTML entities in the URL failing to be converted.
- Fixed issue on Windows 98 with plugins causing a DOS window
to appear when processing a file.
- More bug fixes and changes.
Version 4.1.1003
(11 Jul 2005)
- Fixed bugs with Indexer failing to remember window
positions and "show wizard on startup" settings.
- Fixed bug with Indexer failing to locate default
template file, when "Custom script source path" is enabled.
- Fixed CGI bugs for searching UTF-8 text on a
Windows server and some potential crash issues.
- Fixed ASP script for searching the last entry
in the dictionary.
Version 4.1.1002
(1 Jul 2005)
- Optimized the CGI search function, most significantly
improving performance for large sites of around 100,000 pages
or more.
- Fixed timezone bug in the CGI version which caused
some dates to be shown incorrectly.
- Changed meta descriptions in results to show
"..." only if it has been truncated
- Improved detection of content-type headers -
fixes problems with "Unsupported content-type" for some
file formats.
- Added error messages if FTP username or password
is too long.
- Fixed some memory leaks in the Indexer.
- Updated Italian language file.
Version 4.1.1001 (27
Jun 2005)
- Fixed ASP search script which was incorrectly
encoding the query in HTML, and causing exact phrase matching
to not work.
- Added detection of non-supported Content-Type
headers, which allow Zoom to ignore download scripts which may
serve binary files such as .exe or images instead of indexing
them as text.
- Added decoding of URL escape characters (eg.
%20) in filenames when they are used as a page title (eg. PDF
files with no meta title information or .desc file).
Version 4.1.1000 (22
Jun 2005)
Version 4.1 offers many improvements based on user
requested features, further adding to the power of Zoom, while maintaining
the ease of use and search speed inherent in 4.0. With the new weighting
options, it is also now capable of producing more accurate and relevant
search results than ever.
- Automatic "content density"
weighting adjustment: automatically adjusts the weighting
of words found in a file depending on the density of its content.
This helps prevent large files (such as PDF documents with 50+
pages) from always appearing as the most relevant result, and
gives greater priority to smaller documents.
- Weighting options for heading tags: increase or decrease
the importance of text found in <H1> and <H2> tags.
- Weighting options for filenames: increase or decrease
the importance of search text found in filenames.
- Support for ASP.NET: you can now create an ASP.NET (.aspx)
search page to integrate Zoom into your ASP.NET website seamlessly.
Find out more here.
- Determine file types by "content-type"
header: This allows Zoom to index PDF, DOC, and other file
types served by a server-side download script (eg. http://www.mysite.com/getfile.php?fileid=123)
- New DjVu plugin
support: allows you to index and search DjVu
files (".djvu")
- Support "Last-Modified" meta date tag: allows
you to specify date/time for dynamically generated pages easily
for the "Sort by date" option.
- Follow "file://" style links: this option allows
spider mode indexing on Intranets to scan and index documents
on the web server as well as the file network.
- Support for multiple category patterns: allows you to
specify multiple patterns for each category to match files by.
- New index log window options: allow you to right click
on a line in the index log and copy the line to the clipboard,
or open the URL in your browser, etc.
- Option to enable or disable the use of cookies for spider mode
indexing
- Import categories from CSV file format
- New "Max words per file" optional limit: allows
you to specify the maximum number of words to index from each
file.
- "Jump to match ..." script (highlight.js) now support
exact phrases and UTF-8.
- Indexer remembers window position and size.
- Added join character options for colon and ampersand characters.
- New language files for Spanish and Norwegian.
- Many bug fixes and other improvements to further ensure reliability
and performance.
Version 4.0.1016 (10
Mar 2005)
- Fixed errors ("Warning: compilation failed",
"Expected ... in regular expression") when results highlighting
is enabled, and search query includes parentheses, braces, and
other non-alphanumeric characters in PHP, ASP, and JS scripts
- Fixed bug with corrupting spider URL when user
edits start points
- Fixed crash bug when importing start points from
a text file with blank lines
- Fixed a number of potential crash bugs with text
buffers in memory
- Removed extraneous radio button when using the
CGI search function with "Basic" search form
- Updated documentation for "Jump to match
and highlight" instructions
Version 4.0.1014 (8
Mar 2005)
- Added "Jump
to match and highlight within documents" support. See
link for more information.
- Updated language files for Italian and German
with minor corrections
- Updated CGI search function for XHTML 1.0 compliancy
- Removed grey-out for ASP option in the wizard
when "create a search engine for an Intranet/CD-ROM"
is selected.
- Updated Javascript version to encode query in
links
Version 4.0.1013 (3
Mar 2005)
- Improved XHTML 1.0 Transitional compliancy for
default template HTML and search script output for PHP, ASP, and
Javascript.
- Fixed crash bug caused by "index meta author"
option is enabled and a page with a very long meta keyword tag
is encountered. Also increased the maximum size for meta keywords.
- Fixed bug with "Wizard suggestion"
window occasionally popping up at the end of indexing when running
as a scheduled task.
- Fixed double lines appearing in Index Log when
saved to file
- Fixed bug with appending extra slash when base
URL is specified in backslashes (eg. UNC locations)
Version 4.0.1012 (24
Feb 2005)
- Added Swedish language file (Swedish.zlang)
- Indexer no longer aborts indexing a page if it
comes across a blank href values (eg. <a href="">)
- Fixed rare Windows 98 crash bug with certain
PDF files (and other plugin formats) when the filename contains
certain non-standard characters.
- Fixed Javascript issue when there are multiple
spaces before or after the search query (and substring search
is enabled)
- Added error messages in the Indexer log when
it encounters some HTTP errors (eg. "500 - Internal Server
Error")
- Updated CGI search with support for the "results"
CSS class.
- Updated ASP script to support use with HSP
(a 3rd party application) which allows you to run ASP on a CD.
- Removed "charset mismatch" warning
if no charset is found in the HTML template.
- Some minor changes to avoid issues with Zip drives
being browsed in the "Select folder" window, whilst
there is no disk in the drive.
Version 4.0.1011 (16
Feb 2005)
- Added real-time memory estimation in the "Limits"
tab of the configuration window. This provides you with an estimation
of how much memory is required as you change the values.
- Comments added to the settings output file (eg.
settings.php), to warn users not to edit the file manually.
- Fixed Javascript bug where searching with "match
all search words" with the use of wildcards would return
results matching only the wildcard term.
- Fixed bug when user overflows "max file
size" value. "Max file size" value is also now
converted to KB (1024 bytes) rather than kB (1000 bytes). This
means that when you load in existing configurations, this setting
will appear to be slightly lower than before.
- Added a 60 second time out for HTTP requests
if no response is sent by the server.
- Encoded characters (eg. %20) in URLs are no longer
removed or converted. This should help users indexing sites with
foreign characters in the URLs.
- Fixed bug with finding hypertext links broken
up by new line characters.
Version 4.0.1010 (7
Feb 2005)
- Added "Use PASV mode" option to the
FTP and configuration windows.
- Added Italian language file (Italian.zlang) and
some minor updates to the French language file (French.zlang)
- Removed errorneous semi-colon in the default
"search_template.html" CSS. This was causing the ".summary"
style to be ignored in Mozilla browsers previously.
- Added support for HTML entities such as "
to be used inside a meta description.
- Added support for meta tags with attributes specified
by single quotation marks (and containing double quote marks as
the content value) or vice versa.
- Fixed bug where BASE HREF tag enclosed in a HTML
comment block, was not ignored.
Version 4.0.1009 (31
Jan 2005)
- Fixed bug in build 1008 which caused links with
HTML code in uppercase to be skipped by the spider.
Version 4.0.1008 (27
Jan 2005)
- Added "Do not upload search template file"
option, for users of website generators (eg. NetObjects Fusion)
that wish to manage the uploading/publishing of their search template
with their website generator.
- Fixed bug with spider not following some links
or frames with certain names or attribute values.
- Improved spider crawling process when hunting
for links, especially pages with IFrames and embedded frames.
- Offline mode now allows for base URLs that end
with a question mark (eg. "goto.php?"). This allows
you to pass the links to your site's files as parameters to a
script.
- Added warning message when the default category
name is empty. You must specify a default category name when categories
are enabled.
- Added a missing </p> tag in the ASP and
Javascript output.
Version 4.0.1007 (20
Jan 2005)
- Fixed Windows 98 crash bug when using the "Import"
option in the "Additional start points" window
- Fixed issue in Javascript version which displayed
a different year for search results when viewed in Netscape/Mozilla/Firefox
(eg. it would display "105" instead of "2005").
- Fixed bug where the "Save configuration"
function would sometimes save to the default configuration file
rather than the currently loaded configuration file.
- Added error message when no "Indexing options"
are selected in the configuration window.
Version 4.0.1006 (18
Jan 2005)
- Improved support for extracting meta information
from plugin supported file formats.
- Fixed bug in the Preview panel of the "Results
options" tab, which showed context description even when
Javascript is selected.
- Improved support for avoiding sample HTML in
PDF documents being displayed in context descriptions.
- Fixed bugs with indexing HTML code when indexing
some plugin file formats such as PDF, PPT, XLS.
- Fixed bug with not being able to disable titles
from being indexed, when "Title" is unchecked in the
"Index options" tab.
Version 4.0.1005 (14
Jan 2005)
- Added support for RTF file format (requires new
rtf2txt plugin)
- Added support for meta information from PowerPoint
files (requires updated ppthtml plugin)
- Improved support for HTML or sample-HTML code
appearing in context description
- Fixed Windows 98 bug, where the Langauge (zlang)
file dropdown is disabled.
- Fixed a bug with the "Use meta information
from plugin" feature which caused title and description to
not be extracted correctly for display in the search results.
- Fixed bug with the Javascript version not using
the translated text for the "Search" button
- Fixed bug with the Javascript version failing
to retrieve GET parameters when "zoom_query" is not
specified, eg. "search.html?zoom_per_page=50"
Version 4.0.1004 (7
Jan 2005)
- Fixed ASP bug causing "problem: page number
too big" error
- Fixed Windows 98 bugs which caused the index
log window to not display, and crashes after "Save Index
Log to file..." and Load/Save configurations
- Added detection of password protected DOC files
(requires word2txt version 1.3)
- Added support for meta information from DOC files
(requires word2txt version 1.3)
- Included new Dutch and Danish language files
for the search page
- Fixed additional space character appearing in
categories for the CGI/Linux search results page
- Fixed bug with "zoom_plugin.in" appearing
as a page title for XLS files in search results when "use
meta information from plugins" is enabled
- Fixed bug with truncating FTP usernames over
20 characters
- Fixed bug with multiple start points remaining
after loading a different config file
- Fixed bug with "Index author" option
remaining disabled in configuration window
- Fixed bug with scheduler failing to schedule
with configuration files outside of the program directory
Version 4.0.1002 (13
Dec 2004)
- Includes new Croatian language file for the search
page
- Fixed bug in ASP script when searching for single
character words with Spelling Suggestion enabled
- Fixed crash bug with FTP uploading, when an upload
fails (due to incorrect login information, or when a user hits
cancel), and user attempts to upload immediately afterwards.
- Fixed bug with URLs containing two '?' characters,
which caused the Indexer to mis-identify the file extension.
- Fixed crash bug when output directory does not
exist
- Fixed bug with meta description tag overriding
the .desc meta description
- Fixed bug with sometimes forgetting "Wizard
On Startup" setting
Version 4.0.1000 (9
Dec 2004)
Zoom Search Engine Version 4.0 introduces many new
features and an overhaul of the indexer internals to provide more
powerful searching options, better performance (indexes larger sites
with the same hardware, and indexes them faster), and improved usability.
- Multi-threaded downloading:
Dramatically improve the speed of spider mode indexing, by
allowing multiple downloads and indexing to occur simultaneously.
[screenshot]
- Reduced memory usage: Allows users
to index much larger sites than before, with around 30% less memory
used.
- Sort by date or relevance: Option
to sort search results by date and show file dates.
- Spelling suggestions: Provide
users with spelling suggestions or a similar word which is more
popular, when few results are found.
- Built-in FTP uploading: Zoom can now
upload the search files after indexing, directly to your web server.
No need for an external FTP client, and can be configured to automatically
upload after every index. [screenshot]
- Built-in Zoom Scheduler: You
can now schedule Zoom to automatically index, or generate reports
for your site, directly from Zoom, with a few mouse clicks. [screenshot]
- Step-by-step wizard: an easy-to-follow
guide which helps new users get started setting up the Indexer
for their site. Also useful for more experienced users who just
want to get something up and running as soon as possible. [screenshot]
- Shockwave Flash plugin: Index and search
through Shockwave Flash, and FlashPaper files and support spider
crawling Flash-based sites (it will successfully follow links
from Flash-based navigation menus etc). [plugins
page ]
- Frames and IFRAME spider indexing: The spider will now
crawl pages and follow links within frames and IFRAMES.
- Modify or translate search page text: Zoom now allows
you to translate the text generated by the search page (eg. "No
results found", "Search results for....", etc.),
without having to modify the search script. You can translate
your entire search function to a different language, or just customize
the text as you see fit. French and German translation files have
been included with this release.
- More flexible indexing options: Select exactly what you
want to index, from title of page, filename, page content, meta
description, meta keywords and meta author
- More flexible results layout options: Select which element
to display in the search results with a simple check box.
- More boosting features: Give you greater control over
which page gets higher priority in the search results. New ZOOMPAGEBOOST
options allow you to boost a page's importance. Also, more flexible
word boosting (-5 to +5) options for words that appear within
a page title, meta description or keywords.
- Improved international language options: accent characters
will now appear in the context descriptions (in their original
form), even when accent insensitivity is enabled.
- Multiple start points for offline mode
- Multiple base URLs per spider start point: Each start
point can have multiple base URLs which allow links to be considered
"internal". For example, http://www.mysite.com/ can
have both "http://www.mysite.com/" and "http://private.mysite.com/"
be its base URL.
- New statistical report options: List search terms sorted
by popularity, append unique date/time to filenames, and specify
a different folder for reports to be uploaded to.
- Support for retrieving meta information from PDF files (meta
description, document title, document author, etc.)
- Improved template management: Easier to maintain multiple
sites with different search page templates. No longer necessary
to edit the template source file or to make multiple copies. Just
directly modify the search template in the output directory!
- Highlighting now also applies to search result titles.
- Control which messages appear in the Indexer log window. Progress
bars for downloading messages.
- Many other bug fixes and improvements based on users' requests
and beta testers feedback.
Click here for more old
version history (Zoom Version 3.1 or earlier)
Click here for the latest version history
|
|
