Release Notes for Mindbreeze InSpire

Innovations and new features

Support for GSA-feeds/Adaptor protocols

The Mindbreeze InSpire GSA Façade enables feeds from connectors, which are addressed to the Google Search Appliance (GSA), to index with InSpire without adjustments.

Currently, "Web feeds" are supported without taking into account ACLs and plugins from "Adaptor Framework".

Microsoft Outlook Add-in

Which program do you use most often?
Chances are this is Microsoft Outlook. In addition to your e-mails and calendar entries, you can now also display Mindbreeze InSpire results in Microsoft Outlook. This means you can find old e-mails that are in PST files as well as documents from Microsoft SharePoint or any other data source in your company.
More under Outlook-Add-in

Expanded suggestions

In the search field, the search term can now be searched not only in the current tab, but also in all available tabs.

The configuration is carried out under Development of Search Apps - Suggest

HL7V2 connector

With the new HL7V2 connector, HL7V2 documents can be searched.
Details about the configuration can be found at HL7 Connector.

Improvements in alternative search suggestions

The search can now be automatically extended to include alternative search suggestions.

In multi-word searches, for instance searches which include a person's name like John Doe, the word combinations that are also combined in the indexed documents will now increasingly be suggested.

Details on the configuration can be found at Configuration - Alternative Search Suggestions and Automatic Search Expansion

Integrated plugin for Chinese word decomposition

Chinese characters are strung together into sentences without the use of separators. This causes groups of characters to form words which have a meaning. By using a trained model for the recognition of possible word areas, searches can now also be performed in Chinese texts.

Details can be found at Configuration - Chinese Tokenization Plugin Jieba

Improved support for rules with query expansion

Often used spellings which are used interchangeably, can now be expanded using user-defined rules. A query service with a German stemmer plugin automatically expands the search term "Straße" to "Strasse" or "Müller" to "Mueller" and vice versa.

With a simple set of rules, searches can be supported for different spellings. For instance, a search for Codex will also find documents which contain the word Kodex.

Details about the configuration can be found at Mindbreeze Query Expression Transformation

Document metadata retrieval

Document metadata can now be retrieved via the "document:" prefix.
If the title of document is set, it is available under document:title. The creation date according to the document is available via document:CreationDate, assuming that the document contains the metadata. These can be viewed as a property or used as a filter.

With PDFs, the following properties are supported (subject to availability):

• document:title
• document:Author
• document:Subject
• document:Keywords
• document:Creator
• document:Producer
• document:CreationDate
• document:LastModified
• document:Trapped

Processing Microsoft Office hits from Microsoft SharePoint

Microsoft Office documents can be opened and edited by clicking on the search results directly with Microsoft Office. If Microsoft Office is not installed, the results open in Microsoft SharePoint.

Technical upgrades

Program your own custom widgets

Mindbreeze InSpire search applications already offer great flexibility for designing your search application. But if you’d like a different filter or presentation of results, or you need other widgets, you can now program your own. These widgets are available in every search application just like the standard widgets and also in the designer.

More information at Development of Search Apps

Upgrades to the property expression language

The property expression language has been upgraded to support the string operations "to_lower" and "to_upper".

The arithmetic expressions are supported as of this release on an indexing level.

Comparison operators are now supported outside filters, for instance to compare two function evaluations.

Details can be found at Mindbreeze Property Expression Language

Improvements to the Web connector

Faster removal of excluded documents

The option "Cleanup non matching URLs from index" removes documents that are excluded by the rules "URL regex" and "URL Exclude Pattern" from the index already during indexing. This means that a complete re-indexing is no longer necessary.

Improved recognition of file formats

Microsoft Office documents are also recognized correctly if the Webserver specifies them as content type "application/zip".

Form-based login with relative redirects

Websites that use form-based logins with relative redirects are now supported.

Support for sitemaps with references

Analogous to ACL elements, now sitemaps can also contain references to other documents.

Improved proxy support

Via proxy exceptions, it is now possible to configure for which websites the Web crawler should use the configured proxy.

For configuration details about the Web connector, see Web-Connector.

Calculated properties for the indexing time

Property expressions can be used to transform properties at the following times:

a)Before/After applying entity rules

b)Before/After applying CSV transformation rules

For example, the function parse_date is available to create a date from a text.

parse_date(LastModified, "YYYY-MM-dd HH:mm:ss zzz")

For a definition of the syntax and a list of supported functions, see Mindbreeze Property Expression Language.

With the summer release, existing metadata such as the displayed title or the document date can be replaced by property expressions.

Group cache for Microsoft SharePoint

Group memberships can now be picked up periodically for Microsoft SharePoint. In addition to Microsoft Active Directory groups, Microsoft SharePoint groups are now also updated efficiently, without having to perform any indexing.
Details on the configuration can be found at Caching Principal Resolution Service (SharePoint).

Other product improvements

Reduced memory consumption in the index directory

The index request dumps are now disabled by default, and therefore the memory usage in the index directory is reduced.
Details can be found at Installation and Configuration of Mindbreeze InSpire

Tika filter plugin uses the latest version of Tika

The Mindbreeze InSpire Tika filter plugin now uses Version 1.13 of Apache Tika, providing even better text and metadata extraction.