OpenSearchServer

OpenSearchServer is an open source, enterprise-grade, very powerful, freely distributed and high-performance search engine (also known as search server) program that provides a collection of high-powered full text search algorithms and uses a web-based interface.

Supports parsing of numerous file formats

The software supports parsing of numerous document formats, including HTML, XHTML, XML, Adobe PDF with OCR, Microsoft Office documents (Word, PowerPoint, Excel, Publisher, Visio), Word, RTF, Plain Text, OpenOffice documents, OCR over images, torrent files, MP3, MP4, FLAC, AIFF, WAV, as well as Ogg Vorbis.

Offers powerful search functions

OpenSearchServer's search functions includes advanced full-text search features, phonetic search, advanced boolean search with query language, clustered results with collapsing and faceting, filter search using sub-requests, geolocation, spell-checking, relevance customization using algebraic functions, and auto-completion.

Provides state-of-the-art indexation functions

Another interesting feature is the indexation function, which supports 18 languages, automatic classification, automatic language recognition, named entity recognition, expression and word synonyms, fields schema with analyzers for each supported language, exporting of indexed terms with frequencies, as well as various filters, such as stripping diacritic from words, lemmatization, n-gram, and shingle.

Powerful crawlers are also implemented

OpenSearchServer also contains powerful crawlers, such as web crawlers for Internet, Intranet and Extranet, filesystem crawlers for both remote and local files, supporting the FTP, SMB, CIFS, NFS, FTPS and SWIFT protocols, sitemap import, screenshot capture, SQL join, session parameters removal, filter inclusion/exclusion using wildcards, as well as database crawler for all supported JDBC databases, including the well known MySQL, PostgreSQL, Microsoft SQL Server, and Oracle.

Many other amazing features

Among other features, OpenSearchServer includes REST APIs (JSON and XML), SOAP web service, monitoring module, index replication, scheduler for management of periodic tasks, as well as Drupal module and a WordPress plugin. The software is supported on all GNU/Linux operating systems, well as on Microsoft Windows and BSD OSes.

What is new in this release:

• New features:
• GH-202: Named Entity Recognition module
• GH-183: Script REST API
• GH-2: Copy Of: ability to copy from several fields
• GH-1: Improve spell check query: return existing words
• SF-288: Copy content of a field to another one in the schema
• SF-287: Indexation using text file
• SF-286: Automatically create a unique reference for each document
• SF-285: Hunspell integration
• SF-283: Gzip file support in sitemap
• SF-277: Separate query for snippets
• SF-272: Automatically set cookies in the web crawler
• SF-271: Disable nofollow directive
• SF-270: Database based script
• SF-269: Upgrade ZK library (6.5.2)
• Bug fixes:
• GH-171: Highlighted attribute is not set in Search API (XML/HTTP)
• SF-190: Cancel "Edit job" page in scheduler when changing selected index
• SF-189: Do not display error of previous crawl process in "current status"

What is new in version 1.5 Beta 3:

• Many bugs have been fixed.
• Spaces in URLs are correctly handled by the crawl API, and the HTTP connector of Tomcat uses the UTF-8 charset by default.
• Some external libraries have been upgraded.
• The deletion API now returns the number of deleted documents.
• The new MAPI messager parser can index offline Outlook messages.
• The REST crawler has been improved by integrating a callback mechanism.

What is new in version 1.5 Beta 1:

• This Beta release is already pretty stable and used in several production environments.
• Index optimization is no longer required.
• Both file and Web crawlers are updated in realtime.
• The new search request includes new min, max, and count functions using collapsing.
• The REST crawler can index data available as a RESTful Web service using a JSON path.

What is new in version 1.5-dev4:

• This release has been fully repackaged, and the setup is easier than ever.
• New APIs are provided to insert patterns (exclusion and inclusion) in the Web crawler.
• Downloads of compressed sitemaps are now supported.

What is new in version 1.5-dev3:

• A RESTFul Web service and a new REST crawler were added.
• All existing features have been dramatically improved.
• The build system is now based on Maven.
• A Ruby client library is now available.

What is new in version 1.5-dev2:

• This version introduced a new RESTFul Web service and a new REST crawler.
• All existing features have been dramatically improved.
• The build system is now based on Maven.

What is new in version 1.4:

• Phonetic searching, joints, named entities, geo-location, renderer, and more.

What is new in version 1.4 RC4:

• This release candidate improves the performance when loading a classifier with a large set of items.

What is new in version 1.4 RC2:

• This version fixes a few issues from the previous release candidate.
• It is now possible to decide whether all the content of the RSS feed is indexed in only one document or if each item is indexed individually as one document in the index.
• The refresh action in the runtime/system tab panel no longer changes the selected tab.
• The failover mechanism in the parser is protected against infinite loops.
• In case of a parsing error, every kind of error is reported.

What is new in version 1.4 RC1:

• The second release candidate fixes a bug in RC0 which could crash the Java runtime while crawling websites with malformed cookies.

What is new in version 1.4 RC0:

• The first release candidate of the new 1.4 branch fixes a few bugs and includes user interface improvements and new features.
• A new PDF viewer is able to highlight searched keywords on PDF pages.
• This new feature also supports scanned PDF documents (OCR).
• Visio and Publisher documents are supported, and RSS feeds are indexed.
• A report module provides information on the most often-searched words and on which words don't return a document.
• User and group rights are extracted while browsing CIFS/SMB and FTP repositories.
• The Web crawler supports NTLM authentication and canonical links.

What is new in version 1.4 Beta 5:

• New features:
• 260: Add Visio parser and Publisher parser
• 257: Upgrade Apache CXF to 2.7.x
• 251: A report module
• 250: Server time in scheduler interface
• 249: Extraction of user and group rights when crawling CIFS/SMB and FTP
• 248: Upgrade ZK CE to 6.5
• 247: Support of additional characters in LetterOrDigitTokenizer
• 246: Named entity recognition based on a search request
• 245: Highlighting in PDF documents and images (OCR)
• 243: RSS Parser
• Bug fixes:
• 178 Error reading 'fieldOrSnippetList' on type RendererController
• 170: Wrong file path in file crawler
• 169: java.lang.NullPointerException on flush expired contend from cache
• 168: Screenshot function not working
• 167: java.lang.NullPointerException on enabling crawler cache
• 166: Query -> More like This causes Property 'name' not found on type java.lang.String
• 165: Not possible to copy an address in Crawler -> Web -> Pattern List
• 163: Weight is ignored in query scoring item
• 161: ParseException while crawling files

What is new in version 1.3.1:

• New features:
• 244: Exportable autocompletion index
• 242: Be able to run only one crawl session
• 240: Several HTML parser configuration per index
• Bug fixes:
• 158: File crawler infinite loop while crawling FTP directory
• 157: NullPointerException while deleting documents from web crawler
• 156: Cannot forcefully unlock a NativeFSLock error
• 155: Trigger name cannot be null error when clicking optimize
• 153: Failed replication from Windows to Linux because of file separator