Apache UIMA Sandbox v2.3.0 Release Notes

1. What is UIMA?
2. What is the Apache UIMA annotator package?
3. Major Changes in this Release
4. How to Get Involved
5. How to Report Issues
6. List of JIRA Issues Fixed in this Release

1. What is UIMA?

Unstructured Information Management applications are software systems that analyze large volumes of unstructured information in order to discover knowledge that is relevant to an end user. UIMA is a framework and SDK for developing such applications. An example UIM application might ingest plain text and identify entities, such as persons, places, organizations; or relations, such as works-for or located-at. UIMA enables such an application to be decomposed into components, for example "language identification" -> "language specific segmentation" -> "sentence boundary detection" -> "entity detection (person/place names etc.)". Each component must implement interfaces defined by the framework and must provide self-describing metadata via XML descriptor files. The framework manages these components and the data flow between them. Components are written in Java or C++; the data that flows between components is designed for efficient mapping between these languages. UIMA additionally provides capabilities to wrap components as network services, and can scale to very large volumes by replicating processing pipelines over a cluster of networked nodes.

Apache UIMA is an Apache-licensed open source implementation of the UIMA specification (that specification is, in turn, being developed concurrently by a technical committee within OASIS , a standards organization). We invite and encourage you to participate in both the implementation and specification efforts.

UIMA is a component framework for analysing unstructured content such as text, audio and video. It comprises an SDK and tooling for composing and running analytic components written in Java and C++, with some support for Perl, Python and TCL.

2. What is the Apache UIMA annotator package?

The Apache UIMA annotator package is an add-on package for the base UIMA release. The add-on package contains annotator components developed for Apache UIMA. The add-on package fits the Apache UIMA directory structure and adds a directory called "addons/annotator" that contains the following annotator components:
- DictionaryAnnotator
- RegularExpressionAnnotator
- Tagger
- WhitespaceTokenizer
- DictionaryAnnotator
- RegularExpressionAnnotator
- Tagger
- WhitespaceTokenizer
- Bean Scripting Framework (BSF) BSFAnnotator
- ConceptMapper
- ConfigurableFeatureExtractor
- Lucas - an interface to using UIMA with Lucene
- OpenCalaisAnnotator - an sample annotator using the OpenCalais Service
- SnowballAnnotator - an annotator making use of the snowball stemmers
- TikaAnnotator - an annotator using the Tika project text extractors

Additionally the package contains some components to package annotators and for accessing annotators as simple REST service. These are:
- PearPackagingAntTask
- SimpleServer

Finally, there is an addon to the base UIMA: - FsVariables

Each component has a separate LICENSE and NOTICE files; some also have Readme and other documentation (in docs/). Documentation is also available on the UIMA website, in the Sandbox area.

3. Major Changes in this Release

The Apache UIMA annotator package release version 2.3.0 adds the is the first release following components to the previously released ocmponents:
- Bean Scripting Framework (BSF) BSFAnnotator
- ConceptMapper
- ConfigurableFeatureExtractor
- Lucas - an interface to using UIMA with Lucene
- OpenCalaisAnnotator - an sample annotator using the OpenCalais Service
- SnowballAnnotator - an annotator making use of the snowball stemmers
- TikaAnnotator - an annotator using the Tika project text extractors

The PearPackagingMavenPlugin is moved to the base UIMA release package.

The XMLBean support is migrated to version 2.4.0, and all of the projects now use the maven xmlbeans plugin to generate the XML parsers.

Finally, there is an addon to the base UIMA:
- FsVariables

For a list of all JIRA issues fixed with the current Sandbox release, please refer to chapter 6. List of JIRA Issues Fixed in this Release.

Bug

[UIMA-860] - Add source-style LICENSE and NOTICE files at "root"s of UIMA
[UIMA-990] - change POM description for annotator package POMs
[UIMA-1003] - update PearPackagingMavenPlugin dependency scope
[UIMA-1004] - update SimpleServer try out form
[UIMA-1069] - Model file is not loaded correctly if tagger is deployed more than once in same AE
[UIMA-1071] - http connector fails with some Java implementations
[UIMA-1085] - Fix Sandbox NOTICE files
[UIMA-1106] - update OpenCalaisAnnotator with correct encodings
[UIMA-1108] - correct character offset for OpenCalais annotator
[UIMA-1193] - Tagger throws occasional NPE
[UIMA-1264] - Regex annotator goes into infinite loop on patterns that match the empty string
[UIMA-1374] - TikaAnnotator source code does not compile because of incorrect package declarations
[UIMA-1378] - Build of uimaj-examples project fails
[UIMA-1379] - Type system namespace should be org.apache.uima.tika, not just org.apache.uima
[UIMA-1384] - WhitespaceTokenizer pom still references UIMA 2.2.2
[UIMA-1385] - Regex annotator does not close concept file input stream after reading
[UIMA-1392] - OpenCalaisAnnotator's annotations have truncated 'coveredText' field
[UIMA-1403] - Lucas: many test cases fail
[UIMA-1446] - org.apache.uima.simpleserver.config.impl.SimpleFilterImpl.match() can cause a null pointer exception
[UIMA-1447] - Tabulations are annotated as tokens after a space
[UIMA-1456] - SimpleServer: sample config file does not work
[UIMA-1457] - SimpleServer: docs need updating for Tomcat 6
[UIMA-1460] - AnnotationTokenStream.next(Token) should not catch Throwable
[UIMA-1462] - SimpleUimaAsService has checked in SimpleServer libraries as binaries
[UIMA-1464] - SimpleServer NOTICE file missing JSR 173 attribution
[UIMA-1528] - The documentation describes still the UEAStemmer, which was removed from the distribution
[UIMA-1530] - Index naming is not unique in multithreaded scenarios
[UIMA-1533] - Lucas generated test-sources jar missing license, notice, disclaimer
[UIMA-1535] - Lucas POM issues
[UIMA-1546] - Fix sandbox notice and license files
[UIMA-1547] - XML problems with simple server test cases
[UIMA-1551] - Lucas PearPackagingMavenPlugin PEAR classpath is incorrect
[UIMA-1552] - Lucas: does not compile with Java 1.5
[UIMA-1554] - Fix CFE notice and license file
[UIMA-1557] - LuceneCASIndexer.xml should be in src/test/resources
[UIMA-1558] - LuceneCASIndexerTest fails if the created LuceneCASIndexer procsess a CAS
[UIMA-1571] - FsVariables book_name should be FsVariablesUserGuide instead of fsVariablesUserGuide
[UIMA-1572] - Lucas artifactId does not match folder name in svn
[UIMA-1577] - Do pear packaging with dependencies included for CFE
[UIMA-1582] - SimpleServer ConfigTest fails
[UIMA-1595] - Change build of Sandbox RegExpr to use xmlbean maven plugin, and delete its lib dir
[UIMA-1596] - fix sandbox build - use of <profile> for conditional not working for child projects
[UIMA-1597] - in sandbox common build, change needed to use assembly-bin instead of bin
[UIMA-1605] - Fixed Findbugs issues
[UIMA-1609] - binary assembly wrongly including FOP files
[UIMA-1615] - make build-from-sources work
[UIMA-1639] - Fixed bugs which disabled compiled dicts, static dict attributes
[UIMA-1663] - All directories in UIMA distribution should be 755
[UIMA-1676] - Regex: pear install.xml is missing 2 of the 3 jar files
[UIMA-1683] - fix copyright notice in the NOTICE files
[UIMA-1685] - Sandbox licenses has uima-as licenses, but those are now moved out into a separate entity
[UIMA-1686] - add licenses from some Sandbox projects to overall Sandbox license
[UIMA-1687] - correct top-level svn license/notice files

Improvement

[UIMA-991] - fix Sandbox documentation to avoid overflowing the footer on even pages
[UIMA-992] - update dictionary annotator build documentation - how to create XML Beans jar
[UIMA-1005] - create ant build for XMLBeans class generation
[UIMA-1016] - allow URLs as dictionary files
[UIMA-1301] - Update documentation, log problems when dictionary entries don't load, remove diagnostic message during dictionary loading
[UIMA-1336] - allow multiple dictionary entries to match against a single string
[UIMA-1370] - Lucas: add the usual suspects to svnignore
[UIMA-1372] - Improve description of ConceptMapper on UIMA sandbox components web page
[UIMA-1455] - Lucas should not use deprecated lucene API
[UIMA-1486] - Lucas should not depend on google collections snapshot version
[UIMA-1498] - if an exception is rethrown, the original exception is not currently passed through
[UIMA-1501] - more refactoring and updating - parent POMs
[UIMA-1506] - update Bean Scripting Framework Annotator with info about licenses and documentation
[UIMA-1517] - Don't set executable bits on non-executables, when building assemblies
[UIMA-1518] - improve SandboxDistr assembly
[UIMA-1527] - Upgrade Tika Annotator for 2.3.0 release
[UIMA-1529] - Lucas depends on lucene 2.4.0, it should be lucene 2.4.1
[UIMA-1537] - License Notice Disclaimer copying
[UIMA-1538] - Common Build Step: build source Jars for java Jars
[UIMA-1550] - Remove the uni-jena.de repository from lucas pom
[UIMA-1556] - The LucasCasIndexer should be an Analysis Engine and not a CasConsumer
[UIMA-1559] - lucas.xsd exists 3 times in different places
[UIMA-1567] - Maven build: add <prerequisites> to uimaj to specify minimum Maven release level
[UIMA-1583] - Regularize Sandbox builds and assembly
[UIMA-1585] - Run RAT on projects, fix missing licenses, add RAT running to POM, document exclusions
[UIMA-1586] - CFE - XMLBeans - use maven plugin to generate the parser
[UIMA-1587] - replace stax jar with better licensed geronimo version
[UIMA-1589] - CFE - add Readme describing how to regenerate the EMF generated files
[UIMA-1590] - fix extractAndBuild scripts
[UIMA-1594] - make sandbox assembly build like base and uima-as builds
[UIMA-1610] - add a changeVersion build tool to handle changes needed in sandbox release
[UIMA-1613] - run Rat consistently for all maven assemblies
[UIMA-1642] - Regex rule file parameter should allow wildcard expressions when using the datapath to locate rule files
[UIMA-1689] - fix some miscompares between svn export and source-distribution
[UIMA-1691] - change checkout to export in extract and build scripts

New Feature

[UIMA-880] - Make PEAR installation path configurable in web.xml
[UIMA-1021] - implement OpenCalais service wrapper annotator
[UIMA-1033] - ConceptMapper--a highly configurable, token-based dictionary lookup UIMA component
[UIMA-1065] - CFE - configurable feature extrator for UIMA annotation comparison, evaluation, testing, generation of machine learning features
[UIMA-1095] - Implement a Tika Annotator
[UIMA-1299] - Contribution of Lucene CAS Indexer
[UIMA-1371] - Performance improvement: remove reliance on Property class and excess String building to reduce in-memory dictionary size.
[UIMA-1526] - The adaptor of the lucene stop word filter dosn't support the case sensitive flag

Task

[UIMA-1361] - Lucas: Convert documentation into docbook format
[UIMA-1461] - update sandbox POMs to 2.3.0-incubating-SNAPSHOT version
[UIMA-1614] - update 2.3.0-incubating-SNAPSHOT to drop the snapshot in prep for release

Wish

[UIMA-1469] - Add Lucas to the sandbox home page

Apache UIMA Sandbox v2.3.0 Release Notes

Contents