Changes Report

Release History

Version Date Description
2.3.2
2.3.1
2.3.0
2.2.4
2.2.3
2.2.2
2.2
2.1
1.4 final 2008-07-28
1.3 2008-07-28

Get the RSS feed of the last changes

Release 2.3.2 - null

Type Changes By
Lookup rules can fetch terms by URI. Typically, we want to fetch concept URI by label, but sometimes we know the URI and want to fetch terms' labels. omelayenko
SOLR server as a datasource. Data from a running SOLR server can be converted, the same way as databases/SQL, ontologies/SPARQL, and XML files. omelayenko
It is often necessary to add/remove a few statements from the large generated RDF files. New feature OntologySubtractor that allows removing statements from an RDF file (to-be-removed-from) that are stated in another RDF file (statements-to-be-removed). omelayenko
SparqlQueryHelper allows merging a few RDF files (also just-generated conversion results) with an SPARQL statement. Very convenient if the generated file needs to be augmented with some statements from other RDf files. omelayenko
HierarchyTracingFilter reads a file with top concepts, goes through a bunch of RDF data files and lists all direct or indirect skos:narrower concepts in a separate file, thus making an exhaustive list of all children of the top concepts. omelayenko

Release 2.3.1 - null

Type Changes By
Destination PropertiesFile allows storing subject-value pairs in .properties files. This is handy if these files are then loaded for further processing. omelayenko

Release 2.3.0 - null

Type Changes By
Destination RdfGraphSwapNames stored RDF files with swapped parts of the name. Useful if destinations are assigned dynamically omelayenko
Parameter $ANNOCULTOR_DATASOURCE_ITERATOR_10 in a query converts it to ten queries, where $ANNOCULTOR_DATASOURCE_ITERATOR_10 is replaced with 0 to 9. Made for the use case when a query result is too large to fit RAM omelayenko
Keeping previous copy of RDF and compare it to the current. This feature was not used, and DIFF with a version control systems are more useful omelayenko
Multiple queries are allowed in SQL and SPARQL data sources omelayenko
Converter parameters are loaded first from annocultor.properties file then from environment, and then the yet not set properties are completed with default values omelayenko
Record Filters: if conversion result uses a few terms from a vocabulary then it allows creating a shorter version of this vocabulary with the terms that are used omelayenko
Flexible data destination. Possibility to extend the RdfDestinationFile with custom destination formats omelayenko
Flexible datasource. In addition to the built-in XML, SQL, SPARQL, custom implementations of DataSource interface may be plugged-in to work with custom source formats omelayenko
Redesigned (XPath-like) path matching routines to achieve nearly 100-fold speed increase. omelayenko
Redesigned how statistics are collected to make conversion reports. It used to go out of memory on a few Gb of converted files. Now swaps large chunks of statistics data to disk and runs with reasonable memory needs. Prepared for porting statistics to a database omelayenko
Added SPARQL datasource. Can now convert data coming from XML files, SQL or SPARQL queries. omelayenko
Converter for Europeana ESE format, originally meant to run enrichment. omelayenko
Person name disambiguation utilities (see Europeana ESE converter). omelayenko
Place disambiguation utilities (see Europeana ESE converter). omelayenko
Enrichment links are given readable comments. omelayenko
Fixed encoding issue with parsing Geonames singe-file-dump into per-country RDF files. omelayenko
Handling of xml:lang. omelayenko

Release 2.2.4 - null

Type Changes By
Support for XPath attr queries omelayenko

Release 2.2.3 - null

Type Changes By
Move to eu.annocultor packages omelayenko

Release 2.2.2 - null

Type Changes By
Handling cascading xml:lang omelayenko
Moved to Java 6 and the Java 6 built-in way of invoking Java Compiler there omelayenko
Splitting RDF files into volumes of 300.000 triples. Volume number is added to generated RDF file names omelayenko

Release 2.2 - null

Type Changes By
Dynamic load of tools.jar from environment variable omelayenko
Generated URLs. Made them stricter omelayenko
Path expressions. Qualified attributes omelayenko

Release 2.1 - null

Type Changes By
Conversion progress report: records passed, memory used omelayenko
Detector of rubbish terms in vocabulary lookup rules and warning. If unlucky, they may seriously deteriorate system performance. omelayenko
Performance improvements on computing XML paths omelayenko
Major change: XML API: converters are written in XML (with occasional small snippets of Java code) omelayenko

Release 1.4 final - 2008-07-28

Type Changes By
On failure Converter.run returns -1 that can be returned by the main methods to let maven see that it has failed omelayenko
Maven build omelayenko
Separate XML files can be merged and processed with a single converter run omelayenko
Local.build.properties file allows setting build properties for ant build (e.g. -Xmx) omelayenko
Build numbers in both build system and source omelayenko
Local file annocultor.properties may override environment properties and ANNOCULTOR_HOME omelayenko
New conversion report generated into the /doc directory omelayenko

Release 1.3 - 2008-07-28

Type Changes By
streaming conversion, flat memory profile, but vocabularies are stored in memory omelayenko
all tests are automaticallt extracted omelayenko
Path allows expressions like element[attr='value' and attr='value'] in queries omelayenko
Path allows expressions like element@attr in queries returning attribute value omelayenko
Environment stores a number of directories omelayenko
Conversion stores the RDF to be overwritten into /prev, after conversion it does a diff stored in /diff omelayenko