4.3

Incremental improvements for name matching

• Update test cases to point to 20210811-5
• Fix for issue #171
• Fix for issue #179
• Fix for issue #178
• Fix for issue #177

Improvements to merging and index-building

• Try to re-use left- and right-values if possible.
• Do not duplicate output taxon ids
• Detect and resolve invalid parents
• Merging includes distribution processing, locality normalisation for distributions and reference processing
• Remove stupid names from input

4.2

• Move to source data 20210811-3 #154
  • Uses new conservation lists and invasive species lists
  • Emphasises more authoritative lists
  • Improved handling of phrase names
  • Improved handling of miscellaneous literature
  • Improved handling of nim. nud. and other low-value names
• Handle synthetic names supplied from sources #153

4.1

Improvements to handling for some lists tool requirements.

• Provide a machanism for merging Protista/Chromista/Protoza etc when sources do not agree on kingdom
• Ensure addiitonal identifiers are included and can be used in searches. Needed for the list tool
• Fix for #142
• Fix for #140
• Updated test cases for consistently capitalised common names

4.0

This release has two major changes:

• Update lucene library to 8.1.0 #119
  • This includes a fix for autocomplete behaving differently in lucene 8 #132 Future versions will investigate the lucene autosuggest functionality
• Restructure the name matching library into a number of modules that can be incrementally included in other software. #127 In particular:
  • ala-name-matching-model contains the model classes and vocabulary without name matching functions.
    This allows you to include things like the rank enumeration, ALA terms etc in other software without dragging everything else in.
  • ala-name-matching-search contains the name matching functionality

With additional updates:

• Reference name index updated to 20210811
• Bare 'X sp.' synonyms are now eliminated during merging, since it is impossible to reliably decide what is being synonymised #122
• Illegitimate names are deprecated during merging #121
• Use the lucene API for searches, rather than query parsing. This allows different types of fields (eg name or identifier) to be handled.
• Improvements to phrase name handling #117
• Correct handling of rank/synonym issues #118
• More careful handling of aff. and cf. type names #126
• Updated opencsv library to align with ala-namematching-service #131
• Fix synonym/misapplied conflict #100

3.5

A release to allow a stable version of the sds library, ala-namematching-service and ala-sensitive-data-service to be built.

• Allow a default parent taxon for partial taxonomies during merging
• Allow additional authors/abbreviations from the merge configuration
• Allow a concept resolution strategy for merging
• Report clashing ranks during merging
• Handle common-authority merging
• Fix for #79
• Fix for #100

3.4

• Improved handling of autonyms during merging
• Multiple misapplied names that point to the same accepted taxon no longer trigger higher taxon matches. (This can occur because the same misapplication can occur across multiple sources.)
• Matching with additional higher order information and merge priority is used to prioritise both accepted names and synonyms (helps prevent giant land whelks, amongst other things)
• Parent-child synonyms now resolve to the parent, with a flag, since the parent is going to be accurate even if the supplied name refers to what is now a separate subspecies.
• The master branch is now 3.x, with changes from the 2.4.x series included.

2.4.8: Last lucene 4.10.4 version

This is the last version that uses lucene 4.10.4. Future versions will be on the 3.x version trail and will use lucene 5.x or 6.x.

• The taxon merging can be configured to attempt to place a rank on unranked taxa, either by finding a ranked version supplied by some other source or deducing the rank from the name structure.
• If synonym or parent loops are detected, the loops are broken and the results placed under 'The Unknown Taxon' a generated taxon that can act as a placeholder for broken bits and pieces.
• ICN has been added as a botanical nomenclatural code
• Miscellaneous literature and legislated names can be supplied as vernacular names. A 'pseudo-taxon' is set up so that the same names from other sources are gathered and eliminated if they are to be treated as vernacular names. This is a double-edged sword; treating miscellaneous literature names as a sort of synonym in the input data can mean that higher taxon matches are not attempted.

3.3

Fixes for auto complete methods.

2.4.7: Merge pull request #52 from charvolant/master

• Improved reporting for taxon merging
• Use the keyAdjuster to provide rank adjustments

3.1

Added the ability to specify a language for common name searches

Please note, this version will generate indexes capability with Lucene 5 and above.
Also, the API will only work with indexes. It also lags behind the 2.x versions.