Skip to content

Releases: gnames/gnfinder

v0.15.1

27 Aug 09:22
Compare
Choose a tag to compare

CHANGELOG

Fix [#101]: BOM interferes with offsets when -U flag is used.

v0.15.0

25 Aug 14:07
Compare
Choose a tag to compare

CHANGELOG

Add [#99]: add TSV format and make ouput format an option for REST API.
Add: update modules
Add: update Go to 1.17
Add [#98]: an option to return names positions in bytes from the text start
instead of UTF-8 characters.

Fix [#100]: fix csv/tsv fields number for verification

v0.14.2

17 Jul 12:46
Compare
Choose a tag to compare

CHANGELOG

  • Add [#96]: Zenodo DOI for citing GNfinder.

v0.14.1

20 Jun 15:40
Compare
Choose a tag to compare

CHANGELOG

  • Add [#92]: return UTF8-encoded text only.

v0.14.0

19 Jun 00:36
Compare
Choose a tag to compare

CHANGELOG

  • Add [#91]: convert/extract plain texts locally
  • Add [#89]: configuration file and environment variables.
  • Add [#87]: support PDFs, MS Word, Excel, RTF, HTML, UTF16 etc via Apache Tika.
  • Add [#86]: an option to return unique found names.

v0.13.0

14 Jun 15:07
Compare
Choose a tag to compare

CHANGELOG

  • Add [#84]: metadata about file and name-finding duration.

v0.12.1

27 Apr 18:07
Compare
Choose a tag to compare

CHANGELOG

  • Add: gnf.Find now takes string instead of []byte.
  • Fix: veification for REST interface

v0.12.0

25 Apr 16:20
Compare
Choose a tag to compare

CHANGELOG

This release is not backward compatible with command line flags and name verification. It introduces CSV format as a default output and JSON compact and pretty formats as optional. Verification format is now the same as at https://verifier.globalnames.org, and is using this site as verification engine.

  • Add: [#81]: represent new lines in verbatim output as "\n".
  • Add: [#80]: use CSV, JSON, JSON pretty for output.
  • Add: [#79]: adjust prior odds using the density of found names in a text.
  • Add: [#78]: fix Odds value for names with 'grey' genus and species.
  • Add: [#77]: add RESTful interface.
  • Add: [#76]: remove subcommands from CLI.
  • Add: [#75]: update tests, remove ginkgo depencency for tests.
  • Add: [#73]: benchmark and optimize tokenizer.
  • Add: [#71]: use embed introduced in Go v1.16.
  • Add: [#70]: migrate code to use gner tokenizer.
  • Add: [#69]: Output Odds as a log10.
  • Add: [#68]: Refactor the code with interfaces to be consistent withother projects.
  • Add: [#64]: Remove common words from species.
  • Add: [#63]: Remove geo-names as uninomials.
  • Add: [#62]: Remove human names as uninomials.

v0.11.1 Dictionaries update

15 Jun 12:53
Compare
Choose a tag to compare

CHANGELOG

  • Add: Update dictionaries.
  • Fix [#51]: Remove 'Piper' from black list, add new words to dictionaries.

Cleaner protobuf schema and fixes in CLI JSON

06 May 00:54
Compare
Choose a tag to compare

Output formats are not backward compatible

CHANGELOG

  • Add [#49]: Cleanup protobuf and JSON outputs. Introducing backward
    incompatible changes in the output. Standardising CLI JSON
    to camelcase, introducing cardinality instead of string for
    a name type, adding canonical simple and full canonical foms
    for matched and current names. Removing current name unless
    it is a synonym.