Skip to content

Latest commit

 

History

History
103 lines (67 loc) · 3.32 KB

CHANGELOG.md

File metadata and controls

103 lines (67 loc) · 3.32 KB
Raw

Change Log for GeoExtract

The format of this file is based on Keep a Changelog, and this project uses Semantic Versioning.

Unreleased

Fixed

  • NameExtractor previously raised a hard to understand error message when the location list passed to the Pipeline was empty. It now handles this edge case correctly without raising an exception. (#5, with contributions by @konstin)

0.3.0 (2017-07-07)

Added

  • Support for Python 3.4 and later.

Changed

  • The constructor of PatternExtractor now accepts raw (not compiled) regex strings.
  • All definitions from the geoextract.splitters module have been moved to the geoextract module.
  • The old Normalizer class has been renamed to BasicNormalizer. Normalizer is now an abstract base class from which BasicNormalizer derives.
  • KeyFilterPostprocessor now derives from the new abstract base class Postprocessor.
  • WhitespaceSplitter now derives from the new abstract base class Splitter.
  • All abstract base classes for pipeline components (Splitter, Normalizer, Extractor, Validator, Postprocessor) now derive from a common abstract base class Component.
  • BasicNormalizer.normalize now removes leading and trailing whitespace at the end of the normalization process.
  • Normalization, splitting and validation can now be completely disabled by passing False for the corresponding argument of the Pipeline constructor.

Fixed

  • WindowExtractor and its subclasses ignored the first word of some strings.
  • BasicNormalizer (previously called Normalizer) failed when no substitutions were given to the constructor.

Removed

  • PostalExtractor has been removed.

0.2.0 (2017-05-31)

Added

  • Simple web app for providing geo extraction as a web service.

  • geoextract.NameValidator is a simple validator that ensures that the named locations referred to by an extracted location (for example a street street name) actually exist in the location database.

  • geoextract.Normalizer is a versatile string normalizer.

  • geoextract.KeyFilterPostprocessor is a simple postprocessor that only keeps

Changed

  • GeoExtract now uses a pipeline architecture that covers all aspects of the location extraction process.

  • geoextract.NameExtractor doesn't take the target names as a constructor argument anymore, instead they are automatically provided by the pipeline.

  • The module geoextract.preprocessing was renamed to geoextract.splitters, and the geoextract.preprocessing.split_components function has been refactored into the geoextract.splitters.WhitespaceSplitter class.

Removed

  • geoextract.reduce_locations and geoextract.unique_dicts were merged into the pipeline architecture and aren't available as a standalone functions anymore.

0.1.0 (2017-05-02)

First release.