Skip to content

Latest commit

 

History

History
80 lines (63 loc) · 2.67 KB

changelog.md

File metadata and controls

80 lines (63 loc) · 2.67 KB
Raw

Changelog

[Unreleased]

Added

  • Incremental harvesting during a given time window
  • Option to resume from a given token

Removed

  • Support for Python < 3.6

Changed

[3.0.0] - 2017-10-31

Removed

  • Support for Python 2.6

Fixed

  • Log complete traceback on errors
  • Decode binary metadata string returned by pyoai in Python 3
  • Output files as UTF-8

[2.0.0] - 2017-04-19

Added

  • Support for building a development virtualenv and testing builds using Tox
  • Continuous Integration testing with Travis CI

Fixed

  • Working in Python 3 (thanks to @ulikoehler)
  • Allow providers with https transport scheme (thanks to @ulikoehler)

[1.2.0] - 2016-12-30

Fixed

  • Fix build by using ez_setup.py instead of distribute_setup.py for bootstrapping setuptools.

[1.1.5] - 2014-04-28

Added

  • Report an appropriate error when a given URL does not appear to resolve to an OAI-PMH server.

[1.1.4] - 2014-04-02

Added

  • Option to specify a string that, when it occurs in identifiers, is treated as a separator to store harvested records in sub-directories.

Fixed

  • File name escaping on Windows. On Windows do not protect : characters from being escaped.

[1.1.3] - 2013-10-14

Added

  • Option for creating sub-directories based on slashes in identifiers.

[1.1.2] - 2013-07-24

Added

  • Automatically create download directory if missing, thanks to @mhoffman

[1.1.1] - 2013-05-10

Fixed

  • Removed rogue debugging raise statement
  • Update registry with --until option when given
  • Only update registry if harvest completed (i.e. not truncated with --limit=N)

[1.1] - 2013-05-09

Added

  • Handle deletions
  • Selective harvesting by set
  • Limit harvesting to an arbitrary number of records (e.g. to avoid throttling / blacklisting by provider server)

Fixed

  • Don't ERROR when encountering deleted records

[1.0] - 2013-04-15

Added

  • Harvest from a given URL
  • Selective harvesting by modification date
  • Facilitate regular harvesting from 1 or more providers
    • Register parameters - e.g. URL, Schema (metadataPrefix), target directory - for regular harvesting from a named provider
    • Specify provider using registered name when harvesting, or harvest from all registered providers using the reserved name all
    • Harvest only records modified since last harvest (unless overridden)