Skip to content

Releases: matthiasprobst/h5RDMtoolbox

v2.5.1

01 Nov 07:08

Choose a tag to compare

Changes

  • hotfix Organization class

v2.5.0

31 Oct 14:45

Choose a tag to compare

Changes

  • Zenodo Record returns a "dcat:Dataset" object when publish() is called.
  • Zenodo Record provides as_dcat_dataset() method to obtain the "dcat:Dataset" object for a published record.
  • Shacl result object ValidationResult also provides nodes that caused the validation error
  • bugfix HDF serialization for HDF5 datatypes, which were literals and now are URIs

v2.4.0

19 Oct 16:22

Choose a tag to compare

v2.4.0

  • checksum is correctly checked when downloading files
  • added support for SHACL through function shacl_validate

v2.4.0-rc.2

19 Oct 16:21
25aef6d

Choose a tag to compare

v2.4.0-rc.2

  • fix issues with ZenodoRecord
  • add support of Literals in RDF attributes, e.g. h5.frdf["description"].object = rdflib.Literal("An english
    description", "en")
  • removed deprecated methods in ZenodoSandboxDeposit
  • add support of Literals in RDF attributes, e.g.
    h5.frdf["description"].object = rdflib.Literal("An english description", "en")
  • improve serializing HDF5 contextual and structural metadata to RDF-based formats
  • minor bugfixes
  • default dtime format used within h5rdmtoolbox is now ISO 8601 ('%Y-%m-%dT%H:%M:%S%f')

v2.4.0-rc.1

12 Oct 12:04
d0eddbe

Choose a tag to compare

v2.4.0-rc.1

  • allow numpy 2.x versions
  • extend support for python 3.13
  • upgrade pint to 0.25
  • upgrade pint-xarray up to 0.6.0
  • upgrading to ontolutils 0.21.1
  • fixing linting issues
  • RDF-subjects set for dataset and group are modelled with property schema:about to express that a dataset or group is
    described by the RDF subject. This is more in line with semantic web standards.

v2.3.1

09 Sep 20:06

Choose a tag to compare

v2.3.1

  • fixing error in parsing obj name. "/" is saved and will not be converted anymore

v2.3.0

03 Sep 15:39

Choose a tag to compare

v2.3.0

  • hotfix avoiding blank nodes for hdf filter
  • RDF IRIs are encoded correctly, when using special characters or spaces. The issue was that if HDF names (dataset,
    group, attribute names) contained special characters or spaces, the generated RDF IRI was not encoded correctly. Now,
    the HDF names are URL-encoded when generating the RDF IRI.
  • removed deprecated methods download_file and download_files from RepositoryInterface. The files
    property accessor should be used instead.
  • bug-fixing

v2.2.0

06 Aug 18:46

Choose a tag to compare

  • improved semantic serialization of HDF5 content
  • housekeeping

v2.1.0

05 Jun 17:59

Choose a tag to compare

  • upgrade setuptools due to cve and therefore limit minimum python version to 3.9

v2.0.0

19 May 16:57

Choose a tag to compare

  • changed version number to 2.0.0
  • housekeeping