CHANGELOG.md

HDMF Changelog

HDMF 4.0.0 (Upcoming)

Deprecations

• The following classes have been deprecated and removed: Array, AbstractSortedArray, SortedArray, LinSpace, Query, RegionSlicer, ListSlicer, H5RegionSlicer, DataRegion, RegionBuilder. The following methods have been deprecated and removed: fmt_docval_args, call_docval_func, get_container_cls, add_child, set_dataio (now refactored as set_data_io). We have also removed all early development for region references. @mavaylon1, @rly #1998, #1212
• Python 3.8 has been deprecated. Python 3.9 is the new minimum with support for Python 3.13. @mavaylon1 #1209

Enhancements

• Added support for expandable datasets of references for untyped and compound data types. @stephprince #1188
• Improved html representation of data in Container objects. @h-mayorquin #1100
• Added error when using colon for Container name. A colon cannot be used as a group name when writing to Zarr on Windows. @stephprince #1202

Bug fixes

• Fixed inaccurate error message when validating reference data types. @stephprince #1199
• Fixed incorrect dtype conversion of a StrDataset. @stephprince #1205

HDMF 3.14.5 (October 6, 2024)

Enhancements

• Added support for overriding backend configurations of h5py.Dataset objects in Container.set_data_io. @pauladkisson #1172

Bug fixes

• Fixed bug in writing of string arrays to an HDF5 file that were read from an HDF5 file that was introduced in 3.14.4. @rly @stephprince #1189
• Fixed export of scalar datasets with a compound data type. @stephprince #1185
• Fixed mamba-related error in conda-based GitHub Actions. @rly #1194

HDMF 3.14.4 (September 4, 2024)

Enhancements

• Added support to append to a dataset of references for HDMF-Zarr. @mavaylon1 #1157
• Adjusted stacklevel of warnings to point to user code when possible. @rly #1166
• Improved "already exists" error message when adding a container to a MultiContainerInterface. @rly #1165
• Added support to write multidimensional string arrays. @stephprince #1173
• Add support for appending to a dataset of references. @mavaylon1 #1135

Bug fixes

• Fixed issue where scalar datasets with a compound data type were being written as non-scalar datasets @stephprince #1176
• Fixed H5DataIO not exposing maxshape on non-dci dsets. @cboulay #1149
• Fixed generation of classes in an extension that contain attributes or datasets storing references to other types defined in the extension. @rly #1183

HDMF 3.14.3 (July 29, 2024)

Enhancements

• Added new attribute "dimension_labels" on DatasetBuilder which specifies the names of the dimensions used in the dataset based on the shape of the dataset data and the dimension names in the spec for the data type. This attribute is available on build (during the write process), but not on read of a dataset from a file. @rly #1081
• Speed up loading namespaces by skipping register_type when already registered. @magland #1102
• Speed up namespace loading: return a shallow copy rather than a deep copy in build_const_args. @magland #1103

HDMF 3.14.2 (July 7, 2024)

Enhancements

• Warn when unexpected keys are present in specs. @rly #1134
• Support appending to zarr arrays. @mavaylon1 #1136
• Support specifying "value" key in DatasetSpec. @rly #1143
• Add support for numpy 2. @rly #1139

Bug fixes

• Fix iterator increment causing an extra +1 added after the end of completion. @CodyCBakerPhD #1128

HDMF 3.14.1 (June 6, 2024)

Bug fixes

• Excluded unnecessary artifacts from sdist and wheel. @rly #1119
• Fixed issue with resolving attribute specs that have the same name at different levels of a spec hierarchy. @rly #1122

HDMF 3.14.0 (May 20, 2024)

Enhancements

• Updated _field_config to take type_map as an argument for APIs. @mavaylon1 #1094
• Added TypeConfigurator to automatically wrap fields with TermSetWrapper according to a configuration file. @mavaylon1 #1016
• Updated TermSetWrapper to support validating a single field within a compound array. @mavaylon1 #1061
• Updated testing to not install in editable mode and not run coverage by default. @rly #1107
• Add post_init_method parameter when generating classes to perform post-init functionality, i.e., validation. @mavaylon1 #1089
• Exposed aws_region to HDF5IO and downstream passes to h5py.File. @codycbakerphd #1040
• Exposed progress_bar_class to the GenericDataChunkIterator for more custom control over display of progress while iterating. @codycbakerphd #1110
• Updated loading, unloading, and getting the TypeConfigurator to support a TypeMap parameter. @mavaylon1 #1117

Bug Fixes

• Fixed TermSetWrapper warning raised during the setters. @mavaylon1 #1116

HDMF 3.13.0 (March 20, 2024)

Enhancements

• Unwrap TermSetWrapper within the builder to support different backends more efficiently. @mavaylon1 #1070
• Added docs page that lists limitations of support for the HDMF specification language. @rly #1069
• Added warning when using add_row or add_column to add a ragged array to DynamicTable without an index parameter. @stephprince #1066

HDMF 3.12.2 (February 9, 2024)

Bug fixes

• Fixed recursion error in html representation generation in jupyter notebooks. @stephprince #1038

HDMF 3.12.1 (February 5, 2024)

Bug fixes

• Fixed retrieving the correct path for a HERD zip file on read. #1046
• Fixed internal links in docstrings and tutorials. @stephprince #1031
• Fixed issue with creating documentation links to classes in docval arguments. @rly #1036
• Fixed issue with validator not validating against the spec that defines the data type of the builder. @rly #1050

HDMF 3.12.0 (January 16, 2024)

Enhancements

• Add Data.set_data_io(), which allows for setting a DataIO to a data object after-the-fact. @bendichter and @CodyCBakerPhD #1013
• Added add_ref_termset, updated helper methods for HERD, revised add_ref to support validations prior to populating the tables and added add_ref_container. @mavaylon1 #968
• Use stacklevel in most warnings. @rly #1027
• Fixed broken links in documentation and added internal link checking to workflows. @stephprince #1031

Minor Improvements

• Updated __gather_columns to ignore the order of bases when generating columns from the super class. @mavaylon1 #991
• Update get_key to return all the keys if there are multiple within a HERD instance. @mavaylon1 #999
• Improve HTML rendering of tables. @bendichter #998
• Improved issue and PR templates. @rly #1004
• Added check during validation for if a variable length dataset is empty. @bendichter, @oruebel #789

Bug fixes

• Fixed issue with custom class generation when a spec has a name. @rly #1006
• Fixed issue with usage of deprecated ruamel.yaml.safe_load in src/hdmf/testing/validate_spec.py. @rly #1008
• Fixed issue where ElementIdentifiers data could be set to non-integer values. @rly #1009
• Fixed issue where string datasets/attributes with isodatetime-formatted values failed validation against a text spec. @rly #1026

HDMF 3.11.0 (October 30, 2023)

Enhancements

• Added target_tables attribute to DynamicTable to allow users to specify the target table of any predefined DynamicTableRegion columns of a DynamicTable subclass. @rly #971
• Updated TermSet to include _repr_html_ for easy to read notebook representation. @mavaylon1 #967

Minor improvements

• Set up GitHub dependabot to check for updates to GitHub Actions. @rly #977
• Simplify tox configuration. @rly #988
• Add testing for Python 3.12. @rly #988

Bug fixes

• Updated custom class generation to handle specs with fixed values and required names. @rly #800
• Fixed custom class generation of DynamicTable subtypes to set attributes corresponding to column names for correct write. @rly #800
• Added a .mailmap file to correct mapping of names/emails in git logs. @oruebel #976

HDMF 3.10.0 (October 3, 2023)

Since version 3.9.1 should have been released as 3.10.0 but failed to release on PyPI and conda-forge, this release will increase the minor version number to 3.10.0. See the 3.9.1 release notes below for new features.

Bug fixes

• Fixed issue with testing and deployment of releases. @rly #957

HDMF 3.9.1 (September 29, 2023)

Enhancements

• Updated TermSet to be used with TermSetWrapper, allowing for general use of validation for datasets and attributes. This also brings updates to HERD integration and updates on write to easily add references for wrapped datasets/attributes. @mavaylon1 #950

Minor improvements

• Removed warning when namespaces are loaded and the attribute marking where the specs are cached is missing. @bendichter #926

Bug fixes

• Fixed bug allowing DynamicTable to be constructed with empty id column when initializing all columns via AbstractDataChunkIterator objects. @oruebel #953

HDMF 3.9.0 (August 25, 2023)

New features and minor improvements

• Increased raw data chunk cache size for reading HDF5 files from 1 MiB to 32 MiB. @bendichter, @rly #925
• Increased default chunk size for GenericDataChunkIterator from 1 MB to 10 MB. @bendichter, @rly #925
• Added the magic __reduce__ method as well as two private semi-abstract helper methods to enable pickling of the GenericDataChunkIterator. @codycbakerphd #924
• Updated add_ref_termset to add all instances of TermSet within a given root container. @mavaylon1 #935
• Added Dynamic Enumerations and Schemasheets support to TermSet. @mavaylon1 #923
• Updated HERD to support user defined file name for the HERD zip file. @mavaylon1 #941
• Added method Container.set_data_io, which wraps an existing data field in a DataIO. @bendichter #938

HDMF 3.8.1 (July 25, 2023)

Bug fixes

• Fixed error when calling HDF5IO.read twice. @rly #915

HDMF 3.8.0 (July 21, 2023)

New features and minor improvements

• Added the ability to write ExternalResources if the path is provided and the container has a linked instance of ExternalResources. @mavaylon1 #910

Bug fixes

• Fixed bug on add_ref_term_set in which attributes that were not subscribtable returned an error. @mavaylon1 #909

HDMF 3.7.0 (July 10, 2023)

New features and minor improvements

• Updated ExternalResources to have EntityKeyTable with updated tests/documentation and minor bug fix to ObjectKeyTable. @mavaylon1 #872
• Added abstract static method HDMFIO.can_read() and concrete static method HDF5IO.can_read(). @bendichter #875
• Added warning for DynamicTableRegion links that are not added to the same parent as the original container object. @mavaylon1 #891
• Added the TermSet class along with integrated validation methods for any child of AbstractContainer, e.g., VectorData, Data, DynamicTable. @mavaylon1 #880
• Added AbstractContainer.read_io property to be able to retrieve the HDMFIO object used for reading from the container and to ensure the I/O object used for reading is not garbage collected before the container is being deleted. @bendichter @oruebel #882
• Allow for datetime.date to be used instead of datetime.datetime. @bendichter #874
• Updated HDMFIO and HDF5IO to support ExternalResources. @mavaylon1 #895
• Dropped Python 3.7 support. @rly #897
• Added HTML repr to Container objects which displays an interactive tree of a container's values and children in a Jupyter notebook and other HTML representations. @edeno #883
• Update software_process.rst with the correct external links. @mavaylon1 #900

Documentation and tutorial enhancements:

• Added tutorial for the new TermSet class @mavaylon1 #880

Bug fixes

• Fixed CI testing of minimum installation requirements, and removed some gallery tests run on each PR. @rly #877
• Fixed reporting of version when installed using conda. @rly #890

HMDF 3.6.1 (May 18, 2023)

Bug fixes

• Fixed compatibility with hdmf_zarr for converting string arrays from Zarr to HDF5 by adding logic to determine the dtype for object arrays. @oruebel #866

HDMF 3.6.0 (May 12, 2023)

New features and minor improvements

• Updated ExternalResources to have FileTable and new methods to query data. the ResourceTable has been removed along with methods relating to Resource. @mavaylon1 #850
• Updated hdmf-common-schema version to 1.6.0. @mavaylon1 #850
• Added testing of HDMF-Zarr on PR and nightly. @rly #859
• Replaced setup.py with pyproject.toml. @rly #844
• Use ruff instead of flake8. @rly #844
• Replaced versioneer with setuptools_scm for versioning. @rly #844
• Now recommending developers use pre-commit. @rly #844

Bug fixes

• Export cached specs that are loaded in source IO manager but not in export IO manager. @rly #855
• Fixed issue with passing an S3 URL for source in the constructor of HDMFIO @rly #853

Documentation and tutorial enhancements:

• Updated ExternalResources how to tutorial to include the new features. @mavaylon1 #850

HDMF 3.5.6 (April 28, 2023)

Bug fixes

• Removed references to discontinued requires.io service in documentation. @rly #854

HDMF 3.5.5 (April 13, 2023)

Bug fixes

• Fixed error during export where an external link to the same file was created in some situations. @rly #847
• Removed unused, deprecated codecov package from dev installation requirements. @rly #849
• Fixed export with 'link_data': False' not copying datasets in some situations. @rly #848

HDMF 3.5.4 (April 7, 2023)

Bug fixes

• Fixed typo in deploy release workflow. @rly #845

HDMF 3.5.3 (April 7, 2023)

Bug fixes

• Fixed search bar and missing jquery in ReadTheDocs documentation. @rly #840
• Fixed issue with dynamic class generation for a multi-container interface where one or more are required. @rly #839

Minor improvements

• Added support for Python 3.11. @rly #803
• No longer set upper bounds on dependencies. @rly #803
• DynamicTable.to_dataframe() now sets the index (id) column dtype to int64 rather than the OS default (int32 on Windows, int64 on Mac, Linux). @rly #803

HDMF 3.5.2 (March 13, 2023)

Bug fixes

• Fixed issue with conda CI. @rly #823
• Fixed issue with deprecated pkg_resources. @mavaylon1 #822
• Fixed hdmf.common deprecation warning. @mavaylon1 [#826]((#826)

Internal improvements

• A number of typos fixed and Github action running codespell to ensure that no typo sneaks in #825 was added.
• Added additional documentation for __fields__ in AbstactContainer. @mavaylon1 #827
• Updated warning message for broken links. @mavaylon1 #829

HDMF 3.5.1 (January 26, 2023)

Bug fixes

• Fixed bug when closing a HDF5IO on delete via the new HDMFIO.__del__ before HDF5IO.__init__ has been completed. @oruebel #817

Documentation and tutorial enhancements:

• Updated DynamicTable how to tutorial to clarify the status of EnumData. @oruebel #819

HDMF 3.5.0 (January 17, 2023)

Bug fixes

• Fixed an issue with ExternalResources where retrieving an object resources wasn't in the proper format for a Pandas DataFrame. Additionally, a boolean parameter for creating an object when checking the existence of an object was added. @mavaylon1 #790
• Fixed an issue with the tox tool when upgrading to tox 4. @rly #802
• Fixed export of newly added links to existing elements of the exported file. @rly #808
• Added HDMFIO.__del__ to ensure that I/O objects are being closed on delete. @oruebel#811

Minor improvements

• Added support for reading and writing ExternalResources to and from denormalized TSV files. @mavaylon1 #799
• Changed the name of ExternalResources.export_to_sqlite to ExternalResources.to_sqlite. @mavaylon1 #799
• Updated the tutorial for ExternalResources. @mavaylon1 #799
• Added message argument for assert methods defined by hdmf.testing.TestCase to allow developers to include custom error messages with asserts. @oruebel #812
• Clarify the expected chunk shape behavior for DataChunkIterator. @oruebel #813

HDMF 3.4.7 (November 9, 2022)

Minor improvements

• Update CI, versioneer, and some requirements. @rly (#786)

Bug fixes

• Fix an issue where not providing an optional argument to __init__ of an auto-generated MultiContainerInterface class raised an error. @rly (#779)
• Fixed an issue with the data_utils.GenericDataChunkIterator where if the underlying dataset was such that the numpy.product of the maxshape exceeded the range of the default int32, buffer overflow would occur and cause the true buffer shape to exceed available memory. This has been resolved by dropping all numpy operations (which forced casting within the passed data type) in favor of the unlimited precision of Python builtin integer types @codycbakerphd (#780) (#781)

HDMF 3.4.6 (October 4, 2022)

Minor improvements

• When data is not specified in DataIO, 1) require dtype and shape both be specified and 2) determine length from shape. @ajtritt (#771)

Bug fixes

• Fix an issue when autogenerating a class that extends a class where the constructor docval does not include all of the fields (i.e., when the constructor sets some fields to fixed values). @rly (#773)

HDMF 3.4.5 (September 22, 2022)

Minor improvements

• Allow passing arguments through to column class constructor (argument col_cls) when calling DynamicTable.add_column. @ajtritt (#769)

HDMF 3.4.4 (September 20, 2022)

Bug fixes

• Fixed missing dependency "packaging" introduced in 3.4.3. The code has been updated to avoid the dependency. @rly @oruebel (#770)

HDMF 3.4.3 (September 14, 2022)

Minor improvements

• Began to deprecate the use of the testing script test.py in favor of pytest and test_gallery.py. @rly (#760)
• Updated installation and maintainer documentation. @rly (#760)

Bug fixes

• Fixed CI and flake8 issues. @rly (#760)
• Updated uses of pandas.DataFrame.set_index to avoid FutureWarnings for pandas >=1.5.x @oruebel (#762)
• Fixed broken hdmf.common.get_hdf5io function. @rly (#765)

HDMF 3.4.2 (August 26, 2022)

Minor improvements

• Updated MultiContainerInterface.__make_add to clarify the docstring for add_* methods generated by the function. @oruebel (#758)
• Support "allow_none=True" in docval for args with non-None default. @rly (#757)

Bug fixes

• Fixed deploy release CI. @rly (#759)

HDMF 3.4.1 (August 8, 2022)

Bug fixes

• Fixed deploy release CI. @rly (#754)

HDMF 3.4.0 (August 5, 2022)

Minor improvements

• Allow manual triggering of some GitHub Actions. @rly (#744)
• Relaxed input validation of HDF5IO to allow for s3fs support. Existing arguments of HDF5IO are modified as follows: i) mode was given a default value of "r", ii) path was given a default value of None, and iii) file can now accept an S3File type argument. @bendichter (#746)
• Added ability to create and get back handle to empty HDF5 dataset. @ajtritt (#747)
• Added AbstractContainer._in_construct_mode that is set and modified only by the ObjectMapper when constructing an object from a builder read from a file. Subclasses of AbstractContainer can check _in_construct_mode during the initialization phase as part of __init__ to distinguish between actions during construction (i.e., read from disk) vs. creation by the user, e.g., to determine whether to raise a warning or error when encountering invalid data to support reading and correcting data that is invalid while preventing creation of new data that is invalid. @rly (#751)

Bug fixes

• Fixed PyNWB dev CI. @rly (#749)

HDMF 3.3.2 (June 27, 2022)

Bug fixes

• Fix error message when using GenericDataChunkIterator and chunk shape is larger than dataset size. @rly (#743)
• Fix automatic class generation of extension data types that inherit from both another data type and MultiContainerInteface. @rly (#741)

HDMF 3.3.1 (May 20, 2022)

Bug fixes

• Fixed release deployment CI. @rly (#734, #735, #736)

HDMF 3.3.0 (May 18, 2022)

New features

• Added utility functions and classes useful for implementation of I/O backend to ease maintainability and reuse of functionality across I/O backends. @oruebel (#697)
  • Added HDF5IODataChunkIteratorQueue class in hdmf.backends.hdf5.h5_utils as a new helper class for managing the write of DataChunkIterators to HDF5. @oruebel (#697)
  • Added new hdmf/io/utils.py module with new I/O utility classes: 1) WriteStatusTracker as a simple data structure for tracking the write status of Builders, and 2) NamespaceToBuilderHelper to help with converting of a namespace to a Builder for I/O. @oruebel (#697)
• Added get_min_bounds function to hdmf.data_utils.DataChunk. This functionality was originally part of HDF5IO.__write_chunk__() and has been moved here to enable reuse of the code across data backends. @oruebel (#697)
• Added ignore_string_to_byte option for TestCase.assertContainerEqual to allow us to ignore conversion from string to bytes and just compare unicode values, as different backends may store strings differently. @oruebel (#697)
• Allow zarr.Array as array datatype in docval to support conversion data stored in Zarr to HDMF HDF5 data. @oruebel (#721)
• Allow hdmf.common.table.DynamicTable.add_column to accept nested lists as column data and automatically create the VectorIndex and VectorData objects required to represent the ragged array. @oruebel (#728)
• Added a warning when the __init__ of a class mapped to an HDMF-common data type or an autogenerated data type class is passed positional arguments instead of all keyword arguments. @rly (#730)
• Added helper function hdmf.utils.popargs_to_dict for popping docval args into a dict. @rly (#732)

Bug fixes

• Fixed error with modifying files that contain external links to other files (e.g., shallow copies). @rly (#709)
• Fixed opening of files in append mode on Windows when the files contain links to other open files. @rly (#710)
• Updated HDF5IO to always set the location attribute of GroupBuilders, DatasetBuilders, and LinkBuilders on read. @oruebel (#697)
• Updated HDF5IO.get_types to correctly determine the data type for bytes data. @oruebel (#697)

Minor improvements

• Updated HDF5IO to use the new WriteStatusTracker, NamespaceToBuilderHelper, and HDF5IODataChunkIteratorQueue helper classes. @oruebel (#697)
• Updated hdmf.common.sparse.CSRMatrix to avoid direct dependency on h5py as a particular storage backend. @oruebel (#697)
• Improved readability of Container code. @rly (#707)
• Use GitHub Actions for all CI. @rly (#718)
• Allow write_args=None to be passed to HDF5IO.export. @rly (#733)
• Updated requirements and fixed minor documentation issues and warnings. @rly (#731)

Test enhancements

• Moved test functions to ease reuse and updated tests accordingly. @oruebel (#697)
• Moved Baz, BazData, BazCpdData, BazBucket, get_baz_buildmanager test data classes from tests.unit.test_io_hdf5_h5tools to tests.unit.utils to ease reuse and updated tests accordingly. Also _get_baz_manager was renamed to get_baz_buildmanager as part of this move. @oruebel (#697)
• Added numerous tests to tests/unit/common/test_sparse.py to enhance testing of the CSRMatrix type. @oruebel (#697)

Documentation and tutorial enhancements:

• Add copy button to code blocks. @weiglszonja @oruebel (#726)

HDMF 3.2.1 (February 22, 2022)

Bug fixes

• Fixed release CI that prevented distribution from being uploaded to PyPI. @rly (#699)

HDMF 3.2.0 (February 22, 2022)

New features

• Added hdmf.container.Row.__str__ to improve print of rows. @oruebel (#667)
• Added to_dataframe method for hdmf.common.resources.ExternalResource to improve visualization. @oruebel (#667)
• Added export_to_sqlite method for hdmf.common.resources.ExternalResource. @oruebel (#667)
• Added reset_parent method for hdmf.container.Container. @rly (#692)

Minor improvements

• Plotted results in external resources tutorial. @oruebel (#667)
• Added support for Python 3.10. @rly (#679)
• Updated requirements. @rly @TheChymera (#681)
• Improved testing for ExternalResources. @mavaylon1 (#673)
• Improved docs for export. @rly (#674)
• Enhanced data chunk iteration speeds through new GenericDataChunkIterator class. @CodyCBakerPhD (#672)
• Enhanced issue template forms on GitHub. @CodyCBakerPHD (#700)

Bug fixes

• Fixed setup.py not being able to import versioneer when installing in an embedded Python environment. @rly (#662)
• Fixed broken tests in Python 3.10. @rly (#664)
• Fixed broken LaTeX PDF build of the docs. @oruebel (#669)
• Fixed adding containers as a child to a parent container sometimes not marking the parent container as modified. @rly (#683)
• Fixed to_hierarchcial_dataframe failing when a table contains a VectorIndex column as a regular data column. @oruebel (#666)
• Stop testing against base Python error messages because they may change in the future. @rly (#689)

HDMF 3.1.1 (July 29, 2021)

Bug fixes

• Updated the new DynamicTableRegion.get_linked_tables function (added in 3.1.0) to return lists of typing.NamedTuple objects rather than lists of dicts. @oruebel (#660)

HDMF 3.1.0 (July 29, 2021)

New features

• Added several features to simplify interaction with DynamicTable objects that link to other tables via DynamicTableRegion columns. @oruebel (#645)
  • Added DynamicTable.get_foreign_columns to find all columns in a table that are a DynamicTableRegion
  • Added DynamicTable.has_foreign_columns to identify if a DynamicTable contains DynamicTableRegion columns
  • Added DynamicTable.get_linked_tables to retrieve all tables linked to either directly or indirectly from the current table via DynamicTableRegion
  • Implemented the new get_foreign_columns, has_foreign_columns, and get_linked_tables also for AlignedDynamicTable
  • Added new module hdmf.common.hierarchicaltable with helper functions to facilitate conversion of hierarchically nested DynamicTable objects via the following new functions:
    • to_hierarchical_dataframe to merge linked tables into a single consolidated pandas DataFrame.
    • drop_id_columns to remove "id" columns from a DataFrame.
    • flatten_column_index to replace a pandas.MultiIndex with a regular pandas.Index

Bug fixes

• Do not build wheels compatible with Python 2 because HDMF requires Python 3.7. @rly (#642)
• AlignedDynamicTable did not overwrite its get function. When using DynamicTableRegion to reference AlignedDynamicTable this led to cases where the columns of the category subtables where omitted during data access (e.g., conversion to pandas.DataFrame). This fix adds the AlignedDynamicTable.get based on the existing AlignedDynamicTable.__getitem__. @oruebel (#645)
• Fixed #651 to support selection of cells in an AlignedDynamicTable via slicing with [int, (str, str)](and [int, str, str]) to select a single cell, and [int, str] to select a single row of a category table. @oruebel (#645)

Minor improvements

• Updated DynamicTable.to_dataframe() and DynamicTable.get functions to set the .name attribute on generated pandas DataFrame objects. @oruebel (#645)
• Added AlignedDynamicTable.get_colnames(...) to support look-up of the full list of columns as the AlignedDynamicTable.colnames property only includes the columns of the main table for compliance with DynamicTable @oruebel (#645)
• Fix documentation for DynamicTable.get and DynamicTableRegion.get. @rly (#650)
• Allow passing string column name to DynamicTableRegion, i.e., dtr['col_name'] is a shortcut to dtr.table['col_name']. @rly (#657)

HDMF 3.0.1 (July 7, 2021)

Bug fixes

• Fixed release CI that prevented distribution from being uploaded to PyPI. @rly (#641)

HDMF 3.0.0 (July 6, 2021)

New features

• Add support for Python 3.9, drop support for Python 3.6. @rly (#620)
• Add support for h5py 3. @ajtritt (#480)
  • h5py 3 introduced [breaking changes regarding how strings are handled] (https://docs.h5py.org/en/latest/whatsnew/3.0.html#breaking-changes-deprecations), specifically that variable-length UTF-8 strings in datasets are now read as bytes objects instead of str by default. To reduce the impact of this change on HDMF users, when HDMF reads a variable-length UTF-8 string dataset, instead of returning an h5py.Dataset that is read as bytes objects, HDMF will return a hdmf.utils.StrDataset object that extends h5py.Dataset and is read as str objects, which preserves previous behavior. For example, under HDMF 2.x, an HDF5 dataset d with data ['a', 'b'] is read as a h5py.Dataset object, and d[:] returns str objects. Under HDMF 3.x, the same dataset d is read as a hdmf.utils.StrDataset object and d[:] still returns str objects.
• Add RRID to docs. @oruebel (#633)
• Allow passing index=True to DynamicTable.to_dataframe() to support returning DynamicTableRegion columns as indices or Pandas DataFrame. @rly (#579)
• Improve DynamicTable documentation. @rly (#639)
• Updated external resources tutorial. @mavaylon1 (#611)

Breaking changes and deprecations

• Previously, when using DynamicTable.__getitem__ or DynamicTable.get to access a selection of a DynamicTable containing a DynamicTableRegion, new columns with mangled names for the table data referred to by the DynamicTableRegion were added to the returned DataFrame. This did not work properly for ragged DynamicTableRegion, multiple levels of nesting, or multiple rows returned. Now, these methods will by default return columns of indices of the DynamicTableRegion. If index=False is passed to DynamicTable.get, then nested DataFrames will be returned, one DataFrame per row of the original resulting DataFrame. @rly (#579)

Minor improvements

• Updated requirements and tests. @rly (#640)

Bug fixes

• Update the validator to allow extensions to data types which only define data_type_inc. @dsleiter (#609)
• Fix error when validating lazy-loaded datasets containing references. @dsleiter (#609)
• Fix error when using DynamicTable.__getitem__ or DynamicTable.get when table has a ragged DynamicTableRegion. @rly (#579)

HDMF 2.5.8 (June 16, 2021)

• Fix incorrect dtype precision upgrade for VectorIndex (#631)

Minor improvements

• Improve Sphinx documentation. @rly (#627)

Bug fix

• Fix error with representing an indexed table column when the VectorIndex dtype precision is upgraded more than one step, e.g., uint8 to uint32. This can happen when, for example, a single add_row call is used to add more than 65535 elements to an empty indexed column. @rly (#631)

HDMF 2.5.7 (June 4, 2021)

Bug fix

• Fix generation of extension classes that extend MultiContainerInterface and use a custom _fieldsname. @rly (#626)

HDMF 2.5.6 (May 19, 2021)

Bug fix

• Raise minimum version of pandas from 0.23 to 1.0.5 to be compatible with numpy 1.20. @rly (#618)
• Update documentation and update structure of requirements files. @rly (#619)

HDMF 2.5.5 (May 17, 2021)

Bug fix

• Fix incompatibility issue with downstream github-release tool used to deploy releases to GitHub. @rly (#614)

HDMF 2.5.4 (May 17, 2021)

Bug fix

• Fix incompatibility issue with downstream github-release tool used to deploy releases to GitHub. @rly (#607)
• Fix issue where dependencies of included types were not being loaded in namespaces / extensions. @rly (#613)

HDMF 2.5.3 (May 12, 2021)

Bug fix

• Fix issue where tables with multi-indexed columns defined using __columns__ did not have attributes properly set. @rly (#605)

HDMF 2.5.2 (May 11, 2021)

Bug fix

• Add explicit setuptools requirement. @hrnciar (#596)
• Fix issue with generated custom classes that use a custom fields name (e.g., PyNWB uses __nwbfields__ instead of __fields__). @rly (#598)
• Fix issue with Sphinx Gallery. @rly (#601)

HDMF 2.5.1 (April 23, 2021)

Bug fix

• Revert breaking change in TypeMap.get_container_cls. While this function is returned to its original behavior, it will be modified at the next major release. Please use the new TypeMap.get_dt_container_cls instead. @rly (#590)

HDMF 2.5.0 (April 22, 2021)

New features

• DynamicTable can be automatically generated using get_class. Now the HDMF API can read files with extensions that contain a DynamicTable without needing to import the extension first. @rly and @bendichter (#536)
• Add HDF5IO.get_namespaces(path=path, file=file) method which returns a dict of namespace name mapped to the namespace version (the largest one if there are multiple) for each namespace cached in the given HDF5 file. @rly (#527)
• Use HDMF common schema 1.5.0.
  • Add experimental namespace to HDMF common schema. New data types should go in the experimental namespace (hdmf-experimental) prior to being added to the core (hdmf-common) namespace. The purpose of this is to provide a place to test new data types that may break backward compatibility as they are refined. @ajtritt (#545)
  • ExternalResources was changed to support storing both names and URIs for resources. @mavaylon1 (#517, #548)
  • The VocabData data type was replaced by EnumData to provide more flexible support for data from a set of fixed values.
  • Added AlignedDynamicTable, which defines a DynamicTable that supports storing a collection of sub-tables. Each sub-table is itself a DynamicTable that is aligned with the main table by row index. Each sub-table defines a sub-category in the main table effectively creating a table with sub-headings to organize columns.
  • See https://hdmf-common-schema.readthedocs.io/en/latest/format_release_notes.html#april-19-2021 for more details.
• Add EnumData type for storing data that comes from a fixed set of values. This replaces VocabData i.e. VocabData has been removed. VocabData stored vocabulary elements in an attribute, which has a size limit. EnumData now stores elements in a separate dataset, referenced by an attribute stored on the EnumData dataset. @ajtritt (#537)
• Add AlignedDynamicTable type which defines a DynamicTable that supports storing a collection of subtables. Each sub-table is itself a DynamicTable that is aligned with the main table by row index. Each subtable defines a sub-category in the main table effectively creating a table with sub-headings to organize columns. @oruebel (#551)
• Add tutorial for new AlignedDynamicTable type. @oruebel (#571)
• Equality check for DynamicTable now also checks that the name and description of the table are the same. @rly (#566)

Internal improvements

• Update CI and copyright year. @rly (#523, #524)
• Refactor class generation code. @rly (#533, #535)
• Equality check for DynamicTable returns False if the other object is a DynamicTable instead of raising an error. @rly (#566)
• Update ruamel.yaml usage to new API. @rly (#587)
• Remove use of ColoredTestRunner for more readable verbose test output. @rly (#588)

Bug fixes

• Fix CI testing on Python 3.9. @rly (#523)
• Fix certain edge cases where GroupValidator would not validate all of the child groups or datasets attached to a GroupBuilder. @dsleiter (#526)
• Fix bug for generating classes from link specs and ignored 'help' fields. @rly (#535)
• Various fixes for dynamic class generation. @rly (#561)
• Fix generation of classes that extends both MultiContainerInterface and another class that extends MultiContainerInterface. @rly (#567)
• Fix make clean command for docs to clean up sphinx-gallery tutorial files. @oruebel (#571)
• Make sure we cannot set AlignedDynamicTable as a category on an AlignedDynamicTable. @oruebel (#571)
• Fix included data type resolution between HDMF and custom classes that customize the data_type_inc key. @rly (#503)
• Fix classification of attributes as new/overridden. @rly (#503)

HDMF 2.4.0 (February 23, 2021)

New features

• GroupValidator now checks if child groups, datasets, and links have the correct quantity of elements and returns an IncorrectQuantityError for each mismatch. @dsleiter (#500)

Internal improvements

• Update CI. @rly (#432)
• Added driver option for ros3. @bendichter (#506)

Bug fixes

• Allow np.bool_ as a valid bool dtype when validating. @dsleiter (#505)
• Fix building of Data objects where the spec has no dtype and the Data object value is a DataIO wrapping an AbstractDataChunkIterator. @rly (#512)
• Fix TypeError when validating a group with an illegally-linked child. @dsleiter (#515)
• Fix DynamicTable.get for compound type columns. @rly (#518)
• Fix and removed error "Field 'x' cannot be defined in y." when opening files with some extensions. @rly (#519)

HDMF 2.3.0 (December 8, 2020)

New features

• Add methods for automatic creation of MultiContainerInterface classes. @bendichter (#420, #425)
• Add ability to specify a custom class for new columns to a DynamicTable that are not VectorData, DynamicTableRegion, or VocabData using DynamicTable.__columns__ or DynamicTable.add_column(...). @rly (#436)
• Add support for creating and specifying multi-index columns in a DynamicTable using add_column(...). @bendichter, @rly (#430)
• Add capability to add a row to a column after IO. @bendichter (#426)
• Add method AbstractContainer.get_fields_conf. @rly (#441)
• Add functionality for storing external resource references. @ajtritt (#442)
• Add method hdmf.utils.get_docval_macro to get a tuple of the current values for a docval_macro, e.g., 'array_data' and 'scalar_data'. @rly (#446)
• Add SimpleMultiContainer, a data_type for storing a Container and Data objects together. @ajtritt (#449)
• Support pathlib.Path paths in HDMFIO.__init__, HDF5IO.__init__, and HDF5IO.load_namespaces. @dsleiter (#450)
• Use hdmf-common-schema 1.2.1. See https://hdmf-common-schema.readthedocs.io/en/latest/format_release_notes.html for details.
• Block usage of h5py 3+. h5py>=2.9, <3 is supported. @rly (#461)
• Block usage of numpy>=1.19.4 due to a known issue with numpy on some Windows 10 systems. numpy>1.16, <1.19.4 is supported. @rly (#461)
• Add check for correct quantity during the build process in ObjectMapper. @rly (#463, #492)
• Allow passing GroupSpec and DatasetSpec objects for the 'target_type' argument of LinkSpec.__init__(...). @rly (#468)
• Use hdmf-common-schema 1.3.0. @rly, @ajtritt (#486)
  • Changes from hdmf-common-schema 1.2.0:
    • Add data type ExternalResources for storing ontology information / external resource references. NOTE: this data type is in beta testing and is subject to change in a later version.
    • Fix missing data_type_inc and use dtype uint for CSRMatrix. It now has data_type_inc: Container.
    • Add hdmf-schema-language comment at the top of each yaml file.
    • Add SimpleMultiContainer, a Container for storing other Container and Data objects together.

Internal improvements

• Drop support for Python 3.5. @ajtritt (#459)
• Improve warning about cached namespace when loading namespaces from file. @rly (#422)
• Refactor HDF5IO.write_dataset to be more readable. @rly (#428)
• Fix bug in slicing tables with DynamicTableRegions. @ajtritt (#449)
• Add testing for Python 3.9 and using pre-release packages. @ajtritt, @rly (#459, #472)
• Improve contributing guide. @rly (#474)
• Update CI. @rly, @dsleiter (#481, #493, #497)
• Add citation information to documentation and support for duecredit tool. @rly (#477, #488)
• Add type checking and conversion in CSRMatrix. @rly (#485)
• Clean up unreachable validator code. @rly (#483)
• Reformat imports. @bendichter (#469)
• Remove unused or refactored internal builder functions GroupBuilder.add_group, GroupBuilder.add_dataset, GroupBuilder.add_link, GroupBuilder.set_builder, BaseBuilder.deep_update, GroupBuilder.deep_update, DatasetBuilder.deep_update. Make BaseBuilder not instantiable and refactor builder code. @rly (#452)

Bug fixes

• Fix development package dependency issues. @rly (#431)
• Fix handling of empty lists against a spec with text/bytes dtype. @rly (#434)
• Fix handling of 1-element datasets with compound dtype against a scalar spec with text/bytes dtype. @rly (#438)
• Fix convert dtype when writing numpy array from h5py.Dataset. @rly (#427)
• Fix inheritance when non-AbstractContainer is base class. @rly (#444)
• Fix use of hdmf.testing.assertContainerEqual(...) for Data objects. @rly (#445)
• Add missing support for data conversion against spec dtypes "bytes" and "short". @rly (#456)
• Clarify the validator error message when a named data type is missing. @dsleiter (#478)
• Update documentation on validation to indicate that the example command is not implemented @dsleiter (#482)
• Fix generated docval for classes with a LinkSpec. @rly (#487)
• Fix access of DynamicTableRegion of a DynamicTable with column of references. @rly (#491)
• Fix handling of __fields__ for Data subclasses. @rly (#441)
• Fix DynamicTableRegion having duplicate fields conf 'table'. @rly (#441)
• Fix inefficient and sometimes inaccurate build process. @rly (#451)
• Fix garbage collection issue in Python 3.9. @rly (#496)

HDMF 2.2.0 (August 14, 2020)

New features

• Add ability to get list of tuples when indexing a DynamicTable. i.e. disable conversion to pandas.DataFrame. @ajtritt (#418)

Internal improvements

• Improve documentation and index out of bounds error message for DynamicTable. @rly (#419)

Bug fixes:

• Fix error when constructing DynamicTable with DataChunkIterators as columns. @ajtritt (#418)

HDMF 2.1.0 (August 10, 2020)

New features

• Users can now use the MultiContainerInterface class to generate custom API classes that contain collections of containers of a specified type. @bendichter @rly (#399)
  • See the user guide https://hdmf.readthedocs.io/en/stable/tutorials/multicontainerinterface.html for more information.

Internal improvements

• Add ability to pass callable functions to run when adding or removing items from a LabelledDict. An error is now raised when using unsupported functionality in LabelledDict. @rly (#405)
• Raise a warning when building a container that is missing a required dataset. @rly (#413)

HDMF 2.0.1 (July 22, 2020)

Internal improvements

• Add tests for writing table columns with DataIO data, e.g., chunked, compressed data. @rly (#402)
• Add CI to check for breakpoints and print statements. @rly (#403)

Bug fixes:

• Remove breakpoint. @rly (#403)
• Allow passing None for docval enum arguments with default value None. @rly (#409)
• If a file is written with an orphan container, e.g., a link to a container that is not written, then an OrphanContainerBuildError will be raised. This replaces the OrphanContainerWarning that was previously raised. @rly (#407)

HDMF 2.0.0 (July 17, 2020)

New features

• Users can now call HDF5IO.export and HDF5IO.export_io to write data that was read from one source to a new HDF5 file. Developers can implement the export method in classes that extend HDMFIO to customize the export functionality. See https://hdmf.readthedocs.io/en/latest/export.html for more details. @rly (#388)
• Users can use the new export functionality to read data from one source, modify the data in-memory, and then write the modified data to a new file. Modifications can include additions and removals. To facilitate removals, AbstractContainer contains a new _remove_child method and BuildManager contains a new purge_outdated method. @rly (#388)
• Users can now call Container.generate_new_id to generate new object IDs for the container and all of its children. @rly (#401)
• Use hdmf-common-schema 1.2.0. @ajtritt @rly (#397)
  • VectorIndex now extends VectorData instead of Index. This change allows VectorIndex to index other VectorIndex types.
  • The Index data type is now unused and has been removed.
  • Fix missing dtype for VectorIndex.
  • Add new VocabData data type.

Breaking changes

• Builder objects no longer have the written field which was used by HDF5IO to mark the object as written. This is replaced by HDF5IO.get_written. @rly (#381)
• HDMFIO.write and HDMFIO.write_builder no longer have the keyword argument exhaust_dcis. This remains present in HDF5IO.write and HDF5IO.write_builder. @rly (#388)
• The class method HDF5IO.copy_file is no longer supported and may be removed in a future version. Please use the HDF5IO.export method or h5py.File.copy method instead. @rly (#388)

HDMF 1.6.4 (June 26, 2020)

Internal improvements

• Add ability to close open links. @rly (#383)

Bug fixes:

• Fix validation of empty arrays and scalar attributes. @rly (#377)
• Fix issue with constructing DynamicTable with empty array colnames. @rly (#379)
• Fix TestCase.assertContainerEqual passing wrong arguments. @rly (#385)
• Fix 'link_data' argument not being used when writing non-root level datasets. @rly (#384)
• Fix handling of ASCII numpy array. @rly (#387)
• Fix error when optional attribute reference is missing. @rly (#392)
• Improve testing for get_data_shape and fix issue with sets. @rly (#394)
• Fix inability to write references to HDF5 when the root builder is not named "root". @rly (#395)

HDMF 1.6.3 (June 9, 2020)

Internal improvements

• Improve documentation of DynamicTable. @rly (#371)
• Add user guide / tutorial for DynamicTable. @rly (#372)
• Improve logging of build and write processes. @rly (#373)

Bug fixes:

• Fix adding of optional predefined columns to DynamicTable. @rly (#371)
• Use dtype from dataset data_type definition when extended spec lacks dtype. @rly (#364)

HDMF 1.6.2 (May 26, 2020)

Internal improvements:

• Update MacOS in CI. @rly (#310)
• Raise more informative error when adding column to DynamicTable w/ used name. @rly (#307)
• Refactor _init_class_columns for use by DynamicTable subclasses. @rly (#323)
• Add/fix docstrings for DynamicTable. @oruebel, @rly (#304, #353)
• Make docval-decorated functions more debuggable in pdb. @rly (#308)
• Change dtype conversion warning to include path to type. @rly (#311)
• Refactor DynamicTable.add_column to raise error when name is an optional column. @rly (#305)
• Improve unsupported filter error message. @bendichter (#329)
• Add functionality to validate a yaml file against a json schema file. @bendichter (#332)
• Update requirements-min.txt for yaml validator. @bendichter (#333)
• Add allowed value / enum validation in docval. @rly (#335)
• Add logging of build and hdf5 write process. @rly (#336, #349)
• Allow loading namespaces from h5py.File object not backed by file. @rly (#348)
• Add CHANGELOG.md. @rly (#352)
• Fix codecov reports. @rly (#362)
• Make getargs raise an error if the argument name is not found. @rly (#365)
• Improve get_class and docval support for uint. @rly (#361)

Bug fixes:

• Register new child types before new parent type for dynamic class generation. @rly (#322)
• Raise warning not error when adding column with existing attr name. @rly (#324)
• Add __version__. @rly (#345)
• Only write a specific namespace version if it does not exist. @ajtritt (#346)
• Fix documentation formatting for DynamicTable. @rly (#353)

HDMF 1.6.1 (Mar. 2, 2020)

Internal improvements:

• Allow docval to warn about use of positional arguments. @rly (#293)
• Improve efficiency of writing chunks with DataChunkIterator and HDF5. @d-sot, @oruebel (#295)

Bug fixes:

• Flake8 style fixes. @oruebel (#291)
• Handle missing namespace version. @rly (#292)
• Do not raise error when a numeric type with a higher precision is provided for a spec with a lower precision and different base type. Raise a warning when the base type of a given value is converted to the specified base type, regardless of precision level. Add missing support for boolean conversions. @rly (#298, #299)
• Add forgotten validation of links. @t-b, @ajtritt (#286)
• Improve message for "can't change container_source" error. @rly (#302)
• Fix setup.py development status. @rly (#303)
• Refactor missing namespace version handling. @rly, @ajtritt (#297)
• Add print function for DynamicTableRegion. @oruebel, @rly (#290)
• Fix writing of refined RefSpec attribute. @oruebel, @rly (#301)

HDMF 1.6.0 (Jan. 31, 2020)

Internal improvements:

• Allow extending/overwriting attributes on dataset builders. @rly, @ajtritt (#279)
• Allow ASCII data where UTF8 is specified. @rly (#282)
• Add function to convert DynamicTableRegion to a pandas dataframe. @oruebel (#239)
• Add "mode" property to HDF5IO. @t-b (#280)

Bug fixes:

• Fix readthedocs config to include all submodules. @rly (#277)
• Fix test runner double printing in non-verbose mode. @rly (#278)

HDMF 1.5.4 (Jan. 21, 2020)

Bug fixes:

• Upgrade hdmf-common-schema 1.1.2 -> 1.1.3, which includes a bug fix for missing data and shape keys on VectorData, VectorIndex, and DynamicTableRegion data types. @rly (#272)
• Clean up documentation scripts. @rly (#268)
• Fix broken support for pytest testing framework. @rly (#274)
• Fix missing CI testing of minimum requirements on Windows and Mac. @rly (#270)
• Read 1-element datasets as scalar datasets when a scalar dataset is expected by the spec. @rly (#269)
• Fix bug where 'version' was not required for SpecNamespace. @bendichter (#276)

HDMF 1.5.3 (Jan. 14, 2020)

Minor improvements:

• Update and fix documentation. @rly (#267)

Bug fixes:

• Fix ReadTheDocs integration. @rly (#263)
• Fix conda build. @rly (#266)

HDMF 1.5.2 (Jan. 13, 2020)

Minor improvements:

• Add support and testing for Python 3.8. @rly (#247)
• Remove code duplication and make Type/Value Error exceptions more informative. @yarikoptic (#243)
• Streamline CI and add testing of min requirements. @rly (#258)

Bug fixes:

• Update hdmf-common-schema submodule to 1.1.2. @rly (#249, #252)
• Add support for np.array(DataIO) in py38. @rly (#248)
• Fix bug with latest version of coverage. @rly (#251)
• Stop running CI on latest and latest-tmp tags. @rly (#254)
• Remove lingering mentions of PyNWB. @rly (#257, #261)
• Fix and clean up documentation. @rly (#260)

HDMF 1.5.1 (Jan. 8, 2020)

Minor improvements:

• Allow passing HDF5 integer filter ID for dynamically loaded filters. @d-sot (#215)

Bug fixes:

• Fix reference to hdmf-common-schema 1.1.0. @rly (#231)

HDMF 1.5.0 (Jan. 6, 2020)

Minor improvements:

• Improve CI for HDMF to test whether changes in HDMF break PyNWB. #207 (@rly)
• Improve and clean up unit tests. #211, #214, #217 (@rly)
• Refactor code to remove six dependency and separate ObjectMapper into its own file. #213, #221 (@rly)
• Output exception message in ObjectMapper.construct. #220 (@t-b)
• Improve docstrings for VectorData, VectorIndex, and DynamicTableRegion. #226, #227 (@bendichter)
• Remove unused "datetime64" from supported dtype strings. #230 (@bendichter)
• Cache builders by h5py object id not name. #235 (@oruebel)
• Update copyright date and add legal to source distribution. #232 (@rly)
• Allow access to export_spec function from hdmf.spec package. #233 (@bendichter)
• Make calls to docval functions more efficient, resulting in a ~20% overall speedup. #238 (@rly)

Bug fixes:

• Fix wrong reference in ObjectMapper.get_carg_spec. #208 (@rly)
• Fix container source not being set for some references. #219 (@rly)

Python 2.7 is no longer supported.

HDMF 1.4.0 and earlier

Please see the release notes on the HDMF GitHub repo Releases page.