Exporting NWB files

You can use the export feature of PyNWB to create a modified version of an existing NWB file, while preserving the original file.

To do so, first open the NWB file using NWBHDF5IO. Then, read the NWB file into an NWBFile object, modify the NWBFile object or its child objects, and export the modified NWBFile object to a new file path. The modifications will appear in the exported file and not the original file.

These modifications can consist of removals of containers, additions of containers, and changes to container attributes. If container attributes are changed, then NWBFile.set_modified() must be called on the NWBFile before exporting.

with NWBHDF5IO(self.read_path, mode='r') as read_io:
    nwbfile = read_io.read()
    # ...  # modify nwbfile
    nwbfile.set_modified()  # this may be necessary if the modifications are changes to attributes

    with NWBHDF5IO(self.export_path, mode='w') as export_io:
        export_io.export(src_io=read_io, nwbfile=nwbfile)

Note

Modifications to h5py.Dataset objects act directly on the read file on disk. Changes are applied immediately and do not require exporting or writing the file. If you want to modify a dataset only in the new file, than you should replace the whole object with a new array holding the modified data. To prevent unintentional changes to the source file, the source file should be opened with mode='r'.

Note

Moving containers within the same file is currently not supported directly via export. See the following discussion on the NWB Help Desk for details.

Note

After exporting an NWBFile, the object IDs of the NWBFile and its child containers will be identical to the object IDs of the read NWBFile and its child containers. The object ID of a container uniquely identifies the container within a file, but should not be used to distinguish between two different files.

See also

The tutorial Adding/Removing Containers from an NWB File provides additional examples of adding and removing containers from an NWB file.

How do I create a copy of an NWB file with different data layouts (e.g., applying compression)?

Use the h5repack command line tool from the HDF5 Group.

How do I create a copy of an NWB file with different controls over how links are treated and whether copies are deep or shallow?

Use the h5copy command line tool from the HDF5 Group.

How do I generate new object IDs for a newly exported NWB file?

Before calling export, call the method generate_new_id on the NWBFile to generate a new set of object IDs for the NWBFile and all of its children, recursively. Then export the NWBFile. The original NWB file is preserved.

with NWBHDF5IO(self.read_path, manager=manager, mode='r') as read_io:
    nwbfile = read_io.read()
    # ...  # modify nwbfile if desired
    nwbfile.generate_new_id()

    with NWBHDF5IO(self.export_path, mode='w') as export_io:
        export_io.export(src_io=read_io, nwbfile=nwbfile)

My NWB file contains links to datasets in other HDF5 files. How do I create a new NWB file with copies of the datasets?

Pass the keyword argument write_args={'link_data': False} to NWBHDF5IO.export. This is similar to passing the keyword argument link_data=False to NWBHDF5IO.write when writing a file with a copy of externally linked datasets.

For example:

with NWBHDF5IO(self.read_path, mode='r') as read_io:
    nwbfile = read_io.read()
    # nwbfile contains a TimeSeries where the TimeSeries data array is a link to an external dataset
    # in a different HDF5 file than self.read_path

    with NWBHDF5IO(self.export_path, mode='w') as export_io:
        export_io.export(src_io=read_io, nwbfile=nwbfile, write_args={'link_data': False})  # copy linked datasets
        # the written file will contain no links to external datasets

You can also the h5copy command line tool from the HDF5 Group.

How do I write a newly instantiated NWBFile to two different file paths?

PyNWB does not support writing an NWBFile that was not read from a file to two different files. For example, if you instantiate NWBFile A and write it to file path 1, you cannot also write it to file path 2. However, you can first write the NWBFile to file path 1, read the NWBFile from file path 1, and then export it to file path 2.

with NWBHDF5IO(self.filepath1, manager=manager, mode='w') as write_io:
    write_io.write(nwbfile)

with NWBHDF5IO(self.filepath1, manager=manager, mode='r') as read_io:
    read_nwbfile = read_io.read()

    with NWBHDF5IO(self.filepath2, mode='w') as export_io:
        export_io.export(src_io=read_io, nwbfile=nwbfile)