You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I tried to create a workaround for enabling filters in #107 by creating a dataset with HDF5.PInvoke myself and then using the constructor of a ChunkedDataset to use filtered dataset by providing the same groupId and Datasetname.
FirstDataset method does not work since the Datasetname is already in use.
AppendDataset fails since _oldDims is null and size / offset variable calculations depend on it.
In order to be able to use this workaround, either AppendDataset needs to accept _oldDims as null on the first run (by trying to read the current dimensions) or an "OpenIfExists(groupId, datasetname)" method should be introduced to initialize from an already existing dataset. This would also benefit modifing existing hdf5 files with chunked datasets.
The text was updated successfully, but these errors were encountered:
I tried to create a workaround for enabling filters in #107 by creating a dataset with HDF5.PInvoke myself and then using the constructor of a ChunkedDataset to use filtered dataset by providing the same groupId and Datasetname.
FirstDataset method does not work since the Datasetname is already in use.
AppendDataset fails since _oldDims is null and size / offset variable calculations depend on it.
In order to be able to use this workaround, either AppendDataset needs to accept _oldDims as null on the first run (by trying to read the current dimensions) or an "OpenIfExists(groupId, datasetname)" method should be introduced to initialize from an already existing dataset. This would also benefit modifing existing hdf5 files with chunked datasets.
The text was updated successfully, but these errors were encountered: