grumpy 0.2.0
Significant new features
read_npy() gains a new lazy argument. When
lazy = TRUE, raw bytes are read lazily using ALTREP and
mmap against the .npy file on disk, resulting in much
better speed & memory performance. This is following a feature
request from @btraven00 in #11.
read_npz() gains a new arrays argument to
allow reading only a subset of the arrays in the npz container.
- Integers that don’t fit in the
int32 range are now
automatically and silently converted to doubles, in line
with what base R is doing. This is following feature requests from
Robrecht Cannoodt and Gabriel Hoffman in downstream dependencies.
Minor improvements
- Elements of structured datatypes are now named if names were
provided during the dataset creation.
read_npz() now adds names to the list it returns. The
names are the names of the bundled .npy files without the
file extension.
grumpy 0.1.1
- The package title and description have been revised based on CRAN
feedback.
- The return value for
convert_bytes_to_array() is now
documented.
grumpy 0.1.0