0.5.0
Changes in 0.5.0
New in 0.5.0
-
xcube gen2 CONFIG
will generate a cube from a data input store and a user given cube configuration.
It will write the resulting cube in a user defined output store.- Input Stores: CCIODP, CDS, SentinelHub
- Output stores: memory, directory, S3
-
xcube serve CUBE
will now use the last path component ofCUBE
as dataset title. -
xcube serve
can now be run with AWS credentials (#296).- In the form
xcube serve --config CONFIG
, aDatasets
entry inCONFIG
may now contain the two new keysAccessKeyId: ...
andSecretAccessKey: ...
given thatFileSystem: obs
. - In the form
xcube serve --aws-prof PROFILE CUBE
the cube stored in bucket with URLCUBE
will be accessed using the
credentials found in section[PROFILE]
of your~/.aws/credentials
file. - In the form
xcube serve --aws-env CUBE
the cube stored in bucket with URLCUBE
will be accessed using the
credentials found in environment variablesAWS_ACCESS_KEY_ID
and
AWS_SECRET_ACCESS_KEY
.
- In the form
Enhancements in 0.5.0
-
Added possibility to specify packing of variables within the configuration of
xcube gen
(#269). The user now may specify a different packing variables,
which might be useful for reducing the storage size of the datacubes.
Currently it is only implemented for zarr format.
This may be done by passing the parameters for packing as the following:output_writer_params: packing: analysed_sst: scale_factor: 0.07324442274239326 add_offset: -300.0 dtype: 'uint16' _FillValue: 0.65535
-
Example configurations for
xcube gen2
were added.
Fixes
-
From 0.4.1: Fixed time-series performance drop (#299).
-
Fixed
xcube gen
CLI tool to correctly insert time slices into an
existing cube stored as Zarr (#317). -
When creating an ImageGeom from a dataset, correct the height if it would
otherwise give a maximum latitude >90°. -
Disable the display of warnings in the CLI by default, only showing them if
a--warnings
flag is given. -
xcube has been extended by a new Data Store Framework (#307).
It is provided by thexcube.core.store
package.
It's usage is currently documented only in the form of Jupyter Notebook examples,
seeexamples/store/*.ipynb
. -
During the development of the new Data Store Framework, some
utility packages have been added:xcube.util.jsonschema
- classes that represent JSON Schemas for types null, boolean,
number, string, object, and array. Schema instances are used for JSON validation,
and object marshalling.xcube.util.assertions
- numerousassert_*
functions that are used for function
parameter validation. All functions raiseValueError
in case an assertion is not met.xcube.util.ipython
- functions that can be called for better integration of objects with
Jupyter Notebooks.
-
Fixed a regression when running "xcube serve" with cube path as parameter (#314)
-
From 0.4.3: Extended
xcube serve
by reverse URL prefix option. -
From 0.4.1: Fixed time-series performance drop (#299).