1.4.0

Enhancements

• Added new reference filesystem data store to support "kerchunked" NetCDF files in object storage. (#928)

  See also

  • ReferenceFileSystem
  • kerchunk

• Improved xcube Server's STAC API:

  • Provide links for multiple coverages data formats
  • Add crs and crs_storage properties to STAC data
  • Add spatial and temporal grid data to collection descriptions
  • Add a schema endpoint returning a JSON schema of a dataset's data
    variables
  • Add links to domain set, range type, and range schema to collection
    descriptions

• Improved xcube Server's Coverages API:

  • Support scaling parameters scale-factor, scale-axes, and scale-size
  • Improve handling of bbox parameters
  • Handle half-open datetime intervals
  • More robust and standard-compliant parameter parsing and checking
  • More informative responses for incorrect or unsupported parameters
  • Omit unnecessary dimensions in TIFF and PNG coverages
  • Use crs_wkt when determining CRS, if present and needed
  • Change default subsetting and bbox CRS from EPSG:4326 to OGC:CRS84
  • Implement reprojection for bbox
  • Ensure datetime parameters match dataset’s timezone awareness
  • Reimplement subsetting (better standards conformance, cleaner code)
  • Set Content-Bbox and Content-Crs headers in the HTTP response
  • Support safe CURIE syntax for CRS specification

Fixes

• Fixed KeyError: 'lon_bnds' raised occasionally when opening
  (mostly NetCDF) datasets. (#930)
• Make S3 unit tests compatible with moto 5 server. (#922)
• Make some CLI unit tests compatible with pytest 8. (#922)
• Rename some test classes to avoid spurious warnings. (#924)

Other changes

• Require Python >=3.9 (previously >=3.8)

Full Changelog: v1.3.1...v1.4.0

1.3.1

Changes in 1.3.1

• Updated Dockerfile and GitHub workflows; no changes to the xcube codebase itself

1.3.0

Changes in 1.3.0

Enhancements

• Added a basic implementation of the draft version of OGC API - Coverages.
  (#879, #889, #900)
• Adapted the STAC implementation to additionally offer datasets as
  individual collections for better integration with OGC API - Coverages.
  (#889)
• Various minor improvements to STAC implementation. (#900)

Fixes

• Resolved the issue for CRS84 error due to latest version of gdal (#869)
• Fixed incorrect additional variable data in STAC datacube properties. (#889)
• Fixed access of geotiff datasets from public s3 buckets (#893)

Other changes

• update_dataset_attrs can now also handle datasets with CRS other than
  WGS84 and update the metadata according to the
  ESIP Attribute Convention for Data Discovery.
• removed deprecated module xcube edit, which has been deprecated since
  version 0.13.0
• Update "Development process" section of developer guide.
• Updated GitHub workflow to build docker image for GitHub releases only and
  not on each commit to master.

1.2.0

• Added a new, experimental /compute API to xcube server.
  It comprises the following endpoints:

  • GET compute/operations - List available operations.
  • GET compute/operations/{opId} - Get details of a given operation.
  • PUT compute/jobs - Start a new job that executes an operation.
  • GET compute/jobs - Get all jobs.
  • GET compute/jobs/{jobId} - Get details of a given job.
  • DELETE compute/jobs/{jobId} - Cancel a given job.

  The available operations are currently taken from module
  xcube.webapi.compute.operations.

  To disable the new API use the following server configuration:

  api_spec:
    excludes: ["compute"] 
  ...

Other changes

• Added shutdown_on_close=True parameter to coiled params to ensure that the
  clusters are shut down on close. (#881)
• Introduced new parameter region for utility function new_cluster in
  xcube.util.dask which will ensure coiled creates the dask cluster in the
  prefered default region: eu-central-1. (#882)
• Server offers the function add_place_group in places/context.py,
  which allows plugins to add place groups from external sources.

1.1.2

Changes in 1.1.2

Fixes

• Fixed issue where geotiff access from a protected s3 bucket was denied (#863)

Changes in 1.1.1

• Bundled new build of xcube-viewer 1.1.0.1
  that will correctly respect a given xcube server from loaded from the
  viewer configuration.

Changes in 1.1.0

Enhancements

• Bundled xcube-viewer 1.1.0.

• Updated installation instructions (#859)

• Included support for FTP filesystem by adding a new data store ftp.

  These changes will enable access to data cubes (.zarr or .levels)
  in FTP storage as shown here:

  store = new_data_store(
      "ftp",                     # FTP filesystem protocol
      root="path/to/files",      # Path on FTP server
      storage_options= {'host':  'ftp.xxx',  # The url to the ftp server
                        'port': 21           # Port, defaults to 21  
                        # Optionally, use 
                        # 'username': 'xxx'
                        # 'password': 'xxx'}  
  )
  store.list_data_ids()

  Note that there is no anon parameter, as the store will assume no anonymity
  if no username and password are set.

  Same configuration for xcube Server:

  DataStores:
  - Identifier: siec
    StoreId: ftp
    StoreParams:
      root: my_path_on_the_host
      max_depth: 1
      storage_options:
        host: "ftp.xxx"
        port: xxx
        username: "xxx"
        password': "xxx"

• Updated xcube Dataset Specification.
  (addressing #844)

• Added xcube Data Access documentation.

Fixes

• Fixed various issues with the auto-generated Python API documentation.

• Fixed a problem where time series requests may have missed outer values
  of a requested time range. (#860)

  • Introduced query parameter tolerance for
    endpoint /timeseries/{datasetId}/{varName} which is
    the number of seconds by which the given time range is expanded. Its
    default value is one second to overcome rounding problems with
    microsecond fractions. (#860)
  • We now round the time dimension labels for a dataset as
    follows (rounding frequency is 1 second by default):
    • First times stamp: floor(time[0])
    • Last times stamp: ceil(time[-1])
    • In-between time stamps: round(time[1: -1])

Other changes

• Pinned gdal dependency to >=3.0, <3.6.3 due to incompatibilities.

1.1.1

Changes in 1.1.1

• Bundled new build of xcube-viewer 1.1.0 that will correctly respect a given xcube server from loaded from the viewer configuration.

Full Changelog: v1.1.0...v1.1.1

Changes in 1.1.0

Enhancements

• Bundled xcube-viewer 1.1.0.

• Updated installation instructions (#859)

• Included support for FTP filesystem by adding a new data store ftp.

  These changes will enable access to data cubes (.zarr or .levels)
  in FTP storage as shown here:

  store = new_data_store(
      "ftp",                     # FTP filesystem protocol
      root="path/to/files",      # Path on FTP server
      storage_options= {'host':  'ftp.xxx',  # The url to the ftp server
                        'port': 21           # Port, defaults to 21  
                        # Optionally, use 
                        # 'username': 'xxx'
                        # 'password': 'xxx'}  
  )
  store.list_data_ids()

  Note that there is no anon parameter, as the store will assume no anonymity
  if no username and password are set.

  Same configuration for xcube Server:

  DataStores:
  - Identifier: siec
    StoreId: ftp
    StoreParams:
      root: my_path_on_the_host
      max_depth: 1
      storage_options:
        host: "ftp.xxx"
        port: xxx
        username: "xxx"
        password': "xxx"

• Updated xcube Dataset Specification.
  (addressing #844)

• Added xcube Data Access documentation.

Fixes

• Fixed various issues with the auto-generated Python API documentation.

• Fixed a problem where time series requests may have missed outer values
  of a requested time range. (#860)

  • Introduced query parameter tolerance for
    endpoint /timeseries/{datasetId}/{varName} which is
    the number of seconds by which the given time range is expanded. Its
    default value is one second to overcome rounding problems with
    microsecond fractions. (#860)
  • We now round the time dimension labels for a dataset as
    follows (rounding frequency is 1 second by default):
    • First times stamp: floor(time[0])
    • Last times stamp: ceil(time[-1])
    • In-between time stamps: round(time[1: -1])

Other changes

• Pinned gdal dependency to >=3.0, <3.6.3 due to incompatibilities.

Full Changelog: v1.0.5...v1.1.0

1.1.0

Changes in 1.1.0

Enhancements

• Bundled xcube-viewer 1.1.0.

• Updated installation instructions (#859)

• Included support for FTP filesystem by adding a new data store ftp.

  These changes will enable access to data cubes (.zarr or .levels)
  in FTP storage as shown here:

  store = new_data_store(
      "ftp",                     # FTP filesystem protocol
      root="path/to/files",      # Path on FTP server
      storage_options= {'host':  'ftp.xxx',  # The url to the ftp server
                        'port': 21           # Port, defaults to 21  
                        # Optionally, use 
                        # 'username': 'xxx'
                        # 'password': 'xxx'}  
  )
  store.list_data_ids()

  Note that there is no anon parameter, as the store will assume no anonymity
  if no username and password are set.

  Same configuration for xcube Server:

  DataStores:
  - Identifier: siec
    StoreId: ftp
    StoreParams:
      root: my_path_on_the_host
      max_depth: 1
      storage_options:
        host: "ftp.xxx"
        port: xxx
        username: "xxx"
        password': "xxx"

• Updated xcube Dataset Specification.
  (addressing #844)

• Added xcube Data Access documentation.

Fixes

• Fixed various issues with the auto-generated Python API documentation.

• Fixed a problem where time series requests may have missed outer values
  of a requested time range. (#860)

  • Introduced query parameter tolerance for
    endpoint /timeseries/{datasetId}/{varName} which is
    the number of seconds by which the given time range is expanded. Its
    default value is one second to overcome rounding problems with
    microsecond fractions. (#860)
  • We now round the time dimension labels for a dataset as
    follows (rounding frequency is 1 second by default):
    • First times stamp: floor(time[0])
    • Last times stamp: ceil(time[-1])
    • In-between time stamps: round(time[1: -1])

Other changes

• Pinned gdal dependency to >=3.0, <3.6.3 due to incompatibilities.

Full Changelog: v1.0.5...v1.1.0

1.0.6.dev1

Changes in 1.0.6 (in development)

• Bundled xcube-viewer 1.1.0-dev.1.

1.0.5

Changes in 1.0.5

• When running xcube in a JupyterLab, the class xcube.webapi.viewer.Viewer can be used to programmatically launch a xcube Viewer UI. The class now recognizes an environment variable XCUBE_JUPYTER_LAB_URL that contains a JupyterLab's public base URL for a given user. To work properly, the jupyter-server-proxy extension must be installed and enabled.

• Bundled xcube-viewer 1.0.2.1.

1.0.4

Changes in 1.0.4

• Setting a dataset's BoundingBox in the server configuration
  is now recognised when requesting the dataset details. (#845)

• It is now possible to enforce the order of variables reported by
  xcube server. The new server configuration key Variables can be added
  to Datasets configurations. Is a list of wildcard patterns that
  determines the order of variables and the subset of variables to be
  reported. (#835)

• Pinned Pandas dependency to lower than 2.0 because of incompatibility
  with both xarray and xcube
  (see pydata/xarray#7716).
  Therefore, the following xcube deprecations have been introduced:

  • The optional --base/-b of the xcube resample CLI tool.
  • The keyword argument base of the xcube.core.resample.resample_in_time
    function.

• Bundled xcube-viewer 1.0.2.

Full Changelog: v1.0.3...v1.0.4