Releases: tilebox/tilebox-python
Releases · tilebox/tilebox-python
v0.45.0
Added
tilebox-workflows: Addedexecution_statsto theJobobject to provide programmatic access to a job's execution
statistics.tilebox-workflows: Added query filters to theJobClient.querymethod to filter jobs by multiple automation ids,
job state, and job name.tilebox-workflows: Added additional JobState values to indicate a job's current state and progress more accurately.tilebox-workflows: Removed the restriction of64subtasks per task.
Changed
tilebox-workflows: Deprecated thejob.canceledproperty, which will be removed in a future version. The
equivalent expression behavior for the oldjob.canceledis(job.state in (JobState.CANCELED, JobState.FAILED).tilebox-workflows: Deprecated thejob.started_atproperty, which will be removed in a future version. The
equivalent information is available injob.execution_stats.first_task_started_atinstead.tilebox-workflows: TheJobState.QUEUEDenum value has been renamed in favor of the more accurate
JobState.SUBMITTED. TheQUEUEDvalue is still available as an alias forSUBMITTEDfor backwards compatibility,
but will be removed in a future version.tilebox-workflows: Switched to an updated internalTaskSubmissionmessage format that allows for more efficient
submission of a very large number of tasks.
v0.44.0
Added
tilebox-workflows: Addedprogress_indicatorsto the query response ofJobClient.findto provide programmatic
access to a job's progress indicators.tilebox-workflows: Added anipywidgetsbased interactive display for Job objects for interactive environments like
Jupyter notebooks.
v0.43.0
tilebox-workflows: Added progress tracking support to theTaskRunner.
v0.42.0
Added
tilebox-storage: AddedUSGSLandsatStorageClientto download landsat data from the USGS Landsat S3 bucket.tilebox-storage: Storage client now support concurrent downloads of multiple objects, controlled by the
max_concurrent_downloadsparameter.tilebox-storage: Addedquicklookanddownload_quicklookmethods to theCopernicusStorageClientto download and
display preview images for Sentinel data.
Fixed
tilebox-workflows: Registering duplicate task identifiers with a task runner now raises aValueErrorinstead of
overwriting the existing task.tilebox-workflows: Fixed a bug where thedeserialize_taskfunction would fail to deserialize nested dataclasses or
protobuf messages that are wrapped in anOptionalorAnnotatedtype hint.tilebox-workflows: Callingconfigure_otel_tracingandconfigure_otel_loggingmultiple times correctly configures
multiple exporters instead of overwriting the existing ones.
v0.41.0
Added
tilebox-workflows: Task runners now support receiving a suggested idling duration from the workflows API
Fixed
tilebox-workflows: Change task lease extension logging message toDEBUGlevel
v0.40.0
Added
- Added
NetworkErrorandNetworkTimeoutErrorexceptions classes to indicate network related errors in grpc calls.
Changed
- Dropped support for
protobuf v5and below, minimum supported version is nowprotobuf>=6.
Fixed
tilebox-workflows: Fix opentelemetry logging and tracing initialization on Windows, which was relying onos.uname,
a stdlib function that does not exist on Windows python SDKs.
v0.39.0
Added
tilebox-workflows: Addedcancel_jobsargument toAutomationClient.deleteto optionally cancel all currently queued or running jobs of the automation.
Changed
tilebox-datasets: ChangedTimeIntervalandTimeIntervalLikeimports fromtilebox.datasets.datatotilebox.datasets.query.
v0.38.0
Added
tilebox-datasets: Addeddelete_collectionmethod toDatasetClientto delete a collection.tilebox-workflows: Added support for nested dataclasses and protobuf messages inside alist,dictortuple
as task arguments.
v0.37.1
Fixed
tilebox-datasets: Fixed a bug inTimeseriesDatasetChunk.from_messagerelying on incorrect bool assumptions about
missing protobuf fields.tilebox-grpc: More robust parsing of GRPC channel URLs.tilebox-workflows: Fixed a bug in the timeseries interceptor that resulted in an error when accessing a collection.
v0.37.0
Changed
- Dropped support for legacy Tilebox datasets, no longer necessary since all datasets have been migrated on the server.