Datacube-OWS 1.9.0-rc1 pre-release

Datacube-OWS 1.9.0-rc1 pre-release

This is the first release candidate/pre-release version of Datacube-OWS for Open Data Cube 1.9 and includes support for both the legacy postgres index driver and the new postgis index driver and provides support for accessing multiple ODC databases from a single OWS instance.

• Compatibility with datacube-1.9 and postgres index driver (#1009)
• Cleanup (#1012)
• ODC environment configuration (#1015)
• Postgres ranges rebuild (#1017)
• Cherry picks of bug fixes and new features from 1.8 branch (#1018, #1028, #1040)
• Test database build (#1021)
• Index driver abstraction (#1020)
• Postgis driver support (#1032)
• Prepare for 1.9.0-rc1 release (#1044)

This release includes contributions from @SpacemanPaul

Datacube OWS release 1.8.42

Datacube-OWS 1.8.42

Bug fixes and major extensions for defining custom feature info includes - refer to the documentation
for details.

• Fix dockerfile casing warning (#1035)
• Add --version argument to datacube-ows CLI entry point (#1036)
• Auto-add implicit single top-level folder to ensure strict WMS standard compliance (#1036)
• Changes to materialised view definition to prevent errors on databases with WKT CRS (#1037)
• Custom Feature Info enhancements (#1039)
• Miscellaneous cleanup and backported fixes from 1.9 branch (#1042)
• Update default version number and HISTORY.rst for release (#1043)

This release includes contributions from @SpacemanPaul and @pjonsson.

Datacube-OWS release 1.8.41

What's Changed

New Feature! Multi-date handler aggregator functions for colour-ramp type styles can now receive
either the results of the index function, or the raw band data by setting a config option. (Previously
they always received the results of the index function.)

• Improved error messages when creating extents without materialised views (#1016)
• Several minor bug-fixes and improved error handling in WCS code (#1027)
• Automated updates (#1022)
• Allow multi-date handler aggregator functions to receive raw data (#1033)
• Update HISTORY.rst and increment default version for release (#1034)

This release includes contributions from @SpacemanPaul and @whatnick

Full Changelog: 1.8.40...1.8.41

Datacube-OWS release 1.8.40

Bug Fix release.

• Loading now uses skip_broken_datasets=True by default. (#1001)
• Bump base osgeo/gdal docker image version. (#1003)
• Update versions of several upstream packages to avoid known security issues (#1004, #1005, #1008)
• pre-commit autoupdate (#1006)
• Make S3 URL rewriting work with metadata indexed from STAC (#1011)
• Update HISTORY.rst and increment default version for release and some tests. (#1013)

This release includes contributions from @whatnick, @pjonsson, @robbibt , @SpacemanPaul, and various automatic updater bots.

Full Changelog: 1.8.39...1.8.40

Datacube OWS 1.8.39

Emergency release to complete partially implemented new feature in 1.8.38.

The changes to the spatial materialised view introduced in the previous release are now also implemented in the time materialised view.

Please run datacube-ows-update --schema --role <ows_db_username> again with this new release to access the new behaviour. Note that this command should be run as a dba/superuser, while <ows_db_username> should be the read-only user that OWS normally uses - it is used to grant that user read access to the new views. No system downtime is required for this schema upgrade.

• Automatic CI action update (#998)
• Fix materialised view definition to handle all eo3 compatible metadata types (#999)
• Update HISTORY.rst and increment default version for release (#999)

Datacube OWS 1.8.38

Previously the spatial materialised view recognised metadata types by individual name and required manual tweaking
for every new metadata type. From 1.8.38, all metadata types with a name starting with eo3_ will be treated as eo3 compatible. New non-eo3 metadata types will still require special handling.

Run datacube-ows-update --schema --role <ows_db_username> to activate the new view definition. No database downtime is required for the upgrade.

Also includes miscellaneous bug fixes and maintenance.

• Upgrade pypi publish github action from unsupported version (#994)
• Tweak FeatureInfo JSON documents to be compliant geojson (#995)
• Tweak materialised view definition to handle all eo3 compatible metadata types (#996)
• Fix Dimension sections of WMTS Capabilities documents to comply with standard (#996)
• Update HISTORY.rst and increment default version for release (#997)

Datacube OWS 1.8.37

Maintenance release. Includes security updates and a couple of bug fixes around timezone handling.

Full list of changes:

• Fixes to timezone handling (#958, #982)
• Various Github CI improvements (#959, #972, #974)
• Automatic dependency updates (#966, #970, #971, #975, #976, #977, #980, #981, #984, #986, #988, #991, #992)
• Update dependencies to address upstream security issues (#973)
• Label Prometheus metrics by endpoint, not path (#978)
• Update base docker image and remove docker efficiency analysis GHA (#990)
• Update HISTORY.rst and increment default version for release (#993)

Includes contributions from @benjimin, @pjonsson, @SpacemanPaul and @dependabot.

Datacube OWS 1.8.36

Another maintenance release.

Addresses several issues with the Docker build in 1.8.35, and makes the behaviour of the PYDEV_DEBUG environment variable less surprising.

Note that flask-log-request-id has been removed as a dependency as it is now orphaned and no longer supports the latest versions of Flask. This means that the X-REQUEST-ID header is no longer being set in OWS http responses. If you were using this feature in your OWS deployment, please raise an Issue on Github and we will look into alternative methods of setting the header in the next release.

Datacube OWS 1.8.35

Maintenance release.

• Changes to dependencies. Removed unnecessary version pins and replace a deprecated upstream library (#983, #942, #948, #949)
• Update HISTORY.rst and default version number (#950)

Includes contributions from #pindge, @emmaai and @SpacemanPaul

OWS Release 1.8.34

This OWS release includes two significant new features:

1. Timeless mosaic layers.

   The new mosaic_date_func configuration option allows the creation of timeless (i.e. single time-slice)
   layers. The user-provided function returns the start/end date to use for dataset date searches, and a
   mosaic layer is generated, with the most recent available date for any given pixel returned.

2. Enhanced time resolution handling (subday and overlapping summary periods).

   Major refactor of time resolution handling.

   There are now only 3 time resolution options:

   solar: Replaces the old "raw" option. Local-solar-day time model, used for imagery captured in daily swathes
   by satellites in heliosynchronous orbits.

   summary: Replaces the old "day", "month", and "year" options. Only looks at the "start" datetime, and
   so neatly supports products with overlapping or non-exclusive dataset date-ranges. Expects the time portion
   of the start date to always be "midnight UTC". Used for summary/calculated products.

   subday: New option. Used for for products with multiple time values per day (e.g. hourly/minutely data). Uses
   the "start" datetime of the dataset.

   Note that the solar and summary options explicitly ignore the time component of the time query parameter
   passed by the user. If you need the time component to be significant, you must use subday.

   The old "raw", "day", "month", "year" time_resolution options are still supported as aliases for the new
   values above. A deprecation warning will be issued advising you to update your configuration to the new
   values, but the old values will continue to work. You should not actually move your configuration to
   the new values until after all of your deployment environments have been upgraded v1.8.34.

Full list of changes:

• Increment default version number and update version history (#937)
• Enhanced time resolution handling (subday and overlapping summary periods) (#933, #936)
• Add spellcheck to RST documentation (#929, #930)
• Implement timeless mosaic layers (#928)
• Refactor integration tests to use new collection DEA data (#927)
• Bump datacube-core version (#923, #927, #933)
• Miscellaneous cleanup and code-maintenance (#922)
• Pre-commit auto-updates (#920, #926, #932, #934)