v0.11.0

The large number of minor changes in this release are the result of applying ufmt to organize the imports that the top of each module, and the application of pylint to tidy up a variety of code structures.

Added

• image_records.py - Added ccu_temperature and external_temperature columns to record the AD590 sensor data.

Fixed

• get_position.py - The get_position_and_pose functions now have an explicit timeout and throw exceptions if they have difficulties with their connection.

Changed

• requirements.txt - Added version number requirements that reflect the current testing environment.

v0.10.0

Added

• image_records.py - Added various light_on_x columns to record the state of the
  lights in ImageRecord objects.

Changed

• create_raw.py - For the get_lights() function, session is now optional, and by default,
  the get_lights() function now relies on the values in the passed ImageRecord, but if
  given a session, the values in the light_records table will override what is in the
  ImageRecord.

v0.9.2

Fixed

• nss.py - The DataModeler class would incorrectly always return a weh_arr of all-NaN values. This bug has been fixed, and additional safeguards and error reporting have been placed in the class, and tests written.

v0.9.1

Fixed

• anaglyph.py - when used with scikit-image versions greater than 0.21 changed the
  argument signature for phase_cross_correlation(). Fixed call, and should now
  work more globally.

v0.9.0

Added

• get_position.py - Now has support for http basic auth and can now handle returns
  from the REST service that could be either a sequence of elements or a single
  element.
• create_raw.py - Updated to accommodate the icer_minloss and icer_byte_quota properties
  of an ImageRecord.
• image_records.py - Added properties for the new image_nickname and waypoint_id
  properties, and updated the concept of capture_id to be a combination of the
  waypoint_id and the locally unique_capture_id, instead of a combination of the
  locally unique_capture_id and the image_request_id.

Fixed

• pano_check.py - removed a redundant argument.
• create_mmgis_pano.py - outpaths for saving the pano and thumbnail were being
  incorrectly generated.

v0.8.0

Changed

• The instrument_name() function in pid.py now can deal with alias names that are within a given string, not just an exact match.
• Now that pid.instrument_name() is more robust, ImageRecord.init() can handle a wider variety of Yamcs parameter names, should they change in the future.
• create_mmgis_pano.py - create() now takes a thumbsize int or tuple of ints that will control the creation and size of an output thumbnail JPG file, with naming convention set by Yamcs/OpenMCT.
• image_records.py - Added icer_byte_quota and icer_minloss parameters to ImageRecord, as well as a variety of improved handling related to this change and in pid.py.
• pid.py - Changed handling of the compression value to match the kinds of data we'll get from telemetry, so that the letters are assigned to cover an interval of possible compression ratios, and to reflect the new default byte quota value for driving images.

v0.7.0

Changed

• The light_records.py table now just records state changes (from 'on' to 'off' or
  vice-versa).
• create_image.py's create() method now has an imgtype argument that default to TIFF
  (but can be PNG).
• junc_image_req_ldst.py got some additional columns to manage the Science Team
  evaluation of images acquired from Image Requests.
• create_pano.py's make_pano_product() function renamed to make_pano_record().
• nss_simulator.py - nodata defaults for ideal detector count maps updated to more
  realistic values for dry terrain.
• image_statistics.py - Added additional stats in the compute() function, and added
  a pprint() function to format the info nicely.

Added

• pds.Purpose now provides names and explanations for the PDS-allowable values
  for "purpose."
• yamcs_reception_time column added to the image_records.py table.
• Association table junc_image_pano created which provides a many-to-many
  connection between ImageRecords and PanoRecords and added bidirectional
  relationship entries to each table.
• pano_records table now has pan and tilt angle min/max values to indicate
  angular range of panorama coverage.
• ImageRecord objects will now extract an ImageRequest ID from the provided capture_id
  if it is larger than the 16 bit range.
• image_requests.py - "Acquired," "Not Acquired," "Not Planned," and "Not Obtainable"
  statuses added to enum. Also added asdict() method.
• ptu_records.py - Tables to record the pan and tilt of the rover's pan-tilt-unit (PTU).
• create_browse.py - For making browse products from existing image products.
• create_mmgis_pano.py - For making pano products for use in MMGIS.
• create_pano.py - updated to correctly add PanoRecord associations, can now query
  database for ImageRecords.
• create_pano_product.py - takes PanoRecords and makes a PDS Pano Product.
• get_position.py - Gets position and yaw from a REST-based service.
• create_vis_dbs.py - Now also supports spatialite databases, primarily for testing.
• create_raw.py - Added components for adding observational intent and data quality
  to the XML label.
• labelmaker - A program to help build PDS4 bundle and collection labels.
• bundle_install - A program to copy just the files related to a PDS4 bundle into a
  new location. Fundamentally allowing a "make install" for PDS4 bundles.

Removed

• ldst_verification.py because the evaluation activity reflected here was on a per-image
  basis, but it has been revised to be on a per-Image-Request basis.

Version 0.6.1

Mostly a bug-fix release.

Fixed

• validators.validate_dateimte_asutc now properly raises a ValueError if the provided tz-aware datetime has a non-UTC tz offset (before any tz-aware datetime would pass the validator).
• image_records.ImageRecord object now has pgaGain instead of ppaGain (which was surely a typo in the early upstream data.
• create_image.py now correctly imports all of the tables that have a relation to the image_records table so that SQLAlchemy can properly resolve them, and downstream users of the create_image.create() function don't need to worry about sorting that out.
• nss_simulator.py, when asked to produce an output set of maps, uses a zero nodata value rather than whatever nodata value was present in the input burial depth map.
• anaglyph.py needed some minor changes to align with upcoming changes in the scikit-image architecture.

Version 0.6.0

Changed the concept of primarily recording PDS-like "products" in the database and data structures, and changed the concept to capture "records" of various kinds (ImageRecords, LightRecords, and derived things like PanoRecords) in the database and as the primary data unit for use and interaction. And then PDS "products" will be made at a later time via a process that sources one or many records.

Added

• image_records.py (this replaces the concept of raw_products.py)
• create_image.py (this replaces most of the functionality of the previous create_raw.py)
• image_tags.py which contains image tag information.
• image_requests.py defines the table for holding image requests.
• Junction tables to connect LDST information to ImageRequests and ImageTags to ImageRecords.
• ldst.py table which contains LDST information.
• light_records.py which handles information about luminaire state.
• anaglyph.py
• pano_check.py to sort through image records and figure out what would be a good panorama set.

Changed

• header.pga_gain_dict structure now has values that are always floats, which helps to determine whether this transformation needs to be applied.
• image_stats.py is the new raw_stats.py, supporting ImageRecord objects.
• pano_products.py is now pano_records.py, supporting ImageRecord objects.
• create_pano.py now supports ImageRecords and PanoRecords.
• create_raw.py rearchitected to gather information from "records" to build XML PDS labels.
• create_vis_dbs entry point changed to vis_create_dbs to conform with other vis-related entry points.
• pid.VISID now properly sorts the uncompressed "z" state lower (better) than the lossless compressed "a" state.
• pid.VISID now has a best_compression() function to sort out the best compression state
  from an iterable containing may compression states from a single observation.

Version 0.5.0

Added

• PanoID class added to pds.pid
• pano_products and create_pano added, still very preliminary, mostly just mock-ups.
• colorforge program for managing colormaps.
• mypy is now in the development dependencies to support type checking.
• lint/mypy target added to Makefile.
• tri2gpkg now has a --remove_facets option to remove facets with a particular value.

Changed

• Explicit in documentation about developing in Python 3.8 (although earlier versions
  should still be supported).
• Many changes to improve type checking.
• Added numeric instrument aliases and checking for them.
• Added information for procesingInfo and outputMask from Yamcs.
• Upgrade to SQLAlchemy >=2.
• Moved definition of Base class up to vis.db.
• heatmaps.py will now accept value data lists or arrays with np.nan or None values
  which will be appropriate ignored in the density heatmap calculation.

Fixed

• tri2gpkg - if the provided polygons have zero area, issue an error rather than
  making a confusing GeoPackage file.