v3.5.0

Updates with v3.5.0 --

Band and physical filter: in better alignment with the remainder of the project, and to better support use of rubin_scheduler at the summit with "physical filtername" values for the camera, we have moved to naming things more explicitly "band" in all places that should be "band" (aka ugrizy) rather than "filter" or "filtername" (which in much of the remainder of Rubin refers to the physical filtername, such as g_01 or r_03).
At v3.5.0, you will see warning messages about deprecation of "filter" and "filtername" in favor of "band" and "bandname" in Surveys, BasisFunctions and Detailers -- the result will be remapped from 'filter*' into 'band*' but will be deprecated in v4.0. Features only use 'band', but we do not expect users to be modifying configurations this deep into the FBS API.

The upside of this split from "filter" to "band", with the addition of the BandToFilterDetailer is that configuration files for the FBS should always use band -- and then, for use at the summit, the BandToFilterDetailer can be set so that output observations specify "filter" appropriately for the physical filter. The output observations from the FBS will contain both "band" and "filter" -- "filter" can be used to represent the physical filter.

ObservationArrays
The ObservationArray class which replaced the list of observations used prior to v3.0 now adds a concatenate function, as well as a tolist function, for easier use with detailers or when using the observation array in other contexts.
Detailers and Surveys now return multi-row ObservationArrays instead of list of single-element ObservationArrays.

Target_id
In an effort to improve links between requested observations and acquired observations, the ObservationArray now comes with a value for the target id as well as observations are requested. This should make the target_id values in lsst.sal.Scheduler.logevent_target non-zero.

DeltaCoordDitherDetailer
A new dither detailer is available, the DeltaCoordDitherDetailer. This detailer allows you to specify an arbitrary dither pattern of offsets - a single input observation is then dithered to each of the offsets in the "delta_ra" and "delta_dec" arrays provided. This is intended to enable easier use of arbitrary dithers with ad-hoc FBS configurations for commissioning, which could now specify a specific desired dither pattern solely at the ts_fbs_utils or ts_config_ocs levels.

Minor updates
The Footprint class now takes a list of filters, instead of requiring a dictionary of filters with values of arbitrary integers.
Sim_runner respects the end of the simulation time, even if the queue does not fill (bugfix).
The prenight simulation scripts in sim_archive now run using shared directory installations of rubin_scheduler.

What's Changed

• add tolist method and obsarray_concat function by @yoachim in #140
• Tickets/sp 1781 by @yoachim in #142
• Tickets/sp 1786: Adding target_id by @yoachim in #143
• swap Footprint object to take list of filters rather than dict by @yoachim in #144
• Tickets/sp 1797 by @yoachim in #145
• tickets/SP-1773: use shared directories by @ehneilsen in #141
• Tickets/sp 1793 - use "band" instead of "filter" by @yoachim in #146
• have sim_runner stop if skipping ahead in time due to no valid observ… by @yoachim in #148
• SP-1841: Backwards compatible updates for Band/Filter by @rhiannonlynne in #149
• Correct pairs to minimize filter changes by @yoachim in #150
• refactor ddf_presched to make config easier. eliminate magic numbers by @yoachim in #151
• fix FutureWarning and DeprecationWarning by @yoachim in #153
• arbitrary dither pattern detailer by @yoachim in #147
• refactor to eliminate duplicate kwargs min_area and area_required by @yoachim in #154
• SP-1865: Add filter backwards-compatibility to models by @rhiannonlynne in #152
• SP-1865: Additional changes for use with ts_scheduler v2.3 by @rhiannonlynne in #156

Full Changelog: v3.4.0...v3.5.0

v3.4.0

v3.4.0 adds the StartFieldSequenceDetailer detailer, which allows the insertion of a set of observations before the execution of following observations (say, adding an AOS visit before a sequence for a field survey).

In addition, the default start date for the survey is updated from May 1, 2025 to November 1, 2025, to better align simulations with the expected date of the survey.

A bugfix update added the sun's azimuth value to the Conditions object.

What's Changed

• update default start date to mjd=60949 (Oct 1, 2025) by @yoachim in #130
• missing sun azimuth by @yoachim in #137
• Tickets/sp 1756 by @yoachim in #138

Full Changelog: v3.3.1...v3.4.0

v3.3.1

v3.3.1
Bugfix for ComCamGridDitherDetailer.

The update also includes fixes to unit tests to run faster, primarily by reducing redundancy in more time-consuming tests. This update also fixed some bugs in not commonly used code where radians and degrees were not set appropriately, and expanded unit test coverage to run basic tests on all basis functions.

Some additional basis functions were identified as to-be-deprecated and flagged as such.

What's Changed

• Tickets/sp 1739-speed up unit tests by @yoachim in #132
• SP-1751: Patch ComCamGridDitherDetailer by @bechtol in #136

Full Changelog: v3.3.0...v3.3.1

v3.3.0

v3.3.0 adds the ComCamGridDitherDetailer for commissioning observations with comcam in a 2x2 grid.

The opsim output column "scripted_id" is also removed, and the content that was in that column will be placed inside the "scheduler_note" column instead, preserving links between scripted survey visits and their metadata - within the scheduler_note metadata.

What's Changed

• remove scripted_id by @yoachim in #131
• SP-1745: Add ComCamGridDitherDetailer to scheduler config by @bechtol in #135

New Contributors

• @bechtol made their first contribution in #135

Full Changelog: v3.2.1...v3.3.0

v3.2.1

v3.2.1 fixes a bug in the FieldAltAzSurvey to properly set the requested sequence of visits.
It also fixes a bug in the AltAz2RaDecDetailer calculation of the expected RA/Dec value.

The creation of snapshots of the scheduler for simulation testing and debugging purposes from sim_runner is also now supported.

What's Changed

• tickets/SP-1721: add option to save snapshots by @ehneilsen in #133
• FieldAltAzSurvey pass init kwarg properly by @yoachim in #134

Full Changelog: v3.2.0...v3.2.1

v3.2.0

v3.2.0 adds a rotation converter appropriate for use with comcam.
Set "telescope='comcam'" when using a rotation converter to pick up the comcam version.

What's Changed

• OPSIM-1209: Add rotation converter for comcam by @rhiannonlynne in #129

Full Changelog: v3.1.0...v3.2.0

v3.1.0

v3.1.0

• addition of the FieldAltAzSurvey
• update for ToO surveys to split long exposures which may saturate and add target_name
• bugfix for CameraRotSmallDitherDetailer
• support for reading multiple nights from the consdb for conversion to opsim visit format or ObservationArrays
• deprecation warning added to SolarElongMaskBasisFunction (a duplicate basis function)

What's Changed

• tickets/SP-1667: support reading multiple nights by @ehneilsen in #120
• add target_name to observations competed by ToO surveys by @yoachim in #122
• tickets/SP-1685 by @ehneilsen in #121
• Dep warning added to SolarElongMaskBasisFunction by @yoachim in #123
• Make sure ToO surveys won't saturate if nsnaps is set to 1 by @yoachim in #124
• SP-1722: add FieldAltAzSurvey by @yoachim in #125
• tickets/SP-1723: impove support for data from consdb by @ehneilsen in #126
• Fix bug in CameraRotSmallDitherDetailer by @rhiannonlynne in #127
• SP-1737: Fix tab formatting error by @rhiannonlynne in #128

Full Changelog: v3.0.0...v3.1.0

v3.0.0

Major update 3.0.0

V3.0.0 introduces some breaking changes from v2.0.2 -- primarily in deprecation of some keyword arguments and basis functions, where before backwards-compatible shims had been left in place.

The tel_alt_limit and tel_az_limit values in the Conditions object should come from the telescope kinematic model limits. sky_alt_limit and sky_az_limit lists now are the arbitrary masked zones that can come from the Conditions object.

The previously deprecated empty_observation() was removed, and ObservationArray should be used instead.
Previously deprecated kwargs for FieldSurvey have been removed; sequences should be defined via dictionaries of nvisits, exptimes, and nexps for each of the filters in sequence (or a list of ObservationArrays may be passed).
Previously deprecated survey_start_mjd() and set_default_nside() were removed; use SURVEY_START_MJD and DEFAULT_NSIDE instead.

BaseSurveyFeature was made an abstract class, to enforce definition of both add_observation and add_observation_array to subclasses.

The padding method for the AltAzShadowMaskBasisFunction was updated, so that the mask is grown appropriately along the edges of the defined region.

The example_scheduler will now configure and run (while successfully creating reward dataframes) at any nside.

There were also updates in the code used to read the ConsDb and generate and manage pre-night simulations in the archive module.

What's Changed

• DM-45888: Remove mambaforge for miniforge by @roceb in #111
• OPSIM-1196: more cleanup for summit use by @rhiannonlynne in #110
• remove core sched rotator checking. add whole queue kwarg to request_… by @yoachim in #112
• Tickets/opsim 1197 by @yoachim in #113
• Tickets/sp 1619 by @yoachim in #114
• fix healpy mask growth to work beyond small angle approx by @yoachim in #115
• Tickets/sp 1629 by @yoachim in #116
• docstring updating by @yoachim in #118
• tickets/SP-1623: refactor metadata storage in the prenight scheduler simulation archive to improve performance by @ehneilsen in #117
• tickets/SP-1675: accommodate new shape of seq_num in the consdb by @ehneilsen in #119

New Contributors

• @roceb made their first contribution in #111

Full Changelog: v2.0.2...v3.0.0

v2.0.2

v2.0.2 includes a small update to the ModelObservatory, to account for downtimes which could last less than a full night. Previous behavior was to skip the remainder of the night, if a downtime was encountered; now the ModelObservatory will just skip to the end of the downtime.

This behavior was not relevant before v2.0.0, as older downtime (Unscheduled and Scheduled downtimes) only included full nights; with v2.0.0, the additional unscheduled downtime in UnscheduledDowntimeMoreY1Data within the first year includes downtimes that are as short as 1 hour.

What's Changed

• OPSIM-1194: let unscheduled downtimes be less than a day by @yoachim in #109

Full Changelog: v2.0.1...v2.0.2

v2.0.1

v2.0.1 features small bug fixes:

• SURVEY_START_MJD replaces survey_start_mjd() .. the function survey_start_mjd is still available, but will be deprecated.

• DEFAULT_NSIDE replaces set_default_nside() .. the function is still available but will be deprecated.

• The make_rolling_footprint function works more consistently with current expectations, if called with defaults.

• The conditions object will print its values more consistently.

• The two-shutter-motion warning was made appropriately less sensitive.

• The documentation pages were brought up to date with the new schema and archive pages added.

What's Changed

• Tickets/opsim 1191 by @yoachim in #105
• OPSIM-1192: make_rolling_footprint work with default values by @yoachim in #106
• tiickets/OPSIM-1193: add SURVEY_START_MJD and DEFAULT_NSIDE to replace functions by @yoachim in #108
• Bring documentation up to date by @rhiannonlynne in #107

Full Changelog: v2.0.0...v2.0.1