Skip to content

Releases: AstarVienna/ScopeSim

Version 0.11.1

26 Oct 09:27

Choose a tag to compare

What's Changed

API Changes

Bugs fixed

Changes to or addition of Effects

Dependency Changes

Documentation Improvements

  • Fix some docstring formatting, add version info by @teutoburg in #828

Other Changes

Full Changelog: v0.11.0...v0.11.1

Version 0.11.0

21 Sep 15:57

Choose a tag to compare

Large release in preparation of the ScopeSim Workshop happening next week. See below for important changes and deprecations.

Highlights

  • Finally add support for Python 3.13 and Numpy 2.
  • Partially resolve circular dependency between ScopeSim and ScopeSim_Templates
  • Several new effects in preparation for MOSAIC simulations (#796).
  • New and improved plotting methods on some classes for easier and nicer visualisation of results.

What's Changed

Deprecations

Download functionality

  • Deprecate downloading IRDB packages directly from GitHub by @teutoburg in #727

Installing IRDB packages directly from GitHub using download_packages() is deprecated. We're not aware that this functionality saw any significant use by the community. If you do need to install unreleased IRDB packages from the GitHub repository (and you know why you need that!), the same result can be achieved by cloning the repository (or your own fork) and then use sim.link_irdb(<path_to_clone>) to point ScopeSim there. That solution has the added benefit that you can keep thing in sync through git alone, which is more reliable anyway. The relevant functionality will be removed in version 0.12.0, at which point attempting to use download_packages() with a GitHub ref will result in an error.

Passing a list as the first argument to download_example_data() has been deprecated since version 0.8.4 and will now result in a TypeError being raised. This is to catch any remaining uses of the old call signature of this function. From version 0.12.0 onwards, arguments will be silently forwarded to the file retriever, meaning if a list is still passed at that point, it will result in an error further downstream.

Source Fields

  • Deprecate direct attribute access in SourceField by @teutoburg in #733

After the SourceField refactor in version 0.9.0, it was still possible to access any item of the underlying data structur (Table or ImageHDU) directly through the SourceField via a standard dict-like syntax. This was retained for backwards compatibility, because some code still relied on this functionality. It does however have the potential to leade to some hard-to-track bugs and is generally an unclean solution, which is why we have been working behind the scenes to slowly remove all internal code that still uses this feature. Now because this is related to how the Source object might be put together in user code, we're using a more carefuly approach in deprecating this, to avoid breaking existing code for no urgent reason. We're thus activating a PendingDeprecationWarning now, which should give users an advanced warning that we will probably fully deprecate this in the future. If you do encounter this warning, please try to use the .field attribute to explicitly access the underlying data structur before using dict-like lookup on it.

API Changes

Bugs fixed

  • Fix creation of cube source from file name by @teutoburg in #559
  • Fix WCU radiometry by @oczoske in #754
  • Use RegularGridInterpolator for proper extrapolation of small PSF images by @oczoske in #798

Changes to or addition of Effects

  • Functionality for MOSAIC minimum viable product by @oczoske in #796

New Features or Improvements

Dependency Changes

Documentation Improvements

Other Changes

Full Changelog: v0.10.0...v0.11.0

Version 0.10.0

27 Jun 16:41

Choose a tag to compare

Large release before the summer, with many improvements and also breaking changes, please make sure to also update your IRDB packages accordingly!

Highlights

  • Many new features, including the "simplified IFU cube readout" mode.
  • More physical treatment of the "electrons to ADU" process, including different gain values (#604).
  • Extensive refactoring and many many bug fixes.
  • More consistency across different classes with similar attributes.
  • Improved treatment of flux units (still WIP, see #662).
  • Many improvements to visual outputs (especially in notebooks), a few new progress bars in long-running loops.

What's Changed

API Changes

Bugs fixed

Changes to or addition of Effects

New Features or Improvements

Dependency Changes

Documentation Improvements

Other Changes

Full Changelog: v0.9.3...v0.10.0

Version 0.9.3

25 Apr 13:20

Choose a tag to compare

Addition of average exposure effect by @oczoske in #603, otherwise mostly bugfixes, small quality-of-life improvements and some internal refactoring.

What's Changed

Bugs fixed

  • Fix broken flux scaling for extended sources in spectroscopy by @teutoburg in #638

Changes to or addition of Effects

New Features or Improvements

Dependency Changes

Other Changes

  • Remove base class check in Effect.apply_to() by @teutoburg in #620
  • Use .to_value(<unit>) instead of .to(<unit>).value by @teutoburg in #643

Full Changelog: v0.9.2...v0.9.3

Version 0.9.2

13 Mar 15:06

Choose a tag to compare

Mainly the inclusion of METIS WCU effects in #494 plus a few bugfixes and quality-of-life improvements.

What's Changed

Bugs fixed

Changes to or addition of Effects

Dependency Changes

Documentation Improvements

Other Changes

Full Changelog: v0.9.1...v0.9.2

Version 0.9.1

26 Nov 23:01

Choose a tag to compare

Many small but important fixes and improvements that were missing in v0.9.0.

What's Changed

Changes to or addition of Effects

New Features or Improvements

Dependency Changes

Documentation Improvements

Other Changes

Full Changelog: v0.9.0...v0.9.1

Version 0.9.0

10 Nov 21:21
6259648

Choose a tag to compare

Important

The minimum required Python version for this package is now 3.10 (see Dependency Changes).
Python 3.13 is currently not supported as it causes issues on some platforms that are not yet fully understood. We are currently working on fixing 3.13 support.

What's Changed

API Changes

New Features or Improvements

  • Replace NaNs in images, log warning by @teutoburg in #466
  • Make observe work with no source (empty field) by @oczoske in #483
  • Add a new top level to CMDS nested mapping to store current simulation run settings by @teutoburg in #493

Dependency Changes

Documentation Improvements

Other Changes

New Contributors

Full Changelog: v0.8.4...v0.9.0

Version 0.8.4

29 Aug 09:43

Choose a tag to compare

Last version to support Python 3.9

This includes many small and not-so-small fixes and improvements, which are made available here to Python 3.9, while a few more major changes will be released soon as Version 0.9.0, but without support for Python 3.9.

What's Changed

Changes to or addition of Effects

  • Test all cases of DIT & NDIT vs. exptime and AutoExposure by @teutoburg in #428
  • Avoid in-place operations because of dtype conflicts by @teutoburg in #432
  • Split PSF and Electronic effects into subpackages by @teutoburg in #434
  • Improve SVO filter handling by @teutoburg in #454
  • Added function to round the edges of square PSF kernels by @astronomyk in #459

Dependency Changes

Other Changes

Full Changelog: v0.8.3...v0.8.4

Version 0.8.3

02 Jun 21:29
4515c7b

Choose a tag to compare

Mostly a small hotfix to allow changing of exptime in AutoExposure, plus some housekeeping.

What's Changed

Dependency Changes

Other Changes

Full Changelog: v0.8.2...v0.8.3

Version 0.8.2 Unsigned Hero

14 May 21:34
fecdb7c

Choose a tag to compare

What's Changed

New Features or Improvements

  • Cap negative values below 0 before quantifying to an unsigned int. by @hugobuddel in #414

Full Changelog: v0.8.1...v0.8.2