Releases

Change Log

All notable changes to this project will be documented in this file.

The format is based on Keep a Changelog, and this project adheres to Semantic Versioning.

1.2.1 - 2025-07-04

Added

  • Added a disclaimer during comparison processing to inform users that, if the operation is taking too long, they can improve performance by disabling uncertainty computation.

Changed

  • Optimised observation file loading on the comparison page, achieving 1.25x to 1.5x faster performance.

1.2.0 - 2025-06-20

Added

  • Improved export buttons placement: Export buttons have been repositioned to enhance user experience by reducing screen clutter and allowing more space for the output graph and data display.

    • Export Graph and Export CSV are now available directly on the plot toolbar.

    • Export netCDF button has been removed, but the functionality remains accessible via “File → Save as a netCDF file” in the context menu.

  • Responsive Layout for Selenographic Input: When the window is wide enough, the Selenographic input fields are automatically rearranged in 3 columns × 2 rows instead of the previous fixed 2 rows × 3 columns. This improves space efficiency, leaving more room for the output graph and data display.

  • Introduced Angle of Linear Polarisation (AoLP) feature set.

  • Coefficient versions available by default now include 20250608_V01 version.

Changed

  • Polarisation (DoLP and AoLP) graphs and csv exports contain a disclaimer stating that the calculated values are preliminary and subject to changes.

Fixed

  • Closing the LOAD FILE dialog without selecting a new file, after having previously loaded one, cause the GUI to incorrectly show the previous file as still loaded. The file now does remain loaded correctly in such cases.

  • CLI for simulations from geographic coordinates was wrongly interpreting the height parameter

1.1.0 - 2025-04-21

Added

  • GUI more user friendly: Simulation input takes less window space.

  • The ESA satellites available for selection now include METOP first generation: METOP-A, METOP-B and METOP-C. (NFR109)

  • Allow adding satellites through 3LE/TLE or OSF files through the GUI. (To fully comply with NFR109)

  • Allow to visualize the comparison for all spectral channels simultaneously. (FR203) This is done through two new comparison options:

    • By wavelength (mean): Show the temporal average of all the channel/wavelengths in a plot, having the wavelengths on the x-axis and the irradiance and differences on the y-axis. CLI name: CHANNEL_MEAN

    • By wavelength (boxplot): Show the statistics of all the channel/wavelengths in a plot, showing one boxplot per channel. The boxplot are shown having the wavelengths on the x-axis and the irradiance and differences (the box plot) on the y-axis. CLI name: CHANNEL

  • Improve speed of switching between comparison graphs. (NFR208)

  • Geometry information (coordinates, angles, etc.) and timestamps (if available) are now included in all NetCDF and CSV output files. (NFR305)

  • The TBX accepts coefficients that also include data for the 1088 CIMEL photometer’s 2130 nanometer band. (NFR108-A)

  • Automated Build & Packaging (NFR409):

    • Linux: Fully automated build through Docker.

    • Windows: Partially automated build through Docker for Windows. All automated except first step: compiling EO-CFI C Code, which isn’t mandatory for each build.

  • Timeseries input file now also accepts regular format timestamps, not only CSV.

  • Added METOP-A, METOP-B and METOP-C data (TLE/3LE). METOP-B and METOP-C will need to be periodically updated.

  • Updated TLE/3LE data for PLEIADES 1A, PLEIADES 1B and PROBA-V, now covering the period from their launch until 2025-04-14.

  • Updated EO-CFI library dependencies to version 4.28.

Changed

  • Drop support for Python 3.8.

  • Provide support for Python 3.11 and 3.12.

  • Updated library dependencies and versions.

  • Refactored main Python package grouping subpackages in layer architecture based packages.

Fixed

  • Removed forced conversion of error correlation matrices to float32. ASD error correlation matrices in ds_ASD.nc are float64, now they are kept as float64, which ensures they are positive-definite as the conversion to float32 was slightly modifying the values making them not be positive-definite. This eliminates the overhead of computing the closest positive-definite matrix, achieving a 2.5× speedup in uncertainty calculations compared to v1.0.3. (NFR306)

  • Optimized the ASD wavelengths error correlation matrix by converting it from float64 to float32 while preserving its positive semi-definite (PSD) property. This enhancement achieves a mean relative difference of less than 0.001%, ensuring numerical stability. Additionally, this change provides a 2.5× speedup, contributing to an overall 6.25× improvement in uncertainty calculations compared to v1.0.3. (NFR306)

  • Updated LIME reflectance equation to align with the LIME ATBD. Observer’s selenographic longitude and latitude were previously swapped in the implementation. Corresponding coefficients were also swapped, resulting in correct outputs despite the mismatch. Both the equation and the coefficients are now fully consistent with the ATBD.

    • Replaced old coefficients file 20231120_v01 with corrected version 20231120_v02. Outputs of the current toolbox version using the new coefficients remain unchanged compared to the previous toolbox version (1.0.3) using the old coefficients.

  • Fixed minor bugs

1.0.3 - 2024-01-25

Initial version that serves as the baseline for tracking changes in the change log.