[VOTE] Releasing Orekit 13.1 from release candidate 2

Orekit development
1

This is a vote in order to release version 13.1 of the Orekit library.
Version 13.1 is a minor release.

Orekit version 13.1. The main changes are:

  • Improve performance and accuracy of getJd.
  • Take into account mass in dependsOnPositionOnly.
  • Fixed regression on TLE date parsing.
  • Fixed failing tests on Windows.
  • Add model parameters Jacobian update from dynamics switches.
  • Fixed use of wrong class loader.
  • Allow resetting wind-up.
  • Improved shiftedBy in (Field)StateCovariance.
  • Override getPosition in (Field)ShiftingPVCoordinatesProvider.
  • Remove some calls to deprecated methods.
  • Revamped release process, using fully automated scripts and continuous integration.
  • Fix equals in FieldAbsoluteDate.
  • Add durationFrom with TimeStamped in FieldTimeStamped.
  • Add Jacobian matrix to Lambert solver.
  • Move Lambert solver in dedicated package and wrap it for IOD.
  • Add Lambert differential correction.
  • Add STM update from dynamics switches.
  • Add way not to reset (Field)PropagationStepRecorder.
  • Add way to clear matrices harvester in propagator.
  • Fix problem in TriggerDate without mass in STM.
  • Add method to remove AdditionalDataProvider.
  • Made FieldTimeSpanMap consistent with TimeSpanMap.
  • Added automatic expunge to TimeSpanMap.
  • Add dependsOnTimeOnly in (Field)EventDetector and use it in propagation for performance.
  • Add access to mass derivative in ForceModel.
  • Enable TriggerDate with 7x7 STM.
  • Add (Field)BoundedPVCoordinatesProvider.
  • Add linear Keplerian covariance step handler.
  • Add thrust model based on spherical coordinates.
  • Add FieldTimeStampedGeodeticPoint.
  • Add ability to include mass in numerical state transition matrix.
  • Add getter for (Field)EnablingPredicate.
  • Speed up non-Keplerian acceleration in (Field)Orbit.
  • Add STM to Gradient state conversion.
  • Add Gradient state utils.
  • Add reset state saved in (Field)EventsLogger.
  • Restored JB2006 model.
  • Add FieldGroundAtNightDetector.
  • Add TimeIntervalsManeuverTrigger.
  • Add TimeInterval interface and some applications.
  • Change argument order of TimeStampedDouble and TimeStampedDoubleAndDerivative for consistency with the rest of Orekit.
  • Add TimeStampedGeodeticPoint.
  • Add (Field)ExtremumAngularSeparationDetector.
  • Added getVelocity method.
  • Added Herrick-Gibbs initial orbit determination.
  • Fixed OrbitRelativeFrame definition of ambiguous local orbital frames (CCSDS).
  • Fixed NaN with Gradient in Keplerian motion with circular orbit computed in Cartesian coordinates.

The release candidate 2 can be found on the GitLab repository
as tag 13.1-RC2 in the release-13.1 branch:

The maven artifacts are available in the Orekit Nexus repository at:

The generated site is available at:
https://www.orekit.org/site-orekit-13.1/index.html

The vote will be tallied on 2025-08-01T15:49:26Z (UTC time)

1 Like
2

It seems the generated site is not available.
I don’t think it is a deal breaker, so I let the vote continue.

1 Like
3

All good for me, +1. A solid minor version.

Thanks Luc for all the release work.

Cheers,
Romain.

4

Thank you Luc for your work on the script & the release.

Regarding the release, I checked:

  • Checkstyle: no issues
  • SpotBugs: one non-blocking issue
    • Bug Category Details Line Priority Redundant nullcheck of stream, which is known to be non-null in org.orekit.utils.OrekitConfiguration.getOrekitVersion() STYLE RCN_REDUNDANT_NULLCHECK_OF_NONNULL_VALUE 71 Medium
  • Tests: 0 failure
  • Warnings during compilation (jdk 8): total of 6 warnings (unchecked cast + unchecked generic array creation for varargs parameter). Non blocking for the release
  • Warnings during javadoc generation: 1 false alarm, it mentions an error when fetching URL: Overview (Hipparchus 4.0.1 API) but the link works fine ?

Hence, +1 for the release !

Cheers,
Vincent

5

Looks good! +1 from me.

6

Yes, I know about this one. For now, I kept it for the sake of defensive programming. The stream is non-null as long as source tree remains consistent (i.e. as long as the orekit.properties file is found). We could remove the check if we consider removing this properties file would be a programming error.

It seems I cannot find a way to have both checkstyle, the IDE, and SonarQube happy about all of these. It seems sometimes I can remove the warnings with some @supresswarning, but sometimes another tool is not happy because it considers the annotation is unneeded, this is an endless loop.

This one is new to me, I did not notice it. Perhaps some proxy blocking effect that does not appear when following the link with a browser?

1 Like
7

Yes, it’s probably because i’m connected to 2 separate vpn so there must be something preventing the plugin from contacting the website :thinking:.

Also thank you for updating the release guide.

8

+1 from me as well.

9

+1 from me too

10

+1 for the release

11

Impressive list of improvements for a minor version!
Are you sure that the API is not broken versus 13.0, for the following modification: “Move Lambert solver in dedicated package and wrap it for IOD” ?

1 Like
12

Yes, it should be compatible.
It corresponds to merge request 812.
The existing API is still available, and public methods have been declared deprecated (which we allow for non-major releases).

13

Thanks Luc !
+1 for me

14

The site has been generated correctly, but the link to the directory has not been created. This link will be created when the Orekit website will be updated to include the new version.

+1 for the release.

SĂ©bastien

15

+1 for the release.
Thanks for the work.

16

+1 from me too.

17

My own +1 vote

18

This vote passes, with +1 from Romain, Vincent, Mark, Hank, Maxime, Bryan, SĂ©bastien H. SĂ©batien D., Christophe, Yannick, Anne-Olivia, Paul (on the PMC category), and myself.

I will proceed with the release.

4 Likes
19

I tried to update the site, but it seems the CI/CD does not pick up anymore the latest version.
Looking at the download page in the staging site leads to broken links.

I guess the changes I made to the CI/CD had a nasty side effect…

Could someone more fluent than me with this stuff have a look?

1 Like
20

I’m back. I’ll take a look at what’s wrong.

1 Like