

I currently have no time to look into it but if someone can confirm and possibley create a PR that fixes it I’ll review quickly
FD
From:
gtsam...@googlegroups.com <gtsam...@googlegroups.com> on behalf of
成旭元 <hvikt...@gmail.com>
Date: Saturday, June 4, 2022 at 4:13 AM
To: gtsam users <gtsam...@googlegroups.com>
Subject: [GTSAM] Re: Question about the possible ROT3 quaternion bug?
Here are the code above and its result, sorry for the formatting issue.


--
You received this message because you are subscribed to the Google Groups "gtsam users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to
gtsam-users...@googlegroups.com.
To view this discussion on the web visit
https://groups.google.com/d/msgid/gtsam-users/8ccdcae6-024f-4a0c-8f0f-21da7bef5687n%40googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/gtsam-users/BN7PR07MB48358FBE16A39352E337479ECFA09%40BN7PR07MB4835.namprd07.prod.outlook.com.
I think that would be a good solution.


You can put “quaternion” into a deprecated block (search for it in the code base) and at least replace all occurrences within GTSAM? PS, unlike what you do below, assign to a temporary so you don’t call 4 times ?
From:
gtsam...@googlegroups.com <gtsam...@googlegroups.com> on behalf of
成旭元 <hvikt...@gmail.com>
Date: Tuesday, June 7, 2022 at 10:44 AM
To: gtsam users <gtsam...@googlegroups.com>
Subject: Re: [GTSAM] Re: Question about the possible ROT3 quaternion bug?
Thank you very much, Brice. After testing, `toQuaternion` does return the proper quaternion components. I'm trying to create a pull request to modify the `Rot3::quaternion` API
and make it return a quaternion object instead of the Eigen Vector object. However, A major concern may be that there are many downstream codes relying on the original `quaternion` API.


To view this discussion on the web visit https://groups.google.com/d/msgid/gtsam-users/f28e7aed-59d5-4517-b1d2-232df11b6cban%40googlegroups.com.