Gilles, > I was not indicating that the name "EuclideanTransform" would be > better than "AffineTransform", I was wondering about whether the > class itself is redundant.
Oh, I misunderstood. The "EuclideanTransform" interface is important because it adds the "applyVector(Vector)" method, which has different behavior than the standard "apply" method. All transforms in the euclidean packages have this method but it is not present in the core Transform because not all spaces have associated Vector types (eg, spherical). I had renamed it AffineTransform in the previous PR not because it exposed new functionality or behavior that made it affine, but because it was located in a module defining an affine space. What do you suggest for the name here? > My understanding is that "Transform" can be documented as: > ---CUT--- > In Euclidean space, this must be an affine transform. > ---CUT--- That's part of what the documentation now says. -Matt ________________________________ From: Gilles Sadowski <gillese...@gmail.com> Sent: Monday, January 20, 2020 2:28 PM To: Commons Developers List <dev@commons.apache.org> Subject: Re: [geometry] Rename Transform to AffineTransform Hello. Le lun. 20 janv. 2020 à 16:57, Matt Juntunen <matt.juntu...@hotmail.com> a écrit : > > Gilles, > > > From a design POV, I still think that "AffineTransform" is redundant: > > The "AffineTransform" name change has been reverted. It is now the original > "EuclideanTransform". I was not indicating that the name "EuclideanTransform" would be better than "AffineTransform", I was wondering about whether the class itself is redundant. > I've closed PR #58 and created PR #59 with the latest commits squashed. I've not looked yet. But answering below, to hopefully clarify the misunderstanding. > > IIUC, the required (not just "desired") properties should stand out. > > And, for the mathematically-inclined, the relationship to affine > > transforms would illustrate it (for Euclidean spaces). > > I'm not sure what you're saying here. My understanding is that "Transform" can be documented as: ---CUT--- In Euclidean space, this must be an affine transform. ---CUT--- Gilles > The current documentation is the most complete and mathematically accurate. > > -Matt >>> [...] --------------------------------------------------------------------- To unsubscribe, e-mail: dev-unsubscr...@commons.apache.org For additional commands, e-mail: dev-h...@commons.apache.org
