Level of support for hyperbolic trig functions

[d01010101]

Why the following does not work? It does, though, if sin is used instead of sinh.

remap-shape (shape x y z)
    (/ x (sinh z))
    (/ y (sinh z))
    z)

In function sinh:
wrong-type-arg: Wrong type argument in position 1

What is the level of support of sinh? #348 shows "branch has conflicts that must be resolved". Or are there some math difficulties that prevent sinh from being used in remap-shape?

[mkeeter]

They aren't supported at all! You can find the list of operations here.

sinh is trying to call a Guile built-in function (rather than a libfive wrapper), which won't work in the context of remap-shape; that's why you're getting a wrong-type-arg error.

I don't remember the current state of #348 ; it was blocked for a while waiting for Eigen to release a new version, but it may be possible to rebase that branch now.

[d01010101]

There are some functions which I find useful for asymptotes.

#348 says "mkeeter force-pushed the master branch" but the link does not seem to show anything related to hyperbolic functions.

[mkeeter]

The code is here: https://github.com/libfive/libfive/pull/348/files

You'll probably want to rebase this branch onto the master branch to bring it up to date; I can't offer detailed git support, but the internet is full of tutorials.

[d01010101]

I have corrected the patch against the current master branch. However, as it turned out, it is not so trivial to configure a second ssh key for the same site so I do not know how to create a pull request or whatever.

I attach a diff -r -c2 for the moment. If it is not enough tell me.

libfive-hyperbolic.diff.gz

[d01010101]

Any chance it gets incorporated? It is a second hyperbolic trig patch to libfive slowly becoming obsolete... I am sorry for the format, but given the triviality of the patch, maybe it is enough?