Custom Rotations

vortrace supports arbitrary viewing angles using Tait-Bryan Euler angles (yaw, pitch, roll) for grid projections.

Projection planes

The proj parameter in Python provides shortcuts for the six Cartesian projection planes:

proj	Horizontal axis	Vertical axis	Line of sight
'xy' (default)	x	y	z
'xz'	x	z	y
'yz'	y	z	x
'yx'	y	x	z
'zx'	z	x	y
'zy'	z	y	x

Tait-Bryan angles

For viewing angles that don’t align with the coordinate axes, use the yaw, pitch, and roll parameters. These are applied as extrinsic rotations about the z, y, and x axes respectively, centered on the center point.

Python

import vortrace as vt

center = [BoxSize / 2, BoxSize / 2, BoxSize / 2]

# 30-degree yaw rotation
image = pc.grid_projection(
    extent, npix, bounds, center,
    yaw=0.5236,  # radians
)

# Combined rotation
image = pc.grid_projection(
    extent, npix, bounds, center,
    yaw=0.3, pitch=0.1, roll=0.0,
)

C++

In C++, rotations are handled at the ray-grid level. Build your own rotated start/end arrays and pass them to Projection.

#include <vortrace/vortrace.hpp>
#include <cmath>

// Rotate a point (x, y, z) by yaw angle around the z-axis
// centered on (cx, cy, cz)
void rotate_yaw(double& x, double& y, double z,
                double cx, double cy, double yaw) {
    double dx = x - cx, dy = y - cy;
    x = cx + dx * std::cos(yaw) - dy * std::sin(yaw);
    y = cy + dx * std::sin(yaw) + dy * std::cos(yaw);
}

// Build rotated ray grids and pass to Projection as usual
// (see the grid_projection tutorial for the grid setup loop)

For the full Tait-Bryan rotation implementation, see the Python source in vortrace/grid.py.

Note

Angles are in radians. The center parameter must be set when using rotations – it defines the point around which the grid is rotated.