Accessing Ray Spatial Information

It can be helpful to know where an extracted ray is located in relation to its parent dataset. Thankfully, due to it’s reliance on Trident for ray generation, rays written to disk by SALSA contain information about their generation. This information includes the start and end coordinates of the ray as well as (unique to SALSA) the impact parameter used to generate the rays. [1]

To access this information, we must first load a ray with yt Thanks to Trident and SALSA, this dataset will have a special attribute called light_ray_solution. This object is a list of dictionaries containing information about how the ray was generated. [2]

import yt

ds = yt.load("my_rays/ray0.h5")

print(ds.light_ray_solution[0])

{'end': unyt_array([0.52023164, 0.60173583, 0.55577773], 'unitary'),
 'filename': 'DD0044',
 'impact_parameter': unyt_quantity(0.00922918, 'code_length'),
 'redshift': 0.0,
 'start': unyt_array([0.53940663, 0.46652058, 0.48771738], 'unitary'),
 'traversal_box_fraction': unyt_quantity(0.15258786, 'unitary'),
 'unique_identifier': '1382182007'}

We can see that each key in the light_ray_solution dictionary has units attached. These units (such as unitary and code_length) may reference the coordinate system of the simulation from which the light ray was extracted.

[1]

Trident automatically writes start and end coordinates to its rays. SALSA will add the impact parameter to any ray generated with construct_rays(), which includes the two main ray generation mechanisms, generate_lrays() and generate_catalog()

[2]

The light_ray_solution object is a list because Trident supports rays with multiple segments called compound rays. Each segment can have it’s own dictionary in light_ray_solution. SALSA does not use these ray objects (instead only using “simple” rays) so light_ray_solution will only ever contain one dictionary.