comparison src/Grids/mapped_grid.jl @ 1752:c98d9c528a22 feature/grids/curvilinear

Review: Added review comments to Grids.jl and mapped_grid.jl
author Vidar Stiernström <vidar.stiernstrom@gmail.com>
date Fri, 13 Sep 2024 10:27:17 -0700
parents e4353d5e8fc3
children 672897f64d58
comparison
equal deleted inserted replaced
1749:e4353d5e8fc3 1752:c98d9c528a22
36 same_jacobian = jacobian(a) == jacobian(b) 36 same_jacobian = jacobian(a) == jacobian(b)
37 37
38 return same_logicalgrid && same_coordinates && same_jacobian 38 return same_logicalgrid && same_coordinates && same_jacobian
39 end 39 end
40 40
41 # Review: rename function logicalgrid to logical_grid
42 # for consistency with mapped_grid.
41 """ 43 """
42 logicalgrid(g::MappedGrid) 44 logicalgrid(g::MappedGrid)
43 45
44 The logical grid of `g`. 46 The logical grid of `g`.
45 """ 47 """
132 end 134 end
133 # TBD: Should this be changed to calculate sqrt(g) instead? 135 # TBD: Should this be changed to calculate sqrt(g) instead?
134 # This would make it well defined also for n-dim grids embedded in higher dimensions. 136 # This would make it well defined also for n-dim grids embedded in higher dimensions.
135 # TBD: Is there a better name? metric_determinant? 137 # TBD: Is there a better name? metric_determinant?
136 # TBD: Is the best option to delete it? 138 # TBD: Is the best option to delete it?
139 # Review: I don't think we should delete it. Users building their own
140 # curvilinear operators will need the functionality. Also the
141 # determinant of the jacobian (and not its square root) is required
142 # for quadratures on mapped grids right? For that reason I think we should
143 # keep the function as is. We could provide a function for the square root
144 # as well if we think it would be helpfull. Regarding naming, perhaps
145 # metric_determinant is better?
137 146
138 """ 147 """
139 metric_tensor(g::MappedGrid) 148 metric_tensor(g::MappedGrid)
140 149
141 The metric tensor of `g` as a grid function. 150 The metric tensor of `g` as a grid function.