beginning of nb
Quality Gate passed
Issues
11 New issues
0 Accepted issues
Measures
0 Security Hotspots
0.0% Coverage on New Code
0.0% Duplication on New Code
✔️ c88b8df56b581aaba48c43f374f32c473c23aeca -> Azure artifacts URL
Logfiles from GitLab pipeline #220196 (:white_check_mark:) have been uploaded here!
Status and direct links:
- :white_check_mark: mac_m1_cmake_build: [cmake, ON, OFF, OFF, address]
- :white_check_mark: spack_setup
- :white_check_mark: build:nmodl
- :white_check_mark: build:neuron:nmodl:intel:legacy
- :white_check_mark: build:neuron:nmodl:intel:shared
- :white_check_mark: build:neuron:nmodl:nvhpc:acc:legacy
- :white_check_mark: build:neuron:nmodl:nvhpc:acc:shared
- :white_check_mark: build:neuron:nmodl:nvhpc:omp:legacy
- :white_check_mark: build:neuron:nmodl:nvhpc:omp
- :white_check_mark: test:neuron:nmodl:intel:legacy
- :white_check_mark: test:neuron:nmodl:intel:shared
- :white_check_mark: test:neuron:nmodl:nvhpc:acc:legacy
- :white_check_mark: test:neuron:nmodl:nvhpc:acc:shared
- :white_check_mark: test:neuron:nmodl:nvhpc:omp:legacy
- :white_check_mark: test:neuron:nmodl:nvhpc:omp
IMO the big advantage of nanobind is when we can get rid of the manual ref-counting, or simplify the code significantly thanks to the new API. If the simplification is not that significant it might be even better to leave as is, due to the risk of regressions.
IMO the big advantage with nanobind would be, for static types (like Section and Segment) we can use full class bindings (similar to pytbind11 bindings). For the dynamic Hoc object, we can hopefully do significant simplifications without a full rewrite using the object api