Luc Grosheintz
Luc Grosheintz
We use: ``` file( GLOB NMODL_PYTHON_FILES RELATIVE "${NMODL_PROJECT_SOURCE_DIR}/python/nmodl/" CONFIGURE_DEPENDS "${NMODL_PROJECT_SOURCE_DIR}/python/nmodl/*.py") foreach(file IN LISTS NMODL_PYTHON_FILES) cpp_cc_build_time_copy(INPUT ${NMODL_PROJECT_SOURCE_DIR}/python/nmodl/${file} OUTPUT ${CMAKE_BINARY_DIR}/lib/nmodl/${file}) list(APPEND nmodl_python_binary_dir_files "${CMAKE_BINARY_DIR}/lib/nmodl/${file}") endforeach() ``` with ``` function(cpp_cc_build_time_copy) cmake_parse_arguments(opt "NO_TARGET" "INPUT;OUTPUT" ""...
For example the following: ``` $ cat two_blocks.mod NEURON { SUFFIX two_blocks RANGE A, B, X, Y NONSPECIFIC_CURRENT il } STATE { X Y A B } ASSIGNED { il...
The naming conventions in `solveCrout` need fixing. `y` isn't a good name for a pivot vector, `p` is more canonical, but `p` is something completely different, unless it's used through...
We're occasionally seeing: ``` Filters: "Scenario: Compare Crout solver with Eigen" Randomness seeded to: 370099071 eigen_x_ColMajor vs crout_x_ColMajor (issue) / seed = 665289716 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ testcrout is a Catch2 v3.5.3 host...
While printing code we have the following pattern: ``` void print_net_receive() { printing_net_receive = true; // ... printing_net_receive = false; } ``` The problem is that: 1. We need to...
If there's not global variables, we don't need the struct that gathers up all global variables; and don't need to print it.
Currently, one must call `set_token` in `nmodl.yy` to have information about where the token appears in the MOD file. Since, it's manual, it's rarely used, and many nodes of the...
The script for checking the `ELECTRODE_CURRENT` is a copy-paste from somewhere else, likely NONSPECIFIC_CURRENT. This is needless duplication that can be removed.
The functions that serve as "entry points" into the MOD file, need to set up some variables that we then use, e.g. ``` auto _lmr = ...; auto inst =...