Bart Janssens
Bart Janssens
The directory at https://root.cern.ch/download/cling/ seems to be empty. I am specifically looking for Ubuntu 16.04 binaries for use with our jupyterhub Docker image. The 14.04 binaries from the Github releases...
The following currently doesn't work: ```julia cxx""" template struct LikeEigen { T val = 5.0; }; """ const LikeEigen{T,R} = cxxt"LikeEigen" ``` This example seems to work with the changes...
This allows defining a type using `cxxt` when the C++ template resolves to a `TemplateSpecializationType`, which seems to happen when there are default arguments.
In `CxxWrap.jl` there is a test to measure function call overhead in a tight loop. I added a `Cxx.jl` version for comparison: https://github.com/barche/CxxWrap.jl/blob/master/test/functions.jl#L99-L106 The test loops over a large array,...
The objective of this PR is to get some better cooperation between the Julia parallel processing system and MPI. * `WindowIO` allows point-to-point communication between MPI processes using a Julia...
This basically skips an stdlib in `deps_graph` if it is no longer an stdlib in the current Julia version (and therefore doesn't exist). I'm not entirely sure I know what...
A CI run on x86 reveals that x86 linux broke down again at some point: https://github.com/JuliaInterop/CxxWrap.jl/runs/4725312279?check_suite_focus=true#step:5:93 Relevant log extract: ``` Running tests from basic_types.jl... signal (11): Segmentation fault in expression...
Test errors here: https://github.com/JuliaInterop/CxxWrap.jl/runs/1724658562?check_suite_focus=true#step:5:82 Hypothesis: this is because we use `jlbacktrace`, and it is not re-exported in the new libjulia. Still need to verify this locally on Windows.
Leaving this here so I don't forget it, in the Types test: ``` julia> CppTypes.greet_cref(C_NULL) ERROR: MethodError: greet_cref(::Ptr{Nothing}) is ambiguous. Candidates: greet_cref(arg1::Union{Ptr{Nothing}, ConstCxxPtr{#s16} where #s16
Allow using VS2015 and newer instead of hard-coding for VS2015 only.