nvim-dap + netcoredbg: Unit Test Breakpoints not getting hit (sometimes)
Im currently using
Here's the weird thing, I have the exact same identical nvim configuration (version controlled from github) on two devices.
The first is my work laptop, running Ubuntu on WSL.
The second is my personal machine, running normal Debian.
The WSL Ubuntu machine does hit breakpoints inside of unit tests when I
The Debian machine, however, does not.
Both are running the same SDK (8.0203), latest netcoredbg (3.1.0-1), and same nvim config.
However I have set a bunch of different env vars over time on both machines, and I know there are random sporadic ones that matter for dotnet, so its distinctly possible I have one floating around messing with the debian machine, or, I have a "magic sauce" env var set on the WSL machine that is making stuff work.
I have tried this with both nunit and xunit, neither have worked.
The extra weird part is even though the breakpoints dont get hit on the Debian machine, it still does attach successfully. Dotnet waits and only proceeds with the tests after I attach, and the debugger disconnects and goes back to normal edit mode when the tests finish running.
So
Breakpoints do get hit if I run normal "run the process with netcoredbg" via just running a plain ole net core app.
However this doesnt of course work with unit tests since its a totally different command and would require a bunch of bespoke configuration if I wanna do stuff like pass in
Especially since I know it clearly can work on one machine, I just cant figure out what secret sauce I am missing >_>;
nvim-dap running netcoredbg, in order to debug net core apps.Here's the weird thing, I have the exact same identical nvim configuration (version controlled from github) on two devices.
The first is my work laptop, running Ubuntu on WSL.
The second is my personal machine, running normal Debian.
The WSL Ubuntu machine does hit breakpoints inside of unit tests when I
dotnet test with VSTEST_RUNNER_DEBUG=1 and I use "Attach to Process"The Debian machine, however, does not.
Both are running the same SDK (8.0203), latest netcoredbg (3.1.0-1), and same nvim config.
However I have set a bunch of different env vars over time on both machines, and I know there are random sporadic ones that matter for dotnet, so its distinctly possible I have one floating around messing with the debian machine, or, I have a "magic sauce" env var set on the WSL machine that is making stuff work.
I have tried this with both nunit and xunit, neither have worked.
The extra weird part is even though the breakpoints dont get hit on the Debian machine, it still does attach successfully. Dotnet waits and only proceeds with the tests after I attach, and the debugger disconnects and goes back to normal edit mode when the tests finish running.
So
dotnet test is detecting it got attached to... It's just specifically not wiring into the breakpoints?Breakpoints do get hit if I run normal "run the process with netcoredbg" via just running a plain ole net core app.
However this doesnt of course work with unit tests since its a totally different command and would require a bunch of bespoke configuration if I wanna do stuff like pass in
--filter or --logger or etc to dotnet test so Im not super interested in going down that alley...Especially since I know it clearly can work on one machine, I just cant figure out what secret sauce I am missing >_>;
