Closed mathstuf closed 10 years ago
Hmm... I've got a hunch something isn't handling relative paths correctly somewhere along the line. Will investigate.
Your code should still compile in spite of this warning, though. Is this actually halting compilation, or just spitting messages at you?
It compiles, but Ninja is (or at least seems so; it complains about circular dependencies) hopelessly confused on subsequent runs.
Well, my hunch was completely wrong, but I did manage to track down and fix these bugs, I think. I can't reproduce those errors anymore, anyway.
Thanks! I'll try and test it out tomorrow (not sure how much time I'll have in the evening).
Looks good here. Thanks.
I wonder if I wasn't using an old build when I tried earlier:
DDeps Warning: /home/boeckb/code/other/group-games/group-abagames/mu-cade/src/src/abagames/util/ode/odeactor.d could not be found by the name "abagames/util/ode/odeactor" in any of the following directories:
Ensure your import directories are correct,and that the module name matches the filename.
/home/boeckb/code/other/group-games/group-abagames/mu-cade/src/src/abagames/util/ode/odeactor.d will be not be scanned for dependencies.
I have yet to manage to reproduce this.
Can you confirm that you are on branch d_support3
?
So I debugged this some more. I have -fdebug
on the command line, but it looks like DDeps is expecting -fdebug=VALUE
and when the parser sees that it is missing a value, it bails out before it gets to the -I
modules, but there's no error checking, so it then complains about not finding the module in the paths it didn't look at. Seems to come from Modules/Compiler/GNU-D.cmake
's setting of CMAKE_D_FLAGS_DEBUG_INIT
to -g -fdebug
.
I'm getting errors such as:
from the latest code.