Open pat-s opened 1 month ago
Thanks for reporting. You compile on an ARM CPU? Anyways, I can't thoroughly investigate right now. As a quick fix I would recommend assigning 255 or 0xFF to "substChar" to make the build pass. I will take a closer look on Wednesday and see if our code should be modified for machines where char is unsigned by default.
You compile on an ARM CPU?
Yes, sorry for missing that initially.
As a quick fix I would recommend assigning 255 or 0xFF to "substChar" to make the build pass.
I am building binaries from CRAN sources, so making local modifications is not an option. Not urgent, but would be great to resolve this error in the build chain :)
Thank you for confirming the platform. I will look into getting the reported compilation failure resolved with the next GDX release. It might take a couple days as I am afk until Wednesday but @abhosekar and I should be able to quickly deploy afterwards.
Thank you for reporting again. I will also comb the rest of GDX (and other projects I maintain) for usages of char that were assuming it to always be signed and test compile with GCC on arm, to make sure you won't get another follow up error.
I am not 100% sure why this didn't show up before on the arm64 macOS builds we do with AppleClang. Something else to look into. 😀
I am not 100% sure why this didn't show up before on the arm64 macOS builds we do with AppleClang. Something else to look into. 😀
arm64 macOS uses clang
instead of gcc
and both operate differently in some ways, the reported cased being one of them.
Thanks for fixing it!
Any idea?