On 2020.05.20 22:17, Thomas Schmitt wrote:
Pete, if you can checkout branch "win32_first_tno_fix" (*) then it would be nice to get a confirmation that i got my version of win32.c on the second try by copy+paste of your corrected code. (First commit of the branch was my faulty proposal by mistake. Hrmpf.)
Tested the branch and everything looks good to me.
After successful test, the branch would be ready for merging.
I would advise to combine the 2 commits into a single one before merging, because you don't want to willingly leave a commit that is known to break compilation, as is the case here.
This may not look like a big issue at first glance, but it actually is when you are hunting for a regression using git bisect. That's because, when using bisect, you cannot choose the individual commits that it may end up using, meaning that you could run into the breaking one (which of course defeats the whole idea of using bisect).
In other words, every single commit should leave the repo in a state where the code can be built in full, even if the commit is part of a series where the very next one fixes the breakage.
Apart from that, the branch looks fine. Regards, /Pete
