have a lot of zig code that's still in 0.5.0-land, are you sure you're ready for 0.6.0? have you done your research and found out how energy intensitve it is to replace every warn() call? have i got the tool for you!
zig made a large (only if you're an std.debug.warn/print-debugging addict like me) breaking change that removed varargs and replaced them by tuples throughout all of std.fmt some time ago between 0.5 and 0.6, and i got too lazy to fix every instance of warn(), so i made a tool for it
// old (0.5.0)
warn("test {} {} {}", a, b, c);
// new (0.6.0)
warn("test {} {} {}", .{a, b, c});it might be ok for 3 warns in a bit of code but then you look at me writing a toy compiler and you have as many warns as there are instructions in x86
selling points:
- python 3
- very bad code, unmaintainable code, to the public domain, for everyone
- this won't get all warn calls. manual tinkering required
- generated code needs a zig fmt pass
python3 ./tupleinator.py path/to/unenlightened/code.zig > enlightened_code.zig
# profit
mv enlightened_code.zig path/to/unenlightened/code.zig