Make zig-ecs compatible with current zig version (0.11.0-dev) about zig-ecs HOT 8 CLOSED

Ok, after reading the Language changes in release v0.10.0 I was able to fix a couple of compilation errors, see commit 73d190f of my fork.

However, after running example "simple" I realized, there's still a lot of work to do. For me, it currently prints the following (obviously changed print format string to satisfy zig v0.10.0 compiler):

entity: 1, pos: (x = 10, y = 10), vel: (x = 15, y = 17)
---- resetting iter

I think the output should be something like this:

entity: 1, pos: Position{ .x = 1.0e+01, .y = 1.0e+01 }, vel: Velocity{ .x = 1.5e+01, .y = 1.7e+01 }
entity: 0, pos: Position{ .x = 0.0e+00, .y = 0.0e+00 }, vel: Velocity{ .x = 5.0e+00, .y = 7.0e+00 }
---- resetting iter
entity: 1, pos: Position{ .x = 2.5e+01, .y = 2.7e+01 }, vel: Velocity{ .x = 1.5e+01, .y = 1.7e+01 }
entity: 0, pos: Position{ .x = 5.0e+00, .y = 7.0e+00 }, vel: Velocity{ .x = 5.0e+00, .y = 7.0e+00 }

from zig-ecs.

Hi! I haven't actually updated to v0.10 for many of my libs. There was quite a storm of changes with functions all needing to be * const and a slew of other changes. I was holding out until perhaps zig fmt could do some of the fixing and I also feared that there were still a lot more breaking changes coming (local functions are on their way amongst other things).

That being said, it sounds like you are pretty close. If you do want to send in a PR I'll for sure pull it in. Looking at your fork nothing specifically stands out. It seems you got the * const for all the fns and the unused vars. I'll give it a read after work fully and see if something pops out.

from zig-ecs.

Hey @prime31,

Hi! I haven't actually updated to v0.10 for many of my libs. There was quite a storm of changes with functions all needing to be * const and a slew of other changes. I was holding out until perhaps zig fmt could do some of the fixing and I also feared that there were still a lot more breaking changes coming (local functions are on their way amongst other things).

Yes you are right, there have been a ton of changes. The release notes for v0.10.0 are lengthy 😅

That being said, it sounds like you are pretty close. If you do want to send in a PR I'll for sure pull it in. Looking at your fork nothing specifically stands out. It seems you got the * const for all the fns and the unused vars. I'll give it a read after work fully and see if something pops out.

To be honest, I don't think I'm that close yet. There is still a lot of stuff broken (like Iterators). Also there are one or two compilation errors, that cause building the tests to fail.
Unfortunately, mit zig knowledge isn't really worth mentioning, so it could take some time to figure out what needs to be done 😅

Best regards
Stefan

from zig-ecs.

Hey @prime31,

I managed to fix a couple of errors. Also the Iterator implementation in MultiView should work now.
All tests seem to run successfully. And at least the simple example now shows expected output:

entity: 1, pos: (x = 10, y = 10), vel: (x = 15, y = 17)
entity: 0, pos: (x = 0, y = 0), vel: (x = 5, y = 7)
---- resetting iter
entity: 1, pos: (x = 25, y = 27), vel: (x = 15, y = 17)
entity: 0, pos: (x = 5, y = 7), vel: (x = 5, y = 7)

If you could review my latest commits, I'd be more than happy :)

EDIT: Latest fixes also work for zig compiler v0.10.1. I think upgrading to zig compiler v0.11.0-dev is only a matter of updating the build.zig.

Best regards
Stefan

from zig-ecs.

Ok, I upgraded to zig v0.11.0-dev.1580+a5b34a61a and fixed some compilation errors in this commit.
So it is now compatible with zig v0.11.0.

Best regards
Stefan

from zig-ecs.

That all looks good to me!

from zig-ecs.

Great! I'll create a PR for this ;)

from zig-ecs.

Hey @prime31, thanks for the merge. This issue resolved now :)

from zig-ecs.