Made some changes to complex.h and assert.h docs and a few improved styles #3
Conversation
|
|
||
| For the following C code, which will immediately abort the program if the condition is false. | ||
|
|
||
| ```c |
There was a problem hiding this comment.
I would add an #include <assert.h> here, just used to seeing it in manpages, same thing for importing std in the zig code example. A bit redundant but it allows for people to copy paste code
There was a problem hiding this comment.
True. Will do that.
Also I will add the following scripts for testing Zig & C code so that any future code reference will be always correct.
https://github.com/Sobeston/ziglearn/blob/master/do_tests.zig
https://github.com/Sobeston/ziglearn/blob/master/test-out.zig
Will also write a similar zig "script" for C code too in future too.
| assert(x >= 0.0); // undefined behaviour | ||
| ``` | ||
|
|
||
| This will invoke an undefined behaviour when the condition is `false`. Checkout [std.debug.assert](https://ziglang.org/documentation/master/std/#std;debug.assert) |
There was a problem hiding this comment.
This is a great spot to explain undefined behavior in zig and how it's caught in debug and release-safe, but optimized out of release-fast and release-small, there's a good explanation here https://github.com/ziglang/zig/blob/6fc822a9481c0d2c8b37f26f41be603dd77ab5a4/lib/std/debug.zig#L222
it's so simple that we might want to include it's implementation too.
I would also move the static assertion code up here and add a comptime assert example. (static_assert is also requires C11 iirc so that's important too)
There was a problem hiding this comment.
got it. unreachable to be introduced and more on static_assert & comptime assert
|
|
||
| ## [Zig Complex Type][21] | ||
|
|
||
| In Zig complex numbers are represented as `Complex` type. |
There was a problem hiding this comment.
should note what the underlying number type is used in Complex
There was a problem hiding this comment.
Complex types take in any number type (f64, f32) i guess as a generic. Didn't really check the source code. But sure, I will add it.
| const z2 = Complex { re: 0, im: 2.0 }; | ||
|
|
||
| test "complex multiplication" { | ||
| const zSquared: Complex = complex.mul(z1, z2); |
There was a problem hiding this comment.
we shouldn't need to declare the zSquared type here
There was a problem hiding this comment.
oh. type inference. cool. will remove this!
| ```zig | ||
| const assert = std.debug.assert; | ||
| // ... | ||
| const x: double = -1.0; |
There was a problem hiding this comment.
there's no double in zig, instead it would be f64 (same thing for int used elsewhere)
There was a problem hiding this comment.
oh. sorry about that. i will fix that.
| std.testing.expectApproxEqAbs(x1, x2, 0.0000001); // floating point number comparison with tolerance | ||
|
|
||
| const x3: int = 1; | ||
| const x4: int = 2; |
| !> [cexpf][3] & [cexpl][3] are two variants in C that take `float` and `long double` respectively in C. In Zig those variants | ||
| are handled quite well by Zig generics. | ||
|
|
||
| <!-- C reference links --> |
There was a problem hiding this comment.
love all the reference links
|
Thanks for the review. I will change the stuff mentioned above. Make a PR again tomorrow. |
|
@Himujjal how's it coming? |
|
Hey. Totally got out of my mind, this project. I will put up the changes sure. |
Hey, can you check out this and comment on if the doc format (examples over function signatures) style is good.
https://zig-community.github.io/libc-to-zig/#/headers/complex_h