3 ms·

>The code will compile just fine. The Zig compiler will generate a new number for each unique 'error.*' This is wild. I assume there's at least the tooling to

>The code will compile just fine. The Zig compiler will generate a new number for each unique 'error.*' This is wild. I assume there's at least the tooling to catch this kind of errors right?

TUSF 1y ago
There's no special tooling to catch this, because nobody catches an error with if-else—it's simply not idiomatic. Everyone uses switch statements in the catch block, and I've never seen anybody using anything other than switch, when catching an error.
- veber-alex 1y ago
  But why? If I just need to check for 1 specific error and do something why do I need a switch? In Rust you have both "match" (like switch) and "if let" which just pattern matches one variant but both are properly checked by the compiler to have only valid values.
- arwalk 1y ago
  The real problem is not about the if-else, its that he's comparing to the global error set, and not to the FileError error set he created specifically to define AccessDenied.
  - love2read 1y ago
    The real problem is that this compiles without error
- love2read 1y ago
  “There’s no specific tooling to catch this, because nobody[…]”. So? This is a standard library/language feature, which is usually the first place people go for features in the language. To say that nobody uses it is stupid.
- dminik 1y ago
  Both the Zig standard library as well as several third party projects do check errors like this. I already commented on Zig compiler/stdlib code itself, but here's Tigerbeetle and Bun, the two biggest(?) Zig codebases: https://github.com/search?q=repo%3Atigerbeetle%2Ftigerbeetle%20%22%3D%3D%20error.%22&type=code https://github.com/search?q=repo%3Atigerbeetle%2Ftigerbeetle... https://github.com/search?q=repo%3Aoven-sh%2Fbun%20%22%3D%3D%20error.%22&type=code https://github.com/search?q=repo%3Aoven-sh%2Fbun%20%22%3D%3D...
  - dminik 1y ago
    Ok, while it's cool that the TigerBeetle link now shows no matches (down from two) the comment now feels wrong. Anyways, you guys left in the `!= error.` checks, so here's some snapshots that hopefully won't be invalidated :P https://github.com/tigerbeetle/tigerbeetle/blob/b173fdc8270016e8c29da06640c726523e8acb90/src/vsr/multi_batch.zig#L490 https://github.com/tigerbeetle/tigerbeetle/blob/b173fdc82700... https://github.com/tigerbeetle/tigerbeetle/blob/b173fdc8270016e8c29da06640c726523e8acb90/src/state_machine_fuzz.zig#L104 https://github.com/tigerbeetle/tigerbeetle/blob/b173fdc82700... (different file, same check.)
    - matklad 1y ago
      Good catch of != thanks!
arwalk 1y ago
The example is a bit dubious. Sure, it compiles just fine, because the author is not using errors properly in zig. Here, he uses the global error set with `error. AccessDenid`, and as stated, it compiles just fine because when you reach the global error set, it's integers all the way down. If the author had written `FileError.AccessDenid`, this would not have compiled, as it would be comparing with the `FileError` error set. The global error set is pretty much never used, except when you want to allow a user to provide his own errors, so you allow the method to return `anyerror`.
- veber-alex 1y ago
  Why does the compiler decay the FileError and the global error set to an integer? If they were unique types, the if statement would not have compiled.
  - arwalk 1y ago
    `FileError.AccessDenied` is a unique value in a unique error set. `error.AccessDenid` has not been defined anywhere and hence is just given an integer value at some point by the compiler. As I stated before, this error wouldn't even exist in the first place in no codebase ever: look how the method that fails returns a `FileError` and not an `anyerror` It could be rightly argued that it still shouldn't compile though.
- empath75 1y ago
  All examples of this type come down to "the user made a mistake", but that is kind of the the entire point. It is not possible to make that mistake in rust.
  - arwalk 1y ago
    I'm not saying that zig has the same level of safety than rust, i'm just saying that grabbing a knife by the blade is not an argument for using a spoon. The error presented in this example would not be written by any zig developer. Heck, before this example i didn't even knew that you could compare directly to the global error set, and i maintain a small library. zig and rust do not have the same scope. I honestly do not think they should be compared. Zig is better compared to C, and rust is better compared to C++.
matklad 1y ago
This is compiler bug, the code should not compile (see issue #25046)

>The code will compile just fine. The Zig compiler will generate a new number for each unique 'error.*' This is wild. I assume there's at least the tooling to catch this kind of errors right?

TUSF 1y ago

There's no special tooling to catch this, because nobody catches an error with if-else—it's simply not idiomatic. Everyone uses switch statements in the catch block, and I've never seen anybody using anything other than switch, when catching an error.

veber-alex 1y ago
But why? If I just need to check for 1 specific error and do something why do I need a switch? In Rust you have both "match" (like switch) and "if let" which just pattern matches one variant but both are properly checked by the compiler to have only valid values.
arwalk 1y ago
The real problem is not about the if-else, its that he's comparing to the global error set, and not to the FileError error set he created specifically to define AccessDenied.
- love2read 1y ago
  The real problem is that this compiles without error
love2read 1y ago
“There’s no specific tooling to catch this, because nobody[…]”. So? This is a standard library/language feature, which is usually the first place people go for features in the language. To say that nobody uses it is stupid.
dminik 1y ago
Both the Zig standard library as well as several third party projects do check errors like this. I already commented on Zig compiler/stdlib code itself, but here's Tigerbeetle and Bun, the two biggest(?) Zig codebases: https://github.com/search?q=repo%3Atigerbeetle%2Ftigerbeetle%20%22%3D%3D%20error.%22&type=code https://github.com/search?q=repo%3Atigerbeetle%2Ftigerbeetle... https://github.com/search?q=repo%3Aoven-sh%2Fbun%20%22%3D%3D%20error.%22&type=code https://github.com/search?q=repo%3Aoven-sh%2Fbun%20%22%3D%3D...
- dminik 1y ago
  Ok, while it's cool that the TigerBeetle link now shows no matches (down from two) the comment now feels wrong. Anyways, you guys left in the `!= error.` checks, so here's some snapshots that hopefully won't be invalidated :P https://github.com/tigerbeetle/tigerbeetle/blob/b173fdc8270016e8c29da06640c726523e8acb90/src/vsr/multi_batch.zig#L490 https://github.com/tigerbeetle/tigerbeetle/blob/b173fdc82700... https://github.com/tigerbeetle/tigerbeetle/blob/b173fdc8270016e8c29da06640c726523e8acb90/src/state_machine_fuzz.zig#L104 https://github.com/tigerbeetle/tigerbeetle/blob/b173fdc82700... (different file, same check.)
  - matklad 1y ago
    Good catch of != thanks!

arwalk 1y ago

The example is a bit dubious. Sure, it compiles just fine, because the author is not using errors properly in zig. Here, he uses the global error set with `error. AccessDenid`, and as stated, it compiles just fine because when you reach the global error set, it's integers all the way down. If the author had written `FileError.AccessDenid`, this would not have compiled, as it would be comparing with the `FileError` error set. The global error set is pretty much never used, except when you want to allow a user to provide his own errors, so you allow the method to return `anyerror`.

veber-alex 1y ago
Why does the compiler decay the FileError and the global error set to an integer? If they were unique types, the if statement would not have compiled.
- arwalk 1y ago
  `FileError.AccessDenied` is a unique value in a unique error set. `error.AccessDenid` has not been defined anywhere and hence is just given an integer value at some point by the compiler. As I stated before, this error wouldn't even exist in the first place in no codebase ever: look how the method that fails returns a `FileError` and not an `anyerror` It could be rightly argued that it still shouldn't compile though.
empath75 1y ago
All examples of this type come down to "the user made a mistake", but that is kind of the the entire point. It is not possible to make that mistake in rust.
- arwalk 1y ago
  I'm not saying that zig has the same level of safety than rust, i'm just saying that grabbing a knife by the blade is not an argument for using a spoon. The error presented in this example would not be written by any zig developer. Heck, before this example i didn't even knew that you could compare directly to the global error set, and i maintain a small library. zig and rust do not have the same scope. I honestly do not think they should be compared. Zig is better compared to C, and rust is better compared to C++.

matklad 1y ago

This is compiler bug, the code should not compile (see issue #25046)