-
Notifications
You must be signed in to change notification settings - Fork 56
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Refactor error type to use thiserror
#117
Comments
Hey @ShadowJonathan, Which version of heed are you using? I wrote the version 0.11 to try using real error types when casting bytes by using the bytemuck library instead of the zerocopy one. This had great benefits as the However, I miss something important: The Using the latest version of LMDB makes it work correctly on windows. I also applied a "patch" to the LMDB version we are using, I introduced a new If you want to enjoy all of this stuff you can use the tag v0.12.1 or newer. If you want to continue using the 0.11 version with that returns real parsing errors maybe we can release a v0.13 version with the patch I described above ( |
I'm using @timokoesters's fork, which is based on Edit: going ahead with the |
I think that the PR that @timokoesters has done is to be merged on the new v0.12 that isn't published on crates.io for now.
If you want you could also stop using the new 0.11 and use v0.12.1 version instead, only this version has the LMDB latest version that works better for Windows. |
Two questions;
|
Hey @ShadowJonathan,
I just sent you an email with my Discord handle 😃
In fact, the We are currently waiting for some internal subject to be cleared to publish this |
Alright, I still got some questions about Back to the topic at hand; I think I'll probably make a PR which closes this, and makes error handling a bit less tedious (and more ergonomic rust) with |
Currently, when trying to convert or use heed for
?
-error purposes, i get the following;The error type is a composite enum, which is fine, but it argues it cannot be sent between threads safely because the trait isn't marked, or valid.
I could also ask to wrangle the autotrait in, but i think the error type will also benefit from a bit more readability and maintainance if it's implemented with
thiserror
, for example like hereThe text was updated successfully, but these errors were encountered: