-
-
Notifications
You must be signed in to change notification settings - Fork 507
Compile issues on nightly #369
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
Comments
I've the same problem (unresolved import |
This issue looks like a regression on the nightly compiler. However, it does compile correctly on my machine (MacOS) with the latest nightly: rustc 1.30.0-nightly (33b923fd4 2018-08-18) @Lecrapouille What project exactly are you compiling? Can you try cloning the nalgebra repo and compiling it alone? I have the same compiler version as you but manage to compile nalgebra just fine, so I am not sure how to reproduce the errors. |
@sebcrozet I cloned your master branch and compiled it with success :) So sorry for the confusion, what I wanted to said is I got failure with this project then I saw this issue with the same error concerning geometry::Isometry. I've just started learning Rust since yesterday so maybe I made something wrong. |
I still have not found the cause of those errors. @BaxterEaves What package depending on nalgebra did you try to compile? @Lecrapouille I can reproduce the compile errors by compiling citybounds. That looks like a regression on the rust compiler but I failed to identify it source. Here is the dependency graph of citybound generated by I tried to compile individually the I guess the best workaround for now is to just revert the compiler version like @BaxterEaves did… |
@sebcrozet just to let you know that I followed their .travis.yml file and the compilation passed :) |
@sebcrozet Here is the dependency graph for my |
@BaxterEaves in my previous post I did not thank you for your fast answer: so thank you ! This is just a noob remark, but if these dependencies are similar to C/C++ with their #includes, having your dependencies looking a graph is not a good thing to have because depending on how to traverse the graph, the order of inclusion can be different and therefore getting problem like defining struct after another (instead of before as you were expected first). Cannot you reduce your graph to a minimum spanning tree ? (for example removing the link rv --> serve, num -->num-bigint). |
Just some extra info: I tried to clone and compile master with most recent nightly. |
@Lecrapouille Rust dependencies are not similar to C/C++ includes so having a graph does not cause any issue of redefinition. In fact the Rust compiler doesn't really care about on what order things are declared on a single crate. @BaxterEaves Thank you for your inputs. The failure only when serde is enabled is troubling though. |
My crate produced these errors as well when using nightly-2018-09-02 and the serde feature. Disabling the serde feature makes it work. |
It appears the errors at the root of this problem are similar to:
which can be reproduced with: use serde;
#[repr(C)]
#[derive(Debug)]
#[cfg_attr(feature = "serde-serialize", derive(Serialize, Deserialize))]
#[serde(bound(serialize = ""))]
#[serde(bound(deserialize = ""))]
pub struct Isometry
{
} This is likely related to rust-lang/rust#53675 which is a known but not-yet-fixed regression. A workaround seems to avoid the |
Uh oh!
There was an error while loading. Please reload this page.
Updated nightly this morning and rebuilt a package that has nalgebra as a dependency and I got a lot of
not found
type compile errors:...and so on. I've left a lot out because it was super long. I reverted back to the previous toolchain that i was using,
nightly-2018-08-15-x86_64-apple-darwin
, and it worked.NOTE: I successfully compiled the code on my gitlab CI, which is using the nightly image, so it might be a problem with the apple toolchain.
EDIT: Actually, the nightly docker image hadn't been pushed yet. Now my CI is broken as well.
The text was updated successfully, but these errors were encountered: