Unshadow device re-exports #54
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
In the `device` module, all public items from `cl3::device` are re-exported.
They even [show up as public in the documentation]. Nonetheless they aren't
actually public.
To reproduce this issue, try to import `opencl3::device::CL_UUID_SIZE_KHR`,
you will get an error like:
error[E0603]: constant `CL_UUID_SIZE_KHR` is private
--> src/device.rs:2209:30
|
2209 | opencl3::device::CL_UUID_SIZE_KHR,
| ^^^^^^^^^^^^^^^^ private constant
|
note: the constant `CL_UUID_SIZE_KHR` is defined here
--> /home/vmx/src/pl/filecoin/upstream/opencl3/src/device.rs:45:23
|
45 | CL_LUID_SIZE_KHR, CL_UUID_SIZE_KHR,
| ^^^^^^^^^^^^^^^^
The private imports at
https://github.com/kenba/opencl3/blob/d004f1cf5f6ad78d158c1f67bb96ce1fed0190dc/src/device.rs#L18
shadow the previous public re-exports
https://github.com/kenba/opencl3/blob/d004f1cf5f6ad78d158c1f67bb96ce1fed0190dc/src/device.rs#L15
As most of the private imports are a subset of the public re-exports, we can
just import the missing ones, which in this case is only
`cl_device_feature_capabilities_intel`.
In order verify the subset, you can copy the list of [imports from `cl3::ext`]
and the [`cl3::device` re-exports from `opencl-sys`] into two fiels and run
` tr ',' '\n|awk '{$1=$1;print}'|sort` on each of them and finally diff the
results.
[show up as public in the documentation]: https://docs.rs/opencl3/0.9.0/opencl3/device/constant.CL_UUID_SIZE_KHR.html
[imports from `cl3::ext`]: https://github.com/kenba/opencl3/blob/d004f1cf5f6ad78d158c1f67bb96ce1fed0190dc/src/device.rs#L19-L45
[`cl3::device` re-exports from `opencl-sys`]: https://github.com/kenba/cl3/blob/9c97521ab27dd709e2ec9482855023ef0c5d4838/src/device.rs#L21-L106
Normally I wouldn't use glob imports, but it made sense to me to import all the definitions from the |
|
@vmx this change has now been published in version 0.9.1 of the crate. |
|
Thanks for the quick turn-around time, it fixes the issues I was seeing! |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
In the
devicemodule, all public items fromcl3::deviceare re-exported. They even show up as public in the documentation. Nonetheless they aren't actually public.To reproduce this issue, try to import
opencl3::device::CL_UUID_SIZE_KHR, you will get an error like:The private imports at
opencl3/src/device.rs
Line 18 in d004f1c
opencl3/src/device.rs
Line 15 in d004f1c
As most of the private imports are a subset of the public re-exports, we can just import the missing ones, which in this case is only
cl_device_feature_capabilities_intel.In order verify the subset, you can copy the list of imports from
cl3::extand thecl3::devicere-exports fromopencl-sysinto two fiels and runtr ',' '\n|awk '{$1=$1;print}'|sorton each of them and finally diff the results.I've never been a fan of glob imports, but I wasn't aware of that limitation in Rust. rust-lang/rust#31337 and https://rust-lang.github.io/rfcs/1560-name-resolution.html touches those topics in case someone is interested in all the details.
This is a regression, in earlier versions of
opencl3opencl3::device::CL_UUID_SIZE_KHR.used to be public.