windows: basic support for GetUserProfileDirectoryW #3502
Conversation
| )?; | ||
| // The Windows docs just say that this is written on failure. But std | ||
| // seems to rely on it always being written. | ||
| this.write_scalar(Scalar::from_u32(len.try_into().unwrap()), &size)?; |
There was a problem hiding this comment.
@ChrisDenton something seems to be odd here, or I am misunderstanding something. The docs say about the size pointer
If the buffer specified by lpProfileDir is not large enough or lpProfileDir is NULL, the function fails and this parameter receives the necessary buffer size, including the terminating null character.
So in particular, if the buffer is large enough, it doesn't say that the pointer is changed at all.
But the way std calls this function, it assumes that on success, sz is updated to the actual size (including the null terminator).
There was a problem hiding this comment.
You're right. We shouldn't be doing that if the behaviour isn't documented.
Given that's how it's worked in practice and Rust has apparently been relying on this for a long time, I do think updating the API docs is warranted. However, unless or until that actually happens we should act as if such a documentation change would be rejected.
There was a problem hiding this comment.
Thanks for confirming! I will then land this PR (if I stopped updating the size the tests would fail). I opened an issue to track this: rust-lang/rust#124325.
RalfJung
force-pushed
the
GetUserProfileDirectoryW
branch
3 times, most recently
from
5cd7832 to
817ee4b
Compare
|
☔ The latest upstream changes (presumably #3504) made this pull request unmergeable. Please resolve the merge conflicts. |
…et's require that in general
…a non-default function
RalfJung
force-pushed
the
GetUserProfileDirectoryW
branch
from
817ee4b to
10e8bb8
Compare
|
@bors r+ |
|
☀️ Test successful - checks-actions |
Fixes #3499