Report infer ty errors during hir ty lowering #142420
Conversation
rustbot
added
S-waiting-on-review
| impl ForeignItemKind<'_> { | ||
| pub fn descr(&self) -> &'static str { | ||
| match self { | ||
| ForeignItemKind::Fn(..) => "function", | ||
| ForeignItemKind::Static(..) => "static variable", | ||
| ForeignItemKind::Type => "type", | ||
| } | ||
| } | ||
| } |
There was a problem hiding this comment.
we could probably just use tcx.def_descr, but it causes a lot of diffs across the test suite because it calls associated functions associated functions or methods respectively
| //~^ ERROR: the placeholder `_` is not allowed within types on item signatures for functions | ||
| //~^ ERROR: the placeholder `_` is not allowed within types on item signatures for opaque types | ||
| //~| ERROR: the placeholder `_` is not allowed within types on item signatures for opaque types | ||
| //~| ERROR: the placeholder `_` is not allowed within types on item signatures for opaque types | ||
| //~| ERROR: the placeholder `_` is not allowed within types on item signatures for opaque types |
There was a problem hiding this comment.
This one is a bit dumb, but I could special case Synthetic things separately handle opaque types to also check if they are part of an afit
oli-obk
force-pushed
the
infer-ty-coalescing
branch
from
17b63c8 to
993b819
Compare
| impl Drop for ItemCtxt<'_> { | ||
| fn drop(&mut self) { | ||
| _ = self.check_tainted_by_errors(); | ||
| } | ||
| } |
There was a problem hiding this comment.
Only type_of invokes check_tainted_by_errors as it's the only ItemCtxt using query that actually has a way to report errors
oli-obk
force-pushed
the
infer-ty-coalescing
branch
from
993b819 to
b4492d2
Compare
This comment has been minimized.
This comment has been minimized.
|
☔ The latest upstream changes (presumably #142438) made this pull request unmergeable. Please resolve the merge conflicts. |
oli-obk
force-pushed
the
infer-ty-coalescing
branch
from
b4492d2 to
e7c1ba8
Compare
|
This PR changes a file inside |
|
based on #142449 as the crashes tests were "fixed" by this PR only in as much as that they stopped ICEing, but the error message was referring to |
This centralizes the placeholder type error reporting in one location, but it also exposes the granularity at which we convert things from hir to ty more. E.g. previously infer types in where bounds were errored together with the function signature, but now they are independent.
oli-obk
force-pushed
the
infer-ty-coalescing
branch
from
e7c1ba8 to
fb8cfca
Compare
This centralizes the placeholder type error reporting in one location, but it also exposes the granularity at which we convert things from hir to ty more. E.g. previously infer types in where bounds were errored together with the function signature, but now they are independent.
r? @compiler-errors