Refactor error reporting of constants #51636
Merged
Conversation
This file contains hidden or 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
|
(rust_highfive has picked a reviewer for you, use r? to override) |
rust-highfive
added
the
S-waiting-on-review
eddyb
reviewed
Jun 22, 2018
src/librustc/mir/interpret/error.rs
Outdated
| @@ -150,6 +150,9 @@ pub enum EvalErrorKind<'tcx, O> { | |||
| UnimplementedTraitSelection, | |||
| /// Abort in case type errors are reached | |||
| TypeckError, | |||
| /// Resolution can fail if we are in a too generic context | |||
| ResolutionFailed, | |||
There was a problem hiding this comment.
I'd rather call this TooGeneric or something more obvious.
eddyb
reviewed
Jun 22, 2018
src/librustc/middle/const_val.rs
Outdated
| TypeckError, | ||
| CheckMatchError, | ||
| Miri(::mir::interpret::EvalError<'tcx>, Vec<FrameInfo>), | ||
| pub data: Lrc<(::mir::interpret::EvalError<'tcx>, Vec<FrameInfo>)>, |
There was a problem hiding this comment.
Can this struct be moved to mir::interpret? Also, maybe put the Lrc around the struct, that way you can name the fields that are currently in a tuple.
eddyb
reviewed
Jun 22, 2018
src/librustc/traits/fulfill.rs
Outdated
| ProcessResult::Error( | ||
| CodeSelectionError(ConstEvalFailure(ConstEvalErr { | ||
| span: obligation.cause.span, | ||
| kind: ErrKind::CouldNotResolve.into(), | ||
| data: (err, Vec::new()).into(), |
pietroalbini
added
S-waiting-on-author
oli-obk
force-pushed
the
const_diagnostics
branch
from
8bf8763 to
c93a80c
Compare
|
Addressed all review comments |
eddyb
reviewed
Jun 25, 2018
| --> $DIR/const-len-underflow-separate-spans.rs:21:17 | ||
| | | ||
| LL | let a: [i8; LEN] = unimplemented!(); | ||
| | ^^^ referenced constant has errors |
There was a problem hiding this comment.
This seems weird, since the constant is not in an Operand::Const, but rather in a type.
There was a problem hiding this comment.
This note means to say that LEN has errors, which is a constant reference right where the span points.
There was a problem hiding this comment.
This is an "error: this expression will panic at runtime" inside non-runtime code though?
There was a problem hiding this comment.
Oh... Looks like const prop is running on array lengths?! I wonder how that happened.
oli-obk
force-pushed
the
const_diagnostics
branch
from
ec32bcb to
8b957ca
Compare
oli-obk
commented
Jun 26, 2018
|
|
||
| use super::{EvalResult, Pointer, PointerArithmetic, Allocation}; | ||
|
|
||
| /// Represents a constant value in Rust. ByVal and ScalarPair are optimizations which | ||
| /// matches Value's optimizations for easy conversions between these two types | ||
| #[derive(Clone, Copy, Debug, Eq, PartialEq, PartialOrd, Ord, RustcEncodable, RustcDecodable, Hash)] | ||
| #[derive(Copy, Clone, Debug, Eq, PartialEq, PartialOrd, Ord, RustcEncodable, RustcDecodable, Hash)] | ||
| pub enum ConstValue<'tcx> { |
|
The job Click to expand the log.I'm a bot! I can only do what humans tell me to, so if this was not helpful or you have suggestions for improvements, please ping or otherwise contact |
oli-obk
force-pushed
the
const_diagnostics
branch
2 times, most recently
from
4064f09 to
c190b3a
Compare
|
r? @eddyb (since I'm still on leave) |
|
☔ The latest upstream changes (presumably #51538) made this pull request unmergeable. Please resolve the merge conflicts. |
oli-obk
force-pushed
the
const_diagnostics
branch
from
a155b37 to
4d47330
Compare
|
The job Click to expand the log.I'm a bot! I can only do what humans tell me to, so if this was not helpful or you have suggestions for improvements, please ping or otherwise contact |
|
@bors r+ |
|
📌 Commit 2a55d2f has been approved by |
bors
added
S-waiting-on-bors
Mark-Simulacrum
added a commit
to Mark-Simulacrum/rust
that referenced
this pull request
Jun 28, 2018
Refactor error reporting of constants cc @eddyb This PR should not change any behaviour. It solely simplifies the internal handling of the errors
bors
added a commit
that referenced
this pull request
Jun 28, 2018
Rollup of 6 pull requests Successful merges: - #51636 (Refactor error reporting of constants) - #51765 (Use assert_eq! in copy_from_slice) - #51822 (Provide existing ref suggestions for more E0308 errors) - #51839 (Detect overflows of non u32 shifts) - #51868 (Remove process::id from 'Stabilized APIs' in 1.27.0 release notes) - #51875 (Explicitely disable WASM code generation for Emscripten) Failed merges: r? @ghost
# for free
to join this conversation on GitHub.
Already have an account?
# 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.
cc @eddyb
This PR should not change any behaviour. It solely simplifies the internal handling of the errors