Avoid a bunch of booleans in favor of Result<(), ErrorGuaranteed> as that more robustly proves that an error has been emitted by oli-obk · Pull Request #126317 · rust-lang/rust

oli-obk · 2024-06-12T10:22:29Z

pulled out of #126316

This PR cannot have any effect on compilation.
All it does is shift a Ty::new_misc_error to a span_delayed_bug and preserve the ErrorGuaranteed in all other cases

…s that more robustly proves that an error has been emitted

rustbot · 2024-06-12T10:22:37Z

r? @TaKO8Ki

rustbot has assigned @TaKO8Ki.
They will have a look at your PR within the next two weeks and either review your PR or reassign to another reviewer.

Use r? to explicitly pick a reviewer

…ng it.

workingjubilee · 2024-06-12T10:27:55Z

This PR cannot have any effect on compilation.

hm, if so then
@bors rollup

compiler-errors · 2024-06-12T12:36:55Z

compiler/rustc_hir_typeck/src/expr.rs

+                if segment.ident.name == kw::Empty {
+                    span_bug!(rcvr.span, "empty method name")
+                } else {
+                    match self.report_method_error(expr.hir_id, rcvr_t, error, expected, false) {


please make report_method_error return an ErrorGuaranteed. there's no reason to return Result<Diag, ErrorGuaranteed> when the only two callsites just call diag.emit().

compiler-errors · 2024-06-12T12:38:50Z

compiler/rustc_hir_typeck/src/fn_ctxt/checks.rs

        };
-        if has_error {
+        if let Err(guar) = has_error {
            let err_inputs = self.err_args(args_no_rcvr.len());


make err_args take the ErrorGuaranteed

compiler-errors · 2024-06-12T12:40:19Z

compiler/rustc_hir_typeck/src/method/suggest.rs

        expected: Expectation<'tcx>,
        trait_missing_method: bool,
-    ) -> Option<Diag<'_>> {
+    ) -> Result<Diag<'_>, ErrorGuaranteed> {


Suggested change

) -> Result<Diag<'_>, ErrorGuaranteed> {

) -> ErrorGuaranteed {

compiler-errors · 2024-06-12T12:41:54Z

compiler/rustc_hir_typeck/src/method/suggest.rs

        // We could pass the file for long types into these two, but it isn't strictly necessary
        // given how targeted they are.
-        if self.suggest_wrapping_range_with_parens(
+        self.suggest_wrapping_range_with_parens(


These probably shouldn't be named suggest_*, since they're emitting their own error. Could you think of a better name for this?

compiler-errors

r=me when ci is green

oli-obk · 2024-06-12T15:20:32Z

@bors r=compiler-errors

bors · 2024-06-12T15:20:35Z

📌 Commit 6299071 has been approved by compiler-errors

It is now in the queue for this repository.

…mpiler-errors Rollup of 8 pull requests Successful merges: - rust-lang#125869 (Add `target_env = "p1"` to the `wasm32-wasip1` target) - rust-lang#126019 (Add TODO comment to unsafe env modification) - rust-lang#126036 (Migrate `run-make/short-ice` to `rmake`) - rust-lang#126276 (Detect pub structs never constructed even though they impl pub trait with assoc constants) - rust-lang#126282 (Ensure self-contained linker is only enabled on dev/nightly ) - rust-lang#126317 (Avoid a bunch of booleans in favor of Result<(), ErrorGuaranteed> as that more robustly proves that an error has been emitted) - rust-lang#126324 (Adjust LoongArch64 data layouts for LLVM update) - rust-lang#126340 (Fix outdated predacates_of.rs comments) r? `@ghost` `@rustbot` modify labels: rollup

Rollup merge of rust-lang#126317 - oli-obk:recursive_rpit4, r=compiler-errors Avoid a bunch of booleans in favor of Result<(), ErrorGuaranteed> as that more robustly proves that an error has been emitted pulled out of rust-lang#126316 This PR cannot have any effect on compilation. All it does is shift a `Ty::new_misc_error` to a `span_delayed_bug` and preserve the `ErrorGuaranteed` in all other cases

RalfJung · 2024-08-10T13:27:28Z

compiler/rustc_log/src/lib.rs


+    let lines = match cfg.lines {
+        Ok(v) => &v == "1",
+        Err(_) => false,


I think a surprising side-effect of this change was to turn off the lines by default?

Even when they are turned on, they look very different... much harder than before to just "move up" and find earlier things on the same level of nesting, since the line is now just a snake that moves left and right with the indentation, it no longer properly shows what nesting level we are getting back to.

lines were already turned off in an earlier PR by someone else because they broke code folding which allowed you to fold entire sub-spans. After I switched tracing-tree to use the snaky line, you get both the lines (which I find useful) and the code folding, so I made it opt-in. Locally I was replacing | manually with spaces in the last year just to get code folding in IDEs.

What's "code folding"?

When you hover over the line numbers in vscode, folding thingies like for directories show up. These are indentation based I think, but in code they may be squirly bracket based

RalfJung · 2024-08-11T05:59:08Z

Ah I see. I often go through these logs in a terminal and I find the new format a lot harder to read there than the old one.

oli-obk added 2 commits June 12, 2024 10:11

Bump tracing-tree and allow rendering lines again

185a48d

Avoid a bunch of booleans in favor of Result<(), ErrorGuaranteed> a…

3da9289

…s that more robustly proves that an error has been emitted

rustbot assigned TaKO8Ki Jun 12, 2024

rustbot added S-waiting-on-review Status: Awaiting review from the assignee but also interested parties. T-compiler Relevant to the compiler team, which will review and decide on the PR/issue. labels Jun 12, 2024

Upgrade a span_delayed_bug into a span_bug, as we have no tests hitti…

be464b8

…ng it.

compiler-errors suggested changes Jun 12, 2024

View reviewed changes

compiler-errors assigned compiler-errors and unassigned TaKO8Ki Jun 12, 2024

oli-obk added 3 commits June 12, 2024 12:54

Eagerly emit the diagnostic instead of leaving it to all callers

4d72c42

Rename some functions

9065889

Avoid a Ty::new_misc_error when an ErrorGuaranteed is available

6299071

oli-obk requested a review from compiler-errors June 12, 2024 14:26

compiler-errors approved these changes Jun 12, 2024

View reviewed changes

bors added S-waiting-on-bors Status: Waiting on bors to run and complete tests. Bors will change the label on completion. and removed S-waiting-on-review Status: Awaiting review from the assignee but also interested parties. labels Jun 12, 2024

compiler-errors mentioned this pull request Jun 12, 2024

Rollup of 8 pull requests #126345

Merged

bors merged commit 7276f34 into rust-lang:master Jun 12, 2024

rustbot added this to the 1.81.0 milestone Jun 12, 2024

RalfJung reviewed Aug 10, 2024

View reviewed changes

oli-obk deleted the recursive_rpit4 branch August 10, 2024 20:07

	) -> Result<Diag<'_>, ErrorGuaranteed> {
	) -> ErrorGuaranteed {

Uh oh!

Conversation

oli-obk commented Jun 12, 2024

Uh oh!

rustbot commented Jun 12, 2024

Uh oh!

workingjubilee commented Jun 12, 2024

Uh oh!

compiler-errors Jun 12, 2024 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

compiler-errors Jun 12, 2024

Choose a reason for hiding this comment

Uh oh!

compiler-errors Jun 12, 2024

Choose a reason for hiding this comment

Uh oh!

compiler-errors Jun 12, 2024

Choose a reason for hiding this comment

Uh oh!

compiler-errors left a comment

Choose a reason for hiding this comment

Uh oh!

oli-obk commented Jun 12, 2024

Uh oh!

bors commented Jun 12, 2024

Uh oh!

RalfJung Aug 10, 2024

Choose a reason for hiding this comment

Uh oh!

oli-obk Aug 10, 2024

Choose a reason for hiding this comment

Uh oh!

RalfJung Aug 10, 2024

Choose a reason for hiding this comment

Uh oh!

oli-obk Aug 10, 2024

Choose a reason for hiding this comment

Uh oh!

RalfJung commented Aug 11, 2024 via email

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

7 participants

compiler-errors Jun 12, 2024 •

edited

Loading