Rollup merge of #120727 - Nadrieril:tweak-int-reporting, r=compiler-errors

exhaustiveness: Prefer "`0..MAX` not covered" to "`_` not covered"

There was an exception when reporting integer ranges as missing, it's been there for as long as I can remember. This PR removes it. I think it's nicer to report "`0..MAX` not covered" than "`_` not covered". This also makes it consistent with enums, where we report individual enum variants in this case (as showcased in the rest of the `empty-match.rs` test).

r? ``@estebank``

1 files changed, 3 insertions, 5 deletions

diff --git a/compiler/rustc_pattern_analysis/src/usefulness.rs b/compiler/rustc_pattern_analysis/src/usefulness.rs
index 576005b2c7f..80a807b4f27 100644
--- a/compiler/rustc_pattern_analysis/src/usefulness.rs
+++ b/compiler/rustc_pattern_analysis/src/usefulness.rs
@@ -1520,11 +1520,9 @@ fn compute_exhaustiveness_and_usefulness<'a, 'p, Cx: TypeCx>(
         split_ctors.push(Constructor::Missing);
     }
 
-    // Decide what constructors to report.
-    let is_integers = matches!(ctors_for_ty, ConstructorSet::Integers { .. });
-    let always_report_all = place.is_scrutinee && !is_integers;
-    // Whether we should report "Enum::A and Enum::C are missing" or "_ is missing".
-    let report_individual_missing_ctors = always_report_all || !all_missing;
+    // Whether we should report "Enum::A and Enum::C are missing" or "_ is missing". At the top
+    // level we prefer to list all constructors.
+    let report_individual_missing_ctors = place.is_scrutinee || !all_missing;
     // Which constructors are considered missing. We ensure that `!missing_ctors.is_empty() =>
     // split_ctors.contains(Missing)`. The converse usually holds except when
     // `!place_validity.is_known_valid()`.