Rollup merge of #120259 - HTGAzureX1212:HTGAzureX1212/split-diagnostics-uncommon-codepoints, r=Manishearth

Split Diagnostics for Uncommon Codepoints: Add List to Display Characters Involved

This Pull Request adds a list of the uncommon codepoints involved in the `uncommon_codepoints` lint, as outlined as a first step in #120228.

Example rendered diagnostic:
```
error: identifier contains an uncommon Unicode codepoint: 'µ'
  --> $DIR/lint-uncommon-codepoints.rs:3:7
   |
LL | const µ: f64 = 0.000001;
   |       ^
   |
note: the lint level is defined here
  --> $DIR/lint-uncommon-codepoints.rs:1:9
   |
LL | #![deny(uncommon_codepoints)]
   |         ^^^^^^^^^^^^^^^^^^^
```

(Retrying #120258.)

compiler/rustc_errors/src/diagnostic_impls.rs

@@ -110,6 +110,14 @@ fn into_diagnostic_arg(self) -> DiagnosticArgValue<'static> {
     }
 }
 
+impl IntoDiagnosticArg for Vec<char> {
+    fn into_diagnostic_arg(self) -> DiagnosticArgValue<'static> {
+        DiagnosticArgValue::StrListSepByAnd(
+            self.into_iter().map(|c| Cow::Owned(format!("{c:?}"))).collect(),
+        )
+    }
+}
+
 impl IntoDiagnosticArg for Symbol {
     fn into_diagnostic_arg(self) -> DiagnosticArgValue<'static> {
         self.to_ident_string().into_diagnostic_arg()

compiler/rustc_lint/messages.ftl

@@ -240,7 +240,10 @@ lint_hidden_unicode_codepoints = unicode codepoint changing visible direction of
 
 lint_identifier_non_ascii_char = identifier contains non-ASCII characters
 
-lint_identifier_uncommon_codepoints = identifier contains uncommon Unicode codepoints
+lint_identifier_uncommon_codepoints = identifier contains {$codepoints_len ->
+    [one] an uncommon Unicode codepoint
+    *[other] uncommon Unicode codepoints
+}: {$codepoints}
 
 lint_ignored_unless_crate_specified = {$level}({$name}) is ignored unless specified at crate level
 

compiler/rustc_lint/src/lints.rs

@@ -1107,7 +1107,10 @@ pub struct MultipleSupertraitUpcastable {
 
 #[derive(LintDiagnostic)]
 #[diag(lint_identifier_uncommon_codepoints)]
-pub struct IdentifierUncommonCodepoints;
+pub struct IdentifierUncommonCodepoints {
+    pub codepoints: Vec<char>,
+    pub codepoints_len: usize,
+}
 
 #[derive(LintDiagnostic)]
 #[diag(lint_confusable_identifier_pair)]

compiler/rustc_lint/src/non_ascii_idents.rs

@@ -190,7 +190,17 @@ fn check_crate(&mut self, cx: &EarlyContext<'_>, _: &ast::Crate) {
             if check_uncommon_codepoints
                 && !symbol_str.chars().all(GeneralSecurityProfile::identifier_allowed)
             {
-                cx.emit_span_lint(UNCOMMON_CODEPOINTS, sp, IdentifierUncommonCodepoints);
+                let codepoints: Vec<_> = symbol_str
+                    .chars()
+                    .filter(|c| !GeneralSecurityProfile::identifier_allowed(*c))
+                    .collect();
+                let codepoints_len = codepoints.len();
+
+                cx.emit_span_lint(
+                    UNCOMMON_CODEPOINTS,
+                    sp,
+                    IdentifierUncommonCodepoints { codepoints, codepoints_len },
+                );
             }
         }
 

tests/ui/lexer/lex-emoji-identifiers.rs

@@ -4,7 +4,7 @@ fn invalid_emoji_usages() {
     let wireless🛜 = "basic emoji"; //~ ERROR: identifiers cannot contain emoji
     // FIXME
     let key1️⃣ = "keycap sequence"; //~ ERROR: unknown start of token
-                                    //~^ WARN: identifier contains uncommon Unicode codepoints
+                                    //~^ WARN: identifier contains an uncommon Unicode codepoint
     let flag🇺🇳 = "flag sequence"; //~ ERROR: identifiers cannot contain emoji
     let wales🏴 = "tag sequence"; //~ ERROR: identifiers cannot contain emoji
     let folded🙏🏿 = "modifier sequence"; //~ ERROR: identifiers cannot contain emoji

tests/ui/lexer/lex-emoji-identifiers.stderr

@@ -40,7 +40,7 @@ error: identifiers cannot contain emoji: `folded🙏🏿`
 LL |     let folded🙏🏿 = "modifier sequence";
    |         ^^^^^^^^^^
 
-warning: identifier contains uncommon Unicode codepoints
+warning: identifier contains an uncommon Unicode codepoint: '\u{fe0f}'
   --> $DIR/lex-emoji-identifiers.rs:6:9
    |
 LL |     let key1️⃣ = "keycap sequence";

tests/ui/lint/rfc-2457-non-ascii-idents/lint-uncommon-codepoints.rs

@@ -1,9 +1,9 @@
 #![deny(uncommon_codepoints)]
 
-const µ: f64 = 0.000001; //~ ERROR identifier contains uncommon Unicode codepoints
+const µ: f64 = 0.000001; //~ ERROR identifier contains an uncommon Unicode codepoint
 //~| WARNING should have an upper case name
 
-fn dijkstra() {} //~ ERROR identifier contains uncommon Unicode codepoints
+fn dijkstra() {} //~ ERROR identifier contains an uncommon Unicode codepoint
 
 fn main() {
     let ㇻㇲㇳ = "rust"; //~ ERROR identifier contains uncommon Unicode codepoints

tests/ui/lint/rfc-2457-non-ascii-idents/lint-uncommon-codepoints.stderr

@@ -1,4 +1,4 @@
-error: identifier contains uncommon Unicode codepoints
+error: identifier contains an uncommon Unicode codepoint: 'µ'
   --> $DIR/lint-uncommon-codepoints.rs:3:7
    |
 LL | const µ: f64 = 0.000001;
@@ -10,13 +10,13 @@ note: the lint level is defined here
 LL | #![deny(uncommon_codepoints)]
    |         ^^^^^^^^^^^^^^^^^^^
 
-error: identifier contains uncommon Unicode codepoints
+error: identifier contains an uncommon Unicode codepoint: 'ij'
   --> $DIR/lint-uncommon-codepoints.rs:6:4
    |
 LL | fn dijkstra() {}
    |    ^^^^^^^
 
-error: identifier contains uncommon Unicode codepoints
+error: identifier contains uncommon Unicode codepoints: 'ㇻ', 'ㇲ', and 'ㇳ'
   --> $DIR/lint-uncommon-codepoints.rs:9:9
    |
 LL |     let ㇻㇲㇳ = "rust";