Auto merge of #87640 - JohnTitor:rollup-yq24nq5, r=JohnTitor

Rollup of 9 pull requests

Successful merges:

 - #86072 (Cross compiling rustc_llvm on Darwin requires zlib.)
 - #87385 (Make `SEMICOLON_IN_EXPRESSIONS_FROM_MACROS` warn by default)
 - #87547 (Add missing examples for NonNull)
 - #87557 (Fix issue with autofix for ambiguous associated function from Rust 2021 prelude when struct is generic)
 - #87559 (Tweak borrowing suggestion in `for` loop)
 - #87596 (Add warning when whitespace is not skipped after an escaped newline)
 - #87606 (Add some TAIT-related regression tests)
 - #87609 (Add docs about performance and `Iterator::map` to `[T; N]::map`)
 - #87616 (Fix missing word in rustdoc book)

Failed merges:

r? `@ghost`
`@rustbot` modify labels: rollup

Author: bors

compiler/rustc_ast/src/util/literal.rs

@@ -63,7 +63,11 @@ impl LitKind {
                         unescape_literal(&s, Mode::Str, &mut |_, unescaped_char| {
                             match unescaped_char {
                                 Ok(c) => buf.push(c),
-                                Err(_) => error = Err(LitError::LexerError),
+                                Err(err) => {
+                                    if err.is_fatal() {
+                                        error = Err(LitError::LexerError);
+                                    }
+                                }
                             }
                         });
                         error?;
@@ -83,7 +87,11 @@ impl LitKind {
                         unescape_literal(&s, Mode::RawStr, &mut |_, unescaped_char| {
                             match unescaped_char {
                                 Ok(c) => buf.push(c),
-                                Err(_) => error = Err(LitError::LexerError),
+                                Err(err) => {
+                                    if err.is_fatal() {
+                                        error = Err(LitError::LexerError);
+                                    }
+                                }
                             }
                         });
                         error?;
@@ -100,7 +108,11 @@ impl LitKind {
                 unescape_byte_literal(&s, Mode::ByteStr, &mut |_, unescaped_byte| {
                     match unescaped_byte {
                         Ok(c) => buf.push(c),
-                        Err(_) => error = Err(LitError::LexerError),
+                        Err(err) => {
+                            if err.is_fatal() {
+                                error = Err(LitError::LexerError);
+                            }
+                        }
                     }
                 });
                 error?;
@@ -114,7 +126,11 @@ impl LitKind {
                     unescape_byte_literal(&s, Mode::RawByteStr, &mut |_, unescaped_byte| {
                         match unescaped_byte {
                             Ok(c) => buf.push(c),
-                            Err(_) => error = Err(LitError::LexerError),
+                            Err(err) => {
+                                if err.is_fatal() {
+                                    error = Err(LitError::LexerError);
+                                }
+                            }
                         }
                     });
                     error?;

compiler/rustc_expand/src/mbe/macro_rules.rs

@@ -45,6 +45,8 @@ crate struct ParserAnyMacro<'a> {
     lint_node_id: NodeId,
     is_trailing_mac: bool,
     arm_span: Span,
+    /// Whether or not this macro is defined in the current crate
+    is_local: bool,
 }
 
 crate fn annotate_err_with_kind(
@@ -124,6 +126,7 @@ impl<'a> ParserAnyMacro<'a> {
             lint_node_id,
             arm_span,
             is_trailing_mac,
+            is_local,
         } = *self;
         let snapshot = &mut parser.clone();
         let fragment = match parse_ast_fragment(parser, kind) {
@@ -138,13 +141,15 @@ impl<'a> ParserAnyMacro<'a> {
         // `macro_rules! m { () => { panic!(); } }` isn't parsed by `.parse_expr()`,
         // but `m!()` is allowed in expression positions (cf. issue #34706).
         if kind == AstFragmentKind::Expr && parser.token == token::Semi {
-            parser.sess.buffer_lint_with_diagnostic(
-                SEMICOLON_IN_EXPRESSIONS_FROM_MACROS,
-                parser.token.span,
-                lint_node_id,
-                "trailing semicolon in macro used in expression position",
-                BuiltinLintDiagnostics::TrailingMacro(is_trailing_mac, macro_ident),
-            );
+            if is_local {
+                parser.sess.buffer_lint_with_diagnostic(
+                    SEMICOLON_IN_EXPRESSIONS_FROM_MACROS,
+                    parser.token.span,
+                    lint_node_id,
+                    "trailing semicolon in macro used in expression position",
+                    BuiltinLintDiagnostics::TrailingMacro(is_trailing_mac, macro_ident),
+                );
+            }
             parser.bump();
         }
 
@@ -162,6 +167,7 @@ struct MacroRulesMacroExpander {
     lhses: Vec<mbe::TokenTree>,
     rhses: Vec<mbe::TokenTree>,
     valid: bool,
+    is_local: bool,
 }
 
 impl TTMacroExpander for MacroRulesMacroExpander {
@@ -183,6 +189,7 @@ impl TTMacroExpander for MacroRulesMacroExpander {
             input,
             &self.lhses,
             &self.rhses,
+            self.is_local,
         )
     }
 }
@@ -210,6 +217,7 @@ fn generic_extension<'cx>(
     arg: TokenStream,
     lhses: &[mbe::TokenTree],
     rhses: &[mbe::TokenTree],
+    is_local: bool,
 ) -> Box<dyn MacResult + 'cx> {
     let sess = &cx.sess.parse_sess;
 
@@ -311,6 +319,7 @@ fn generic_extension<'cx>(
                     lint_node_id: cx.current_expansion.lint_node_id,
                     is_trailing_mac: cx.current_expansion.is_trailing_mac,
                     arm_span,
+                    is_local,
                 });
             }
             Failure(token, msg) => match best_failure {
@@ -544,6 +553,9 @@ pub fn compile_declarative_macro(
         lhses,
         rhses,
         valid,
+        // Macros defined in the current crate have a real node id,
+        // whereas macros from an external crate have a dummy id.
+        is_local: def.id != DUMMY_NODE_ID,
     }))
 }
 

compiler/rustc_lexer/src/unescape.rs

@@ -7,7 +7,7 @@ use std::str::Chars;
 #[cfg(test)]
 mod tests;
 
-/// Errors that can occur during string unescaping.
+/// Errors and warnings that can occur during string unescaping.
 #[derive(Debug, PartialEq, Eq)]
 pub enum EscapeError {
     /// Expected 1 char, but 0 were found.
@@ -56,6 +56,20 @@ pub enum EscapeError {
     NonAsciiCharInByte,
     /// Non-ascii character in byte string literal.
     NonAsciiCharInByteString,
+
+    /// After a line ending with '\', the next line contains whitespace
+    /// characters that are not skipped.
+    UnskippedWhitespaceWarning,
+}
+
+impl EscapeError {
+    /// Returns true for actual errors, as opposed to warnings.
+    pub fn is_fatal(&self) -> bool {
+        match self {
+            EscapeError::UnskippedWhitespaceWarning => false,
+            _ => true,
+        }
+    }
 }
 
 /// Takes a contents of a literal (without quotes) and produces a
@@ -283,7 +297,7 @@ where
                         // if unescaped '\' character is followed by '\n'.
                         // For details see [Rust language reference]
                         // (https://doc.rust-lang.org/reference/tokens.html#string-literals).
-                        skip_ascii_whitespace(&mut chars);
+                        skip_ascii_whitespace(&mut chars, start, callback);
                         continue;
                     }
                     _ => scan_escape(first_char, &mut chars, mode),
@@ -297,13 +311,25 @@ where
         callback(start..end, unescaped_char);
     }
 
-    fn skip_ascii_whitespace(chars: &mut Chars<'_>) {
+    fn skip_ascii_whitespace<F>(chars: &mut Chars<'_>, start: usize, callback: &mut F)
+    where
+        F: FnMut(Range<usize>, Result<char, EscapeError>),
+    {
         let str = chars.as_str();
         let first_non_space = str
             .bytes()
             .position(|b| b != b' ' && b != b'\t' && b != b'\n' && b != b'\r')
             .unwrap_or(str.len());
-        *chars = str[first_non_space..].chars()
+        let tail = &str[first_non_space..];
+        if let Some(c) = tail.chars().nth(0) {
+            // For error reporting, we would like the span to contain the character that was not
+            // skipped.  The +1 is necessary to account for the leading \ that started the escape.
+            let end = start + first_non_space + c.len_utf8() + 1;
+            if c.is_whitespace() {
+                callback(start..end, Err(EscapeError::UnskippedWhitespaceWarning));
+            }
+        }
+        *chars = tail.chars();
     }
 }
 

compiler/rustc_lexer/src/unescape/tests.rs

@@ -98,6 +98,25 @@ fn test_unescape_char_good() {
     check(r"\u{1F63b}", '😻');
 }
 
+#[test]
+fn test_unescape_str_warn() {
+    fn check(literal: &str, expected: &[(Range<usize>, Result<char, EscapeError>)]) {
+        let mut unescaped = Vec::with_capacity(literal.len());
+        unescape_literal(literal, Mode::Str, &mut |range, res| unescaped.push((range, res)));
+        assert_eq!(unescaped, expected);
+    }
+
+    check(
+        "\\\n \u{a0} x",
+        &[
+            (0..5, Err(EscapeError::UnskippedWhitespaceWarning)),
+            (3..5, Ok('\u{a0}')),
+            (5..6, Ok(' ')),
+            (6..7, Ok('x')),
+        ],
+    );
+}
+
 #[test]
 fn test_unescape_str_good() {
     fn check(literal_text: &str, expected: &str) {

compiler/rustc_lint_defs/src/builtin.rs

@@ -2799,7 +2799,7 @@ declare_lint! {
     /// [issue #79813]: https://github.com/rust-lang/rust/issues/79813
     /// [future-incompatible]: ../index.md#future-incompatible-lints
     pub SEMICOLON_IN_EXPRESSIONS_FROM_MACROS,
-    Allow,
+    Warn,
     "trailing semicolon in macro body used as expression",
     @future_incompatible = FutureIncompatibleInfo {
         reference: "issue #79813 <https://github.com/rust-lang/rust/issues/79813>",

compiler/rustc_llvm/build.rs

@@ -182,7 +182,7 @@ fn main() {
     } else if target.contains("windows-gnu") {
         println!("cargo:rustc-link-lib=shell32");
         println!("cargo:rustc-link-lib=uuid");
-    } else if target.contains("netbsd") || target.contains("haiku") {
+    } else if target.contains("netbsd") || target.contains("haiku") || target.contains("darwin") {
         println!("cargo:rustc-link-lib=z");
     }
     cmd.args(&components);

compiler/rustc_mir/src/borrow_check/diagnostics/move_errors.rs

@@ -2,7 +2,8 @@ use rustc_errors::{Applicability, DiagnosticBuilder};
 use rustc_middle::mir::*;
 use rustc_middle::ty;
 use rustc_span::source_map::DesugaringKind;
-use rustc_span::{sym, Span};
+use rustc_span::{sym, Span, DUMMY_SP};
+use rustc_trait_selection::traits::type_known_to_meet_bound_modulo_regions;
 
 use crate::borrow_check::diagnostics::UseSpans;
 use crate::borrow_check::prefixes::PrefixSet;
@@ -384,36 +385,44 @@ impl<'a, 'tcx> MirBorrowckCtxt<'a, 'tcx> {
                 }
             }
         };
-        if let Ok(snippet) = self.infcx.tcx.sess.source_map().span_to_snippet(span) {
-            let def_id = match *move_place.ty(self.body, self.infcx.tcx).ty.kind() {
-                ty::Adt(self_def, _) => self_def.did,
-                ty::Foreign(def_id)
-                | ty::FnDef(def_id, _)
-                | ty::Closure(def_id, _)
-                | ty::Generator(def_id, ..)
-                | ty::Opaque(def_id, _) => def_id,
-                _ => return err,
+        let ty = move_place.ty(self.body, self.infcx.tcx).ty;
+        let def_id = match *ty.kind() {
+            ty::Adt(self_def, _) => self_def.did,
+            ty::Foreign(def_id)
+            | ty::FnDef(def_id, _)
+            | ty::Closure(def_id, _)
+            | ty::Generator(def_id, ..)
+            | ty::Opaque(def_id, _) => def_id,
+            _ => return err,
+        };
+        let is_option = self.infcx.tcx.is_diagnostic_item(sym::option_type, def_id);
+        let is_result = self.infcx.tcx.is_diagnostic_item(sym::result_type, def_id);
+        if (is_option || is_result) && use_spans.map_or(true, |v| !v.for_closure()) {
+            err.span_suggestion_verbose(
+                span.shrink_to_hi(),
+                &format!(
+                    "consider borrowing the `{}`'s content",
+                    if is_option { "Option" } else { "Result" }
+                ),
+                ".as_ref()".to_string(),
+                Applicability::MaybeIncorrect,
+            );
+        } else if matches!(span.desugaring_kind(), Some(DesugaringKind::ForLoop(_))) {
+            let suggest = match self.infcx.tcx.get_diagnostic_item(sym::IntoIterator) {
+                Some(def_id) => type_known_to_meet_bound_modulo_regions(
+                    &self.infcx,
+                    self.param_env,
+                    self.infcx.tcx.mk_imm_ref(self.infcx.tcx.lifetimes.re_erased, ty),
+                    def_id,
+                    DUMMY_SP,
+                ),
+                _ => false,
             };
-            let is_option = self.infcx.tcx.is_diagnostic_item(sym::option_type, def_id);
-            let is_result = self.infcx.tcx.is_diagnostic_item(sym::result_type, def_id);
-            if (is_option || is_result) && use_spans.map_or(true, |v| !v.for_closure()) {
-                err.span_suggestion(
-                    span,
-                    &format!(
-                        "consider borrowing the `{}`'s content",
-                        if is_option { "Option" } else { "Result" }
-                    ),
-                    format!("{}.as_ref()", snippet),
-                    Applicability::MaybeIncorrect,
-                );
-            } else if matches!(span.desugaring_kind(), Some(DesugaringKind::ForLoop(_)))
-                && self.infcx.tcx.is_diagnostic_item(sym::vec_type, def_id)
-            {
-                // FIXME: suggest for anything that implements `IntoIterator`.
-                err.span_suggestion(
-                    span,
-                    "consider iterating over a slice of the `Vec<_>`'s content",
-                    format!("&{}", snippet),
+            if suggest {
+                err.span_suggestion_verbose(
+                    span.shrink_to_lo(),
+                    &format!("consider iterating over a slice of the `{}`'s content", ty),
+                    "&".to_string(),
                     Applicability::MaybeIncorrect,
                 );
             }

compiler/rustc_parse/src/lexer/unescape_error_reporting.rs

@@ -253,6 +253,12 @@ pub(crate) fn emit_unescape_error(
             let msg = "invalid trailing slash in literal";
             handler.struct_span_err(span, msg).span_label(span, msg).emit();
         }
+        EscapeError::UnskippedWhitespaceWarning => {
+            let (c, char_span) = last_char();
+            let msg =
+                format!("non-ASCII whitespace symbol '{}' is not skipped", c.escape_unicode());
+            handler.struct_span_warn(span, &msg).span_label(char_span, &msg).emit();
+        }
     }
 }
 

compiler/rustc_typeck/src/check/method/prelude2021.rs

@@ -5,7 +5,7 @@ use rustc_ast::Mutability;
 use rustc_errors::Applicability;
 use rustc_hir as hir;
 use rustc_middle::ty::subst::InternalSubsts;
-use rustc_middle::ty::{Ref, Ty};
+use rustc_middle::ty::{Adt, Ref, Ty};
 use rustc_session::lint::builtin::RUST_2021_PRELUDE_COLLISIONS;
 use rustc_span::symbol::kw::Underscore;
 use rustc_span::symbol::{sym, Ident};
@@ -255,16 +255,31 @@ impl<'a, 'tcx> FnCtxt<'a, 'tcx> {
                 method_name.name
             ));
 
-            let self_ty = self
+            let self_ty_name = self
                 .sess()
                 .source_map()
                 .span_to_snippet(self_ty_span)
                 .unwrap_or_else(|_| self_ty.to_string());
 
+            let self_ty_generics_count = match self_ty.kind() {
+                // Get the number of generics the self type has (if an Adt) unless we can determine that
+                // the user has written the self type with generics already which we (naively) do by looking
+                // for a "<" in `self_ty_name`.
+                Adt(def, _) if !self_ty_name.contains("<") => self.tcx.generics_of(def.did).count(),
+                _ => 0,
+            };
+            let self_ty_generics = if self_ty_generics_count > 0 {
+                format!("<{}>", vec!["_"; self_ty_generics_count].join(", "))
+            } else {
+                String::new()
+            };
             lint.span_suggestion(
                 span,
                 "disambiguate the associated function",
-                format!("<{} as {}>::{}", self_ty, trait_name, method_name.name,),
+                format!(
+                    "<{}{} as {}>::{}",
+                    self_ty_name, self_ty_generics, trait_name, method_name.name,
+                ),
                 Applicability::MachineApplicable,
             );