Remove NtItem and NtStmt.

This involves replacing `nt_pretty_printing_compatibility_hack` with
`stream_pretty_printing_compatibility_hack`.

The handling of statements in `transcribe` is slightly different to
other nonterminal kinds, due to the lack of `from_ast` implementation
for empty statements.

Notable test changes:
- `tests/ui/macros/nonterminal-matching.rs` now passes. Removal of
  `Token::Interpolated` will remove the "captured metavariables except
  for `:tt`, `:ident` and `:lifetime` cannot be compared to other
  tokens" restriction.

- `tests/ui/proc-macro/expand-to-derive.rs`: the diff looks large but
  the only difference is the insertion of a single invisible-delimited
  group.

Author: nnethercote

compiler/rustc_ast/src/ast_traits.rs

@@ -231,8 +231,6 @@ impl HasTokens for Attribute {
 impl HasTokens for Nonterminal {
     fn tokens(&self) -> Option<&LazyAttrTokenStream> {
         match self {
-            Nonterminal::NtItem(item) => item.tokens(),
-            Nonterminal::NtStmt(stmt) => stmt.tokens(),
             Nonterminal::NtExpr(expr) | Nonterminal::NtLiteral(expr) => expr.tokens(),
             Nonterminal::NtMeta(attr_item) => attr_item.tokens(),
             Nonterminal::NtPath(path) => path.tokens(),
@@ -241,8 +239,6 @@ impl HasTokens for Nonterminal {
     }
     fn tokens_mut(&mut self) -> Option<&mut Option<LazyAttrTokenStream>> {
         match self {
-            Nonterminal::NtItem(item) => item.tokens_mut(),
-            Nonterminal::NtStmt(stmt) => stmt.tokens_mut(),
             Nonterminal::NtExpr(expr) | Nonterminal::NtLiteral(expr) => expr.tokens_mut(),
             Nonterminal::NtMeta(attr_item) => attr_item.tokens_mut(),
             Nonterminal::NtPath(path) => path.tokens_mut(),

compiler/rustc_ast/src/mut_visit.rs

@@ -819,19 +819,7 @@ pub fn visit_token<T: MutVisitor>(t: &mut Token, vis: &mut T) {
 // multiple items there....
 fn visit_nonterminal<T: MutVisitor>(nt: &mut token::Nonterminal, vis: &mut T) {
     match nt {
-        token::NtItem(item) => visit_clobber(item, |item| {
-            // This is probably okay, because the only visitors likely to
-            // peek inside interpolated nodes will be renamings/markings,
-            // which map single items to single items.
-            vis.flat_map_item(item).expect_one("expected visitor to produce exactly one item")
-        }),
         token::NtBlock(block) => vis.visit_block(block),
-        token::NtStmt(stmt) => visit_clobber(stmt, |stmt| {
-            // See reasoning above.
-            stmt.map(|stmt| {
-                vis.flat_map_stmt(stmt).expect_one("expected visitor to produce exactly one item")
-            })
-        }),
         token::NtExpr(expr) => vis.visit_expr(expr),
         token::NtLiteral(expr) => vis.visit_expr(expr),
         token::NtMeta(item) => {

compiler/rustc_ast/src/token.rs

@@ -939,9 +939,7 @@ impl PartialEq<TokenKind> for Token {
 #[derive(Clone, Encodable, Decodable)]
 /// For interpolation during macro expansion.
 pub enum Nonterminal {
-    NtItem(P<ast::Item>),
     NtBlock(P<ast::Block>),
-    NtStmt(P<ast::Stmt>),
     NtExpr(P<ast::Expr>),
     NtLiteral(P<ast::Expr>),
     /// Stuff inside brackets for attributes
@@ -1043,9 +1041,7 @@ impl fmt::Display for NonterminalKind {
 impl Nonterminal {
     pub fn use_span(&self) -> Span {
         match self {
-            NtItem(item) => item.span,
             NtBlock(block) => block.span,
-            NtStmt(stmt) => stmt.span,
             NtExpr(expr) | NtLiteral(expr) => expr.span,
             NtMeta(attr_item) => attr_item.span(),
             NtPath(path) => path.span,
@@ -1054,9 +1050,7 @@ impl Nonterminal {
 
     pub fn descr(&self) -> &'static str {
         match self {
-            NtItem(..) => "item",
             NtBlock(..) => "block",
-            NtStmt(..) => "statement",
             NtExpr(..) => "expression",
             NtLiteral(..) => "literal",
             NtMeta(..) => "attribute",
@@ -1078,9 +1072,7 @@ impl PartialEq for Nonterminal {
 impl fmt::Debug for Nonterminal {
     fn fmt(&self, f: &mut fmt::Formatter<'_>) -> fmt::Result {
         match *self {
-            NtItem(..) => f.pad("NtItem(..)"),
             NtBlock(..) => f.pad("NtBlock(..)"),
-            NtStmt(..) => f.pad("NtStmt(..)"),
             NtExpr(..) => f.pad("NtExpr(..)"),
             NtLiteral(..) => f.pad("NtLiteral(..)"),
             NtMeta(..) => f.pad("NtMeta(..)"),

compiler/rustc_ast/src/tokenstream.rs

@@ -13,7 +13,7 @@
 //! and a borrowed `TokenStream` is sufficient to build an owned `TokenStream` without taking
 //! ownership of the original.
 
-use crate::ast::{AttrStyle, StmtKind};
+use crate::ast::AttrStyle;
 use crate::ast_traits::{HasAttrs, HasSpan, HasTokens};
 use crate::token::{self, Delimiter, InvisibleOrigin, Nonterminal, Token, TokenKind};
 use crate::AttrVec;
@@ -466,13 +466,7 @@ impl TokenStream {
 
     pub fn from_nonterminal_ast(nt: &Nonterminal) -> TokenStream {
         match nt {
-            Nonterminal::NtItem(item) => TokenStream::from_ast(item),
             Nonterminal::NtBlock(block) => TokenStream::from_ast(block),
-            Nonterminal::NtStmt(stmt) if let StmtKind::Empty = stmt.kind => {
-                // FIXME: Properly collect tokens for empty statements.
-                TokenStream::token_alone(token::Semi, stmt.span)
-            }
-            Nonterminal::NtStmt(stmt) => TokenStream::from_ast(stmt),
             Nonterminal::NtMeta(attr) => TokenStream::from_ast(attr),
             Nonterminal::NtPath(path) => TokenStream::from_ast(path),
             Nonterminal::NtExpr(expr) | Nonterminal::NtLiteral(expr) => TokenStream::from_ast(expr),

compiler/rustc_expand/src/base.rs

@@ -5,7 +5,7 @@ use crate::module::DirOwnership;
 
 use rustc_ast::attr::MarkedAttrs;
 use rustc_ast::ptr::P;
-use rustc_ast::token::Nonterminal;
+use rustc_ast::token::NonterminalKind;
 use rustc_ast::tokenstream::TokenStream;
 use rustc_ast::visit::{AssocCtxt, Visitor};
 use rustc_ast::{self as ast, AttrVec, Attribute, HasAttrs, Item, NodeId, PatKind};
@@ -15,7 +15,8 @@ use rustc_data_structures::sync::{self, Lrc};
 use rustc_errors::{DiagCtxtHandle, ErrorGuaranteed, PResult};
 use rustc_feature::Features;
 use rustc_lint_defs::{BufferedEarlyLint, RegisteredTools};
-use rustc_parse::{parser::Parser, MACRO_ARGUMENTS};
+use rustc_parse::parser::{ParseNtResult, Parser};
+use rustc_parse::MACRO_ARGUMENTS;
 use rustc_session::config::CollapseMacroDebuginfo;
 use rustc_session::{parse::ParseSess, Limit, Session};
 use rustc_span::def_id::{CrateNum, DefId, LocalDefId};
@@ -1377,13 +1378,13 @@ pub fn parse_macro_name_and_helper_attrs(
 /// If this item looks like a specific enums from `rental`, emit a fatal error.
 /// See #73345 and #83125 for more details.
 /// FIXME(#73933): Remove this eventually.
-fn pretty_printing_compatibility_hack(item: &Item, sess: &Session) {
+fn pretty_printing_compatibility_hack(item: &Item, psess: &ParseSess) {
     let name = item.ident.name;
     if name == sym::ProceduralMasqueradeDummyType
         && let ast::ItemKind::Enum(enum_def, _) = &item.kind
         && let [variant] = &*enum_def.variants
         && variant.ident.name == sym::Input
-        && let FileName::Real(real) = sess.source_map().span_to_filename(item.ident.span)
+        && let FileName::Real(real) = psess.source_map().span_to_filename(item.ident.span)
         && let Some(c) = real
             .local_path()
             .unwrap_or(Path::new(""))
@@ -1403,15 +1404,15 @@ fn pretty_printing_compatibility_hack(item: &Item, sess: &Session) {
         if crate_matches {
             // FIXME: make this translatable
             #[allow(rustc::untranslatable_diagnostic)]
-            sess.dcx().emit_fatal(errors::ProcMacroBackCompat {
+            psess.dcx().emit_fatal(errors::ProcMacroBackCompat {
                 crate_name: "rental".to_string(),
                 fixed_version: "0.5.6".to_string(),
             });
         }
     }
 }
 
-pub(crate) fn ann_pretty_printing_compatibility_hack(ann: &Annotatable, sess: &Session) {
+pub(crate) fn ann_pretty_printing_compatibility_hack(ann: &Annotatable, psess: &ParseSess) {
     let item = match ann {
         Annotatable::Item(item) => item,
         Annotatable::Stmt(stmt) => match &stmt.kind {
@@ -1420,17 +1421,34 @@ pub(crate) fn ann_pretty_printing_compatibility_hack(ann: &Annotatable, sess: &S
         },
         _ => return,
     };
-    pretty_printing_compatibility_hack(item, sess)
+    pretty_printing_compatibility_hack(item, psess)
 }
 
-pub(crate) fn nt_pretty_printing_compatibility_hack(nt: &Nonterminal, sess: &Session) {
-    let item = match nt {
-        Nonterminal::NtItem(item) => item,
-        Nonterminal::NtStmt(stmt) => match &stmt.kind {
-            ast::StmtKind::Item(item) => item,
-            _ => return,
-        },
+pub(crate) fn stream_pretty_printing_compatibility_hack(
+    kind: NonterminalKind,
+    stream: &TokenStream,
+    psess: &ParseSess,
+) {
+    let item = match kind {
+        NonterminalKind::Item => {
+            let mut parser = Parser::new(psess, stream.clone(), None);
+            let Ok(ParseNtResult::Item(item)) = parser.parse_nonterminal(kind) else {
+                panic!("failed to reparse");
+            };
+            item
+        }
+        NonterminalKind::Stmt => {
+            // njn: reparsing and then checking for StmtKind::Item sucks, hmm
+            let mut parser = Parser::new(psess, stream.clone(), None);
+            let Ok(ParseNtResult::Stmt(stmt)) = parser.parse_nonterminal(kind) else {
+                panic!("failed to reparse");
+            };
+            match &stmt.kind {
+                ast::StmtKind::Item(item) => item.clone(),
+                _ => return,
+            }
+        }
         _ => return,
     };
-    pretty_printing_compatibility_hack(item, sess)
+    pretty_printing_compatibility_hack(&item, psess)
 }

compiler/rustc_expand/src/mbe/transcribe.rs

@@ -9,6 +9,7 @@ use rustc_ast::mut_visit::{self, MutVisitor};
 use rustc_ast::token::IdentIsRaw;
 use rustc_ast::token::{self, Delimiter, InvisibleOrigin, NonterminalKind, Token, TokenKind};
 use rustc_ast::tokenstream::{DelimSpacing, DelimSpan, Spacing, TokenStream, TokenTree};
+use rustc_ast::StmtKind;
 use rustc_data_structures::fx::FxHashMap;
 use rustc_errors::{pluralize, Diag, DiagCtxtHandle, PResult};
 use rustc_parse::parser::ParseNtResult;
@@ -296,6 +297,18 @@ pub(super) fn transcribe<'a>(
                             let kind = token::NtLifetime(*ident);
                             TokenTree::token_alone(kind, sp)
                         }
+                        MatchedSingle(ParseNtResult::Item(ref item)) => {
+                            mk_delimited(NonterminalKind::Item, TokenStream::from_ast(item))
+                        }
+                        MatchedSingle(ParseNtResult::Stmt(ref stmt)) => {
+                            let stream = if let StmtKind::Empty = stmt.kind {
+                                // FIXME: Properly collect tokens for empty statements.
+                                TokenStream::token_alone(token::Semi, stmt.span)
+                            } else {
+                                TokenStream::from_ast(stmt)
+                            };
+                            mk_delimited(NonterminalKind::Stmt, stream)
+                        }
                         MatchedSingle(ParseNtResult::PatParam(ref pat, inferred)) => mk_delimited(
                             NonterminalKind::PatParam { inferred: *inferred },
                             TokenStream::from_ast(pat),

compiler/rustc_expand/src/proc_macro.rs

@@ -123,7 +123,7 @@ impl MultiItemModifier for DeriveProcMacro {
         // We had a lint for a long time, but now we just emit a hard error.
         // Eventually we might remove the special case hard error check
         // altogether. See #73345.
-        crate::base::ann_pretty_printing_compatibility_hack(&item, &ecx.sess);
+        crate::base::ann_pretty_printing_compatibility_hack(&item, &ecx.sess.psess);
         let input = item.to_tokens();
         let stream = {
             let _timer =

compiler/rustc_expand/src/proc_macro_server.rs

@@ -114,11 +114,25 @@ impl FromInternal<(TokenStream, &mut Rustc<'_, '_>)> for Vec<TokenTree<TokenStre
 
         while let Some(tree) = cursor.next() {
             let (Token { kind, span }, joint) = match tree.clone() {
-                tokenstream::TokenTree::Delimited(span, _, delim, tts) => {
-                    let delimiter = pm::Delimiter::from_internal(delim);
+                tokenstream::TokenTree::Delimited(span, _, delim, stream) => {
+                    // We used to have an alternative behaviour for crates that
+                    // needed it: a hack used to pass AST fragments to
+                    // attribute and derive macros as a single nonterminal
+                    // token instead of a token stream. Such token needs to be
+                    // "unwrapped" and not represented as a delimited group. We
+                    // had a lint for a long time, but now we just emit a hard
+                    // error. Eventually we might remove the special case hard
+                    // error check altogether. See #73345.
+                    if let Delimiter::Invisible(InvisibleOrigin::MetaVar(kind)) = delim {
+                        crate::base::stream_pretty_printing_compatibility_hack(
+                            kind,
+                            &stream,
+                            rustc.psess(),
+                        );
+                    }
                     trees.push(TokenTree::Group(Group {
-                        delimiter,
-                        stream: Some(tts),
+                        delimiter: pm::Delimiter::from_internal(delim),
+                        stream: Some(stream),
                         span: DelimSpan {
                             open: span.open,
                             close: span.close,
@@ -277,15 +291,6 @@ impl FromInternal<(TokenStream, &mut Rustc<'_, '_>)> for Vec<TokenTree<TokenStre
 
                 Interpolated(nt) => {
                     let stream = TokenStream::from_nonterminal_ast(&nt);
-                    // We used to have an alternative behaviour for crates that
-                    // needed it: a hack used to pass AST fragments to
-                    // attribute and derive macros as a single nonterminal
-                    // token instead of a token stream. Such token needs to be
-                    // "unwrapped" and not represented as a delimited group. We
-                    // had a lint for a long time, but now we just emit a hard
-                    // error. Eventually we might remove the special case hard
-                    // error check altogether. See #73345.
-                    crate::base::nt_pretty_printing_compatibility_hack(&nt, rustc.ecx.sess);
                     trees.push(TokenTree::Group(Group {
                         delimiter: pm::Delimiter::None,
                         stream: Some(stream),

compiler/rustc_parse/src/parser/item.rs

@@ -1,11 +1,12 @@
 use super::diagnostics::{dummy_arg, ConsumeClosingDelim};
 use super::ty::{AllowPlus, RecoverQPath, RecoverReturnSign};
 use super::{
-    AttrWrapper, FollowedByType, ForceCollect, Parser, PathStyle, Trailing, TrailingToken,
+    AttrWrapper, FollowedByType, ForceCollect, ParseNtResult, Parser, PathStyle, Recovered,
+    Trailing, TrailingToken,
 };
 use crate::errors::{self, MacroExpandsToAdtField};
 use crate::fluent_generated as fluent;
-use crate::maybe_whole;
+use crate::maybe_reparse_metavar_seq;
 use ast::token::IdentIsRaw;
 use rustc_ast::ast::*;
 use rustc_ast::ptr::P;
@@ -123,10 +124,16 @@ impl<'a> Parser<'a> {
         fn_parse_mode: FnParseMode,
         force_collect: ForceCollect,
     ) -> PResult<'a, Option<Item>> {
-        maybe_whole!(self, NtItem, |item| {
+        if let Some(mut item) = maybe_reparse_metavar_seq!(
+            self,
+            NonterminalKind::Item,
+            NonterminalKind::Item,
+            ParseNtResult::Item(item),
+            item
+        ) {
             attrs.prepend_to_nt_inner(&mut item.attrs);
-            Some(item.into_inner())
-        });
+            return Ok(Some(item.into_inner()));
+        }
 
         let item =
             self.collect_tokens_trailing_token(attrs, force_collect, |this: &mut Self, attrs| {

compiler/rustc_parse/src/parser/mod.rs

@@ -1701,6 +1701,8 @@ pub enum ParseNtResult {
     Tt(TokenTree),
     Ident(Ident, IdentIsRaw),
     Lifetime(Ident),
+    Item(P<ast::Item>),
+    Stmt(P<ast::Stmt>),
     PatParam(P<ast::Pat>, /* inferred */ bool),
     PatWithOr(P<ast::Pat>),
     Ty(P<ast::Ty>),