[gcc r14-8110] gccrs: Fix token lexed as a float literal
Arthur Cohen
cohenarthur@gcc.gnu.org
Tue Jan 16 18:19:30 GMT 2024
https://gcc.gnu.org/g:311121156e27e156c71bbbe2f12e35b1618ed100
commit r14-8110-g311121156e27e156c71bbbe2f12e35b1618ed100
Author: Pierre-Emmanuel Patry <pierre-emmanuel.patry@embecosm.com>
Date: Tue Oct 17 13:24:54 2023 +0200
gccrs: Fix token lexed as a float literal
The lexer cannot distinguish the difference between a float literal and a
tuple index in some cases. This means we should fix this while parsing
depending on the context.
gcc/rust/ChangeLog:
* expand/rust-macro-invoc-lexer.cc (MacroInvocLexer::split_current_token):
Add implementation for multiple token split.
* expand/rust-macro-invoc-lexer.h: Add function prototype.
* expand/rust-proc-macro-invoc-lexer.cc (ProcMacroInvocLexer::split_current_token):
Add implementation for 2+ token split for procedural macros.
* expand/rust-proc-macro-invoc-lexer.h: Add function prototype.
* lex/rust-lex.cc (Lexer::split_current_token): Add function to split a
token in multiple other tokens.
* lex/rust-lex.h: Add function prototype for split_current_token.
* parse/rust-parse-impl.h (Parser::left_denotation): Handle float tuple
index identified as a float literal.
Signed-off-by: Pierre-Emmanuel Patry <pierre-emmanuel.patry@embecosm.com>
Diff:
---
gcc/rust/expand/rust-macro-invoc-lexer.cc | 16 ++++++++++++++++
gcc/rust/expand/rust-macro-invoc-lexer.h | 2 ++
gcc/rust/expand/rust-proc-macro-invoc-lexer.cc | 15 +++++++++++++++
gcc/rust/expand/rust-proc-macro-invoc-lexer.h | 2 ++
gcc/rust/lex/rust-lex.cc | 12 ++++++++++++
gcc/rust/lex/rust-lex.h | 2 ++
gcc/rust/parse/rust-parse-impl.h | 22 ++++++++++++++++++++++
7 files changed, 71 insertions(+)
diff --git a/gcc/rust/expand/rust-macro-invoc-lexer.cc b/gcc/rust/expand/rust-macro-invoc-lexer.cc
index aa1e1ff2ebd..003c87d62ff 100644
--- a/gcc/rust/expand/rust-macro-invoc-lexer.cc
+++ b/gcc/rust/expand/rust-macro-invoc-lexer.cc
@@ -49,6 +49,22 @@ MacroInvocLexer::split_current_token (TokenId new_left, TokenId new_right)
std::unique_ptr<AST::Token> (new AST::Token (l_tok)));
}
+void
+MacroInvocLexer::split_current_token (std::vector<TokenPtr> new_tokens)
+{
+ rust_assert (new_tokens.size () > 0);
+
+ auto current_pos = token_stream.begin () + offs;
+
+ token_stream.erase (current_pos);
+
+ for (size_t i = 1; i < new_tokens.size (); i++)
+ {
+ token_stream.insert (current_pos + i, std::unique_ptr<AST::Token> (
+ new AST::Token (new_tokens[i])));
+ }
+}
+
std::vector<std::unique_ptr<AST::Token>>
MacroInvocLexer::get_token_slice (size_t start_idx, size_t end_idx) const
{
diff --git a/gcc/rust/expand/rust-macro-invoc-lexer.h b/gcc/rust/expand/rust-macro-invoc-lexer.h
index a65816b5ef5..4dc6d691c11 100644
--- a/gcc/rust/expand/rust-macro-invoc-lexer.h
+++ b/gcc/rust/expand/rust-macro-invoc-lexer.h
@@ -67,6 +67,8 @@ public:
// this will only work with "simple" tokens like punctuation.
void split_current_token (TokenId new_left, TokenId new_right);
+ void split_current_token (std::vector<TokenPtr> new_tokens);
+
std::vector<std::unique_ptr<AST::Token>>
get_token_slice (size_t start_idx, size_t end_idx) const;
};
diff --git a/gcc/rust/expand/rust-proc-macro-invoc-lexer.cc b/gcc/rust/expand/rust-proc-macro-invoc-lexer.cc
index 5990dec4325..1c370c10b29 100644
--- a/gcc/rust/expand/rust-proc-macro-invoc-lexer.cc
+++ b/gcc/rust/expand/rust-proc-macro-invoc-lexer.cc
@@ -47,4 +47,19 @@ ProcMacroInvocLexer::split_current_token (TokenId new_left, TokenId new_right)
token_stream.insert (current_pos, r_tok);
}
+void
+ProcMacroInvocLexer::split_current_token (std::vector<TokenPtr> new_tokens)
+{
+ rust_assert (new_tokens.size () > 0);
+
+ auto current_pos = token_stream.begin () + offs;
+
+ token_stream.erase (current_pos);
+
+ for (size_t i = 1; i < new_tokens.size (); i++)
+ {
+ token_stream.insert (current_pos + i, new_tokens[i]);
+ }
+}
+
} // namespace Rust
diff --git a/gcc/rust/expand/rust-proc-macro-invoc-lexer.h b/gcc/rust/expand/rust-proc-macro-invoc-lexer.h
index 5a11a4c111c..94eb6d5bfe2 100644
--- a/gcc/rust/expand/rust-proc-macro-invoc-lexer.h
+++ b/gcc/rust/expand/rust-proc-macro-invoc-lexer.h
@@ -40,6 +40,8 @@ public:
// closes (i.e. T<U<X>> where >> is wrongly lexed as one token). Note that
// this will only work with "simple" tokens like punctuation.
void split_current_token (TokenId new_left, TokenId new_right);
+
+ void split_current_token (std::vector<TokenPtr> new_tokens);
};
} // namespace Rust
diff --git a/gcc/rust/lex/rust-lex.cc b/gcc/rust/lex/rust-lex.cc
index 3883f2379fc..2cfc9cec5d4 100644
--- a/gcc/rust/lex/rust-lex.cc
+++ b/gcc/rust/lex/rust-lex.cc
@@ -2529,6 +2529,18 @@ Lexer::split_current_token (TokenId new_left, TokenId new_right)
token_queue.insert (1, std::move (new_right_tok));
}
+void
+Lexer::split_current_token (std::vector<TokenPtr> new_tokens)
+{
+ rust_assert (new_tokens.size () > 0);
+ token_queue.replace_current_value (new_tokens[0]);
+
+ for (size_t i = 1; i < new_tokens.size (); i++)
+ {
+ token_queue.insert (i, new_tokens[i]);
+ }
+}
+
void
Lexer::start_line (int current_line, int current_column)
{
diff --git a/gcc/rust/lex/rust-lex.h b/gcc/rust/lex/rust-lex.h
index 683e8c67a89..feaa3e05fc9 100644
--- a/gcc/rust/lex/rust-lex.h
+++ b/gcc/rust/lex/rust-lex.h
@@ -202,6 +202,8 @@ public:
* this will only work with "simple" tokens like punctuation. */
void split_current_token (TokenId new_left, TokenId new_right);
+ void split_current_token (std::vector<TokenPtr> new_tokens);
+
Linemap *get_line_map () { return line_map; }
std::string get_filename () { return std::string (input.get_filename ()); }
diff --git a/gcc/rust/parse/rust-parse-impl.h b/gcc/rust/parse/rust-parse-impl.h
index 4738393fd42..eb0310c888f 100644
--- a/gcc/rust/parse/rust-parse-impl.h
+++ b/gcc/rust/parse/rust-parse-impl.h
@@ -12895,6 +12895,28 @@ Parser<ManagedTokenSource>::left_denotation (const_TokenPtr tok,
std::move (outer_attrs),
restrictions);
}
+ else if (next_tok->get_id () == FLOAT_LITERAL)
+ {
+ // Lexer has misidentified a tuple index as a float literal
+ // eg: `(x, (y, z)).1.0` -> 1.0 has been identified as a float
+ // literal. This means we should split it into three new separate
+ // tokens, the first tuple index, the dot and the second tuple
+ // index.
+ auto current_loc = next_tok->get_locus ();
+ auto str = next_tok->get_str ();
+ auto dot_pos = str.find (".");
+ auto prefix = str.substr (0, dot_pos);
+ auto suffix = str.substr (dot_pos + 1);
+ lexer.split_current_token (
+ {Token::make_int (current_loc, std::move (prefix),
+ CORETYPE_PURE_DECIMAL),
+ Token::make (DOT, current_loc + 1),
+ Token::make_int (current_loc + 2, std::move (suffix),
+ CORETYPE_PURE_DECIMAL)});
+ return parse_tuple_index_expr (tok, std::move (left),
+ std::move (outer_attrs),
+ restrictions);
+ }
else if (next_tok->get_id () == IDENTIFIER
&& lexer.peek_token (1)->get_id () != LEFT_PAREN
&& lexer.peek_token (1)->get_id () != SCOPE_RESOLUTION)
More information about the Gcc-cvs
mailing list