public inbox for gcc-cvs@sourceware.org
help / color / mirror / Atom feed
* [gcc r13-5390] c++: Reject UDLs in certain contexts [PR105300]
@ 2023-01-26 15:08 Marek Polacek
  0 siblings, 0 replies; only message in thread
From: Marek Polacek @ 2023-01-26 15:08 UTC (permalink / raw)
  To: gcc-cvs

https://gcc.gnu.org/g:9f353b0c1dc9385ba8b8a64b65d66d5452383c11

commit r13-5390-g9f353b0c1dc9385ba8b8a64b65d66d5452383c11
Author: Marek Polacek <polacek@redhat.com>
Date:   Fri Nov 11 17:59:30 2022 -0500

    c++: Reject UDLs in certain contexts [PR105300]
    
    In this PR, we are crashing because we've encountered a UDL where a
    string-literal is expected.  This patch makes the parser reject string
    and character UDLs in all places where the grammar requires a
    string-literal and not a user-defined-string-literal.
    
    I've introduced two new wrappers; the existing cp_parser_string_literal
    was renamed to cp_parser_string_literal_common and should not be called
    directly.  finish_userdef_string_literal is renamed from
    cp_parser_userdef_string_literal.
    
            PR c++/105300
    
    gcc/c-family/ChangeLog:
    
            * c-pragma.cc (handle_pragma_message): Warn for CPP_STRING_USERDEF.
    
    gcc/cp/ChangeLog:
    
            * parser.cc: Remove unnecessary forward declarations.
            (cp_parser_string_literal): New wrapper.
            (cp_parser_string_literal_common): Renamed from
            cp_parser_string_literal.  Add a bool parameter.  Give an error when
            UDLs are not permitted.
            (cp_parser_userdef_string_literal): New wrapper.
            (finish_userdef_string_literal): Renamed from
            cp_parser_userdef_string_literal.
            (cp_parser_primary_expression): Call cp_parser_userdef_string_literal
            instead of cp_parser_string_literal.
            (cp_parser_linkage_specification): Move a variable declaration closer
            to its first use.
            (cp_parser_static_assert): Likewise.
            (cp_parser_operator): Call cp_parser_userdef_string_literal instead of
            cp_parser_string_literal.
            (cp_parser_asm_definition): Move a variable declaration closer to its
            first use.
            (cp_parser_asm_specification_opt): Move variable declarations closer to
            their first use.
            (cp_parser_asm_operand_list): Likewise.
            (cp_parser_asm_clobber_list): Likewise.
    
    gcc/testsuite/ChangeLog:
    
            * g++.dg/cpp0x/udlit-error1.C: New test.

Diff:
---
 gcc/c-family/c-pragma.cc                  |   3 +
 gcc/cp/parser.cc                          | 133 ++++++++++++++++++++----------
 gcc/testsuite/g++.dg/cpp0x/udlit-error1.C |  21 +++++
 3 files changed, 113 insertions(+), 44 deletions(-)

diff --git a/gcc/c-family/c-pragma.cc b/gcc/c-family/c-pragma.cc
index 68d1c8bda13..0d2b333cebb 100644
--- a/gcc/c-family/c-pragma.cc
+++ b/gcc/c-family/c-pragma.cc
@@ -1390,6 +1390,9 @@ handle_pragma_message (cpp_reader *)
     }
   else if (token == CPP_STRING)
     message = x;
+  else if (token == CPP_STRING_USERDEF)
+    GCC_BAD ("string literal with user-defined suffix is invalid in this "
+	     "context");
   else
     GCC_BAD ("expected a string after %<#pragma message%>");
 
diff --git a/gcc/cp/parser.cc b/gcc/cp/parser.cc
index e0c46d2d748..4cdc1cd472f 100644
--- a/gcc/cp/parser.cc
+++ b/gcc/cp/parser.cc
@@ -2227,16 +2227,8 @@ pop_unparsed_function_queues (cp_parser *parser)
 
 /* Lexical conventions [gram.lex]  */
 
-static cp_expr cp_parser_identifier
-  (cp_parser *);
-static cp_expr cp_parser_string_literal
-  (cp_parser *, bool, bool, bool);
-static cp_expr cp_parser_userdef_char_literal
-  (cp_parser *);
-static tree cp_parser_userdef_string_literal
+static tree finish_userdef_string_literal
   (tree);
-static cp_expr cp_parser_userdef_numeric_literal
-  (cp_parser *);
 
 /* Basic concepts [gram.basic]  */
 
@@ -4408,11 +4400,15 @@ cp_parser_identifier (cp_parser* parser)
     return error_mark_node;
 }
 
-/* Parse a sequence of adjacent string constants.  Returns a
+/* Worker for cp_parser_string_literal and cp_parser_userdef_string_literal.
+   Do not call this directly; use either of the above.
+
+   Parse a sequence of adjacent string constants.  Return a
    TREE_STRING representing the combined, nul-terminated string
    constant.  If TRANSLATE is true, translate the string to the
    execution character set.  If WIDE_OK is true, a wide string is
-   invalid here.
+   valid here.  If UDL_OK is true, a string literal with user-defined
+   suffix can be used in this context.
 
    C++98 [lex.string] says that if a narrow string literal token is
    adjacent to a wide string literal token, the behavior is undefined.
@@ -4422,9 +4418,11 @@ cp_parser_identifier (cp_parser* parser)
    This code is largely lifted from lex_string() in c-lex.cc.
 
    FUTURE: ObjC++ will need to handle @-strings here.  */
+
 static cp_expr
-cp_parser_string_literal (cp_parser *parser, bool translate, bool wide_ok,
-			  bool lookup_udlit = true)
+cp_parser_string_literal_common (cp_parser *parser, bool translate,
+				 bool wide_ok, bool udl_ok,
+				 bool lookup_udlit)
 {
   tree value;
   size_t count;
@@ -4449,6 +4447,12 @@ cp_parser_string_literal (cp_parser *parser, bool translate, bool wide_ok,
 
   if (cpp_userdef_string_p (tok->type))
     {
+      if (!udl_ok)
+	{
+	  error_at (loc, "string literal with user-defined suffix "
+		    "is invalid in this context");
+	  return error_mark_node;
+	}
       string_tree = USERDEF_LITERAL_VALUE (tok->u.value);
       curr_type = cpp_userdef_string_remove_type (tok->type);
       curr_tok_is_userdef_p = true;
@@ -4539,6 +4543,12 @@ cp_parser_string_literal (cp_parser *parser, bool translate, bool wide_ok,
 	  tok = cp_lexer_peek_token (parser->lexer);
 	  if (cpp_userdef_string_p (tok->type))
 	    {
+	      if (!udl_ok)
+		{
+		  error_at (loc, "string literal with user-defined suffix "
+			    "is invalid in this context");
+		  return error_mark_node;
+		}
 	      string_tree = USERDEF_LITERAL_VALUE (tok->u.value);
 	      curr_type = cpp_userdef_string_remove_type (tok->type);
 	      curr_tok_is_userdef_p = true;
@@ -4608,7 +4618,7 @@ cp_parser_string_literal (cp_parser *parser, bool translate, bool wide_ok,
 	  tree literal = build_userdef_literal (suffix_id, value,
 						OT_NONE, NULL_TREE);
 	  if (lookup_udlit)
-	    value = cp_parser_userdef_string_literal (literal);
+	    value = finish_userdef_string_literal (literal);
 	  else
 	    value = literal;
 	}
@@ -4626,6 +4636,37 @@ cp_parser_string_literal (cp_parser *parser, bool translate, bool wide_ok,
   return cp_expr (value, loc);
 }
 
+/* Parse a sequence of adjacent string constants.  Return a TREE_STRING
+   representing the combined, nul-terminated string constant.  If
+   TRANSLATE is true, translate the string to the execution character set.
+   If WIDE_OK is true, a wide string is valid here.
+
+   This function issues an error if a user defined string literal is
+   encountered; use cp_parser_userdef_string_literal if UDLs are allowed.  */
+
+static inline cp_expr
+cp_parser_string_literal (cp_parser *parser, bool translate, bool wide_ok)
+{
+  return cp_parser_string_literal_common (parser, translate, wide_ok,
+					  /*udl_ok=*/false,
+					  /*lookup_udlit=*/false);
+}
+
+/* Parse a string literal or user defined string literal.
+
+   user-defined-string-literal :
+     string-literal ud-suffix
+
+   If LOOKUP_UDLIT, perform a lookup for a suitable template function.  */
+
+static inline cp_expr
+cp_parser_userdef_string_literal (cp_parser *parser, bool lookup_udlit)
+{
+  return cp_parser_string_literal_common (parser, /*translate=*/true,
+					  /*wide_ok=*/true, /*udl_ok=*/true,
+					  lookup_udlit);
+}
+
 /* Look up a literal operator with the name and the exact arguments.  */
 
 static tree
@@ -4923,7 +4964,7 @@ cp_parser_userdef_numeric_literal (cp_parser *parser)
    as arguments.  */
 
 static tree
-cp_parser_userdef_string_literal (tree literal)
+finish_userdef_string_literal (tree literal)
 {
   tree suffix_id = USERDEF_LITERAL_SUFFIX_ID (literal);
   tree name = cp_literal_operator_id (IDENTIFIER_POINTER (suffix_id));
@@ -5663,10 +5704,15 @@ cp_parser_primary_expression (cp_parser *parser,
       /* ??? Should wide strings be allowed when parser->translate_strings_p
 	 is false (i.e. in attributes)?  If not, we can kill the third
 	 argument to cp_parser_string_literal.  */
-      return (cp_parser_string_literal (parser,
-					parser->translate_strings_p,
-					true)
-	      .maybe_add_location_wrapper ());
+      if (parser->translate_strings_p)
+	return (cp_parser_userdef_string_literal (parser,
+						  /*lookup_udlit=*/true)
+		.maybe_add_location_wrapper ());
+      else
+	return (cp_parser_string_literal (parser,
+					  /*translate=*/false,
+					  /*wide_ok=*/true)
+		.maybe_add_location_wrapper ());
 
     case CPP_OPEN_PAREN:
       /* If we see `( { ' then we are looking at the beginning of
@@ -16222,15 +16268,14 @@ cp_parser_function_specifier_opt (cp_parser* parser,
 static void
 cp_parser_linkage_specification (cp_parser* parser, tree prefix_attr)
 {
-  tree linkage;
-
   /* Look for the `extern' keyword.  */
   cp_token *extern_token
     = cp_parser_require_keyword (parser, RID_EXTERN, RT_EXTERN);
 
   /* Look for the string-literal.  */
   cp_token *string_token = cp_lexer_peek_token (parser->lexer);
-  linkage = cp_parser_string_literal (parser, false, false);
+  tree linkage = cp_parser_string_literal (parser, /*translate=*/false,
+					   /*wide_ok=*/false);
 
   /* Transform the literal into an identifier.  If the literal is a
      wide-character string, or contains embedded NULs, then we can't
@@ -16360,9 +16405,8 @@ cp_parser_static_assert(cp_parser *parser, bool member_p)
       cp_parser_require (parser, CPP_COMMA, RT_COMMA);
 
       /* Parse the string-literal message.  */
-      message = cp_parser_string_literal (parser,
-                                	  /*translate=*/false,
-                                	  /*wide_ok=*/true);
+      message = cp_parser_string_literal (parser, /*translate=*/false,
+					  /*wide_ok=*/true);
 
       /* A `)' completes the static assertion.  */
       if (!parens.require_close (parser))
@@ -17410,7 +17454,6 @@ cp_parser_operator (cp_parser* parser, location_t start_loc)
     case CPP_STRING16_USERDEF:
     case CPP_STRING32_USERDEF:
       {
-	cp_expr str;
 	tree string_tree;
 	int sz, len;
 
@@ -17418,8 +17461,8 @@ cp_parser_operator (cp_parser* parser, location_t start_loc)
 	  maybe_warn_cpp0x (CPP0X_USER_DEFINED_LITERALS);
 
 	/* Consume the string.  */
-	str = cp_parser_string_literal (parser, /*translate=*/true,
-				      /*wide_ok=*/true, /*lookup_udlit=*/false);
+	cp_expr str = cp_parser_userdef_string_literal (parser,
+							/*lookup_udlit=*/false);
 	if (str == error_mark_node)
 	  return error_mark_node;
 	else if (TREE_CODE (str) == USERDEF_LITERAL)
@@ -22072,7 +22115,6 @@ cp_parser_using_directive (cp_parser* parser)
 static void
 cp_parser_asm_definition (cp_parser* parser)
 {
-  tree string;
   tree outputs = NULL_TREE;
   tree inputs = NULL_TREE;
   tree clobbers = NULL_TREE;
@@ -22180,7 +22222,8 @@ cp_parser_asm_definition (cp_parser* parser)
   if (!cp_parser_require (parser, CPP_OPEN_PAREN, RT_OPEN_PAREN))
     return;
   /* Look for the string.  */
-  string = cp_parser_string_literal (parser, false, false);
+  tree string = cp_parser_string_literal (parser, /*translate=*/false,
+					  /*wide_ok=*/false);
   if (string == error_mark_node)
     {
       cp_parser_skip_to_closing_parenthesis (parser, true, false,
@@ -28655,11 +28698,8 @@ cp_parser_yield_expression (cp_parser* parser)
 static tree
 cp_parser_asm_specification_opt (cp_parser* parser)
 {
-  cp_token *token;
-  tree asm_specification;
-
   /* Peek at the next token.  */
-  token = cp_lexer_peek_token (parser->lexer);
+  cp_token *token = cp_lexer_peek_token (parser->lexer);
   /* If the next token isn't the `asm' keyword, then there's no
      asm-specification.  */
   if (!cp_parser_is_keyword (token, RID_ASM))
@@ -28672,7 +28712,9 @@ cp_parser_asm_specification_opt (cp_parser* parser)
   parens.require_open (parser);
 
   /* Look for the string-literal.  */
-  asm_specification = cp_parser_string_literal (parser, false, false);
+  tree asm_specification = cp_parser_string_literal (parser,
+						     /*translate=*/false,
+						     /*wide_ok=*/false);
 
   /* Look for the `)'.  */
   parens.require_close (parser);
@@ -28705,8 +28747,6 @@ cp_parser_asm_operand_list (cp_parser* parser)
 
   while (true)
     {
-      tree string_literal;
-      tree expression;
       tree name;
 
       if (cp_lexer_next_token_is (parser->lexer, CPP_OPEN_SQUARE))
@@ -28724,13 +28764,15 @@ cp_parser_asm_operand_list (cp_parser* parser)
       else
 	name = NULL_TREE;
       /* Look for the string-literal.  */
-      string_literal = cp_parser_string_literal (parser, false, false);
+      tree string_literal = cp_parser_string_literal (parser,
+						      /*translate=*/false,
+						      /*wide_ok=*/false);
 
       /* Look for the `('.  */
       matching_parens parens;
       parens.require_open (parser);
       /* Parse the expression.  */
-      expression = cp_parser_expression (parser);
+      tree expression = cp_parser_expression (parser);
       /* Look for the `)'.  */
       parens.require_close (parser);
 
@@ -28770,10 +28812,10 @@ cp_parser_asm_clobber_list (cp_parser* parser)
 
   while (true)
     {
-      tree string_literal;
-
       /* Look for the string literal.  */
-      string_literal = cp_parser_string_literal (parser, false, false);
+      tree string_literal = cp_parser_string_literal (parser,
+						      /*translate=*/false,
+						      /*wide_ok=*/false);
       /* Add it to the list.  */
       clobbers = tree_cons (NULL_TREE, string_literal, clobbers);
       /* If the next token is not a `,', then the list is
@@ -46372,7 +46414,9 @@ cp_parser_omp_context_selector (cp_parser *parser, tree set, bool has_parms_p)
 		      cp_lexer_consume_token (parser->lexer);
 		    }
 		  else if (cp_lexer_next_token_is (parser->lexer, CPP_STRING))
-		    value = cp_parser_string_literal (parser, false, false);
+		    value = cp_parser_string_literal (parser,
+						      /*translate=*/false,
+						      /*wide_ok=*/false);
 		  else
 		    {
 		      cp_parser_error (parser, "expected identifier or "
@@ -49394,7 +49438,8 @@ pragma_lex (tree *value, location_t *loc)
   if (ret == CPP_PRAGMA_EOL)
     ret = CPP_EOF;
   else if (ret == CPP_STRING)
-    *value = cp_parser_string_literal (the_parser, false, false);
+    *value = cp_parser_string_literal (the_parser, /*translate=*/false,
+				       /*wide_ok=*/false);
   else
     {
       if (ret == CPP_KEYWORD)
diff --git a/gcc/testsuite/g++.dg/cpp0x/udlit-error1.C b/gcc/testsuite/g++.dg/cpp0x/udlit-error1.C
new file mode 100644
index 00000000000..66e300e350f
--- /dev/null
+++ b/gcc/testsuite/g++.dg/cpp0x/udlit-error1.C
@@ -0,0 +1,21 @@
+// PR c++/105300
+// { dg-do compile { target c++11 } }
+
+void operator""_x(const char *, decltype(sizeof(0)));
+
+#include ""_x		  // { dg-error "include expects" }
+#line ""_x		  // { dg-error "not a positive integer" }
+#if __has_include(""_x)	  // { dg-error "requires a header-name" }
+#endif
+
+#pragma message "hi"_x	  // { dg-warning "string literal with user-defined suffix is invalid in this context" }
+
+extern "C"_x { void g(); } // { dg-error "before user-defined string literal" }
+static_assert(true, "foo"_x); // { dg-error "string literal with user-defined suffix is invalid in this context|expected" }
+
+[[deprecated("oof"_x)]]
+void
+lol () // { dg-error "not a string" }
+{
+  asm (""_x); // { dg-error "string literal with user-defined suffix is invalid in this context" }
+}

^ permalink raw reply	[flat|nested] only message in thread

only message in thread, other threads:[~2023-01-26 15:08 UTC | newest]

Thread overview: (only message) (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2023-01-26 15:08 [gcc r13-5390] c++: Reject UDLs in certain contexts [PR105300] Marek Polacek

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).