public inbox for gcc-patches@gcc.gnu.org
 help / color / mirror / Atom feed
From: Marek Polacek <polacek@redhat.com>
To: GCC Patches <gcc-patches@gcc.gnu.org>, Jason Merrill <jason@redhat.com>
Subject: [PATCH] c++: Reject UDLs in certain contexts [PR105300]
Date: Sat, 12 Nov 2022 11:53:31 -0500	[thread overview]
Message-ID: <20221112165331.349041-1-polacek@redhat.com> (raw)

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.

Bootstrapped/regtested on x86_64-pc-linux-gnu, ok for trunk?

	PR c++/105300

gcc/c-family/ChangeLog:

	* c-pragma.cc (handle_pragma_message): Warn for CPP_STRING_USERDEF.

gcc/cp/ChangeLog:

	* parser.cc (cp_parser_string_literal): Add a bool parameter.
	Give an error when UDLs are not permitted.
	(cp_parser_primary_expression): Adjust the call to
	cp_parser_string_literal.
	(cp_parser_linkage_specification): Likewise.
	(cp_parser_static_assert): Likewise.
	(cp_parser_operator): Likewise.
	(cp_parser_asm_definition): Likewise.
	(cp_parser_asm_specification_opt): Likewise.
	(cp_parser_asm_operand_list): Likewise.
	(cp_parser_asm_clobber_list): Likewise.
	(cp_parser_omp_context_selector): Likewise.
	(pragma_lex): Likewise.

gcc/testsuite/ChangeLog:

	* g++.dg/cpp0x/udlit-error1.C: New test.
---
 gcc/c-family/c-pragma.cc                  |  3 +
 gcc/cp/parser.cc                          | 69 ++++++++++++++---------
 gcc/testsuite/g++.dg/cpp0x/udlit-error1.C | 21 +++++++
 3 files changed, 65 insertions(+), 28 deletions(-)
 create mode 100644 gcc/testsuite/g++.dg/cpp0x/udlit-error1.C

diff --git a/gcc/c-family/c-pragma.cc b/gcc/c-family/c-pragma.cc
index 142a46441ac..49f405b605b 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 e4021835ed5..ae2798e2a33 100644
--- a/gcc/cp/parser.cc
+++ b/gcc/cp/parser.cc
@@ -2226,7 +2226,7 @@ pop_unparsed_function_queues (cp_parser *parser)
 static cp_expr cp_parser_identifier
   (cp_parser *);
 static cp_expr cp_parser_string_literal
-  (cp_parser *, bool, bool, bool);
+  (cp_parser *, bool, bool, bool, bool);
 static cp_expr cp_parser_userdef_char_literal
   (cp_parser *);
 static tree cp_parser_userdef_string_literal
@@ -4402,7 +4402,8 @@ cp_parser_identifier (cp_parser* parser)
    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.
@@ -4414,7 +4415,7 @@ cp_parser_identifier (cp_parser* parser)
    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)
+			  bool udl_ok, bool lookup_udlit = true)
 {
   tree value;
   size_t count;
@@ -4439,6 +4440,12 @@ 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;
@@ -5655,7 +5662,7 @@ cp_parser_primary_expression (cp_parser *parser,
 	 argument to cp_parser_string_literal.  */
       return (cp_parser_string_literal (parser,
 					parser->translate_strings_p,
-					true)
+					/*wide_ok=*/true, /*udl_ok=*/true)
 	      .maybe_add_location_wrapper ());
 
     case CPP_OPEN_PAREN:
@@ -16161,15 +16168,14 @@ 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, /*udl_ok=*/false);
 
   /* Transform the literal into an identifier.  If the literal is a
      wide-character string, or contains embedded NULs, then we can't
@@ -16300,8 +16306,8 @@ cp_parser_static_assert(cp_parser *parser, bool member_p)
 
       /* Parse the string-literal message.  */
       message = cp_parser_string_literal (parser,
-                                	  /*translate=*/false,
-                                	  /*wide_ok=*/true);
+					  /*translate=*/false, /*wide_ok=*/true,
+					  /*udl_ok=*/false);
 
       /* A `)' completes the static assertion.  */
       if (!parens.require_close (parser))
@@ -17349,7 +17355,6 @@ 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;
 
@@ -17357,8 +17362,10 @@ 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_string_literal (parser, /*translate=*/true,
+						/*wide_ok=*/true,
+						/*udl_ok=*/true,
+						/*lookup_udlit=*/false);
 	if (str == error_mark_node)
 	  return error_mark_node;
 	else if (TREE_CODE (str) == USERDEF_LITERAL)
@@ -21975,7 +21982,6 @@ 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;
@@ -22083,7 +22089,8 @@ 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, /*udl_ok=*/false);
   if (string == error_mark_node)
     {
       cp_parser_skip_to_closing_parenthesis (parser, true, false,
@@ -28485,11 +28492,8 @@ 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))
@@ -28502,7 +28506,10 @@ 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,
+						     /*udl_ok=*/false);
 
   /* Look for the `)'.  */
   parens.require_close (parser);
@@ -28535,8 +28542,6 @@ 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))
@@ -28554,13 +28559,16 @@ 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,
+						      /*udl_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);
 
@@ -28600,10 +28608,11 @@ 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,
+						      /*udl_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
@@ -45761,7 +45770,10 @@ 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,
+						      /*udl_ok=*/false);
 		  else
 		    {
 		      cp_parser_error (parser, "expected identifier or "
@@ -48783,7 +48795,8 @@ 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, /*udl_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" }
+}

base-commit: f232715d15618e91c90eb210e23de10909590944
-- 
2.38.1


             reply	other threads:[~2022-11-12 16:53 UTC|newest]

Thread overview: 13+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-11-12 16:53 Marek Polacek [this message]
2022-11-15 23:58 ` Jason Merrill
2022-11-16  0:35   ` Marek Polacek
2022-11-16 13:22     ` Jason Merrill
2022-11-17  1:12       ` Marek Polacek
2022-11-18  0:06         ` Jason Merrill
2022-11-18 23:52           ` [PATCH v2] " Marek Polacek
2022-11-19  1:39             ` Jason Merrill
2022-12-02 23:58               ` [PATCH v3] " Marek Polacek
2022-12-03 19:58                 ` Jason Merrill
2023-01-13 23:22                   ` [PATCH v4] " Marek Polacek
2023-01-24 22:49                     ` Marek Polacek
2023-01-25 19:36                     ` Jason Merrill

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20221112165331.349041-1-polacek@redhat.com \
    --to=polacek@redhat.com \
    --cc=gcc-patches@gcc.gnu.org \
    --cc=jason@redhat.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
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).