bug#58847: Patch to properly parse c++11 multiline strings

[Eli Zaretskii]

> Date: Fri, 28 Oct 2022 16:13:42 -0400
> From:  Jan Stranik via "Bug reports for GNU Emacs,
>  the Swiss army knife of text editors" <bug-gnu-emacs@gnu.org>
> 
> Hello - 
> I’m happy user of emacs and ebrowse feature. Recently I noticed that ebrowse 
> does not work for multi-line strings in c++. 
> The r-string parsing is on also for c files, but it does not matter since c 
> does not have r strings. 

Thanks.  Gerd, any comments?

> EBROWSE: parse c++11 rstrings
> 
> C++11 allows definition of  multi-line stirngs. This patch makes ebrowse 
> propely parse these strings.
> 
> Example of test multi-line string:
> 
> repro.cxx:
> ----------
> struct Foo {
>     using STR = const char*;
>     STR rstrprefix = R"prefix(is is a C++11 multi
>     line string
> )prefix";
> 
>     STR rstr = R"(
> multiline string without a prefix
> )";
> 
>     STR rstr_test = R"prefix(
> )prefix not at end
> )prefixtoolong"
> )pref" to short
> 
> string still continues
> )prefix";
> 
>     const char* str = "a regular string";
> 
>     void func() {
>     }
> };
> ----------
> 
> ~/project/test/lit_repro $ c++ -std=c++10 -c repro.cxx     # repro.cxx 
> compiles
> 
> ~/project/test/lit_repro $ ebrowse repro.cxx               # current ebrowse 
> chokes on file and produces wrong symbols
> repro.cxx:3: newline in string constant
> repro.cxx:4: newline in string constant
> repro.cxx:7: newline in string constant
> repro.cxx:8: newline in string constant
> repro.cxx:11: newline in string constant
> repro.cxx:12: newline in string constant
> repro.cxx:14: newline in string constant
> repro.cxx:15: newline in string constant
> repro.cxx:16: newline in string constant
> ~/project/test/lit_repro $ cat BROWSE
> [ebrowse-hs "ebrowse 5.0" " -x" () ()][ebrowse-ts [ebrowse-cs "Foo" () 
> 0"repro.cxx" "struct Foo {" 12"repro.cxx" ]
> ()([ebrowse-ms "R" () 0 () "multiline string without a prefix
> )\";" 175 0  () () 0]
> [ebrowse-ms "pref" () 0 () ")prefix\";" 291 0  () () 0]
> [ebrowse-ms "str" () 0 () "    const char* str = \"a regular string\";" 334 0 
>  () () 0]
> )
> ([ebrowse-ms "func" () 0 () "    void func()" 351 0  () "    void func()" 351]
> )
> ~/project/test/lit_repro $  ~/Downloads/emacs-master/lib-src/ebrowse 
> repro.cxx    # patch properly parses source and generates symbols
> ~/project/test/lit_repro $ cat BROWSE
> [ebrowse-hs "ebrowse 5.0" " -x" () ()][ebrowse-ts [ebrowse-cs "Foo" () 
> 0"repro.cxx" "struct Foo {" 12"repro.cxx" ]
> ()([ebrowse-ms "rstr" () 0 () "multiline string without a prefix
> )\";" 175 0  () () 0]
> [ebrowse-ms "rstr_test" () 0 () ")prefix\";" 291 0  () () 0]
> [ebrowse-ms "rstrprefix" () 0 () ")prefix\";" 117 0  () () 0]
> [ebrowse-ms "str" () 0 () "    const char* str = \"a regular string\";" 334 0 
>  () () 0]
> )
> ([ebrowse-ms "func" () 0 () "    void func()" 351 0  () "    void func()" 351]
> )
> Index: emacs-master/lib-src/ebrowse.c
> ===================================================================
> --- emacs-master.orig/lib-src/ebrowse.c
> +++ emacs-master/lib-src/ebrowse.c
> @@ -1574,6 +1574,51 @@ yylex (void)
>  
>          end_string:
>            return end_char == '\'' ? CCHAR : CSTRING;
> +     case 'R':
> +       if (GET (c) == '"') {
> +         /* c++11 rstrings */
> +
> +            #define RSTRING_EOF_CHECK do {if (c=='\0') { 
> yyerror("unterminated c++11 rstring", NULL); UNGET(); return 
> CSTRING;}}while(0)
> +         char *rstring_prefix_start = in;
> +
> +         while (GET (c) != '(') {
> +           RSTRING_EOF_CHECK;
> +           if (c == '"')
> +             {
> +               yyerror ("malformed c++11 rstring", NULL);
> +               return CSTRING;
> +             }
> +         }
> +         char *rstring_prefix_end = in - 1;
> +         while (TRUE) {
> +           switch(GET (c)) {
> +           default:
> +             RSTRING_EOF_CHECK;
> +             break;
> +           case '\n':
> +             INCREMENT_LINENO;
> +             break;
> +           case ')':
> +             {
> +               char *in_saved = in;
> +               char *prefix = rstring_prefix_start;
> +               while (prefix != rstring_prefix_end && GET (c) == *prefix) {
> +                 RSTRING_EOF_CHECK;
> +                 prefix++;
> +               }
> +               if (prefix == rstring_prefix_end) {
> +                 if (GET(c) == '"')
> +                   return CSTRING;
> +                 RSTRING_EOF_CHECK;
> +               }
> +               in = in_saved;
> +             }
> +           }
> +         }
> +       }
> +
> +          UNGET ();
> +          /* fall through to ident */
>  
>          case 'a': case 'b': case 'c': case 'd': case 'e': case 'f': case 'g':
>          case 'h': case 'i': case 'j': case 'k': case 'l': case 'm': case 'n':
> @@ -1581,7 +1626,7 @@ yylex (void)
>          case 'v': case 'w': case 'x': case 'y': case 'z':
>          case 'A': case 'B': case 'C': case 'D': case 'E': case 'F': case 'G':
>          case 'H': case 'I': case 'J': case 'K': case 'L': case 'M': case 'N':
> -        case 'O': case 'P': case 'Q': case 'R': case 'S': case 'T': case 'U':
> +        case 'O': case 'P': case 'Q': case 'S': case 'T': case 'U':
>          case 'V': case 'W': case 'X': case 'Y': case 'Z': case '_':
>            {
>              /* Identifier and keywords.  */
>