clang/test/Preprocessor/ucn-pp-identifier.c

   1 // RUN: %clang_cc1 %s -fsyntax-only -std=c99 -pedantic -verify=expected,ext -Wundef -DTRIGRAPHS=1
   2 // RUN: %clang_cc1 %s -fsyntax-only -std=c2x -pedantic -verify=expected,ext -Wundef -DTRIGRAPHS=1
   3 // RUN: %clang_cc1 %s -fsyntax-only -x c++ -pedantic -verify=expected,ext -Wundef -fno-trigraphs
   4 // RUN: %clang_cc1 %s -fsyntax-only -x c++ -std=c++23 -pedantic -ftrigraphs -DTRIGRAPHS=1 -verify=expected,cxx23 -Wundef -Wpre-c++23-compat
   5 // RUN: %clang_cc1 %s -fsyntax-only -x c++ -pedantic -verify=expected,ext -Wundef -ftrigraphs -DTRIGRAPHS=1
   6 // RUN: not %clang_cc1 %s -fsyntax-only -std=c99 -pedantic -Wundef 2>&1 | FileCheck -strict-whitespace %s
   7
   8 #define \u00FC
   9 #define a\u00FD() 0
  10 #ifndef \u00FC
  11 #error "This should never happen"
  12 #endif
  13
  14 #if a\u00FD()
  15 #error "This should never happen"
  16 #endif
  17
  18 #if a\U000000FD()
  19 #error "This should never happen"
  20 #endif
  21
  22 #if a\u{FD}() // ext-warning {{extension}} cxx23-warning {{before C++23}}
  23 #error "This should never happen"
  24 #endif
  25
  26 #if \uarecool // expected-warning{{incomplete universal character name; treating as '\' followed by identifier}} expected-error {{invalid token at start of a preprocessor expression}}
  27 #endif
  28 #if \uwerecool // expected-warning{{\u used with no following hex digits; treating as '\' followed by identifier}} expected-error {{invalid token at start of a preprocessor expression}}
  29 #endif
  30 #if \U0001000  // expected-warning{{incomplete universal character name; treating as '\' followed by identifier}} expected-error {{invalid token at start of a preprocessor expression}}
  31 #endif
  32
  33 // Make sure we reject disallowed UCNs
  34 #define \ufffe // expected-error {{macro name must be an identifier}}
  35 #define \U10000000      // expected-error {{macro name must be an identifier}}
  36 #define \u0061          // expected-error {{character 'a' cannot be specified by a universal character name}} expected-error {{macro name must be an identifier}}
  37 #define \u{fffe}        // expected-error {{macro name must be an identifier}} \
  38                         // ext-warning {{extension}} cxx23-warning {{before C++23}}
  39 #define \N{ALERT}       // expected-error {{universal character name refers to a control character}} \
  40                    // expected-error {{macro name must be an identifier}} \
  41                    // ext-warning {{extension}} cxx23-warning {{before C++23}}
  42 #define \N{WASTEBASKET} // expected-error {{macro name must be an identifier}} \
  43                         // ext-warning {{extension}} cxx23-warning {{before C++23}}
  44 #define a\u0024a  // expected-error {{character '$' cannot be specified by a universal character name}} \
  45                   // expected-warning {{requires whitespace after the macro name}}
  46
  47 #if \u0110 // expected-warning {{is not defined, evaluates to 0}}
  48 #endif
  49
  50
  51 #define \u0110 1 / 0
  52 #if \u0110 // expected-error {{division by zero in preprocessor expression}}
  53 #endif
  54
  55 #define STRINGIZE(X) # X
  56
  57 extern int check_size[sizeof(STRINGIZE(\u0112)) == 3 ? 1 : -1];
  58
  59 // Check that we still diagnose disallowed UCNs in #if 0 blocks.
  60 // C99 5.1.1.2p1 and C++11 [lex.phases]p1 dictate that preprocessor tokens are
  61 // formed before directives are parsed.
  62 // expected-error@+4 {{character 'a' cannot be specified by a universal character name}}
  63 #if 0
  64 #define \ufffe // okay
  65 #define \U10000000 // okay
  66 #define \u0061 // error, but -verify only looks at comments outside #if 0
  67 #endif
  68
  69
  70 // A UCN formed by token pasting is undefined in both C99 and C++.
  71 // Right now we don't do anything special, which causes us to coincidentally
  72 // accept the first case below but reject the second two.
  73 #define PASTE(A, B) A ## B
  74 extern int PASTE(\, u00FD);
  75 extern int PASTE(\u, 00FD); // expected-warning{{\u used with no following hex digits}}
  76 extern int PASTE(\u0, 0FD); // expected-warning{{incomplete universal character name}}
  77 #ifdef __cplusplus
  78 // expected-error@-3 {{expected unqualified-id}}
  79 // expected-error@-3 {{expected unqualified-id}}
  80 #else
  81 // expected-error@-6 {{expected identifier}}
  82 // expected-error@-6 {{expected identifier}}
  83 #endif
  84
  85
  86 // A UCN produced by line splicing is valid in C99 but undefined in C++.
  87 // Since undefined behavior can do anything including working as intended,
  88 // we just accept it in C++ as well.;
  89 #define newline_1_\u00F\
  90 C 1
  91 #define newline_2_\u00\
  92 F\
  93 C 1
  94 #define newline_3_\u\
  95 00\
  96 FC 1
  97 #define newline_4_\\
  98 u00FC 1
  99 #define newline_5_\\
 100 u\
 101 \
 102 0\
 103 0\
 104 F\
 105 C 1
 106
 107 #if (newline_1_\u00FC && newline_2_\u00FC && newline_3_\u00FC && \
 108      newline_4_\u00FC && newline_5_\u00FC)
 109 #else
 110 #error "Line splicing failed to produce UCNs"
 111 #endif
 112
 113
 114 #define capital_u_\U00FC
 115 // expected-warning@-1 {{incomplete universal character name}} expected-note@-1 {{did you mean to use '\u'?}} expected-warning@-1 {{whitespace}}
 116 // CHECK: note: did you mean to use '\u'?
 117 // CHECK-NEXT: {{^  .* | #define capital_u_\U00FC}}
 118 // CHECK-NEXT: {{^      |                    \^}}
 119 // CHECK-NEXT: {{^      |                    u}}
 120
 121 #define \u{}           // expected-warning {{empty delimited universal character name; treating as '\' 'u' '{' '}'}} expected-error {{macro name must be an identifier}}
 122 #define \u1{123}       // expected-warning {{incomplete universal character name; treating as '\' followed by identifier}} expected-error {{macro name must be an identifier}}
 123 #define \u{123456789}  // expected-error {{hex escape sequence out of range}} expected-error {{macro name must be an identifier}}
 124 #define \u{            // expected-warning {{incomplete delimited universal character name; treating as '\' 'u' '{' identifier}} expected-error {{macro name must be an identifier}}
 125 #define \u{fgh}        // expected-warning {{incomplete delimited universal character name; treating as '\' 'u' '{' identifier}} expected-error {{macro name must be an identifier}}
 126 #define \N{
 127 // expected-warning@-1 {{incomplete delimited universal character name; treating as '\' 'N' '{' identifier}}
 128 // expected-error@-2 {{macro name must be an identifier}}
 129 #define \N{}           // expected-warning {{empty delimited universal character name; treating as '\' 'N' '{' '}'}} expected-error {{macro name must be an identifier}}
 130 #define \N{NOTATHING}  // expected-error {{'NOTATHING' is not a valid Unicode character name}} \
 131                        // expected-error {{macro name must be an identifier}}
 132 #define \NN            // expected-warning {{incomplete universal character name; treating as '\' followed by identifier}} expected-error {{macro name must be an identifier}}
 133 #define \N{GREEK_SMALL-LETTERALPHA}  // expected-error {{'GREEK_SMALL-LETTERALPHA' is not a valid Unicode character name}} \
 134                                      // expected-note {{characters names in Unicode escape sequences are sensitive to case and whitespaces}}
 135 #define \N{🤡}  // expected-error {{'🤡' is not a valid Unicode character name}} \
 136                 // expected-error {{macro name must be an identifier}}
 137
 138 #define CONCAT(A, B) A##B
 139 int CONCAT(\N{GREEK
 140 , CAPITALLETTERALPHA});
 141 // expected-error@-2 {{expected}} \
 142 // expected-warning@-2 {{incomplete delimited universal character name}}
 143
 144 int \N{\
 145 LATIN CAPITAL LETTER A WITH GRAVE};
 146 //ext-warning@-2 {{extension}} cxx23-warning@-2 {{before C++23}}
 147
 148 #ifdef TRIGRAPHS
 149 int \N??<GREEK CAPITAL LETTER ALPHA??> = 0; // cxx23-warning {{before C++23}} \
 150                                             //ext-warning {{extension}}\
 151                                             // expected-warning 2{{trigraph converted}}
 152
 153 int a\N{LATIN CAPITAL LETTER A WITH GRAVE??>; // expected-warning {{trigraph converted}}
 154 #endif
 155
 156 #ifndef TRIGRAPHS
 157 int a\N{LATIN CAPITAL LETTER A WITH GRAVE??>;
 158 // expected-warning@-1 {{trigraph ignored}}\
 159 // expected-warning@-1 {{incomplete}}\
 160 // expected-error@-1 {{expected unqualified-id}}
 161 #endif
 162
 163 // GH64161
 164 int A\N{LEFT-TO-RIGHT OVERRIDE}; // expected-error {{character <U+202D> not allowed in an identifier}}