clang/test/Lexer/unicode.c

   1 // RUN: %clang_cc1 -fsyntax-only -verify -x c -std=c11 %s
   2 // RUN: %clang_cc1 -fsyntax-only -verify -x c++ -std=c++11 %s
   3 // RUN: %clang_cc1 -std=c99 -E -DPP_ONLY=1 %s | FileCheck %s --strict-whitespace
   4 // RUN: %clang_cc1 -E -DPP_ONLY=1 %s | FileCheck %s --strict-whitespace
   5
   6 // This file contains Unicode characters; please do not "fix" them!
   7
   8 extern int x; // expected-warning {{treating Unicode character as whitespace}}
   9 extern int　x; // expected-warning {{treating Unicode character as whitespace}}
  10
  11 // CHECK: extern int {{x}}
  12 // CHECK: extern int　{{x}}
  13
  14 #pragma mark ¡Unicode!
  15
  16 #define COPYRIGHT Copyright © 2012
  17 #define XSTR(X) #X
  18 #define STR(X) XSTR(X)
  19
  20 static const char *copyright = STR(COPYRIGHT); // no-warning
  21 // CHECK: static const char *copyright = "Copyright © {{2012}}";
  22
  23 #if PP_ONLY
  24 COPYRIGHT
  25 // CHECK: Copyright © {{2012}}
  26 CHECK : The preprocessor should not complain about Unicode characters like ©.
  27 #endif
  28
  29         int _;
  30
  31 #ifdef __cplusplus
  32
  33 extern int ༀ;
  34 extern int 𑩐;
  35 extern int 𐠈;
  36 extern int ꙮ;
  37 extern int  \u1B4C;     // BALINESE LETTER ARCHAIC JNYA - Added in Unicode 14
  38 extern int  \U00016AA2; // TANGSA LETTER GA - Added in Unicode 14
  39 // This character doesn't have the XID_Start property
  40 extern int  \U00016AC0; // TANGSA DIGIT ZERO  // expected-error {{expected unqualified-id}}
  41 extern int _\U00016AC0; // TANGSA DIGIT ZERO
  42
  43 extern int 🌹; // expected-error {{unexpected character <U+1F339>}} \
  44                   expected-warning {{declaration does not declare anything}}
  45
  46 extern int 👷; // expected-error {{unexpected character <U+1F477>}} \
  47                   expected-warning {{declaration does not declare anything}}
  48
  49 extern int 👷‍♀; // expected-warning {{declaration does not declare anything}} \
  50                   expected-error {{unexpected character <U+1F477>}} \
  51                   expected-error {{unexpected character <U+200D>}} \
  52                   expected-error {{unexpected character <U+2640>}}
  53 #else
  54
  55 // A 🌹 by any other name....
  56 extern int 🌹;
  57 int 🌵(int 🌻) { return 🌻+ 1; }
  58 int main (void) {
  59   int 🌷 = 🌵(🌹);
  60   return 🌷;
  61 }
  62
  63 int n; = 3; // expected-warning {{treating Unicode character <U+037E> as identifier character rather than as ';' symbol}}
  64 int *n꞉꞉v = &n;; // expected-warning 2{{treating Unicode character <U+A789> as identifier character rather than as ':' symbol}}
  65                  // expected-warning@-1 {{treating Unicode character <U+037E> as identifier character rather than as ';' symbol}}
  66 int v＝［＝］（auto）｛return～x；｝（）; // expected-warning 12{{treating Unicode character}}
  67
  68 int ⁠xx‍;
  69 // expected-warning@-1 {{identifier contains Unicode character <U+2060> that is invisible in some environments}}
  70 // expected-warning@-2 {{identifier contains Unicode character <U+FEFF> that is invisible in some environments}}
  71 // expected-warning@-3 {{identifier contains Unicode character <U+200D> that is invisible in some environments}}
  72 int foobar = 0; // expected-warning {{identifier contains Unicode character <U+200B> that is invisible in some environments}}
  73 int x = foobar; // expected-error {{undeclared identifier}}
  74
  75 int ∣foo; // expected-error {{unexpected character <U+2223>}}
  76 #ifndef PP_ONLY
  77 #define ∶ x // expected-error {{macro name must be an identifier}}
  78 #endif
  79
  80 #endif