gcc/go/gofrontend/go-encode-id.cc

   1 // go-encode-id.cc -- Go identifier encoding hooks
   2
   3 // Copyright 2016 The Go Authors. All rights reserved.
   4 // Use of this source code is governed by a BSD-style
   5 // license that can be found in the LICENSE file.
   6
   7 #include "go-location.h"
   8 #include "go-linemap.h"
   9 #include "go-encode-id.h"
  10
  11 // Return whether the character c is OK to use in the assembler.
  12
  13 static bool
  14 char_needs_encoding(char c)
  15 {
  16   switch (c)
  17     {
  18     case 'A': case 'B': case 'C': case 'D': case 'E': case 'F':
  19     case 'G': case 'H': case 'I': case 'J': case 'K': case 'L':
  20     case 'M': case 'N': case 'O': case 'P': case 'Q': case 'R':
  21     case 'S': case 'T': case 'U': case 'V': case 'W': case 'X':
  22     case 'Y': case 'Z':
  23     case 'a': case 'b': case 'c': case 'd': case 'e': case 'f':
  24     case 'g': case 'h': case 'i': case 'j': case 'k': case 'l':
  25     case 'm': case 'n': case 'o': case 'p': case 'q': case 'r':
  26     case 's': case 't': case 'u': case 'v': case 'w': case 'x':
  27     case 'y': case 'z':
  28     case '0': case '1': case '2': case '3': case '4':
  29     case '5': case '6': case '7': case '8': case '9':
  30     case '_': case '.': case '$': case '/':
  31       return false;
  32     default:
  33       return true;
  34     }
  35 }
  36
  37 // Return whether the identifier needs to be translated because it
  38 // contains non-ASCII characters.
  39
  40 bool
  41 go_id_needs_encoding(const std::string& str)
  42 {
  43   for (std::string::const_iterator p = str.begin();
  44        p != str.end();
  45        ++p)
  46     if (char_needs_encoding(*p))
  47       return true;
  48   return false;
  49 }
  50
  51 // Pull the next UTF-8 character out of P and store it in *PC.  Return
  52 // the number of bytes read.
  53
  54 static size_t
  55 fetch_utf8_char(const char* p, unsigned int* pc)
  56 {
  57   unsigned char c = *p;
  58   if ((c & 0x80) == 0)
  59     {
  60       *pc = c;
  61       return 1;
  62     }
  63   size_t len = 0;
  64   while ((c & 0x80) != 0)
  65     {
  66       ++len;
  67       c <<= 1;
  68     }
  69   unsigned int rc = *p & ((1 << (7 - len)) - 1);
  70   for (size_t i = 1; i < len; i++)
  71     {
  72       unsigned int u = p[i];
  73       rc <<= 6;
  74       rc |= u & 0x3f;
  75     }
  76   *pc = rc;
  77   return len;
  78 }
  79
  80 // Encode an identifier using ASCII characters.
  81
  82 std::string
  83 go_encode_id(const std::string &id)
  84 {
  85   std::string ret;
  86   const char* p = id.c_str();
  87   const char* pend = p + id.length();
  88   while (p < pend)
  89     {
  90       unsigned int c;
  91       size_t len = fetch_utf8_char(p, &c);
  92       if (len == 1 && !char_needs_encoding(c))
  93         ret += c;
  94       else
  95         {
  96           ret += "$U";
  97           char buf[30];
  98           snprintf(buf, sizeof buf, "%x", c);
  99           ret += buf;
 100           ret += "$";
 101         }
 102       p += len;
 103     }
 104   return ret;
 105 }
 106
 107 std::string
 108 go_selectively_encode_id(const std::string &id)
 109 {
 110   if (go_id_needs_encoding(id))
 111     return go_encode_id(id);
 112   return std::string();
 113 }