stringencode.cc

本人收集整理的一份c/c++跨平台网络库

#ifdef WIN32
#include <malloc.h>
#endif  // WIN32
#ifdef POSIX
#include <alloca.h>
#define _alloca alloca
#endif  // POSIX

#include "basictypes.h"
#include "common.h"
#include "stringencode.h"
#include "stringutils.h"

namespace utils_base {

/////////////////////////////////////////////////////////////////////////////
// String Encoding Utilities
/////////////////////////////////////////////////////////////////////////////

static const char HEX[] = "0123456789abcdef";

char hex_encode(unsigned char val) {
  ASSERT(val < 16);
  return (val < 16) ? HEX[val] : '!';
}

unsigned char hex_decode(char ch) {
  char lower = tolower(ch);
  ASSERT(((ch >= '0') && (ch <= '9')) || ((lower >= 'a') && (lower <= 'z')));
  return (ch <= '9') ? (ch - '0') : ((lower - 'a') + 10);
}

size_t escape(char * buffer, size_t buflen,
              const char * source, size_t srclen,
              const char * illegal, char escape) {
  ASSERT(NULL != buffer);  // TODO: estimate output size
  if (buflen <= 0)
    return 0;

  size_t srcpos = 0, bufpos = 0;
  while ((srcpos < srclen) && (bufpos + 1 < buflen)) {
    char ch = source[srcpos++];
    if ((ch == escape) || ::strchr(illegal, ch)) {
      if (bufpos + 2 >= buflen)
        break;
      buffer[bufpos++] = escape;
    }
    buffer[bufpos++] = ch;
  }

  buffer[bufpos] = '\0';
  return bufpos;
}

size_t unescape(char * buffer, size_t buflen,
                const char * source, size_t srclen,
                char escape) {
  ASSERT(NULL != buffer);  // TODO: estimate output size
  if (buflen <= 0)
    return 0;

  size_t srcpos = 0, bufpos = 0;
  while ((srcpos < srclen) && (bufpos + 1 < buflen)) {
    char ch = source[srcpos++];
    if ((ch == escape) && (srcpos < srclen)) {
      ch = source[srcpos++];
    }
    buffer[bufpos++] = ch;
  }
  buffer[bufpos] = '\0';
  return bufpos;
}

size_t encode(char * buffer, size_t buflen,
              const char * source, size_t srclen,
              const char * illegal, char escape) {
  ASSERT(NULL != buffer);  // TODO: estimate output size
  if (buflen <= 0)
    return 0;

  size_t srcpos = 0, bufpos = 0;
  while ((srcpos < srclen) && (bufpos + 1 < buflen)) {
    char ch = source[srcpos++];
    if ((ch != escape) && !::strchr(illegal, ch)) {
      buffer[bufpos++] = ch;
    } else if (bufpos + 3 >= buflen) {
      break;
    } else {
      buffer[bufpos+0] = escape;
      buffer[bufpos+1] = hex_encode((static_cast<unsigned char>(ch) >> 4) & 0xF);
      buffer[bufpos+2] = hex_encode((static_cast<unsigned char>(ch)     ) & 0xF);
      bufpos += 3;
    }
  }
  buffer[bufpos] = '\0';
  return bufpos;
}

size_t decode(char * buffer, size_t buflen,
              const char * source, size_t srclen,
              char escape) {
  if (buflen <= 0)
    return 0;

  size_t srcpos = 0, bufpos = 0;
  while ((srcpos < srclen) && (bufpos + 1 < buflen)) {
    char ch = source[srcpos++];
    if ((ch == escape) && (srcpos + 1 < srclen)) {
      buffer[bufpos++] = (hex_decode(source[srcpos]) << 4)
                         | hex_decode(source[srcpos+1]);
      srcpos += 2;
    } else {
      buffer[bufpos++] = ch;
    }
  }
  buffer[bufpos] = '\0';
  return bufpos;
}

const char* unsafe_filename_characters() {
  // It might be better to have a single specification which is the union of
  // all operating systems, unless one system is overly restrictive.
#ifdef WIN32
  return "\\/:*?\"<>|";
#else  // !WIN32
  // TODO
#endif  // !WIN23
}

const unsigned char URL_UNSAFE  = 0x1; // 0-33 "#$%&+,/:;<=>?@[\]^`{|} 127
const unsigned char XML_UNSAFE  = 0x2; // "&'<>
const unsigned char HTML_UNSAFE = 0x2; // "&'<>

//  ! " # $ % & ' ( ) * + , - . / 0 1 2 3 4 6 5 7 8 9 : ; < = > ?
//@ A B C D E F G H I J K L M N O P Q R S T U V W X Y Z [ \ ] ^ _
//` a b c d e f g h i j k l m n o p q r s t u v w x y z { | } ~ 

const unsigned char ASCII_CLASS[128] = {
  1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,
  1,0,3,1,1,1,3,2,0,0,0,1,1,0,0,1,0,0,0,0,0,0,0,0,0,0,1,1,3,1,3,1,
  1,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,1,1,1,1,0,
  1,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,1,1,1,0,1,
};

size_t url_encode(char * buffer, size_t buflen,
                  const char * source, size_t srclen) {
  if (NULL == buffer)
    return srclen * 3 + 1;
  if (buflen <= 0)
    return 0;

  size_t srcpos = 0, bufpos = 0;
  while ((srcpos < srclen) && (bufpos + 1 < buflen)) {
    unsigned char ch = source[srcpos++];
    if ((ch < 128) && (ASCII_CLASS[ch] & URL_UNSAFE)) {
      if (bufpos + 3 >= buflen) {
        break;
      }
      buffer[bufpos+0] = '%';
      buffer[bufpos+1] = hex_encode((ch >> 4) & 0xF);
      buffer[bufpos+2] = hex_encode((ch     ) & 0xF);
      bufpos += 3;
    } else {
      buffer[bufpos++] = ch;
    }
  }
  buffer[bufpos] = '\0';
  return bufpos;
}

size_t url_decode(char * buffer, size_t buflen,
                  const char * source, size_t srclen) {
  if (NULL == buffer)
    return srclen + 1;
  if (buflen <= 0)
    return 0;

  size_t srcpos = 0, bufpos = 0;
  while ((srcpos < srclen) && (bufpos + 1 < buflen)) {
    unsigned char ch = source[srcpos++];
    if (ch == '+') {
      buffer[bufpos++] = ' ';
    } else if ((ch == '%') && (srcpos + 1 < srclen)) {
      buffer[bufpos++] = (hex_decode(source[srcpos]) << 4)
                         | hex_decode(source[srcpos+1]);
      srcpos += 2;
    } else {
      buffer[bufpos++] = ch;
    }
  }
  buffer[bufpos] = '\0';
  return bufpos;
}

size_t utf8_decode(const char* source, size_t srclen, unsigned long* value) {
  const unsigned char* s = reinterpret_cast<const unsigned char*>(source);
  if ((s[0] & 0x80) == 0x00) {                    // Check s[0] == 0xxxxxxx
    *value = s[0];
    return 1;
  }
  if ((srclen < 2) || ((s[1] & 0xC0) != 0x80)) {  // Check s[1] != 10xxxxxx
    return 0;
  }
  // Accumulate the trailer byte values in value16, and combine it with the
  // relevant bits from s[0], once we've determined the sequence length.
  unsigned long value16 = (s[1] & 0x3F);
  if ((s[0] & 0xE0) == 0xC0) {                    // Check s[0] == 110xxxxx
    *value = ((s[0] & 0x1F) << 6) | value16;
    return 2;
  }
  if ((srclen < 3) || ((s[2] & 0xC0) != 0x80)) {  // Check s[2] != 10xxxxxx
    return 0;
  }
  value16 = (value16 << 6) | (s[2] & 0x3F);
  if ((s[0] & 0xF0) == 0xE0) {                    // Check s[0] == 1110xxxx 
    *value = ((s[0] & 0x0F) << 12) | value16;
    return 3;
  }
  if ((srclen < 4) || ((s[3] & 0xC0) != 0x80)) {  // Check s[3] != 10xxxxxx
    return 0;
  }
  value16 = (value16 << 6) | (s[3] & 0x3F);
  if ((s[0] & 0xF8) == 0xF0) {                    // Check s[0] == 11110xxx 
    *value = ((s[0] & 0x07) << 18) | value16;
    return 4;
  }
  return 0;
}

size_t utf8_encode(char* buffer, size_t buflen, unsigned long value) {
  if ((value <= 0x7F) && (buflen >= 1)) {
    buffer[0] = static_cast<unsigned char>(value);
    return 1;
  }
  if ((value <= 0x7FF) && (buflen >= 2)) {
    buffer[0] = 0xC0 | static_cast<unsigned char>(value >> 6);
    buffer[1] = 0x80 | static_cast<unsigned char>(value & 0x3F);
    return 2;
  }
  if ((value <= 0xFFFF) && (buflen >= 3)) {
    buffer[0] = 0xE0 | static_cast<unsigned char>(value >> 12);
    buffer[1] = 0x80 | static_cast<unsigned char>((value >> 6) & 0x3F);
    buffer[2] = 0x80 | static_cast<unsigned char>(value & 0x3F);
    return 3;
  }
  if ((value <= 0x1FFFFF) && (buflen >= 4)) {
    buffer[0] = 0xF0 | static_cast<unsigned char>(value >> 18);
    buffer[1] = 0x80 | static_cast<unsigned char>((value >> 12) & 0x3F);
    buffer[2] = 0x80 | static_cast<unsigned char>((value >> 6) & 0x3F);
    buffer[3] = 0x80 | static_cast<unsigned char>(value & 0x3F);
    return 4;
  }
  return 0;
}

size_t html_encode(char * buffer, size_t buflen,
                   const char * source, size_t srclen) {
  ASSERT(NULL != buffer);  // TODO: estimate output size
  if (buflen <= 0)
    return 0;

  size_t srcpos = 0, bufpos = 0;
  while ((srcpos < srclen) && (bufpos + 1 < buflen)) {
    unsigned char ch = source[srcpos];
    if (ch < 128) {
      srcpos += 1;
      if (ASCII_CLASS[ch] & HTML_UNSAFE) {
        const char * escseq = 0;
        size_t esclen = 0;
        switch (ch) {
          case '<':  escseq = "&lt;";   esclen = 4; break;
          case '>':  escseq = "&gt;";   esclen = 4; break;
          case '\'': escseq = "&#39;";  esclen = 5; break;
          case '\"': escseq = "&quot;"; esclen = 6; break;
          case '&':  escseq = "&amp;";  esclen = 5; break;
          default: ASSERT(false);
        }