Language Namespace Reference

Mnemonic language related namespace. More...

Classes
class	Chinese_Simplified
class	Dutch
class	English
class	EnglishOld
class	Esperanto
class	French
class	German
class	Italian
class	Japanese
struct	WordHash
struct	WordEqual
class	Base
	A base language class which all languages have to inherit from for Polymorphism. More...
class	Lojban
class	Portuguese
class	Russian
class	Singleton
	Single helper class. More...
class	Spanish

Functions
template<typename T>
T	utf8prefix (const T &s, size_t count)
	Returns a string made of (at most) the first count characters in s. Assumes well formedness. No check is made for this.
template<typename T>
T	utf8canonical (const T &s)

Detailed Description

Mnemonic language related namespace.

Function Documentation

utf8canonical()

template<typename T>

T Language::utf8canonical ( const T & s )

inline

Definition at line 78 of file language_base.h.

  {
    T sc = "";
    size_t avail = s.size();
    const char *ptr = s.data();
    wint_t cp = 0;
    int bytes = 1;
    char wbuf[8], *wptr;
    while (avail--)
    {
      if ((*ptr & 0x80) == 0)
      {
        cp = *ptr++;
        bytes = 1;
      }
      else if ((*ptr & 0xe0) == 0xc0)
      {
        if (avail < 1)
          throw std::runtime_error("Invalid UTF-8");
        cp = (*ptr++ & 0x1f) << 6;
        cp |= *ptr++ & 0x3f;
        --avail;
        bytes = 2;
      }
      else if ((*ptr & 0xf0) == 0xe0)
      {
        if (avail < 2)
          throw std::runtime_error("Invalid UTF-8");
        cp = (*ptr++ & 0xf) << 12;
        cp |= (*ptr++ & 0x3f) << 6;
        cp |= *ptr++ & 0x3f;
        avail -= 2;
        bytes = 3;
      }
      else if ((*ptr & 0xf8) == 0xf0)
      {
        if (avail < 3)
          throw std::runtime_error("Invalid UTF-8");
        cp = (*ptr++ & 0x7) << 18;
        cp |= (*ptr++ & 0x3f) << 12;
        cp |= (*ptr++ & 0x3f) << 6;
        cp |= *ptr++ & 0x3f;
        avail -= 3;
        bytes = 4;
      }
      else
        throw std::runtime_error("Invalid UTF-8");
 
      cp = std::towlower(cp);
      wptr = wbuf;
      switch (bytes)
      {
        case 1: *wptr++ = cp; break;
        case 2: *wptr++ = 0xc0 | (cp >> 6); *wptr++ = 0x80 | (cp & 0x3f); break;
        case 3: *wptr++ = 0xe0 | (cp >> 12); *wptr++ = 0x80 | ((cp >> 6) & 0x3f); *wptr++ = 0x80 | (cp & 0x3f); break;
        case 4: *wptr++ = 0xf0 | (cp >> 18); *wptr++ = 0x80 | ((cp >> 12) & 0x3f); *wptr++ = 0x80 | ((cp >> 6) & 0x3f); *wptr++ = 0x80 | (cp & 0x3f); break;
        default: throw std::runtime_error("Invalid UTF-8");
      }
      *wptr = 0;
      sc += T(wbuf, bytes);
      cp = 0;
      bytes = 1;
    }
    return sc;
  }

Here is the caller graph for this function:

utf8prefix()

template<typename T>

T Language::utf8prefix ( const T & s, size_t count )

inline

Returns a string made of (at most) the first count characters in s. Assumes well formedness. No check is made for this.

Parameters

s	The string from which to return the first count characters.
count	How many characters to return.

Returns

A string consisting of the first count characters in s.

Definition at line 60 of file language_base.h.

  {
    T prefix = "";
    size_t avail = s.size();
    const char *ptr = s.data();
    while (count-- && avail--)
    {
      prefix += *ptr++;
      while (avail && ((*ptr) & 0xc0) == 0x80)
      {
        prefix += *ptr++;
        --avail;
      }
    }
    return prefix;
  }

Here is the caller graph for this function: