ccutil/unichar.h

Go to the documentation of this file.
00001 
00020 #ifndef THIRD_PARTY_TESSERACT_CCUTIL_UNICHAR_H__
00021 #define THIRD_PARTY_TESSERACT_CCUTIL_UNICHAR_H__
00022 
00023 #include <memory.h>
00024 
00030 #define UNICHAR_LEN 4
00031 
00041 class UNICHAR {
00042  public:
00043   UNICHAR() {
00044     memset(chars, 0, UNICHAR_LEN);
00045   }
00046 
00047   // Construct from a utf8 string. If len<0 then the string is null terminated.
00048   // If the string is too long to fit in the UNICHAR then it takes only what
00049   // will fit.
00050   UNICHAR(const char* utf8_str, int len);
00051 
00052   // Construct from a single UCS4 character.
00053   explicit UNICHAR(int unicode);
00054 
00055   // Default copy constructor and operator= are OK.
00056 
00057   // Get the first character as UCS-4.
00058   int first_uni() const;
00059 
00060   // Get the length of the UTF8 string.
00061   int utf8_len() const {
00062     int len = chars[UNICHAR_LEN - 1];
00063     return len >=0 && len < UNICHAR_LEN ? len : UNICHAR_LEN;
00064   }
00065 
00066   // Get a UTF8 string, but NOT NULL terminated.
00067   const char* utf8() const {
00068     return chars;
00069   }
00070 
00071   // Get a terminated UTF8 string: Must delete[] it after use.
00072   char* utf8_str() const;
00073 
00074   // Get the number of bytes in the first character of the given utf8 string.
00075   static int utf8_step(const char* utf8_str);
00076 
00077  private:
00078   // A UTF-8 representation of 1 or more Unicode characters.
00079   // The last element (chars[UNICHAR_LEN - 1]) is a length if
00080   // its value < UNICHAR_LEN, otherwise it is a genuine character.
00081   char chars[UNICHAR_LEN];
00082 };
00083 
00084 #endif  // THIRD_PARTY_TESSERACT_CCUTIL_UNICHAR_H__

Generated on Wed Feb 28 19:49:09 2007 for Tesseract by  doxygen 1.5.1