I have written some time ago a little C program that generates such a list for all of Unicode, not just the UCS-2 subset.
It uses macros from a few ICU header files, but does not need the compiled ICU library. On Unixes, you may need to runConfigure, but on Windows it will work out of the box.
For supplementary code points on the non-UTF-8-side, it will write either 6-digit code points (default) or two 4-digit surrogates (have a command line argument, anything).
For ICU see http://oss.software.ibm.com/icu/
markus
----------------------- utf8map.c ------------
#include <stdio.h>
#include "unicode/utypes.h"
extern int
main(int argc, const char *argv[]) {
uint16_t utf16[2];
uint8_t utf8[4];
uint32_t c, i, j;
for(c=0; c<=0x7f; ++c) {
printf("%04lx %02x\n", c, c);
}
for(; c<=0x7ff; ++c) {
j=0;
UTF8_APPEND_CHAR_UNSAFE(utf8, j, c);
printf("%04lx %02x%02x\n", c, utf8[0], utf8[1]);
}
for(; c<=0xffff; ++c) {
j=0;
UTF8_APPEND_CHAR_UNSAFE(utf8, j, c);
printf("%04lx %02x%02x%02x\n", c, utf8[0], utf8[1], utf8[2]);
}
if(argc==1) {
/* no argument: write code points */
for(; c<=0xfffff; ++c) {
j=0;
UTF8_APPEND_CHAR_UNSAFE(utf8, j, c);
printf("%lx %02x%02x%02x%02x\n", c, utf8[0], utf8[1], utf8[2], utf8[3]);
}
/* 6 digits for the code points, align nicely */
for(; c<=0x10ffff; ++c) {
j=0;
UTF8_APPEND_CHAR_UNSAFE(utf8, j, c);
printf("%lx %02x%02x%02x%02x\n", c, utf8[0], utf8[1], utf8[2], utf8[3]);
}
} else {
/* there is an argument: write surrogate pairs */
for(; c<=0x10ffff; ++c) {
i=j=0;
UTF16_APPEND_CHAR_UNSAFE(utf16, i, c);
UTF8_APPEND_CHAR_UNSAFE(utf8, j, c);
printf("%x%x %02x%02x%02x%02x\n", utf16[0], utf16[1], utf8[0], utf8[1], utf8[2], utf8[3]);
}
}
return 0;
}
This archive was generated by hypermail 2.1.2 : Wed Sep 19 2001 - 14:49:40 EDT