mirror of
https://github.com/GothenburgBitFactory/taskwarrior.git
synced 2025-06-26 10:54:26 +02:00
Code Migration
- Migrated taskd JSON parser into task, to provide encode/decode capability to Task::composeJSON. - Migrated taskd utf8 code, replacing old unused code. - Added unit tests or JSON. - Migrated Tree updates from taskd.
This commit is contained in:
Paul Beckingham
parent
2f4e0d9e17
commit
57c1983e07
13 changed files with 717 additions and 607 deletions
640
src/utf8.cpp
640
src/utf8.cpp
|
|
@ -1,486 +1,7 @@
|
|||
#ifdef NOPE
|
||||
/*
|
||||
Basic UTF-8 manipulation routines
|
||||
by Jeff Bezanson
|
||||
placed in the public domain Fall 2005
|
||||
|
||||
This code is designed to provide the utilities you need to manipulate
|
||||
UTF-8 as an internal string encoding. These functions do not perform the
|
||||
error checking normally needed when handling UTF-8 data, so if you happen
|
||||
to be from the Unicode Consortium you will want to flay me alive.
|
||||
I do this because error checking can be performed at the boundaries (I/O),
|
||||
with these routines reserved for higher performance on data known to be
|
||||
valid.
|
||||
*/
|
||||
#include <stdlib.h>
|
||||
#include <stdio.h>
|
||||
#include <string.h>
|
||||
#include <stdarg.h>
|
||||
#ifdef WIN32
|
||||
#include <malloc.h>
|
||||
#else
|
||||
#include <alloca.h>
|
||||
#endif
|
||||
|
||||
#include "utf8.h"
|
||||
|
||||
static const u_int32_t offsetsFromUTF8[6] = {
|
||||
0x00000000UL, 0x00003080UL, 0x000E2080UL,
|
||||
0x03C82080UL, 0xFA082080UL, 0x82082080UL
|
||||
};
|
||||
|
||||
static const char trailingBytesForUTF8[256] = {
|
||||
0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,
|
||||
0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,
|
||||
0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,
|
||||
0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,
|
||||
0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,
|
||||
0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,
|
||||
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,
|
||||
2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2, 3,3,3,3,3,3,3,3,4,4,4,4,5,5,5,5
|
||||
};
|
||||
|
||||
/* returns length of next utf-8 sequence */
|
||||
int u8_seqlen(char *s)
|
||||
{
|
||||
return trailingBytesForUTF8[(unsigned int)(unsigned char)s[0]] + 1;
|
||||
}
|
||||
|
||||
/* conversions without error checking
|
||||
only works for valid UTF-8, i.e. no 5- or 6-byte sequences
|
||||
srcsz = source size in bytes, or -1 if 0-terminated
|
||||
sz = dest size in # of wide characters
|
||||
|
||||
returns # characters converted
|
||||
dest will always be L'\0'-terminated, even if there isn't enough room
|
||||
for all the characters.
|
||||
if sz = srcsz+1 (i.e. 4*srcsz+4 bytes), there will always be enough space.
|
||||
*/
|
||||
int u8_toucs(u_int32_t *dest, int sz, char *src, int srcsz)
|
||||
{
|
||||
u_int32_t ch;
|
||||
char *src_end = src + srcsz;
|
||||
int nb;
|
||||
int i=0;
|
||||
|
||||
while (i < sz-1) {
|
||||
nb = trailingBytesForUTF8[(unsigned char)*src];
|
||||
if (srcsz == -1) {
|
||||
if (*src == 0)
|
||||
goto done_toucs;
|
||||
}
|
||||
else {
|
||||
if (src + nb >= src_end)
|
||||
goto done_toucs;
|
||||
}
|
||||
ch = 0;
|
||||
switch (nb) {
|
||||
/* these fall through deliberately */
|
||||
case 3: ch += (unsigned char)*src++; ch <<= 6;
|
||||
case 2: ch += (unsigned char)*src++; ch <<= 6;
|
||||
case 1: ch += (unsigned char)*src++; ch <<= 6;
|
||||
case 0: ch += (unsigned char)*src++;
|
||||
}
|
||||
ch -= offsetsFromUTF8[nb];
|
||||
dest[i++] = ch;
|
||||
}
|
||||
done_toucs:
|
||||
dest[i] = 0;
|
||||
return i;
|
||||
}
|
||||
|
||||
/* srcsz = number of source characters, or -1 if 0-terminated
|
||||
sz = size of dest buffer in bytes
|
||||
|
||||
returns # characters converted
|
||||
dest will only be '\0'-terminated if there is enough space. this is
|
||||
for consistency; imagine there are 2 bytes of space left, but the next
|
||||
character requires 3 bytes. in this case we could NUL-terminate, but in
|
||||
general we can't when there's insufficient space. therefore this function
|
||||
only NUL-terminates if all the characters fit, and there's space for
|
||||
the NUL as well.
|
||||
the destination string will never be bigger than the source string.
|
||||
*/
|
||||
int u8_toutf8(char *dest, int sz, u_int32_t *src, int srcsz)
|
||||
{
|
||||
u_int32_t ch;
|
||||
int i = 0;
|
||||
char *dest_end = dest + sz;
|
||||
|
||||
while (srcsz<0 ? src[i]!=0 : i < srcsz) {
|
||||
ch = src[i];
|
||||
if (ch < 0x80) {
|
||||
if (dest >= dest_end)
|
||||
return i;
|
||||
*dest++ = (char)ch;
|
||||
}
|
||||
else if (ch < 0x800) {
|
||||
if (dest >= dest_end-1)
|
||||
return i;
|
||||
*dest++ = (ch>>6) | 0xC0;
|
||||
*dest++ = (ch & 0x3F) | 0x80;
|
||||
}
|
||||
else if (ch < 0x10000) {
|
||||
if (dest >= dest_end-2)
|
||||
return i;
|
||||
*dest++ = (ch>>12) | 0xE0;
|
||||
*dest++ = ((ch>>6) & 0x3F) | 0x80;
|
||||
*dest++ = (ch & 0x3F) | 0x80;
|
||||
}
|
||||
else if (ch < 0x110000) {
|
||||
if (dest >= dest_end-3)
|
||||
return i;
|
||||
*dest++ = (ch>>18) | 0xF0;
|
||||
*dest++ = ((ch>>12) & 0x3F) | 0x80;
|
||||
*dest++ = ((ch>>6) & 0x3F) | 0x80;
|
||||
*dest++ = (ch & 0x3F) | 0x80;
|
||||
}
|
||||
i++;
|
||||
}
|
||||
if (dest < dest_end)
|
||||
*dest = '\0';
|
||||
return i;
|
||||
}
|
||||
|
||||
int u8_wc_toutf8(char *dest, u_int32_t ch)
|
||||
{
|
||||
if (ch < 0x80) {
|
||||
dest[0] = (char)ch;
|
||||
return 1;
|
||||
}
|
||||
if (ch < 0x800) {
|
||||
dest[0] = (ch>>6) | 0xC0;
|
||||
dest[1] = (ch & 0x3F) | 0x80;
|
||||
return 2;
|
||||
}
|
||||
if (ch < 0x10000) {
|
||||
dest[0] = (ch>>12) | 0xE0;
|
||||
dest[1] = ((ch>>6) & 0x3F) | 0x80;
|
||||
dest[2] = (ch & 0x3F) | 0x80;
|
||||
return 3;
|
||||
}
|
||||
if (ch < 0x110000) {
|
||||
dest[0] = (ch>>18) | 0xF0;
|
||||
dest[1] = ((ch>>12) & 0x3F) | 0x80;
|
||||
dest[2] = ((ch>>6) & 0x3F) | 0x80;
|
||||
dest[3] = (ch & 0x3F) | 0x80;
|
||||
return 4;
|
||||
}
|
||||
return 0;
|
||||
}
|
||||
|
||||
/* charnum => byte offset */
|
||||
int u8_offset(char *str, int charnum)
|
||||
{
|
||||
int offs=0;
|
||||
|
||||
while (charnum > 0 && str[offs]) {
|
||||
(void)(isutf(str[++offs]) || isutf(str[++offs]) ||
|
||||
isutf(str[++offs]) || ++offs);
|
||||
charnum--;
|
||||
}
|
||||
return offs;
|
||||
}
|
||||
|
||||
/* byte offset => charnum */
|
||||
int u8_charnum(char *s, int offset)
|
||||
{
|
||||
int charnum = 0, offs=0;
|
||||
|
||||
while (offs < offset && s[offs]) {
|
||||
(void)(isutf(s[++offs]) || isutf(s[++offs]) ||
|
||||
isutf(s[++offs]) || ++offs);
|
||||
charnum++;
|
||||
}
|
||||
return charnum;
|
||||
}
|
||||
|
||||
/* number of characters */
|
||||
int u8_strlen(char *s)
|
||||
{
|
||||
int count = 0;
|
||||
int i = 0;
|
||||
|
||||
while (u8_nextchar(s, &i) != 0)
|
||||
count++;
|
||||
|
||||
return count;
|
||||
}
|
||||
|
||||
/* reads the next utf-8 sequence out of a string, updating an index */
|
||||
u_int32_t u8_nextchar(char *s, int *i)
|
||||
{
|
||||
u_int32_t ch = 0;
|
||||
int sz = 0;
|
||||
|
||||
do {
|
||||
ch <<= 6;
|
||||
ch += (unsigned char)s[(*i)++];
|
||||
sz++;
|
||||
} while (s[*i] && !isutf(s[*i]));
|
||||
ch -= offsetsFromUTF8[sz-1];
|
||||
|
||||
return ch;
|
||||
}
|
||||
|
||||
void u8_inc(char *s, int *i)
|
||||
{
|
||||
(void)(isutf(s[++(*i)]) || isutf(s[++(*i)]) ||
|
||||
isutf(s[++(*i)]) || ++(*i));
|
||||
}
|
||||
|
||||
void u8_dec(char *s, int *i)
|
||||
{
|
||||
(void)(isutf(s[--(*i)]) || isutf(s[--(*i)]) ||
|
||||
isutf(s[--(*i)]) || --(*i));
|
||||
}
|
||||
|
||||
int octal_digit(char c)
|
||||
{
|
||||
return (c >= '0' && c <= '7');
|
||||
}
|
||||
|
||||
int hex_digit(char c)
|
||||
{
|
||||
return ((c >= '0' && c <= '9') ||
|
||||
(c >= 'A' && c <= 'F') ||
|
||||
(c >= 'a' && c <= 'f'));
|
||||
}
|
||||
|
||||
/* assumes that src points to the character after a backslash
|
||||
returns number of input characters processed */
|
||||
int u8_read_escape_sequence(char *str, u_int32_t *dest)
|
||||
{
|
||||
u_int32_t ch;
|
||||
char digs[9]="\0\0\0\0\0\0\0\0";
|
||||
int dno=0, i=1;
|
||||
|
||||
ch = (u_int32_t)str[0]; /* take literal character */
|
||||
if (str[0] == 'n')
|
||||
ch = L'\n';
|
||||
else if (str[0] == 't')
|
||||
ch = L'\t';
|
||||
else if (str[0] == 'r')
|
||||
ch = L'\r';
|
||||
else if (str[0] == 'b')
|
||||
ch = L'\b';
|
||||
else if (str[0] == 'f')
|
||||
ch = L'\f';
|
||||
else if (str[0] == 'v')
|
||||
ch = L'\v';
|
||||
else if (str[0] == 'a')
|
||||
ch = L'\a';
|
||||
else if (octal_digit(str[0])) {
|
||||
i = 0;
|
||||
do {
|
||||
digs[dno++] = str[i++];
|
||||
} while (octal_digit(str[i]) && dno < 3);
|
||||
ch = strtol(digs, NULL, 8);
|
||||
}
|
||||
else if (str[0] == 'x') {
|
||||
while (hex_digit(str[i]) && dno < 2) {
|
||||
digs[dno++] = str[i++];
|
||||
}
|
||||
if (dno > 0)
|
||||
ch = strtol(digs, NULL, 16);
|
||||
}
|
||||
else if (str[0] == 'u') {
|
||||
while (hex_digit(str[i]) && dno < 4) {
|
||||
digs[dno++] = str[i++];
|
||||
}
|
||||
if (dno > 0)
|
||||
ch = strtol(digs, NULL, 16);
|
||||
}
|
||||
else if (str[0] == 'U') {
|
||||
while (hex_digit(str[i]) && dno < 8) {
|
||||
digs[dno++] = str[i++];
|
||||
}
|
||||
if (dno > 0)
|
||||
ch = strtol(digs, NULL, 16);
|
||||
}
|
||||
*dest = ch;
|
||||
|
||||
return i;
|
||||
}
|
||||
|
||||
/* convert a string with literal \uxxxx or \Uxxxxxxxx characters to UTF-8
|
||||
example: u8_unescape(mybuf, 256, "hello\\u220e")
|
||||
note the double backslash is needed if called on a C string literal */
|
||||
int u8_unescape(char *buf, int sz, char *src)
|
||||
{
|
||||
int c=0, amt;
|
||||
u_int32_t ch;
|
||||
char temp[4];
|
||||
|
||||
while (*src && c < sz) {
|
||||
if (*src == '\\') {
|
||||
src++;
|
||||
amt = u8_read_escape_sequence(src, &ch);
|
||||
}
|
||||
else {
|
||||
ch = (u_int32_t)*src;
|
||||
amt = 1;
|
||||
}
|
||||
src += amt;
|
||||
amt = u8_wc_toutf8(temp, ch);
|
||||
if (amt > sz-c)
|
||||
break;
|
||||
memcpy(&buf[c], temp, amt);
|
||||
c += amt;
|
||||
}
|
||||
if (c < sz)
|
||||
buf[c] = '\0';
|
||||
return c;
|
||||
}
|
||||
|
||||
int u8_escape_wchar(char *buf, int sz, u_int32_t ch)
|
||||
{
|
||||
if (ch == L'\n')
|
||||
return snprintf(buf, sz, "\\n");
|
||||
else if (ch == L'\t')
|
||||
return snprintf(buf, sz, "\\t");
|
||||
else if (ch == L'\r')
|
||||
return snprintf(buf, sz, "\\r");
|
||||
else if (ch == L'\b')
|
||||
return snprintf(buf, sz, "\\b");
|
||||
else if (ch == L'\f')
|
||||
return snprintf(buf, sz, "\\f");
|
||||
else if (ch == L'\v')
|
||||
return snprintf(buf, sz, "\\v");
|
||||
else if (ch == L'\a')
|
||||
return snprintf(buf, sz, "\\a");
|
||||
else if (ch == L'\\')
|
||||
return snprintf(buf, sz, "\\\\");
|
||||
else if (ch < 32 || ch == 0x7f)
|
||||
return snprintf(buf, sz, "\\x%hhX", (unsigned char)ch);
|
||||
else if (ch > 0xFFFF)
|
||||
return snprintf(buf, sz, "\\U%.8X", (u_int32_t)ch);
|
||||
else if (ch >= 0x80 && ch <= 0xFFFF)
|
||||
return snprintf(buf, sz, "\\u%.4hX", (unsigned short)ch);
|
||||
|
||||
return snprintf(buf, sz, "%c", (char)ch);
|
||||
}
|
||||
|
||||
int u8_escape(char *buf, int sz, char *src, int escape_quotes)
|
||||
{
|
||||
int c=0, i=0, amt;
|
||||
|
||||
while (src[i] && c < sz) {
|
||||
if (escape_quotes && src[i] == '"') {
|
||||
amt = snprintf(buf, sz - c, "\\\"");
|
||||
i++;
|
||||
}
|
||||
else {
|
||||
amt = u8_escape_wchar(buf, sz - c, u8_nextchar(src, &i));
|
||||
}
|
||||
c += amt;
|
||||
buf += amt;
|
||||
}
|
||||
if (c < sz)
|
||||
*buf = '\0';
|
||||
return c;
|
||||
}
|
||||
|
||||
char *u8_strchr(char *s, u_int32_t ch, int *charn)
|
||||
{
|
||||
int i = 0, lasti=0;
|
||||
u_int32_t c;
|
||||
|
||||
*charn = 0;
|
||||
while (s[i]) {
|
||||
c = u8_nextchar(s, &i);
|
||||
if (c == ch) {
|
||||
return &s[lasti];
|
||||
}
|
||||
lasti = i;
|
||||
(*charn)++;
|
||||
}
|
||||
return NULL;
|
||||
}
|
||||
|
||||
char *u8_memchr(char *s, u_int32_t ch, size_t sz, int *charn)
|
||||
{
|
||||
int i = 0, lasti=0;
|
||||
u_int32_t c;
|
||||
int csz;
|
||||
|
||||
*charn = 0;
|
||||
while (i < sz) {
|
||||
c = csz = 0;
|
||||
do {
|
||||
c <<= 6;
|
||||
c += (unsigned char)s[i++];
|
||||
csz++;
|
||||
} while (i < sz && !isutf(s[i]));
|
||||
c -= offsetsFromUTF8[csz-1];
|
||||
|
||||
if (c == ch) {
|
||||
return &s[lasti];
|
||||
}
|
||||
lasti = i;
|
||||
(*charn)++;
|
||||
}
|
||||
return NULL;
|
||||
}
|
||||
|
||||
int u8_is_locale_utf8(char *locale)
|
||||
{
|
||||
/* this code based on libutf8 */
|
||||
const char* cp = locale;
|
||||
|
||||
for (; *cp != '\0' && *cp != '@' && *cp != '+' && *cp != ','; cp++) {
|
||||
if (*cp == '.') {
|
||||
const char* encoding = ++cp;
|
||||
for (; *cp != '\0' && *cp != '@' && *cp != '+' && *cp != ','; cp++)
|
||||
;
|
||||
if ((cp-encoding == 5 && !strncmp(encoding, "UTF-8", 5))
|
||||
|| (cp-encoding == 4 && !strncmp(encoding, "utf8", 4)))
|
||||
return 1; /* it's UTF-8 */
|
||||
break;
|
||||
}
|
||||
}
|
||||
return 0;
|
||||
}
|
||||
|
||||
int u8_vprintf(char *fmt, va_list ap)
|
||||
{
|
||||
int cnt, sz=0;
|
||||
char *buf;
|
||||
u_int32_t *wcs;
|
||||
|
||||
sz = 512;
|
||||
buf = (char*)alloca(sz);
|
||||
try_print:
|
||||
cnt = vsnprintf(buf, sz, fmt, ap);
|
||||
if (cnt >= sz) {
|
||||
buf = (char*)alloca(cnt - sz + 1);
|
||||
sz = cnt + 1;
|
||||
goto try_print;
|
||||
}
|
||||
wcs = (u_int32_t*)alloca((cnt+1) * sizeof(u_int32_t));
|
||||
cnt = u8_toucs(wcs, cnt+1, buf, cnt);
|
||||
printf("%ls", (wchar_t*)wcs);
|
||||
return cnt;
|
||||
}
|
||||
|
||||
int u8_printf(char *fmt, ...)
|
||||
{
|
||||
int cnt;
|
||||
va_list args;
|
||||
|
||||
va_start(args, fmt);
|
||||
|
||||
cnt = u8_vprintf(fmt, args);
|
||||
|
||||
va_end(args);
|
||||
return cnt;
|
||||
}
|
||||
#endif
|
||||
|
||||
////////////////////////////////////////////////////////////////////////////////
|
||||
// taskwarrior - a command line task list manager.
|
||||
//
|
||||
// Copyright 2006 - 2011, Paul Beckingham, Federico Hernandez.
|
||||
// Copyright 2010 - 2011, Paul Beckingham, Federico Hernandez.
|
||||
// All rights reserved.
|
||||
//
|
||||
// This program is free software; you can redistribute it and/or modify it under
|
||||
|
|
@ -504,42 +25,145 @@ int u8_printf(char *fmt, ...)
|
|||
//
|
||||
////////////////////////////////////////////////////////////////////////////////
|
||||
|
||||
#include "utf8.h"
|
||||
|
||||
static const u_int32_t offsetsFromUTF8[6] =
|
||||
{
|
||||
0x00000000UL, 0x00003080UL, 0x000E2080UL,
|
||||
0x03C82080UL, 0xFA082080UL, 0x82082080UL,
|
||||
};
|
||||
#include <string>
|
||||
#include <utf8.h>
|
||||
|
||||
////////////////////////////////////////////////////////////////////////////////
|
||||
// number of characters in a string.
|
||||
int utf8_length (const std::string& s)
|
||||
{
|
||||
int count = 0;
|
||||
int i = 0;
|
||||
|
||||
while (utf8_nextchar (s.c_str (), &i) != 0)
|
||||
count++;
|
||||
|
||||
return count;
|
||||
}
|
||||
// Converts '0' -> 0
|
||||
// '9' -> 9
|
||||
// 'a'/'A' -> 10
|
||||
// 'f'/'F' -> 15
|
||||
#define XDIGIT(x) ((x) >= '0' && (x) <= '9' ? ((x) - '0') : \
|
||||
(x) >= 'a' && (x) <= 'f' ? ((x) + 10 - 'a') : \
|
||||
(x) >= 'A' && (x) <= 'F' ? ((x) + 10 - 'A') : 0)
|
||||
|
||||
////////////////////////////////////////////////////////////////////////////////
|
||||
// reads the next utf-8 sequence out of a string, updating an index.
|
||||
u_int32_t utf8_nextchar (const char* s, int* i)
|
||||
// Note: Assumes 4-digit hex codepoints:
|
||||
// xxxx
|
||||
// \uxxxx
|
||||
// U+xxxx
|
||||
unsigned int utf8_codepoint (const std::string& input)
|
||||
{
|
||||
u_int32_t ch = 0;
|
||||
int sz = 0;
|
||||
unsigned int codepoint = 0;
|
||||
int length = input.length ();
|
||||
|
||||
do
|
||||
// U+xxxx, \uxxxx
|
||||
if (length >= 6 &&
|
||||
((input[0] == 'U' && input[1] == '+') ||
|
||||
(input[0] == '\\' && input[1] == 'u')))
|
||||
{
|
||||
ch <<= 6;
|
||||
ch += (unsigned char) s[(*i)++];
|
||||
sz++;
|
||||
codepoint = XDIGIT (input[2]) << 12 |
|
||||
XDIGIT (input[3]) << 8 |
|
||||
XDIGIT (input[4]) << 4 |
|
||||
XDIGIT (input[5]);
|
||||
}
|
||||
while (s[*i] && ! isutf (s[*i]));
|
||||
else if (length >= 4)
|
||||
{
|
||||
codepoint = XDIGIT (input[0]) << 12 |
|
||||
XDIGIT (input[1]) << 8 |
|
||||
XDIGIT (input[2]) << 4 |
|
||||
XDIGIT (input[3]);
|
||||
}
|
||||
else
|
||||
throw std::string ("Invalid codepoint representation.");
|
||||
|
||||
return ch - offsetsFromUTF8[sz - 1];
|
||||
return codepoint;
|
||||
}
|
||||
|
||||
////////////////////////////////////////////////////////////////////////////////
|
||||
// Iterates along a UTF8 string.
|
||||
unsigned int utf8_next_char (const std::string& input, std::string::size_type& i)
|
||||
{
|
||||
// How many bytes in the sequence?
|
||||
int length = utf8_sequence (input[i]);
|
||||
|
||||
// 0xxxxxxx -> 0xxxxxxx
|
||||
if (length == 1)
|
||||
return input[i++];
|
||||
|
||||
// 110yyyyy 10xxxxxx -> 00000yyy yyxxxxxx
|
||||
if (length == 2)
|
||||
return ((input[i++] & 0x1F) << 6) +
|
||||
(input[i++] & 0x3F);
|
||||
|
||||
// 1110zzzz 10yyyyyy 10xxxxxx -> zzzzyyyy yyxxxxxx
|
||||
if (length == 3)
|
||||
return ((input[i++] & 0xF) << 12) +
|
||||
((input[i++] & 0x3F) << 6) +
|
||||
(input[i++] & 0x3F);
|
||||
|
||||
// 11110www 10zzzzzz 10yyyyyy 10xxxxxx -> 000wwwzz zzzzyyyy yyxxxxxx
|
||||
if (length == 4)
|
||||
return ((input[i++] & 0x7) << 18) +
|
||||
((input[i++] & 0x3F) << 12) +
|
||||
((input[i++] & 0x3F) << 6) +
|
||||
(input[i++] & 0x3F);
|
||||
|
||||
|
||||
// Default: pretend as though it's a single character.
|
||||
// TODO Or should this throw?
|
||||
return input[i++];
|
||||
}
|
||||
|
||||
////////////////////////////////////////////////////////////////////////////////
|
||||
// http://en.wikipedia.org/wiki/UTF-8
|
||||
std::string utf8_character (unsigned int codepoint)
|
||||
{
|
||||
char sequence[5];
|
||||
|
||||
// 0xxxxxxx -> 0xxxxxxx
|
||||
if (codepoint < 0x80)
|
||||
{
|
||||
sequence[0] = codepoint;
|
||||
sequence[1] = 0;
|
||||
}
|
||||
|
||||
// 00000yyy yyxxxxxx -> 110yyyyy 10xxxxxx
|
||||
else if (codepoint < 0x800)
|
||||
{
|
||||
sequence[0] = 0xC0 | (codepoint & 0x7C0) >> 6;
|
||||
sequence[1] = 0x80 | (codepoint & 0x3F);
|
||||
sequence[2] = 0;
|
||||
}
|
||||
|
||||
// zzzzyyyy yyxxxxxx -> 1110zzzz 10yyyyyy 10xxxxxx
|
||||
else if (codepoint < 0x10000)
|
||||
{
|
||||
sequence[0] = 0xE0 | (codepoint & 0xF000) >> 12;
|
||||
sequence[1] = 0x80 | (codepoint & 0xFC0) >> 6;
|
||||
sequence[2] = 0x80 | (codepoint & 0x3F);
|
||||
sequence[3] = 0;
|
||||
}
|
||||
|
||||
// 000wwwzz zzzzyyyy yyxxxxxx -> 11110www 10zzzzzz 10yyyyyy 10xxxxxx
|
||||
else if (codepoint < 0x110000)
|
||||
{
|
||||
sequence[0] = 0xF0 | (codepoint & 0x1C0000) >> 18;
|
||||
sequence[1] = 0x80 | (codepoint & 0x03F000) >> 12;
|
||||
sequence[2] = 0x80 | (codepoint & 0x0FC0) >> 6;
|
||||
sequence[3] = 0x80 | (codepoint & 0x3F);
|
||||
sequence[4] = 0;
|
||||
}
|
||||
else
|
||||
throw std::string ("Invalid Unicode codepoint.");
|
||||
|
||||
sequence[4] = '\0';
|
||||
return std::string (sequence);
|
||||
}
|
||||
|
||||
////////////////////////////////////////////////////////////////////////////////
|
||||
int utf8_sequence (unsigned int character)
|
||||
{
|
||||
if ((character & 0xE0) == 0xC0)
|
||||
return 2;
|
||||
|
||||
if ((character & 0xF0) == 0xE0)
|
||||
return 3;
|
||||
|
||||
if ((character & 0xF8) == 0xF0)
|
||||
return 4;
|
||||
|
||||
return 1;
|
||||
}
|
||||
|
||||
////////////////////////////////////////////////////////////////////////////////
|
||||
|
|
|
|||
Loading…
Add table
Add a link
Reference in a new issue