v/vlib/builtin/utf8.v

209 lines
4.3 KiB
V
Raw Normal View History

2020-02-03 05:00:36 +01:00
// Copyright (c) 2019-2020 Alexander Medvednikov. All rights reserved.
2019-06-23 04:21:30 +02:00
// Use of this source code is governed by an MIT license
// that can be found in the LICENSE file.
2019-06-22 20:20:28 +02:00
module builtin
2019-09-15 14:36:05 +02:00
pub fn utf8_char_len(b byte) int {
2019-12-19 21:52:45 +01:00
return ((0xe5000000>>((b>>3) & 0x1e)) & 3) + 1
2019-06-22 20:20:28 +02:00
}
// Convert utf32 to utf8
// utf32 == Codepoint
pub fn utf32_to_str(code u32) string {
2019-12-19 21:52:45 +01:00
icode := int(code) // Prevents doing casts everywhere
mut buffer := malloc(5)
if icode <= 127/* 0x7F */ {
buffer[0] = icode
return tos(buffer, 1)
}
2019-12-19 21:52:45 +01:00
if (icode <= 2047/* 0x7FF */) {
buffer[0] = 192/*0xC0*/ | (icode>>6)/* 110xxxxx */
buffer[1] = 128/*0x80*/ | (icode & 63/*0x3F*/)/* 10xxxxxx */
return tos(buffer, 2)
}
2019-12-19 21:52:45 +01:00
if (icode <= 65535/* 0xFFFF */) {
buffer[0] = 224/*0xE0*/ | (icode>>12)/* 1110xxxx */
buffer[1] = 128/*0x80*/ | ((icode>>6) & 63/*0x3F*/)/* 10xxxxxx */
buffer[2] = 128/*0x80*/ | (icode & 63/*0x3F*/)/* 10xxxxxx */
return tos(buffer, 3)
}
2019-12-19 21:52:45 +01:00
if (icode <= 1114111/* 0x10FFFF */) {
buffer[0] = 240/*0xF0*/ | (icode>>18)/* 11110xxx */
buffer[1] = 128/*0x80*/ | ((icode>>12) & 63/*0x3F*/)/* 10xxxxxx */
buffer[2] = 128/*0x80*/ | ((icode>>6) & 63/*0x3F*/)/* 10xxxxxx */
buffer[3] = 128/*0x80*/ | (icode & 63/*0x3F*/)/* 10xxxxxx */
return tos(buffer, 4)
}
2019-06-22 20:20:28 +02:00
return ''
}
// TODO copypasta
pub fn utf32_to_str_no_malloc(code u32, buf voidptr) string {
2019-12-19 21:52:45 +01:00
icode := int(code) // Prevents doing casts everywhere
mut buffer := byteptr(buf)
if icode <= 127/* 0x7F */ {
buffer[0] = icode
return tos(buffer, 1)
}
2019-12-19 21:52:45 +01:00
if (icode <= 2047/* 0x7FF */) {
buffer[0] = 192/*0xC0*/ | (icode>>6)/* 110xxxxx */
buffer[1] = 128/*0x80*/ | (icode & 63/*0x3F*/)/* 10xxxxxx */
return tos(buffer, 2)
}
2019-12-19 21:52:45 +01:00
if (icode <= 65535/* 0xFFFF */) {
buffer[0] = 224/*0xE0*/ | (icode>>12)/* 1110xxxx */
buffer[1] = 128/*0x80*/ | ((icode>>6) & 63/*0x3F*/)/* 10xxxxxx */
buffer[2] = 128/*0x80*/ | (icode & 63/*0x3F*/)/* 10xxxxxx */
return tos(buffer, 3)
}
2019-12-19 21:52:45 +01:00
if (icode <= 1114111/* 0x10FFFF */) {
buffer[0] = 240/*0xF0*/ | (icode>>18)/* 11110xxx */
buffer[1] = 128/*0x80*/ | ((icode>>12) & 63/*0x3F*/)/* 10xxxxxx */
buffer[2] = 128/*0x80*/ | ((icode>>6) & 63/*0x3F*/)/* 10xxxxxx */
buffer[3] = 128/*0x80*/ | (icode & 63/*0x3F*/)/* 10xxxxxx */
return tos(buffer, 4)
}
2019-06-22 20:20:28 +02:00
return ''
}
// Convert utf8 to utf32
pub fn (_rune string) utf32_code() int {
2019-06-22 20:20:28 +02:00
if _rune.len == 0 {
return 0
}
// save ASC symbol as is
if _rune.len == 1 {
return int(_rune[0])
}
mut b := byte(int(_rune[0]))
2019-06-22 20:20:28 +02:00
// TODO should be
// res := int( rune[0] << rune.len)
2019-12-19 21:52:45 +01:00
b = b<<_rune.len
mut res := int(b)
2019-06-22 20:20:28 +02:00
mut shift := 6 - _rune.len
for i := 1; i < _rune.len; i++ {
c := int(_rune[i])
2019-12-19 21:52:45 +01:00
res = res<<shift
res |= c & 63 // 0x3f
2019-06-22 20:20:28 +02:00
shift = 6
}
return res
}
2019-07-24 12:16:45 +02:00
const (
2019-12-19 21:52:45 +01:00
CP_UTF8 = 65001
)
2019-07-24 12:16:45 +02:00
pub fn (_str string) to_wide() &u16 {
2019-11-16 00:30:50 +01:00
$if windows {
2019-12-07 14:58:43 +01:00
num_chars := (C.MultiByteToWideChar(CP_UTF8, 0, _str.str, _str.len, 0, 0))
2019-11-16 00:30:50 +01:00
mut wstr := &u16(malloc((num_chars + 1) * 2)) // sizeof(wchar_t)
if !isnil(wstr) {
C.MultiByteToWideChar(CP_UTF8, 0, _str.str, _str.len, wstr, num_chars)
C.memset(&byte(wstr) + num_chars * 2, 0, 2)
}
return wstr
} $else {
return 0
}
2019-07-24 12:16:45 +02:00
}
pub fn string_from_wide(_wstr &u16) string {
2019-12-19 21:52:45 +01:00
$if windows {
wstr_len := C.wcslen(_wstr)
return string_from_wide2(_wstr, wstr_len)
} $else {
return ''
}
2019-07-24 12:16:45 +02:00
}
pub fn string_from_wide2(_wstr &u16, len int) string {
2019-12-19 21:52:45 +01:00
$if windows {
num_chars := C.WideCharToMultiByte(CP_UTF8, 0, _wstr, len, 0, 0, 0, 0)
mut str_to := malloc(num_chars + 1)
if !isnil(str_to) {
C.WideCharToMultiByte(CP_UTF8, 0, _wstr, len, str_to, num_chars, 0, 0)
C.memset(str_to + num_chars, 0, 1)
}
return tos2(str_to)
} $else {
return ''
}
2019-07-24 12:16:45 +02:00
}
// Calculate length to read from the first byte
fn utf8_len(c byte) int {
2019-12-19 21:52:45 +01:00
mut b := 0
mut x := c
if ((x & 240) != 0) {
// 0xF0
x >>= 4
}
else {
b += 4
}
if ((x & 12) != 0) {
// 0x0C
x >>= 2
}
else {
b += 2
}
if ((x & 2) == 0) {
// 0x02
b++
}
return b
}
// Reads an utf8 character from standard input
pub fn utf8_getchar() int {
2019-12-19 21:52:45 +01:00
c := C.getchar()
len := utf8_len(~c)
if c < 0 {
return 0
}
else if len == 0 {
return c
}
else if len == 1 {
return -1
}
else {
mut uc := c & ((1<<(7 - len)) - 1)
for i := 0; i + 1 < len; i++ {
c2 := C.getchar()
if c2 != -1 && (c2>>6) == 2 {
uc <<= 6
uc |= (c2 & 63)
}
else if c2 == -1 {
return 0
}
else {
return -1
}
}
return uc
}
}
2019-12-19 21:52:45 +01:00