Convert UTF-8 to 32-bit character, process single character input. Based on stb_from_utf8() from github.com/nothings/stb/ We handle UTF-8 decoding error by skipping forward.
| 1552 | // Based on stb_from_utf8() from github.com/nothings/stb/ |
| 1553 | // We handle UTF-8 decoding error by skipping forward. |
| 1554 | int ImTextCharFromUtf8(unsigned int* out_char, const char* in_text, const char* in_text_end) |
| 1555 | { |
| 1556 | unsigned int c = (unsigned int)-1; |
| 1557 | const unsigned char* str = (const unsigned char*)in_text; |
| 1558 | if (!(*str & 0x80)) |
| 1559 | { |
| 1560 | c = (unsigned int)(*str++); |
| 1561 | *out_char = c; |
| 1562 | return 1; |
| 1563 | } |
| 1564 | if ((*str & 0xe0) == 0xc0) |
| 1565 | { |
| 1566 | *out_char = IM_UNICODE_CODEPOINT_INVALID; // will be invalid but not end of string |
| 1567 | if (in_text_end && in_text_end - (const char*)str < 2) return 1; |
| 1568 | if (*str < 0xc2) return 2; |
| 1569 | c = (unsigned int)((*str++ & 0x1f) << 6); |
| 1570 | if ((*str & 0xc0) != 0x80) return 2; |
| 1571 | c += (*str++ & 0x3f); |
| 1572 | *out_char = c; |
| 1573 | return 2; |
| 1574 | } |
| 1575 | if ((*str & 0xf0) == 0xe0) |
| 1576 | { |
| 1577 | *out_char = IM_UNICODE_CODEPOINT_INVALID; // will be invalid but not end of string |
| 1578 | if (in_text_end && in_text_end - (const char*)str < 3) return 1; |
| 1579 | if (*str == 0xe0 && (str[1] < 0xa0 || str[1] > 0xbf)) return 3; |
| 1580 | if (*str == 0xed && str[1] > 0x9f) return 3; // str[1] < 0x80 is checked below |
| 1581 | c = (unsigned int)((*str++ & 0x0f) << 12); |
| 1582 | if ((*str & 0xc0) != 0x80) return 3; |
| 1583 | c += (unsigned int)((*str++ & 0x3f) << 6); |
| 1584 | if ((*str & 0xc0) != 0x80) return 3; |
| 1585 | c += (*str++ & 0x3f); |
| 1586 | *out_char = c; |
| 1587 | return 3; |
| 1588 | } |
| 1589 | if ((*str & 0xf8) == 0xf0) |
| 1590 | { |
| 1591 | *out_char = IM_UNICODE_CODEPOINT_INVALID; // will be invalid but not end of string |
| 1592 | if (in_text_end && in_text_end - (const char*)str < 4) return 1; |
| 1593 | if (*str > 0xf4) return 4; |
| 1594 | if (*str == 0xf0 && (str[1] < 0x90 || str[1] > 0xbf)) return 4; |
| 1595 | if (*str == 0xf4 && str[1] > 0x8f) return 4; // str[1] < 0x80 is checked below |
| 1596 | c = (unsigned int)((*str++ & 0x07) << 18); |
| 1597 | if ((*str & 0xc0) != 0x80) return 4; |
| 1598 | c += (unsigned int)((*str++ & 0x3f) << 12); |
| 1599 | if ((*str & 0xc0) != 0x80) return 4; |
| 1600 | c += (unsigned int)((*str++ & 0x3f) << 6); |
| 1601 | if ((*str & 0xc0) != 0x80) return 4; |
| 1602 | c += (*str++ & 0x3f); |
| 1603 | // utf-8 encodings of values used in surrogate pairs are invalid |
| 1604 | if ((c & 0xFFFFF800) == 0xD800) return 4; |
| 1605 | *out_char = c; |
| 1606 | return 4; |
| 1607 | } |
| 1608 | *out_char = 0; |
| 1609 | return 0; |
| 1610 | } |
| 1611 |
no outgoing calls
no test coverage detected