Improve width estimation (#2033)

This commit is contained in:
Victor Zverovich
2021-02-13 09:24:39 -08:00
parent 13b117b5bc
commit 7e72673d87
3 changed files with 25 additions and 3 deletions

View File

@@ -695,7 +695,28 @@ FMT_CONSTEXPR inline size_t compute_width(string_view s) {
// It is not a lambda for compatibility with C++14.
struct count_code_points {
size_t* count;
FMT_CONSTEXPR void operator()(uint32_t, int) const { ++*count; }
FMT_CONSTEXPR void operator()(uint32_t cp, int error) const {
*count +=
1 +
(error == 0 && cp >= 0x1100 &&
(cp <= 0x115f || // Hangul Jamo init. consonants
cp == 0x2329 || // LEFT-POINTING ANGLE BRACKET〈
cp == 0x232a || // RIGHT-POINTING ANGLE BRACKET 〉
// CJK ... Yi except Unicode Character “〿”:
(cp >= 0x2e80 && cp <= 0xa4cf && cp != 0x303f) ||
(cp >= 0xac00 && cp <= 0xd7a3) || // Hangul Syllables
(cp >= 0xf900 && cp <= 0xfaff) || // CJK Compatibility Ideographs
(cp >= 0xfe10 && cp <= 0xfe19) || // Vertical Forms
(cp >= 0xfe30 && cp <= 0xfe6f) || // CJK Compatibility Forms
(cp >= 0xff00 && cp <= 0xff60) || // Fullwidth Forms
(cp >= 0xffe0 && cp <= 0xffe6) || // Fullwidth Forms
(cp >= 0x20000 && cp <= 0x2fffd) || // CJK
(cp >= 0x30000 && cp <= 0x3fffd) ||
// Miscellaneous Symbols and Pictographs + Emoticons:
(cp >= 0x1f300 && cp <= 0x1f64f) ||
// Supplemental Symbols and Pictographs:
(cp >= 0x1f900 && cp <= 0x1f9ff)));
}
};
for_each_codepoint(s, count_code_points{&num_code_points});
return num_code_points;