Add u8wcnt_human() and u8wnext_human()

3 files changed, 39 insertions, 0 deletions

diff --git a/include/unicode/string.h b/include/unicode/string.h
index 9ac029a..887a216 100644
--- a/include/unicode/string.h
+++ b/include/unicode/string.h
@@ -19,9 +19,11 @@ enum [[clang::flag_enum]] caseflags {
 
 [[nodiscard]] size_t u8gcnt(const char8_t *, size_t);
 [[nodiscard]] size_t u8wcnt(const char8_t *, size_t);
+[[nodiscard]] size_t u8wcnt_human(const char8_t *, size_t);
 
 size_t u8gnext(struct u8view *, const char8_t **, size_t *);
 size_t u8wnext(struct u8view *, const char8_t **, size_t *);
+size_t u8wnext_human(struct u8view *, const char8_t **, size_t *);
 
 #define mlib_warn_trunc nodiscard("don’t forget to check for truncation")
 [[mlib_warn_trunc]] size_t u8casefold(char8_t *restrict, size_t,
diff --git a/lib/unicode/string/u8wcnt_human.c b/lib/unicode/string/u8wcnt_human.c
new file mode 100644
index 0000000..6e70398
--- /dev/null
+++ b/lib/unicode/string/u8wcnt_human.c
@@ -0,0 +1,10 @@
+#include "unicode/string.h"
+
+size_t
+u8wcnt_human(const char8_t *s, size_t n)
+{
+	size_t m = 0;
+	while (u8wnext_human(nullptr, &s, &n))
+		m++;
+	return m;
+}
diff --git a/lib/unicode/string/u8wnext_human.c b/lib/unicode/string/u8wnext_human.c
new file mode 100644
index 0000000..d85abf1
--- /dev/null
+++ b/lib/unicode/string/u8wnext_human.c
@@ -0,0 +1,27 @@
+#include "macros.h"
+#include "mbstring.h"
+#include "unicode/prop.h"
+#include "unicode/string.h"
+
+size_t
+u8wnext_human(struct u8view *dst, const char8_t **s, size_t *n)
+{
+	ASSUME(n != nullptr);
+	ASSUME(s != nullptr);
+	ASSUME(*s != nullptr);
+
+	struct u8view w;
+	while (u8wnext(&w, s, n)) {
+		rune ch;
+		struct u8view cpy = w;
+		while (u8next(&ch, U8_ARGSP(cpy))) {
+			if (uprop_get_gc(ch) & (GC_L | GC_N)) {
+				if (dst != nullptr)
+					*dst = w;
+				return w.len;
+			}
+		}
+	}
+
+	return 0;
+}