1 files changed, 79 insertions, 13 deletions

diff --git a/txt.c b/txt.c
index 9f29f84..8fcafe9 100644
--- a/txt.c
+++ b/txt.c
@@ -9,6 +9,7 @@
 #include "wrmr.h"
 #include "dynarr.h"
 #include "txt.h"
+#include "utf8.h"
 
 void txt_replace_piece(Txt *b, u32 pi, TxtBufIdx buf, u32 ofs, u32 n) { 
 	b->ptbl.v[pi] = (TxtPiece) { buf, ofs, n };
@@ -39,16 +40,6 @@ u32 txt_split_piece(Txt *b, u32 pi, u32 i) {
 	return pi + 1;
 }
 
-TxtLoc txt_at(Txt *b, u32 cur) {
-	for (u32 i = 0; i < b->ptbl.n; i++) {
-		if (cur < b->ptbl.v[i].n) {
-			return (TxtLoc) { i, cur };
-		}
-		cur -= b->ptbl.v[i].n;
-	}
-	return (TxtLoc) { b->ptbl.n - 1, b->ptbl.v[b->ptbl.n - 1].n };
-}
-
 static void txt_buf_fit(Txt *b, TxtBufIdx bi, u32 sz) {
 	TxtBuf *buf = &b->buf[bi];
 	if (sz > buf->c) {
@@ -98,9 +89,11 @@ u32 txt_insert(Txt *b, u32 cur, const char *s, u32 n) {
 	return cur + n;
 }
 
-u32 txt_insert_c(Txt *b, u32 cur, char ch) {
-	/* TODO: utf-8 char */
-	return txt_insert(b, cur, &ch, 1);
+u32 txt_insert_c(Txt *b, u32 cur, u32 ch) {
+	char buf[6];
+	u32 n = utf8_encode_len(&ch, 1);
+	utf8_encode(buf, &ch, 1);
+	return txt_insert(b, cur, buf, n);
 }
 
 static int txt_are_pieces_adjacent(Txt *t, u32 a, u32 b) {
@@ -144,6 +137,13 @@ u32 txt_delete(Txt *b, u32 cur, u32 n) {
 	return cur;
 }
 
+u32 txt_delete_c(Txt *b, u32 cur) {
+	while (cur > 0 && (txt_byte(b, txt_at(b, cur-1)) & 0xc0) == 0x80) 
+		cur = txt_delete(b, cur, 1);
+	cur = txt_delete(b, cur, 1);
+	return cur;
+}
+
 int txt_load(Txt *b, const char *path) {
 	struct stat sb;
 	int fd = open(path, O_RDONLY);
@@ -192,3 +192,69 @@ void txt_free(Txt *t) {
 	}
 	free(t->ptbl.v);
 }
+
+/* reading individual chars */
+
+u8 txt_byte(Txt *t, TxtLoc l) {
+	TxtPiece *p = &t->ptbl.v[l.p];
+	TxtBuf *b = &t->buf[p->buf];
+	if (l.p < t->ptbl.n && l.i == p->n) p++, l.i = 0;
+	if (p->ofs + l.i >= b->n) return 0;
+	return b->s[p->ofs + l.i];
+}
+
+u32 txt_chr(Txt *t, TxtLoc l) {
+	TxtPiece *p = &t->ptbl.v[l.p];
+	TxtBuf *b = &t->buf[p->buf];
+	if (l.p < t->ptbl.n && l.i == p->n) p++, l.i = 0;
+	return utf8_decode_at(b->s, p->ofs + l.i, b->n);
+}
+
+/* cursor manipulation */
+
+TxtLoc txt_at(Txt *b, u32 cur) {
+	for (u32 i = 0; i < b->ptbl.n; i++) {
+		if (cur < b->ptbl.v[i].n) {
+			return (TxtLoc) { i, cur };
+		}
+		cur -= b->ptbl.v[i].n;
+	}
+	return (TxtLoc) { b->ptbl.n - 1, b->ptbl.v[b->ptbl.n - 1].n };
+}
+
+u32 txt_ofs(Txt *b, TxtLoc l) {
+	u32 r = 0;
+	for (u32 i = 0; i < l.p; i++) {
+		r += b->ptbl.v[i].n;
+	}
+	return r + l.i;
+}
+
+TxtLoc txt_next(Txt *b, TxtLoc l) {
+	TxtPiece *p = &b->ptbl.v[l.p];
+	if (l.i+1 < p->n) return (TxtLoc) { l.p, l.i + 1 };
+	if (l.p+1 < b->ptbl.n) {
+		return (TxtLoc) { l.p + 1, l.i };
+	} else {
+		return (TxtLoc) { l.p, p->n };
+	}
+}
+
+TxtLoc txt_prev(Txt *b, TxtLoc l) {
+	if (l.i > 0) {
+		return (TxtLoc) { l.p, l.i - 1 };
+	} else if (l.p > 0) {
+		return (TxtLoc) { l.p - 1, b->ptbl.v[l.p - 1].n - 1 };
+	} else {
+		return (TxtLoc) { 0, 0 };
+	}
+}
+
+int txt_at_start(Txt *b, TxtLoc l) {
+	(void)b;
+	return l.p == 0 && l.i == 0;
+}
+
+int txt_at_end(Txt *b, TxtLoc l) {
+	return l.p + 1 == b->ptbl.n && l.i == b->ptbl.v[l.p].n;
+}