Add support for utf-8 display width

Supports wide characters and combining characters

Signed-off-by: Steve Bennett <steveb@workware.net.au>

diff --git a/utf8.c b/utf8.c
index db8f7c11f6dfafdbf0dccb72583606fbc9ca96b6..966ca854847efc30887f9887f42602e8a01ac630 100644 (file)
--- a/utf8.c
+++ b/utf8.c
@@ -74,6 +74,19 @@ int utf8_strlen(const char *str, int bytelen)
     return charlen;
 }
 
+int utf8_strwidth(const char *str, int charlen)
+{
+    int width = 0;
+    while (charlen) {
+        int c;
+        int l = utf8_tounicode(str, &c);
+        width += utf8_width(c);
+        str += l;
+        charlen--;
+    }
+    return width;
+}
+
 int utf8_index(const char *str, int index)
 {
     const char *s = str;
@@ -116,4 +129,95 @@ int utf8_tounicode(const char *str, int *uc)
     return 1;
 }
 
+struct utf8range {
+    int lower;     /* lower inclusive */
+    int upper;     /* upper exclusive */
+};
+
+/* From http://unicode.org/Public/UNIDATA/UnicodeData.txt */
+static const struct utf8range unicode_range_combining[] = {
+        { 0x0300, 0x0370 },     { 0x0483, 0x048a },     { 0x0591, 0x05be },     { 0x05bf, 0x05c0 },
+        { 0x05c1, 0x05c3 },     { 0x05c4, 0x05c6 },     { 0x05c7, 0x05d0 },     { 0x0610, 0x061b },
+        { 0x064b, 0x0660 },     { 0x0670, 0x0671 },     { 0x06d6, 0x06dd },     { 0x06de, 0x06e5 },
+        { 0x06e7, 0x06e9 },     { 0x06ea, 0x06ee },     { 0x0711, 0x0712 },     { 0x0730, 0x074d },
+        { 0x07a6, 0x07b1 },     { 0x07eb, 0x07f4 },     { 0x0816, 0x081a },     { 0x081b, 0x0824 },
+        { 0x0825, 0x0828 },     { 0x0829, 0x0830 },     { 0x0900, 0x0904 },     { 0x093c, 0x093d },
+        { 0x093e, 0x0950 },     { 0x0951, 0x0958 },     { 0x0962, 0x0964 },     { 0x0981, 0x0985 },
+        { 0x09bc, 0x09bd },     { 0x09be, 0x09ce },     { 0x09d7, 0x09dc },     { 0x09e2, 0x09e6 },
+        { 0x0a01, 0x0a05 },     { 0x0a3c, 0x0a59 },     { 0x0a70, 0x0a72 },     { 0x0a75, 0x0a85 },
+        { 0x0abc, 0x0abd },     { 0x0abe, 0x0ad0 },     { 0x0ae2, 0x0ae6 },     { 0x0b01, 0x0b05 },
+        { 0x0b3c, 0x0b3d },     { 0x0b3e, 0x0b5c },     { 0x0b62, 0x0b66 },     { 0x0b82, 0x0b83 },
+        { 0x0bbe, 0x0bd0 },     { 0x0bd7, 0x0be6 },     { 0x0c01, 0x0c05 },     { 0x0c3e, 0x0c58 },
+        { 0x0c62, 0x0c66 },     { 0x0c82, 0x0c85 },     { 0x0cbc, 0x0cbd },     { 0x0cbe, 0x0cde },
+        { 0x0ce2, 0x0ce6 },     { 0x0d02, 0x0d05 },     { 0x0d3e, 0x0d60 },     { 0x0d62, 0x0d66 },
+        { 0x0d82, 0x0d85 },     { 0x0dca, 0x0df4 },     { 0x0e31, 0x0e32 },     { 0x0e34, 0x0e3f },
+        { 0x0e47, 0x0e4f },     { 0x0eb1, 0x0eb2 },     { 0x0eb4, 0x0ebd },     { 0x0ec8, 0x0ed0 },
+        { 0x0f18, 0x0f1a },     { 0x0f35, 0x0f36 },     { 0x0f37, 0x0f38 },     { 0x0f39, 0x0f3a },
+        { 0x0f3e, 0x0f40 },     { 0x0f71, 0x0f85 },     { 0x0f86, 0x0f88 },     { 0x0f90, 0x0fbe },
+        { 0x0fc6, 0x0fc7 },     { 0x102b, 0x103f },     { 0x1056, 0x105a },     { 0x105e, 0x1061 },
+        { 0x1062, 0x1065 },     { 0x1067, 0x106e },     { 0x1071, 0x1075 },     { 0x1082, 0x108e },
+        { 0x108f, 0x1090 },     { 0x109a, 0x109e },     { 0x135f, 0x1360 },     { 0x1712, 0x1720 },
+        { 0x1732, 0x1735 },     { 0x1752, 0x1760 },     { 0x1772, 0x1780 },     { 0x17b6, 0x17d4 },
+        { 0x17dd, 0x17e0 },     { 0x180b, 0x180e },     { 0x18a9, 0x18aa },     { 0x1920, 0x1940 },
+        { 0x19b0, 0x19c1 },     { 0x19c8, 0x19d0 },     { 0x1a17, 0x1a1e },     { 0x1a55, 0x1a80 },
+        { 0x1b00, 0x1b05 },     { 0x1b34, 0x1b45 },     { 0x1b6b, 0x1b74 },     { 0x1b80, 0x1b83 },
+        { 0x1ba1, 0x1bae },     { 0x1c24, 0x1c3b },     { 0x1cd0, 0x1cd3 },     { 0x1cd4, 0x1ce9 },
+        { 0x1ced, 0x1cee },     { 0x1cf2, 0x1d00 },     { 0x1dc0, 0x1e00 },     { 0x20d0, 0x2100 },
+        { 0x2cef, 0x2cf9 },     { 0x2de0, 0x2e00 },     { 0x302a, 0x3030 },     { 0x3099, 0x309b },
+        { 0xa66f, 0xa673 },     { 0xa67c, 0xa67e },     { 0xa6f0, 0xa6f2 },     { 0xa802, 0xa803 },
+        { 0xa806, 0xa807 },     { 0xa80b, 0xa80c },     { 0xa823, 0xa828 },     { 0xa880, 0xa882 },
+        { 0xa8b4, 0xa8ce },     { 0xa8e0, 0xa8f2 },     { 0xa926, 0xa92e },     { 0xa947, 0xa95f },
+        { 0xa980, 0xa984 },     { 0xa9b3, 0xa9c1 },     { 0xaa29, 0xaa40 },     { 0xaa43, 0xaa44 },
+        { 0xaa4c, 0xaa50 },     { 0xaa7b, 0xaa80 },     { 0xaab0, 0xaab1 },     { 0xaab2, 0xaab5 },
+        { 0xaab7, 0xaab9 },     { 0xaabe, 0xaac0 },     { 0xaac1, 0xaac2 },     { 0xabe3, 0xabeb },
+        { 0xabec, 0xabf0 },     { 0xfb1e, 0xfb1f },     { 0xfe00, 0xfe10 },     { 0xfe20, 0xfe30 },
+};
+
+static const struct utf8range unicode_range_wide[] = {
+        { 0x1100, 0x115f },     { 0x2329, 0x232a },     { 0x2e80, 0x2e99 },     { 0x2e9b, 0x2ef3 },
+        { 0x2f00, 0x2fd5 },     { 0x2ff0, 0x2ffb },     { 0x3000, 0x303e },     { 0x3041, 0x3096 },
+        { 0x3099, 0x30ff },     { 0x3105, 0x312d },     { 0x3131, 0x318e },     { 0x3190, 0x31ba },
+        { 0x31c0, 0x31e3 },     { 0x31f0, 0x321e },     { 0x3220, 0x3247 },     { 0x3250, 0x4dbf },
+        { 0x4e00, 0xa48c },     { 0xa490, 0xa4c6 },     { 0xa960, 0xa97c },     { 0xac00, 0xd7a3 },
+        { 0xf900, 0xfaff },     { 0xfe10, 0xfe19 },     { 0xfe30, 0xfe52 },     { 0xfe54, 0xfe66 },
+        { 0xfe68, 0xfe6b },     { 0xff01, 0xffe6 },     { 0x1b000, 0x1b001 },   { 0x1f200, 0x1f202 },
+        { 0x1f210, 0x1f23a },   { 0x1f240, 0x1f248 },   { 0x1f250, 0x1f251 },   { 0x20000, 0x3fffd },
+};
+
+#define ARRAYSIZE(A) sizeof(A) / sizeof(*(A))
+
+static int cmp_range(const void *key, const void *cm)
+{
+    const struct utf8range *range = (const struct utf8range *)cm;
+    int ch = *(int *)key;
+    if (ch < range->lower) {
+        return -1;
+    }
+    if (ch >= range->upper) {
+        return 1;
+    }
+    return 0;
+}
+
+static int utf8_in_range(const struct utf8range *range, int num, int ch)
+{
+    const struct utf8range *r =
+        bsearch(&ch, range, num, sizeof(*range), cmp_range);
+
+    if (r) {
+        return 1;
+    }
+    return 0;
+}
+
+int utf8_width(int ch)
+{
+    if (utf8_in_range(unicode_range_combining, ARRAYSIZE(unicode_range_combining), ch)) {
+        return 0;
+    }
+    if (utf8_in_range(unicode_range_wide, ARRAYSIZE(unicode_range_wide), ch)) {
+        return 2;
+    }
+    return 1;
+}
 #endif

diff --git a/utf8.h b/utf8.h
index 5aee96b2a296d4be8bf80a638448a87e95d0d438..d4fb206bf27747e1c8cd17d9ed2ce5193b46f28b 100644 (file)
--- a/utf8.h
+++ b/utf8.h
@@ -18,9 +18,11 @@ extern "C" {
 
 /* No utf-8 support. 1 byte = 1 char */
 #define utf8_strlen(S, B) ((B) < 0 ? (int)strlen(S) : (B))
+#define utf8_strwidth(S, B) utf8_strlen((S), (B))
 #define utf8_tounicode(S, CP) (*(CP) = (unsigned char)*(S), 1)
 #define utf8_index(C, I) (I)
 #define utf8_charlen(C) 1
+ #define utf8_width(C) 1
 
 #else
 /**
@@ -53,6 +55,12 @@ int utf8_charlen(int c);
  */
 int utf8_strlen(const char *str, int bytelen);
 
+/**
+ * Calculates the display width of the first 'charlen' characters in 'str'.
+ * See utf8_width()
+ */
+int utf8_strwidth(const char *str, int charlen);
+
 /**
  * Returns the byte index of the given character in the utf-8 string.
  *
@@ -79,6 +87,12 @@ int utf8_index(const char *str, int charindex);
  */
 int utf8_tounicode(const char *str, int *uc);
 
+/**
+ * Returns the width (in characters) of the given unicode codepoint.
+ * This is 1 for normal letters and 0 for combining characters and 2 for wide characters.
+ */
+int utf8_width(int ch);
+
 #endif
 
 #ifdef __cplusplus