Merging the UTF8 patch, consisting of O'ksi'd s original 1.1.6 patch and additions by Ian. PLEASE BE AWARE that the patch in its current incarnation is a regression in many aspects and further work is required before we can announce Unicode support.

git-svn-id: file:///fltk/svn/fltk/branches/branch-1.3@6212 ea41ed52-d2ee-0310-a9c1-e6b18d33e121

1 files changed, 110 insertions, 0 deletions

diff --git a/src/xutf8/lcUniConv/utf8.h b/src/xutf8/lcUniConv/utf8.h
new file mode 100644
index 000000000..fe189cc2b
--- /dev/null
+++ b/src/xutf8/lcUniConv/utf8.h
@@ -0,0 +1,110 @@
+/* $XFree86: xc/lib/X11/lcUniConv/utf8.h,v 1.3 2000/11/28 18:50:07 dawes Exp $ */
+
+/*
+ * UTF-8
+ */
+
+/* Specification: RFC 2279 */
+
+static int
+utf8_mbtowc (conv_t conv, ucs4_t *pwc, const unsigned char *s, int n)
+{
+  unsigned char c = s[0];
+
+  if (c < 0x80) {
+    *pwc = c;
+    return 1;
+  } else if (c < 0xc2) {
+    return RET_ILSEQ;
+  } else if (c < 0xe0) {
+    if (n < 2)
+      return RET_TOOFEW(0);
+    if (!((s[1] ^ 0x80) < 0x40))
+      return RET_ILSEQ;
+    *pwc = ((ucs4_t) (c & 0x1f) << 6)
+           | (ucs4_t) (s[1] ^ 0x80);
+    return 2;
+  } else if (c < 0xf0) {
+    if (n < 3)
+      return RET_TOOFEW(0);
+    if (!((s[1] ^ 0x80) < 0x40 && (s[2] ^ 0x80) < 0x40
+          && (c >= 0xe1 || s[1] >= 0xa0)))
+      return RET_ILSEQ;
+    *pwc = ((ucs4_t) (c & 0x0f) << 12)
+           | ((ucs4_t) (s[1] ^ 0x80) << 6)
+           | (ucs4_t) (s[2] ^ 0x80);
+    return 3;
+  } else if (c < 0xf8) {
+    if (n < 4)
+      return RET_TOOFEW(0);
+    if (!((s[1] ^ 0x80) < 0x40 && (s[2] ^ 0x80) < 0x40
+          && (s[3] ^ 0x80) < 0x40
+          && (c >= 0xf1 || s[1] >= 0x90)))
+      return RET_ILSEQ;
+    *pwc = ((ucs4_t) (c & 0x07) << 18)
+           | ((ucs4_t) (s[1] ^ 0x80) << 12)
+           | ((ucs4_t) (s[2] ^ 0x80) << 6)
+           | (ucs4_t) (s[3] ^ 0x80);
+    return 4;
+  } else if (c < 0xfc) {
+    if (n < 5)
+      return RET_TOOFEW(0);
+    if (!((s[1] ^ 0x80) < 0x40 && (s[2] ^ 0x80) < 0x40
+          && (s[3] ^ 0x80) < 0x40 && (s[4] ^ 0x80) < 0x40
+          && (c >= 0xf9 || s[1] >= 0x88)))
+      return RET_ILSEQ;
+    *pwc = ((ucs4_t) (c & 0x03) << 24)
+           | ((ucs4_t) (s[1] ^ 0x80) << 18)
+           | ((ucs4_t) (s[2] ^ 0x80) << 12)
+           | ((ucs4_t) (s[3] ^ 0x80) << 6)
+           | (ucs4_t) (s[4] ^ 0x80);
+    return 5;
+  } else if (c < 0xfe) {
+    if (n < 6)
+      return RET_TOOFEW(0);
+    if (!((s[1] ^ 0x80) < 0x40 && (s[2] ^ 0x80) < 0x40
+          && (s[3] ^ 0x80) < 0x40 && (s[4] ^ 0x80) < 0x40
+          && (s[5] ^ 0x80) < 0x40
+          && (c >= 0xfd || s[1] >= 0x84)))
+      return RET_ILSEQ;
+    *pwc = ((ucs4_t) (c & 0x01) << 30)
+           | ((ucs4_t) (s[1] ^ 0x80) << 24)
+           | ((ucs4_t) (s[2] ^ 0x80) << 18)
+           | ((ucs4_t) (s[3] ^ 0x80) << 12)
+           | ((ucs4_t) (s[4] ^ 0x80) << 6)
+           | (ucs4_t) (s[5] ^ 0x80);
+    return 6;
+  } else
+    return RET_ILSEQ;
+}
+
+static int
+utf8_wctomb (conv_t conv, unsigned char *r, ucs4_t wc, int n) /* n == 0 is acceptable */
+{
+  int count;
+  if (wc < 0x80)
+    count = 1;
+  else if (wc < 0x800)
+    count = 2;
+  else if (wc < 0x10000)
+    count = 3;
+  else if (wc < 0x200000)
+    count = 4;
+  else if (wc < 0x4000000)
+    count = 5;
+  else if (wc <= 0x7fffffff)
+    count = 6;
+  else
+    return RET_ILSEQ;
+  if (n < count)
+    return RET_TOOSMALL;
+  switch (count) { /* note: code falls through cases! */
+    case 6: r[5] = 0x80 | (wc & 0x3f); wc = wc >> 6; wc |= 0x4000000;
+    case 5: r[4] = 0x80 | (wc & 0x3f); wc = wc >> 6; wc |= 0x200000;
+    case 4: r[3] = 0x80 | (wc & 0x3f); wc = wc >> 6; wc |= 0x10000;
+    case 3: r[2] = 0x80 | (wc & 0x3f); wc = wc >> 6; wc |= 0x800;
+    case 2: r[1] = 0x80 | (wc & 0x3f); wc = wc >> 6; wc |= 0xc0;
+    case 1: r[0] = wc;
+  }
+  return count;
+}