summaryrefslogtreecommitdiff
path: root/encoding.c
diff options
context:
space:
mode:
Diffstat (limited to 'encoding.c')
-rw-r--r--encoding.c132
1 files changed, 117 insertions, 15 deletions
diff --git a/encoding.c b/encoding.c
index 6c1567a..86c2efd 100644
--- a/encoding.c
+++ b/encoding.c
@@ -514,6 +514,8 @@ int from, to;
if (rmc.font == 0) /* latin1 is the same in unicode */
return mc;
c = rmc.image | (rmc.font << 8);
+ if (from == UTF8)
+ c |= rmc.fontx << 16;
#ifdef DW_CHARS
if (rmc.mbcs)
{
@@ -526,6 +528,8 @@ int from, to;
c = recode_char_to_encoding(c, to);
rmc.image = c & 255;
rmc.font = c >> 8 & 255;
+ if (to == UTF8)
+ rmc.fontx = c >> 16 & 255;
return &rmc;
}
@@ -542,7 +546,7 @@ int from, to;
if (from == to || (from != UTF8 && to != UTF8) || w == 0)
return ml;
- if (ml->font == null && encodings[from].deffont == 0)
+ if (ml->font == null && ml->fontx == null && encodings[from].deffont == 0)
return ml;
if (w > maxlen)
{
@@ -556,7 +560,11 @@ int from, to;
rml[i].font = malloc(w);
else
rml[i].font = realloc(rml[i].font, w);
- if (rml[i].image == 0 || rml[i].font == 0)
+ if (rml[i].fontx == 0)
+ rml[i].fontx = malloc(w);
+ else
+ rml[i].fontx = realloc(rml[i].fontx, w);
+ if (rml[i].image == 0 || rml[i].font == 0 || rml[i].fontx == 0)
{
maxlen = 0;
return ml; /* sorry */
@@ -578,6 +586,12 @@ int from, to;
for (i = 0; i < w; i++)
debug1("%c", "0123456789abcdef"[(ml->font[i] ) & 15]);
debug("\n");
+ for (i = 0; i < w; i++)
+ debug1("%c", "0123456789abcdef"[(ml->fontx[i] >> 4) & 15]);
+ debug("\n");
+ for (i = 0; i < w; i++)
+ debug1("%c", "0123456789abcdef"[(ml->fontx[i] ) & 15]);
+ debug("\n");
rl = rml + last;
rl->attr = ml->attr;
@@ -590,6 +604,8 @@ int from, to;
for (i = 0; i < w; i++)
{
c = ml->image[i] | (ml->font[i] << 8);
+ if (from == UTF8)
+ c |= ml->fontx[i] << 16;
if (from != UTF8 && c < 256)
c |= encodings[from].deffont << 8;
#ifdef DW_CHARS
@@ -603,6 +619,8 @@ int from, to;
i++;
c2 = ml->image[i] | (ml->font[i] << 8);
c = recode_char_dw_to_encoding(c, &c2, to);
+ if (to == UTF8)
+ rl->fontx[i - 1] = c >> 16 & 255;
rl->font[i - 1] = c >> 8 & 255;
rl->image[i - 1] = c & 255;
c = c2;
@@ -613,6 +631,8 @@ int from, to;
c = recode_char_to_encoding(c, to);
rl->image[i] = c & 255;
rl->font[i] = c >> 8 & 255;
+ if (to == UTF8)
+ rl->fontx[i] = c >> 16 & 255;
}
last ^= 1;
debug("recode_mline: to\n");
@@ -628,14 +648,20 @@ int from, to;
for (i = 0; i < w; i++)
debug1("%c", "0123456789abcdef"[(rl->font[i] ) & 15]);
debug("\n");
+ for (i = 0; i < w; i++)
+ debug1("%c", "0123456789abcdef"[(rl->fontx[i] >> 4) & 15]);
+ debug("\n");
+ for (i = 0; i < w; i++)
+ debug1("%c", "0123456789abcdef"[(rl->fontx[i] ) & 15]);
+ debug("\n");
return rl;
}
struct combchar {
- unsigned short c1;
- unsigned short c2;
- unsigned short next;
- unsigned short prev;
+ unsigned int c1;
+ unsigned int c2;
+ unsigned int next;
+ unsigned int prev;
};
struct combchar **combchars;
@@ -649,10 +675,20 @@ int c;
AddUtf8(combchars[c - 0xd800]->c1);
c = combchars[c - 0xd800]->c2;
}
+ if (c >= 0x10000)
+ {
+ if (c >= 0x200000)
+ {
+ AddChar((c & 0x3000000) >> 12 ^ 0xf8);
+ c = (c & 0xffffff) ^ ((0xf0 ^ 0x80) << 18);
+ }
+ AddChar((c & 0x1fc0000) >> 18 ^ 0xf0);
+ c = (c & 0x3ffff) ^ ((0xe0 ^ 0x80) << 12);
+ }
if (c >= 0x800)
{
- AddChar((c & 0xf000) >> 12 | 0xe0);
- c = (c & 0x0fff) | 0x1000;
+ AddChar((c & 0x7f000) >> 12 ^ 0xe0);
+ c = (c & 0x0fff) ^ ((0xc0 ^ 0x80) << 6);
}
if (c >= 0x80)
{
@@ -683,10 +719,24 @@ char *p;
int c;
{
int l = 1;
+ if (c >= 0x10000)
+ {
+ if (c >= 0x200000)
+ {
+ if (p)
+ *p++ = (c & 0x3000000) >> 12 ^ 0xf8;
+ l++;
+ c = (c & 0xffffff) ^ ((0xf0 ^ 0x80) << 18);
+ }
+ if (p)
+ *p++ = (c & 0x1fc0000) >> 18 ^ 0xf0;
+ l++;
+ c = (c & 0x3ffff) ^ ((0xe0 ^ 0x80) << 12);
+ }
if (c >= 0x800)
{
if (p)
- *p++ = (c & 0xf000) >> 12 | 0xe0;
+ *p++ = (c & 0x7f000) >> 12 ^ 0xe0;
l++;
c = (c & 0x0fff) | 0x1000;
}
@@ -758,8 +808,13 @@ int c, *utf8charp;
*utf8charp = utf8char = (c & 0x80000000) ? c : 0;
if (utf8char)
return -1;
+#if 0
if (c & 0xffff0000)
c = UCS_REPL; /* sorry, only know 16bit Unicode */
+#else
+ if (c & 0xff800000)
+ c = UCS_REPL; /* sorry, only know 23bit Unicode */
+#endif
if (c >= 0xd800 && (c <= 0xdfff || c == 0xfffe || c == 0xffff))
c = UCS_REPL; /* illegal code */
return c;
@@ -803,11 +858,13 @@ int encoding;
#else
ml = &p->w_mlines[j];
#endif
- if (ml->font == null && encodings[p->w_encoding].deffont == 0)
+ if (ml->font == null && ml->fontx == 0 && encodings[p->w_encoding].deffont == 0)
continue;
for (i = 0; i < p->w_width; i++)
{
c = ml->image[i] | (ml->font[i] << 8);
+ if (p->w_encoding == UTF8)
+ c |= ml->fontx[i] << 16;
if (p->w_encoding != UTF8 && c < 256)
c |= encodings[p->w_encoding].deffont << 8;
if (c < 256)
@@ -829,8 +886,22 @@ int encoding;
{
int c2;
i++;
- c2 = ml->image[i] | (ml->font[i] << 8);
+ c2 = ml->image[i] | (ml->font[i] << 8) | (ml->fontx[i] << 16);
c = recode_char_dw_to_encoding(c, &c2, encoding);
+ if (encoding == UTF8)
+ {
+ if (c > 0x10000 && ml->fontx == null)
+ {
+ if ((ml->fontx = (unsigned char *)calloc(p->w_width + 1, 1)) == 0)
+ {
+ ml->fontx = null;
+ break;
+ }
+ }
+ ml->fontx[i - 1] = c >> 16 & 255;
+ }
+ else
+ ml->fontx = null;
ml->font[i - 1] = c >> 8 & 255;
ml->image[i - 1] = c & 255;
c = c2;
@@ -841,6 +912,20 @@ int encoding;
c = recode_char_to_encoding(c, encoding);
ml->image[i] = c & 255;
ml->font[i] = c >> 8 & 255;
+ if (encoding == UTF8)
+ {
+ if (c > 0x10000 && ml->fontx == null)
+ {
+ if ((ml->fontx = (unsigned char *)calloc(p->w_width + 1, 1)) == 0)
+ {
+ ml->fontx = null;
+ break;
+ }
+ }
+ ml->fontx[i] = c >> 16 & 255;
+ }
+ else
+ ml->fontx = null;
}
}
p->w_encoding = encoding;
@@ -1039,7 +1124,7 @@ struct mchar *mc;
int root, i, c1;
int isdouble;
- c1 = mc->image | (mc->font << 8);
+ c1 = mc->image | (mc->font << 8) | mc->fontx << 16;
isdouble = c1 >= 0x1100 && utf8_isdouble(c1);
if (!combchars)
{
@@ -1102,6 +1187,7 @@ struct mchar *mc;
combchars[i]->c2 = c;
mc->image = i & 0xff;
mc->font = (i >> 8) + 0xd8;
+ mc->fontx = 0;
debug3("combinig char %x %x -> %x\n", c1, c, i + 0xd800);
comb_tofront(root, i);
}
@@ -1220,6 +1306,15 @@ struct win *p;
p->w_c1 = 0;
}
+/* decoded char: 32-bit <fontx><font><c2><c>
+ * fontx: non-bmp utf8
+ * c2: multi-byte character
+ * font is always zero for utf8
+ * returns: -1 need more bytes
+ * -2 decode error
+ */
+
+
int
DecodeChar(c, encoding, statep)
int c;
@@ -1231,7 +1326,12 @@ int *statep;
debug2("Decoding char %02x for encoding %d\n", c, encoding);
#ifdef UTF8
if (encoding == UTF8)
- return FromUtf8(c, statep);
+ {
+ c = FromUtf8(c, statep);
+ if (c >= 0x10000)
+ c = (c & 0x7f0000) << 8 | (c & 0xffff);
+ return c;
+ }
#endif
if (encoding == SJIS)
{
@@ -1345,7 +1445,7 @@ int *fontp;
}
return 3;
}
- f = c >> 16;
+ f = (c >> 16) & 0xff;
#ifdef UTF8
if (encoding == UTF8)
@@ -1368,8 +1468,10 @@ int *fontp;
}
return ToUtf8(bp, c);
}
- if ((c & 0xff00) && f == 0) /* is_utf8? */
+ if (f == 0 && (c & 0x7f00ff00) != 0) /* is_utf8? */
{
+ if (c >= 0x10000)
+ c = (c & 0x7f0000) >> 8 | (c & 0xffff);
# ifdef DW_CHARS
if (utf8_isdouble(c))
{
generated by cgit v1.2.3 (git 2.39.1) at 2024-06-02 20:53:30 +0000