/* CCL (Code Conversion Language) interpreter.
Copyright (C) 1995, 1997 Electrotechnical Laboratory, JAPAN.
+ Copyright (C) 2001 Free Software Foundation, Inc.
Licensed to the Free Software Foundation.
This file is part of GNU Emacs.
multibyte form later. */ \
extra_bytes++; \
} \
- else \
+ else if (CHAR_VALID_P (ch, 0)) \
dst += CHAR_STRING (ch, dst); \
+ else \
+ CCL_INVALID_CMD; \
+ } \
+ else \
+ CCL_SUSPEND (CCL_STAT_SUSPEND_BY_DST); \
+ } while (0)
+
+/* Encode one character CH to multibyte form and write to the current
+ output buffer. The output bytes always forms a valid multibyte
+ sequence. */
+#define CCL_WRITE_MULTIBYTE_CHAR(ch) \
+ do { \
+ int bytes = CHAR_BYTES (ch); \
+ if (!dst) \
+ CCL_INVALID_CMD; \
+ else if (dst + bytes + extra_bytes < (dst_bytes ? dst_end : src)) \
+ { \
+ if (CHAR_VALID_P ((ch), 0)) \
+ dst += CHAR_STRING ((ch), dst); \
+ else \
+ CCL_INVALID_CMD; \
} \
else \
CCL_SUSPEND (CCL_STAT_SUSPEND_BY_DST); \
{
register int *reg = ccl->reg;
register int ic = ccl->ic;
- register int code, field1, field2;
+ register int code = 0, field1, field2;
register Lisp_Object *ccl_prog = ccl->prog;
unsigned char *src = source, *src_end = src + src_bytes;
unsigned char *dst = destination, *dst_end = dst + dst_bytes;
int jump_address;
- int i, j, op;
+ int i = 0, j, op;
int stack_idx = ccl->stack_idx;
/* Instruction counter of the current CCL code. */
- int this_ic;
+ int this_ic = 0;
/* CCL_WRITE_CHAR will produce 8-bit code of range 0x80..0x9F. But,
each of them will be converted to multibyte form of 2-byte
sequence. For that conversion, we remember how many more bytes
if (ic >= ccl->eof_ic)
ic = CCL_HEADER_MAIN;
- if (ccl->buf_magnification ==0) /* We can't produce any bytes. */
+ if (ccl->buf_magnification == 0) /* We can't produce any bytes. */
dst = NULL;
/* Set mapping stack pointer. */
goto ccl_read_multibyte_character_suspend;
}
+ if (!ccl->multibyte)
+ {
+ int bytes;
+ if (!UNIBYTE_STR_AS_MULTIBYTE_P (src, src_end - src, bytes))
+ {
+ reg[RRR] = CHARSET_8_BIT_CONTROL;
+ reg[rrr] = *src++;
+ break;
+ }
+ }
i = *src++;
if (i == '\n' && ccl->eol_type != CODING_EOL_LF)
{
reg[rrr] = i;
reg[RRR] = CHARSET_ASCII;
}
- else if (i <= MAX_CHARSET_OFFICIAL_DIMENSION1)
- {
- if (src >= src_end)
- goto ccl_read_multibyte_character_suspend;
- reg[RRR] = i;
- reg[rrr] = (*src++ & 0x7F);
- }
else if (i <= MAX_CHARSET_OFFICIAL_DIMENSION2)
{
- if ((src + 1) >= src_end)
+ int dimension = BYTES_BY_CHAR_HEAD (i) - 1;
+
+ if (dimension == 0)
+ {
+ /* `i' is a leading code for an undefined charset. */
+ reg[RRR] = CHARSET_8_BIT_GRAPHIC;
+ reg[rrr] = i;
+ }
+ else if (src + dimension > src_end)
goto ccl_read_multibyte_character_suspend;
- reg[RRR] = i;
- i = (*src++ & 0x7F);
- reg[rrr] = ((i << 7) | (*src & 0x7F));
- src++;
+ else
+ {
+ reg[RRR] = i;
+ i = (*src++ & 0x7F);
+ if (dimension == 1)
+ reg[rrr] = i;
+ else
+ reg[rrr] = ((i << 7) | (*src++ & 0x7F));
+ }
}
else if ((i == LEADING_CODE_PRIVATE_11)
|| (i == LEADING_CODE_PRIVATE_12))
break;
ccl_read_multibyte_character_suspend:
+ if (src <= src_end && !ccl->multibyte && ccl->last_block)
+ {
+ reg[RRR] = CHARSET_8_BIT_CONTROL;
+ reg[rrr] = i;
+ break;
+ }
src--;
if (ccl->last_block)
{
else
i = ((i - 0xE0) << 14) | reg[rrr];
- CCL_WRITE_CHAR (i);
+ CCL_WRITE_MULTIBYTE_CHAR (i);
break;
bcopy (msg, dst, msglen);
dst += msglen;
}
+
+ if (ccl->status == CCL_STAT_INVALID_CMD)
+ {
+#if 0 /* If the remaining bytes contain 0x80..0x9F, copying them
+ results in an invalid multibyte sequence. */
+
+ /* Copy the remaining source data. */
+ int i = src_end - src;
+ if (dst_bytes && (dst_end - dst) < i)
+ i = dst_end - dst;
+ bcopy (src, dst, i);
+ src += i;
+ dst += i;
+#else
+ /* Signal that we've consumed everything. */
+ src = src_end;
+#endif
+ }
}
ccl_finish:
ccl->ic = ic;
ccl->stack_idx = stack_idx;
ccl->prog = ccl_prog;
- if (consumed) *consumed = src - source;
+ ccl->eight_bit_control = (extra_bytes > 0);
+ if (consumed)
+ *consumed = src - source;
return (dst ? dst - destination : 0);
}
"Execute CCL-PROGRAM with registers initialized by REGISTERS.\n\
\n\
CCL-PROGRAM is a CCL program name (symbol)\n\
-or a compiled code generated by `ccl-compile' (for backward compatibility,\n\
-in this case, the overhead of the execution is bigger than the former case).\n\
+or compiled code generated by `ccl-compile' (for backward compatibility.\n\
+In the latter case, the execution overhead is bigger than in the former).\n\
No I/O commands should appear in CCL-PROGRAM.\n\
\n\
REGISTERS is a vector of [R0 R1 ... R7] where RN is an initial value\n\
- of Nth register.\n\
+for the Nth register.\n\
\n\
As side effect, each element of REGISTERS holds the value of\n\
- corresponding register after the execution.\n\
+the corresponding register after the execution.\n\
\n\
-See the documentation of `define-ccl-program' for the detail of CCL program.")
+See the documentation of `define-ccl-program' for a definition of CCL\n\
+programs.")
(ccl_prog, reg)
Lisp_Object ccl_prog, reg;
{
/* Register code conversion map.
A code conversion map consists of numbers, Qt, Qnil, and Qlambda.
- The first element is start code point.
- The rest elements are mapped numbers.
+ The first element is the start code point.
+ The other elements are mapped numbers.
Symbol t means to map to an original number before mapping.
Symbol nil means that the corresponding element is empty.
- Symbol lambda menas to terminate mapping here.
+ Symbol lambda means to terminate mapping here.
*/
DEFUN ("register-code-conversion-map", Fregister_code_conversion_map,
if (EQ (symbol, XCAR (slot)))
{
index = make_number (i);
- XCDR (slot) = map;
+ XSETCDR (slot, map);
Fput (symbol, Qcode_conversion_map, map);
Fput (symbol, Qcode_conversion_map_id, index);
return index;
HCoop / bpt / emacs.git / blobdiff