X-Git-Url: https://code.delx.au/gnu-emacs/blobdiff_plain/3943ed767c4230c9cf7a489124ede96168a4c33b..7087d5e9af41c8835c3a5090bd8e2c6893685466:/src/charset.c
diff --git a/src/charset.c b/src/charset.c
index 9c9395fdc6..052f318683 100644
--- a/src/charset.c
+++ b/src/charset.c
@@ -1,8 +1,8 @@
/* Basic character set support.
- Copyright (C) 2001, 2002, 2003, 2004, 2005,
- 2006, 2007, 2008 Free Software Foundation, Inc.
+ Copyright (C) 2001, 2002, 2003, 2004, 2005, 2006, 2007,
+ 2008, 2009 Free Software Foundation, Inc.
Copyright (C) 1995, 1996, 1997, 1998, 1999, 2000, 2001, 2002, 2003, 2004,
- 2005, 2006, 2007, 2008
+ 2005, 2006, 2007, 2008, 2009
National Institute of Advanced Industrial Science and Technology (AIST)
Registration Number H14PRO021
@@ -12,10 +12,10 @@
This file is part of GNU Emacs.
-GNU Emacs is free software; you can redistribute it and/or modify
+GNU Emacs is free software: you can redistribute it and/or modify
it under the terms of the GNU General Public License as published by
-the Free Software Foundation; either version 3, or (at your option)
-any later version.
+the Free Software Foundation, either version 3 of the License, or
+(at your option) any later version.
GNU Emacs is distributed in the hope that it will be useful,
but WITHOUT ANY WARRANTY; without even the implied warranty of
@@ -23,9 +23,7 @@ MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
GNU General Public License for more details.
You should have received a copy of the GNU General Public License
-along with GNU Emacs; see the file COPYING. If not, write to
-the Free Software Foundation, Inc., 51 Franklin Street, Fifth Floor,
-Boston, MA 02110-1301, USA. */
+along with GNU Emacs. If not, see . */
#include
@@ -76,12 +74,14 @@ Lisp_Object Qascii;
Lisp_Object Qeight_bit;
Lisp_Object Qiso_8859_1;
Lisp_Object Qunicode;
+Lisp_Object Qemacs;
/* The corresponding charsets. */
int charset_ascii;
int charset_eight_bit;
int charset_iso_8859_1;
int charset_unicode;
+int charset_emacs;
/* The other special charsets. */
int charset_jisx0201_roman;
@@ -97,6 +97,10 @@ int charset_unibyte;
/* List of charsets ordered by the priority. */
Lisp_Object Vcharset_ordered_list;
+/* Sub-list of Vcharset_ordered_list that contains all non-preferred
+ charsets. */
+Lisp_Object Vcharset_non_preferred_head;
+
/* Incremented everytime we change Vcharset_ordered_list. This is
unsigned short so that it fits in Lisp_Int and never matches
-1. */
@@ -116,7 +120,10 @@ int iso_charset_table[ISO_MAX_DIMENSION][ISO_MAX_CHARS][ISO_MAX_FINAL];
Lisp_Object Vcharset_map_path;
-Lisp_Object Vchar_unified_charset_table;
+/* If nonzero, don't load charset maps. */
+int inhibit_load_charset_map;
+
+Lisp_Object Vcurrent_iso639_language;
/* Defined in chartab.c */
extern void
@@ -160,7 +167,63 @@ map_char_table_for_charset P_ ((void (*c_function) (Lisp_Object, Lisp_Object),
| (((charset)->code_space[12] + ((idx) / (charset)->code_space[11])) \
<< 24))))
+/* Structure to hold mapping tables for a charset. Used by temacs
+ invoked for dumping. */
+static struct
+{
+ /* The current charset for which the following tables are setup. */
+ struct charset *current;
+
+ /* 1 iff the following table is used for encoder. */
+ short for_encoder;
+
+ /* When the following table is used for encoding, mininum and
+ maxinum character of the current charset. */
+ int min_char, max_char;
+
+ /* A Unicode character correspoinding to the code indice 0 (i.e. the
+ minimum code-point) of the current charset, or -1 if the code
+ indice 0 is not a Unicode character. This is checked when
+ table.encoder[CHAR] is zero. */
+ int zero_index_char;
+
+ union {
+ /* Table mapping code-indices (not code-points) of the current
+ charset to Unicode characters. If decoder[CHAR] is -1, CHAR
+ doesn't belong to the current charset. */
+ int decoder[0x10000];
+ /* Table mapping Unicode characters to code-indices of the current
+ charset. The first 0x10000 elements are for BMP (0..0xFFFF),
+ and the last 0x10000 are for SMP (0x10000..0x1FFFF) or SIP
+ (0x20000..0x2FFFF). Note that there is no charset map that
+ uses both SMP and SIP. */
+ unsigned short encoder[0x20000];
+ } table;
+} *temp_charset_work;
+
+#define SET_TEMP_CHARSET_WORK_ENCODER(C, CODE) \
+ do { \
+ if ((CODE) == 0) \
+ temp_charset_work->zero_index_char = (C); \
+ else if ((C) < 0x20000) \
+ temp_charset_work->table.encoder[(C)] = (CODE); \
+ else \
+ temp_charset_work->table.encoder[(C) - 0x10000] = (CODE); \
+ } while (0)
+
+#define GET_TEMP_CHARSET_WORK_ENCODER(C) \
+ ((C) == temp_charset_work->zero_index_char ? 0 \
+ : (C) < 0x20000 ? (temp_charset_work->table.encoder[(C)] \
+ ? (int) temp_charset_work->table.encoder[(C)] : -1) \
+ : temp_charset_work->table.encoder[(C) - 0x10000] \
+ ? temp_charset_work->table.encoder[(C) - 0x10000] : -1)
+
+#define SET_TEMP_CHARSET_WORK_DECODER(C, CODE) \
+ (temp_charset_work->table.decoder[(CODE)] = (C))
+
+#define GET_TEMP_CHARSET_WORK_DECODER(CODE) \
+ (temp_charset_work->table.decoder[(CODE)])
/* Set to 1 to warn that a charset map is loaded and thus a buffer
@@ -176,16 +239,30 @@ struct charset_map_entries
struct charset_map_entries *next;
};
-/* Load the mapping information for CHARSET from ENTRIES.
+/* Load the mapping information of CHARSET from ENTRIES for
+ initializing (CONTROL_FLAG == 0), decoding (CONTROL_FLAG == 1), and
+ encoding (CONTROL_FLAG == 2).
+
+ If CONTROL_FLAG is 0, setup CHARSET->min_char, CHARSET->max_char,
+ and CHARSET->fast_map.
- If CONTROL_FLAG is 0, setup CHARSET->min_char and CHARSET->max_char.
+ If CONTROL_FLAG is 1, setup the following tables according to
+ CHARSET->method and inhibit_load_charset_map.
- If CONTROL_FLAG is 1, setup CHARSET->min_char, CHARSET->max_char,
- CHARSET->decoder, and CHARSET->encoder.
+ CHARSET->method | inhibit_lcm == 0 | inhibit_lcm == 1
+ ----------------------+--------------------+---------------------------
+ CHARSET_METHOD_MAP | CHARSET->decoder | temp_charset_work->decoder
+ ----------------------+--------------------+---------------------------
+ CHARSET_METHOD_OFFSET | Vchar_unify_table | temp_charset_work->decoder
- If CONTROL_FLAG is 2, setup CHARSET->deunifier and
- Vchar_unify_table. If Vchar_unified_charset_table is non-nil,
- setup it too. */
+ If CONTROL_FLAG is 2, setup the following tables.
+
+ CHARSET->method | inhibit_lcm == 0 | inhibit_lcm == 1
+ ----------------------+--------------------+---------------------------
+ CHARSET_METHOD_MAP | CHARSET->encoder | temp_charset_work->encoder
+ ----------------------+--------------------+--------------------------
+ CHARSET_METHOD_OFFSET | CHARSET->deunifier | temp_charset_work->encoder
+*/
static void
load_charset_map (charset, entries, n_entries, control_flag)
@@ -204,16 +281,55 @@ load_charset_map (charset, entries, n_entries, control_flag)
if (n_entries <= 0)
return;
- if (control_flag > 0)
+ if (control_flag)
{
- int n = CODE_POINT_TO_INDEX (charset, max_code) + 1;
-
- table = Fmake_char_table (Qnil, Qnil);
- if (control_flag == 1)
- vec = Fmake_vector (make_number (n), make_number (-1));
- else if (! CHAR_TABLE_P (Vchar_unify_table))
- Vchar_unify_table = Fmake_char_table (Qnil, Qnil);
+ if (! inhibit_load_charset_map)
+ {
+ if (control_flag == 1)
+ {
+ if (charset->method == CHARSET_METHOD_MAP)
+ {
+ int n = CODE_POINT_TO_INDEX (charset, max_code) + 1;
+ vec = CHARSET_DECODER (charset)
+ = Fmake_vector (make_number (n), make_number (-1));
+ }
+ else
+ {
+ char_table_set_range (Vchar_unify_table,
+ charset->min_char, charset->max_char,
+ Qnil);
+ }
+ }
+ else
+ {
+ table = Fmake_char_table (Qnil, Qnil);
+ if (charset->method == CHARSET_METHOD_MAP)
+ CHARSET_ENCODER (charset) = table;
+ else
+ CHARSET_DEUNIFIER (charset) = table;
+ }
+ }
+ else
+ {
+ if (! temp_charset_work)
+ temp_charset_work = malloc (sizeof (*temp_charset_work));
+ if (control_flag == 1)
+ {
+ memset (temp_charset_work->table.decoder, -1,
+ sizeof (int) * 0x10000);
+ temp_charset_work->for_encoder = 0;
+ }
+ else
+ {
+ memset (temp_charset_work->table.encoder, 0,
+ sizeof (unsigned short) * 0x20000);
+ temp_charset_work->zero_index_char = -1;
+ }
+ temp_charset_work->current = charset;
+ temp_charset_work->for_encoder = (control_flag == 2);
+ control_flag += 2;
+ }
charset_map_loaded = 1;
}
@@ -245,14 +361,48 @@ load_charset_map (charset, entries, n_entries, control_flag)
if (from_index < 0 || to_index < 0)
continue;
- if (control_flag < 2)
- {
- int c;
+ if (to_c > max_char)
+ max_char = to_c;
+ else if (from_c < min_char)
+ min_char = from_c;
- if (to_c > max_char)
- max_char = to_c;
- else if (from_c < min_char)
- min_char = from_c;
+ if (control_flag == 1)
+ {
+ if (charset->method == CHARSET_METHOD_MAP)
+ for (; from_index <= to_index; from_index++, from_c++)
+ ASET (vec, from_index, make_number (from_c));
+ else
+ for (; from_index <= to_index; from_index++, from_c++)
+ CHAR_TABLE_SET (Vchar_unify_table,
+ CHARSET_CODE_OFFSET (charset) + from_index,
+ make_number (from_c));
+ }
+ else if (control_flag == 2)
+ {
+ if (charset->method == CHARSET_METHOD_MAP
+ && CHARSET_COMPACT_CODES_P (charset))
+ for (; from_index <= to_index; from_index++, from_c++)
+ {
+ unsigned code = INDEX_TO_CODE_POINT (charset, from_index);
+
+ if (NILP (CHAR_TABLE_REF (table, from_c)))
+ CHAR_TABLE_SET (table, from_c, make_number (code));
+ }
+ else
+ for (; from_index <= to_index; from_index++, from_c++)
+ {
+ if (NILP (CHAR_TABLE_REF (table, from_c)))
+ CHAR_TABLE_SET (table, from_c, make_number (from_index));
+ }
+ }
+ else if (control_flag == 3)
+ for (; from_index <= to_index; from_index++, from_c++)
+ SET_TEMP_CHARSET_WORK_DECODER (from_c, from_index);
+ else if (control_flag == 4)
+ for (; from_index <= to_index; from_index++, from_c++)
+ SET_TEMP_CHARSET_WORK_ENCODER (from_c, from_index);
+ else /* control_flag == 0 */
+ {
if (ascii_compatible_p)
{
if (! ASCII_BYTE_P (from_c))
@@ -266,70 +416,22 @@ load_charset_map (charset, entries, n_entries, control_flag)
}
}
- for (c = from_c; c <= to_c; c++)
- CHARSET_FAST_MAP_SET (c, fast_map);
-
- if (control_flag == 1)
- {
- unsigned code = from;
-
- if (CHARSET_COMPACT_CODES_P (charset))
- while (1)
- {
- ASET (vec, from_index, make_number (from_c));
- if (NILP (CHAR_TABLE_REF (table, from_c)))
- CHAR_TABLE_SET (table, from_c, make_number (code));
- if (from_index == to_index)
- break;
- from_index++, from_c++;
- code = INDEX_TO_CODE_POINT (charset, from_index);
- }
- else
- for (; from_index <= to_index; from_index++, from_c++)
- {
- ASET (vec, from_index, make_number (from_c));
- if (NILP (CHAR_TABLE_REF (table, from_c)))
- CHAR_TABLE_SET (table, from_c, make_number (from_index));
- }
- }
- }
- else
- {
- unsigned code = from;
-
- while (1)
- {
- int c1 = DECODE_CHAR (charset, code);
-
- if (c1 >= 0)
- {
- CHAR_TABLE_SET (table, from_c, make_number (c1));
- CHAR_TABLE_SET (Vchar_unify_table, c1, make_number (from_c));
- if (CHAR_TABLE_P (Vchar_unified_charset_table))
- CHAR_TABLE_SET (Vchar_unified_charset_table, c1,
- CHARSET_NAME (charset));
- }
- if (from_index == to_index)
- break;
- from_index++, from_c++;
- code = INDEX_TO_CODE_POINT (charset, from_index);
- }
+ for (; from_c <= to_c; from_c++)
+ CHARSET_FAST_MAP_SET (from_c, fast_map);
}
}
- if (control_flag < 2)
+ if (control_flag == 0)
{
CHARSET_MIN_CHAR (charset) = (ascii_compatible_p
? nonascii_min_char : min_char);
CHARSET_MAX_CHAR (charset) = max_char;
- if (control_flag == 1)
- {
- CHARSET_DECODER (charset) = vec;
- CHARSET_ENCODER (charset) = table;
- }
}
- else
- CHARSET_DEUNIFIER (charset) = table;
+ else if (control_flag == 4)
+ {
+ temp_charset_work->min_char = min_char;
+ temp_charset_work->max_char = max_char;
+ }
}
@@ -525,21 +627,31 @@ load_charset_map_from_vector (charset, vec, control_flag)
load_charset_map (charset, head, n_entries, control_flag);
}
+
+/* Load a mapping table for CHARSET. CONTROL-FLAG tells what kind of
+ map it is (see the comment of load_charset_map for the detail). */
+
static void
-load_charset (charset)
+load_charset (charset, control_flag)
struct charset *charset;
+ int control_flag;
{
- if (CHARSET_METHOD (charset) == CHARSET_METHOD_MAP_DEFERRED)
- {
- Lisp_Object map;
+ Lisp_Object map;
- map = CHARSET_MAP (charset);
- if (STRINGP (map))
- load_charset_map_from_file (charset, map, 1);
- else
- load_charset_map_from_vector (charset, map, 1);
- CHARSET_METHOD (charset) = CHARSET_METHOD_MAP;
- }
+ if (inhibit_load_charset_map
+ && temp_charset_work
+ && charset == temp_charset_work->current
+ && (control_flag == 2 == temp_charset_work->for_encoder))
+ return;
+
+ if (CHARSET_METHOD (charset) == CHARSET_METHOD_MAP)
+ map = CHARSET_MAP (charset);
+ else if (CHARSET_UNIFIED_P (charset))
+ map = CHARSET_UNIFY_MAP (charset);
+ if (STRINGP (map))
+ load_charset_map_from_file (charset, map, control_flag);
+ else
+ load_charset_map_from_vector (charset, map, control_flag);
}
@@ -552,6 +664,68 @@ DEFUN ("charsetp", Fcharsetp, Scharsetp, 1, 1, 0,
}
+void map_charset_for_dump P_ ((void (*c_function) (Lisp_Object, Lisp_Object),
+ Lisp_Object function, Lisp_Object arg,
+ unsigned from, unsigned to));
+
+void
+map_charset_for_dump (c_function, function, arg, from, to)
+ void (*c_function) (Lisp_Object, Lisp_Object);
+ Lisp_Object function, arg;
+ unsigned from, to;
+{
+ int from_idx = CODE_POINT_TO_INDEX (temp_charset_work->current, from);
+ int to_idx = CODE_POINT_TO_INDEX (temp_charset_work->current, to);
+ Lisp_Object range;
+ int c, stop;
+ struct gcpro gcpro1;
+
+ range = Fcons (Qnil, Qnil);
+ GCPRO1 (range);
+
+ c = temp_charset_work->min_char;
+ stop = (temp_charset_work->max_char < 0x20000
+ ? temp_charset_work->max_char : 0xFFFF);
+
+ while (1)
+ {
+ int index = GET_TEMP_CHARSET_WORK_ENCODER (c);
+
+ if (index >= from_idx && index <= to_idx)
+ {
+ if (NILP (XCAR (range)))
+ XSETCAR (range, make_number (c));
+ }
+ else if (! NILP (XCAR (range)))
+ {
+ XSETCDR (range, make_number (c - 1));
+ if (c_function)
+ (*c_function) (arg, range);
+ else
+ call2 (function, range, arg);
+ XSETCAR (range, Qnil);
+ }
+ if (c == stop)
+ {
+ if (c == temp_charset_work->max_char)
+ {
+ if (! NILP (XCAR (range)))
+ {
+ XSETCDR (range, make_number (c));
+ if (c_function)
+ (*c_function) (arg, range);
+ else
+ call2 (function, range, arg);
+ }
+ break;
+ }
+ c = 0x1FFFF;
+ stop = temp_charset_work->max_char;
+ }
+ c++;
+ }
+}
+
void
map_charset_chars (c_function, function, arg,
charset, from, to)
@@ -563,20 +737,9 @@ map_charset_chars (c_function, function, arg,
Lisp_Object range;
int partial;
- if (CHARSET_METHOD (charset) == CHARSET_METHOD_MAP_DEFERRED)
- load_charset (charset);
-
partial = (from > CHARSET_MIN_CODE (charset)
|| to < CHARSET_MAX_CODE (charset));
- if (CHARSET_UNIFIED_P (charset)
- && CHAR_TABLE_P (CHARSET_DEUNIFIER (charset)))
- {
- map_char_table_for_charset (c_function, function,
- CHARSET_DEUNIFIER (charset), arg,
- partial ? charset : NULL, from, to);
- }
-
if (CHARSET_METHOD (charset) == CHARSET_METHOD_OFFSET)
{
int from_idx = CODE_POINT_TO_INDEX (charset, from);
@@ -584,6 +747,18 @@ map_charset_chars (c_function, function, arg,
int from_c = from_idx + CHARSET_CODE_OFFSET (charset);
int to_c = to_idx + CHARSET_CODE_OFFSET (charset);
+ if (CHARSET_UNIFIED_P (charset))
+ {
+ if (! CHAR_TABLE_P (CHARSET_DEUNIFIER (charset)))
+ load_charset (charset, 2);
+ if (CHAR_TABLE_P (CHARSET_DEUNIFIER (charset)))
+ map_char_table_for_charset (c_function, function,
+ CHARSET_DEUNIFIER (charset), arg,
+ partial ? charset : NULL, from, to);
+ else
+ map_charset_for_dump (c_function, function, arg, from, to);
+ }
+
range = Fcons (make_number (from_c), make_number (to_c));
if (NILP (function))
(*c_function) (arg, range);
@@ -593,10 +768,13 @@ map_charset_chars (c_function, function, arg,
else if (CHARSET_METHOD (charset) == CHARSET_METHOD_MAP)
{
if (! CHAR_TABLE_P (CHARSET_ENCODER (charset)))
- return;
- map_char_table_for_charset (c_function, function,
- CHARSET_ENCODER (charset), arg,
- partial ? charset : NULL, from, to);
+ load_charset (charset, 2);
+ if (CHAR_TABLE_P (CHARSET_ENCODER (charset)))
+ map_char_table_for_charset (c_function, function,
+ CHARSET_ENCODER (charset), arg,
+ partial ? charset : NULL, from, to);
+ else
+ map_charset_for_dump (c_function, function, arg, from, to);
}
else if (CHARSET_METHOD (charset) == CHARSET_METHOD_SUBSET)
{
@@ -815,7 +993,7 @@ usage: (define-charset-internal ...) */)
charset.max_code = code;
}
- charset.compact_codes_p = charset.max_code < 0x1000000;
+ charset.compact_codes_p = charset.max_code < 0x10000;
val = args[charset_arg_invalid_code];
if (NILP (val))
@@ -904,11 +1082,7 @@ usage: (define-charset-internal ...) */)
{
val = args[charset_arg_map];
ASET (attrs, charset_map, val);
- if (STRINGP (val))
- load_charset_map_from_file (&charset, val, 0);
- else
- load_charset_map_from_vector (&charset, val, 0);
- charset.method = CHARSET_METHOD_MAP_DEFERRED;
+ charset.method = CHARSET_METHOD_MAP;
}
else if (! NILP (args[charset_arg_subset]))
{
@@ -1024,6 +1198,12 @@ usage: (define-charset-internal ...) */)
charset.id = id;
charset_table[id] = charset;
+ if (charset.method == CHARSET_METHOD_MAP)
+ {
+ load_charset (&charset, 0);
+ charset_table[id] = charset;
+ }
+
if (charset.iso_final >= 0)
{
ISO_CHARSET_TABLE (charset.dimension, charset.iso_chars_96,
@@ -1038,7 +1218,7 @@ usage: (define-charset-internal ...) */)
else if (ISO_CHARSET_TABLE (2, 0, 'B') == id)
charset_jisx0208 = id;
}
-
+
if (charset.emacs_mule_id >= 0)
{
emacs_mule_charset[charset.emacs_mule_id] = CHARSET_FROM_ID (id);
@@ -1058,8 +1238,29 @@ usage: (define-charset-internal ...) */)
Vcharset_ordered_list = nconc2 (Vcharset_ordered_list,
Fcons (make_number (id), Qnil));
else
- Vcharset_ordered_list = Fcons (make_number (id),
- Vcharset_ordered_list);
+ {
+ Lisp_Object tail;
+
+ for (tail = Vcharset_ordered_list; CONSP (tail); tail = XCDR (tail))
+ {
+ struct charset *cs = CHARSET_FROM_ID (XINT (XCAR (tail)));
+
+ if (cs->supplementary_p)
+ break;
+ }
+ if (EQ (tail, Vcharset_ordered_list))
+ Vcharset_ordered_list = Fcons (make_number (id),
+ Vcharset_ordered_list);
+ else if (NILP (tail))
+ Vcharset_ordered_list = nconc2 (Vcharset_ordered_list,
+ Fcons (make_number (id), Qnil));
+ else
+ {
+ val = Fcons (XCAR (tail), XCDR (tail));
+ XSETCDR (tail, val);
+ XSETCAR (tail, make_number (id));
+ }
+ }
charset_ordered_list_tick++;
}
@@ -1149,42 +1350,6 @@ DEFUN ("define-charset-alias", Fdefine_charset_alias,
}
-DEFUN ("unibyte-charset", Funibyte_charset, Sunibyte_charset, 0, 0, 0,
- doc: /* Return the unibyte charset (set by `set-unibyte-charset'). */)
- ()
-{
- return CHARSET_NAME (CHARSET_FROM_ID (charset_unibyte));
-}
-
-
-DEFUN ("set-unibyte-charset", Fset_unibyte_charset, Sset_unibyte_charset,
- 1, 1, 0,
- doc: /* Set the unibyte charset to CHARSET.
-This determines how unibyte/multibyte conversion is done. See also
-function `unibyte-charset'. */)
- (charset)
- Lisp_Object charset;
-{
- struct charset *cs;
- int i, c;
-
- CHECK_CHARSET_GET_CHARSET (charset, cs);
- if (! cs->ascii_compatible_p
- || cs->dimension != 1)
- error ("Inappropriate unibyte charset: %s", SDATA (SYMBOL_NAME (charset)));
- charset_unibyte = cs->id;
- memset (unibyte_has_multibyte_table, 1, 128);
- for (i = 128; i < 256; i++)
- {
- c = DECODE_CHAR (cs, i);
- unibyte_to_multibyte_table[i] = (c < 0 ? BYTE8_TO_CHAR (i) : c);
- unibyte_has_multibyte_table[i] = c >= 0;
- }
-
- return Qnil;
-}
-
-
DEFUN ("charset-plist", Fcharset_plist, Scharset_plist, 1, 1, 0,
doc: /* Return the property list of CHARSET. */)
(charset)
@@ -1228,8 +1393,6 @@ Optional third argument DEUNIFY, if non-nil, means to de-unify CHARSET. */)
CHECK_CHARSET_GET_ID (charset, id);
cs = CHARSET_FROM_ID (id);
- if (CHARSET_METHOD (cs) == CHARSET_METHOD_MAP_DEFERRED)
- load_charset (cs);
if (NILP (deunify)
? CHARSET_UNIFIED_P (cs) && ! NILP (CHARSET_DEUNIFIER (cs))
: ! CHARSET_UNIFIED_P (cs))
@@ -1238,18 +1401,21 @@ Optional third argument DEUNIFY, if non-nil, means to de-unify CHARSET. */)
CHARSET_UNIFIED_P (cs) = 0;
if (NILP (deunify))
{
- if (CHARSET_METHOD (cs) != CHARSET_METHOD_OFFSET)
+ if (CHARSET_METHOD (cs) != CHARSET_METHOD_OFFSET
+ || CHARSET_CODE_OFFSET (cs) < 0x110000)
error ("Can't unify charset: %s", SDATA (SYMBOL_NAME (charset)));
if (NILP (unify_map))
unify_map = CHARSET_UNIFY_MAP (cs);
- if (STRINGP (unify_map))
- load_charset_map_from_file (cs, unify_map, 2);
- else if (VECTORP (unify_map))
- load_charset_map_from_vector (cs, unify_map, 2);
- else if (NILP (unify_map))
- error ("No unify-map for charset");
else
- error ("Bad unify-map arg");
+ {
+ if (! STRINGP (unify_map) && ! VECTORP (unify_map))
+ signal_error ("Bad unify-map", unify_map);
+ CHARSET_UNIFY_MAP (cs) = unify_map;
+ }
+ if (NILP (Vchar_unify_table))
+ Vchar_unify_table = Fmake_char_table (Qnil, Qnil);
+ char_table_set_range (Vchar_unify_table,
+ cs->min_char, cs->max_char, charset);
CHARSET_UNIFIED_P (cs) = 1;
}
else if (CHAR_TABLE_P (Vchar_unify_table))
@@ -1268,7 +1434,7 @@ Optional third argument DEUNIFY, if non-nil, means to de-unify CHARSET. */)
DEFUN ("get-unused-iso-final-char", Fget_unused_iso_final_char,
Sget_unused_iso_final_char, 2, 2, 0,
doc: /*
-Return an unused ISO final char for a charset of DIMENISION and CHARS.
+Return an unused ISO final char for a charset of DIMENSION and CHARS.
DIMENSION is the number of bytes to represent a character: 1 or 2.
CHARS is the number of characters in a dimension: 94 or 96.
@@ -1494,6 +1660,41 @@ only `ascii', `eight-bit-control', and `eight-bit-graphic'. */)
+/* Return a unified character code for C (>= 0x110000). VAL is a
+ value of Vchar_unify_table for C; i.e. it is nil, an integer, or a
+ charset symbol. */
+int
+maybe_unify_char (c, val)
+ int c;
+ Lisp_Object val;
+{
+ struct charset *charset;
+
+ if (INTEGERP (val))
+ return XINT (val);
+ if (NILP (val))
+ return c;
+
+ CHECK_CHARSET_GET_CHARSET (val, charset);
+ load_charset (charset, 1);
+ if (! inhibit_load_charset_map)
+ {
+ val = CHAR_TABLE_REF (Vchar_unify_table, c);
+ if (! NILP (val))
+ c = XINT (val);
+ }
+ else
+ {
+ int code_index = c - CHARSET_CODE_OFFSET (charset);
+ int unified = GET_TEMP_CHARSET_WORK_DECODER (code_index);
+
+ if (unified > 0)
+ c = unified;
+ }
+ return c;
+}
+
+
/* Return a character correponding to the code-point CODE of
CHARSET. */
@@ -1508,12 +1709,6 @@ decode_char (charset, code)
if (code < CHARSET_MIN_CODE (charset) || code > CHARSET_MAX_CODE (charset))
return -1;
- if (method == CHARSET_METHOD_MAP_DEFERRED)
- {
- load_charset (charset);
- method = CHARSET_METHOD (charset);
- }
-
if (method == CHARSET_METHOD_SUBSET)
{
Lisp_Object subset_info;
@@ -1556,21 +1751,24 @@ decode_char (charset, code)
decoder = CHARSET_DECODER (charset);
if (! VECTORP (decoder))
- return -1;
- c = XINT (AREF (decoder, char_index));
+ {
+ load_charset (charset, 1);
+ decoder = CHARSET_DECODER (charset);
+ }
+ if (VECTORP (decoder))
+ c = XINT (AREF (decoder, char_index));
+ else
+ c = GET_TEMP_CHARSET_WORK_DECODER (char_index);
}
- else
+ else /* method == CHARSET_METHOD_OFFSET */
{
c = char_index + CHARSET_CODE_OFFSET (charset);
+ if (CHARSET_UNIFIED_P (charset)
+ && c > MAX_UNICODE_CHAR)
+ MAYBE_UNIFY_CHAR (c);
}
}
- if (CHARSET_UNIFIED_P (charset)
- && c >= 0)
- {
- MAYBE_UNIFY_CHAR (c);
- }
-
return c;
}
@@ -1592,16 +1790,27 @@ encode_char (charset, c)
if (CHARSET_UNIFIED_P (charset))
{
Lisp_Object deunifier, deunified;
+ int code_index = -1;
deunifier = CHARSET_DEUNIFIER (charset);
if (! CHAR_TABLE_P (deunifier))
{
- Funify_charset (CHARSET_NAME (charset), Qnil, Qnil);
+ load_charset (charset, 2);
deunifier = CHARSET_DEUNIFIER (charset);
}
- deunified = CHAR_TABLE_REF (deunifier, c);
- if (! NILP (deunified))
- c = XINT (deunified);
+ if (CHAR_TABLE_P (deunifier))
+ {
+ Lisp_Object deunified = CHAR_TABLE_REF (deunifier, c);
+
+ if (INTEGERP (deunified))
+ code_index = XINT (deunified);
+ }
+ else
+ {
+ code_index = GET_TEMP_CHARSET_WORK_ENCODER (c);
+ }
+ if (code_index >= 0)
+ c = CHARSET_CODE_OFFSET (charset) + code_index;
}
if (method == CHARSET_METHOD_SUBSET)
@@ -1642,12 +1851,6 @@ encode_char (charset, c)
|| c < CHARSET_MIN_CHAR (charset) || c > CHARSET_MAX_CHAR (charset))
return CHARSET_INVALID_CODE (charset);
- if (method == CHARSET_METHOD_MAP_DEFERRED)
- {
- load_charset (charset);
- method = CHARSET_METHOD (charset);
- }
-
if (method == CHARSET_METHOD_MAP)
{
Lisp_Object encoder;
@@ -1655,18 +1858,30 @@ encode_char (charset, c)
encoder = CHARSET_ENCODER (charset);
if (! CHAR_TABLE_P (CHARSET_ENCODER (charset)))
- return CHARSET_INVALID_CODE (charset);
- val = CHAR_TABLE_REF (encoder, c);
- if (NILP (val))
- return CHARSET_INVALID_CODE (charset);
- code = XINT (val);
- if (! CHARSET_COMPACT_CODES_P (charset))
- code = INDEX_TO_CODE_POINT (charset, code);
+ {
+ load_charset (charset, 2);
+ encoder = CHARSET_ENCODER (charset);
+ }
+ if (CHAR_TABLE_P (encoder))
+ {
+ val = CHAR_TABLE_REF (encoder, c);
+ if (NILP (val))
+ return CHARSET_INVALID_CODE (charset);
+ code = XINT (val);
+ if (! CHARSET_COMPACT_CODES_P (charset))
+ code = INDEX_TO_CODE_POINT (charset, code);
+ }
+ else
+ {
+ code = GET_TEMP_CHARSET_WORK_ENCODER (c);
+ code = INDEX_TO_CODE_POINT (charset, code);
+ }
}
else /* method == CHARSET_METHOD_OFFSET */
{
- code = c - CHARSET_CODE_OFFSET (charset);
- code = INDEX_TO_CODE_POINT (charset, code);
+ int code_index = c - CHARSET_CODE_OFFSET (charset);
+
+ code = INDEX_TO_CODE_POINT (charset, code_index);
}
return code;
@@ -1680,7 +1895,7 @@ Return nil if CODE-POINT is not valid in CHARSET.
CODE-POINT may be a cons (HIGHER-16-BIT-VALUE . LOWER-16-BIT-VALUE).
Optional argument RESTRICTION specifies a way to map the pair of CCS
-and CODE-POINT to a chracter. Currently not supported and just ignored. */)
+and CODE-POINT to a character. Currently not supported and just ignored. */)
(charset, code_point, restriction)
Lisp_Object charset, code_point, restriction;
{
@@ -1710,7 +1925,7 @@ DEFUN ("encode-char", Fencode_char, Sencode_char, 2, 3, 0,
doc: /* Encode the character CH into a code-point of CHARSET.
Return nil if CHARSET doesn't include CH.
-Optional argument RESTRICTION specifies a way to map CHAR to a
+Optional argument RESTRICTION specifies a way to map CH to a
code-point in CCS. Currently not supported and just ignored. */)
(ch, charset, restriction)
Lisp_Object ch, charset, restriction;
@@ -1822,8 +2037,12 @@ char_charset (c, charset_list, code_return)
Lisp_Object charset_list;
unsigned *code_return;
{
+ int maybe_null = 0;
+
if (NILP (charset_list))
charset_list = Vcharset_ordered_list;
+ else
+ maybe_null = 1;
while (CONSP (charset_list))
{
@@ -1837,17 +2056,22 @@ char_charset (c, charset_list, code_return)
return charset;
}
charset_list = XCDR (charset_list);
+ if (c <= MAX_UNICODE_CHAR
+ && EQ (charset_list, Vcharset_non_preferred_head))
+ return CHARSET_FROM_ID (charset_unicode);
}
- return NULL;
+ return (maybe_null ? NULL
+ : c <= MAX_5_BYTE_CHAR ? CHARSET_FROM_ID (charset_emacs)
+ : CHARSET_FROM_ID (charset_eight_bit));
}
DEFUN ("split-char", Fsplit_char, Ssplit_char, 1, 1, 0,
doc:
- /*Return list of charset and one to four position-codes of CHAR.
+ /*Return list of charset and one to four position-codes of CH.
The charset is decided by the current priority order of charsets.
A position-code is a byte value of each dimension of the code-point of
-CHAR in the charset. */)
+CH in the charset. */)
(ch)
Lisp_Object ch;
{
@@ -1874,15 +2098,41 @@ CHAR in the charset. */)
}
-DEFUN ("char-charset", Fchar_charset, Schar_charset, 1, 1, 0,
- doc: /* Return the charset of highest priority that contains CH. */)
- (ch)
- Lisp_Object ch;
+DEFUN ("char-charset", Fchar_charset, Schar_charset, 1, 2, 0,
+ doc: /* Return the charset of highest priority that contains CH.
+If optional 2nd arg RESTRICTION is non-nil, it is a list of charsets
+from which to find the charset. It may also be a coding system. In
+that case, find the charset from what supported by that coding system. */)
+ (ch, restriction)
+ Lisp_Object ch, restriction;
{
struct charset *charset;
CHECK_CHARACTER (ch);
- charset = CHAR_CHARSET (XINT (ch));
+ if (NILP (restriction))
+ charset = CHAR_CHARSET (XINT (ch));
+ else
+ {
+ Lisp_Object charset_list;
+
+ if (CONSP (restriction))
+ {
+ for (charset_list = Qnil; CONSP (restriction);
+ restriction = XCDR (restriction))
+ {
+ int id;
+
+ CHECK_CHARSET_GET_ID (XCAR (restriction), id);
+ charset_list = Fcons (make_number (id), charset_list);
+ }
+ charset_list = Fnreverse (charset_list);
+ }
+ else
+ charset_list = coding_system_charset_list (restriction);
+ charset = char_charset (XINT (ch), charset_list, NULL);
+ if (! charset)
+ return Qnil;
+ }
return (CHARSET_NAME (charset));
}
@@ -1912,7 +2162,7 @@ Return charset of ISO's specification DIMENSION, CHARS, and FINAL-CHAR.
ISO 2022's designation sequence (escape sequence) distinguishes charsets
by their DIMENSION, CHARS, and FINAL-CHAR,
-where as Emacs distinguishes them by charset symbol.
+whereas Emacs distinguishes them by charset symbol.
See the documentation of the function `charset-info' for the meanings of
DIMENSION, CHARS, and FINAL-CHAR. */)
(dimension, chars, final_char)
@@ -1932,35 +2182,23 @@ DIMENSION, CHARS, and FINAL-CHAR. */)
DEFUN ("clear-charset-maps", Fclear_charset_maps, Sclear_charset_maps,
0, 0, 0,
doc: /*
-Clear encoder and decoder of charsets that are loaded from mapfiles. */)
+Internal use only.
+Clear temporary charset mapping tables.
+It should be called only from temacs invoked for dumping. */)
()
{
int i;
struct charset *charset;
Lisp_Object attrs;
- for (i = 0; i < charset_table_used; i++)
+ if (temp_charset_work)
{
- charset = CHARSET_FROM_ID (i);
- attrs = CHARSET_ATTRIBUTES (charset);
-
- if (CHARSET_METHOD (charset) == CHARSET_METHOD_MAP)
- {
- CHARSET_ATTR_DECODER (attrs) = Qnil;
- CHARSET_ATTR_ENCODER (attrs) = Qnil;
- CHARSET_METHOD (charset) = CHARSET_METHOD_MAP_DEFERRED;
- }
-
- if (CHARSET_UNIFIED_P (charset))
- CHARSET_ATTR_DEUNIFIER (attrs) = Qnil;
+ free (temp_charset_work);
+ temp_charset_work = NULL;
}
- if (CHAR_TABLE_P (Vchar_unified_charset_table))
- {
- Foptimize_char_table (Vchar_unified_charset_table);
- Vchar_unify_table = Vchar_unified_charset_table;
- Vchar_unified_charset_table = Qnil;
- }
+ if (CHAR_TABLE_P (Vchar_unify_table))
+ Foptimize_char_table (Vchar_unify_table, Qnil);
return Qnil;
}
@@ -2009,7 +2247,7 @@ usage: (set-charset-priority &rest charsets) */)
}
}
arglist[0] = Fnreverse (new_head);
- arglist[1] = old_list;
+ arglist[1] = Vcharset_non_preferred_head = old_list;
Vcharset_ordered_list = Fnconc (2, arglist);
charset_ordered_list_tick++;
@@ -2044,9 +2282,18 @@ Return charset identification number of CHARSET. */)
void
init_charset ()
{
- Vcharset_map_path
- = Fcons (Fexpand_file_name (build_string ("charsets"), Vdata_directory),
- Qnil);
+ Lisp_Object tempdir;
+ tempdir = Fexpand_file_name (build_string ("charsets"), Vdata_directory);
+ if (access (SDATA (tempdir), 0) < 0)
+ {
+ dir_warning ("Error: charsets directory (%s) does not exist.\n\
+Emacs will not function correctly without the character map files.\n\
+Please check your installation!\n",
+ tempdir);
+ /* TODO should this be a fatal error? (Bug#909) */
+ }
+
+ Vcharset_map_path = Fcons (tempdir, Qnil);
}
@@ -2082,6 +2329,7 @@ syms_of_charset ()
DEFSYM (Qascii, "ascii");
DEFSYM (Qunicode, "unicode");
+ DEFSYM (Qemacs, "emacs");
DEFSYM (Qeight_bit, "eight-bit");
DEFSYM (Qiso_8859_1, "iso-8859-1");
@@ -2114,15 +2362,10 @@ syms_of_charset ()
xmalloc (sizeof (struct charset) * charset_table_size));
charset_table_used = 0;
- staticpro (&Vchar_unified_charset_table);
- Vchar_unified_charset_table = Fmake_char_table (Qnil, make_number (-1));
-
defsubr (&Scharsetp);
defsubr (&Smap_charset_chars);
defsubr (&Sdefine_charset_internal);
defsubr (&Sdefine_charset_alias);
- defsubr (&Sunibyte_charset);
- defsubr (&Sset_unibyte_charset);
defsubr (&Scharset_plist);
defsubr (&Sset_charset_plist);
defsubr (&Sunify_charset);
@@ -2143,13 +2386,23 @@ syms_of_charset ()
defsubr (&Scharset_id_internal);
DEFVAR_LISP ("charset-map-path", &Vcharset_map_path,
- doc: /* *Lisp of directories to search for charset map files. */);
+ doc: /* *List of directories to search for charset map files. */);
Vcharset_map_path = Qnil;
+ DEFVAR_BOOL ("inhibit-load-charset-map", &inhibit_load_charset_map,
+ doc: /* Inhibit loading of charset maps. Used when dumping Emacs. */);
+ inhibit_load_charset_map = 0;
+
DEFVAR_LISP ("charset-list", &Vcharset_list,
doc: /* List of all charsets ever defined. */);
Vcharset_list = Qnil;
+ DEFVAR_LISP ("current-iso639-language", &Vcurrent_iso639_language,
+ doc: /* ISO639 language mnemonic symbol for the current language environment.
+If the current language environment is for multiple languages (e.g. "Latin-1"),
+the value may be a list of mnemonics. */);
+ Vcurrent_iso639_language = Qnil;
+
charset_ascii
= define_charset_internal (Qascii, 1, "\x00\x7F\x00\x00\x00\x00",
0, 127, 'B', -1, 0, 1, 0, 0);
@@ -2159,9 +2412,12 @@ syms_of_charset ()
charset_unicode
= define_charset_internal (Qunicode, 3, "\x00\xFF\x00\xFF\x00\x10",
0, MAX_UNICODE_CHAR, -1, 0, -1, 1, 0, 0);
+ charset_emacs
+ = define_charset_internal (Qemacs, 3, "\x00\xFF\x00\xFF\x00\x3F",
+ 0, MAX_5_BYTE_CHAR, -1, 0, -1, 1, 1, 0);
charset_eight_bit
= define_charset_internal (Qeight_bit, 1, "\x80\xFF\x00\x00\x00\x00",
- 128, 255, -1, 0, -1, 0, 0,
+ 128, 255, -1, 0, -1, 0, 1,
MAX_5_BYTE_CHAR + 1);
}