/* Basic character set support.
- Copyright (C) 2001-2012 Free Software Foundation, Inc.
- Copyright (C) 1995, 1996, 1997, 1998, 1999, 2000, 2001, 2002, 2003, 2004,
- 2005, 2006, 2007, 2008, 2009, 2010, 2011
- National Institute of Advanced Industrial Science and Technology (AIST)
- Registration Number H14PRO021
- Copyright (C) 2003, 2004
- National Institute of Advanced Industrial Science and Technology (AIST)
- Registration Number H13PRO009
+Copyright (C) 2001-2014 Free Software Foundation, Inc.
+
+Copyright (C) 1995, 1996, 1997, 1998, 1999, 2000, 2001, 2002, 2003, 2004,
+ 2005, 2006, 2007, 2008, 2009, 2010, 2011
+ National Institute of Advanced Industrial Science and Technology (AIST)
+ Registration Number H14PRO021
+
+Copyright (C) 2003, 2004
+ National Institute of Advanced Industrial Science and Technology (AIST)
+ Registration Number H13PRO009
This file is part of GNU Emacs.
#include <config.h>
-#define CHARSET_INLINE EXTERN_INLINE
-
+#include <errno.h>
#include <stdio.h>
#include <unistd.h>
#include <limits.h>
#include <sys/types.h>
-#include <setjmp.h>
#include <c-ctype.h>
#include "lisp.h"
#include "character.h"
/* Set to 1 to warn that a charset map is loaded and thus a buffer
text and a string data may be relocated. */
-int charset_map_loaded;
+bool charset_map_loaded;
struct charset_map_entries
{
{
Lisp_Object vec IF_LINT (= Qnil), table IF_LINT (= Qnil);
unsigned max_code = CHARSET_MAX_CODE (charset);
- int ascii_compatible_p = charset->ascii_compatible_p;
+ bool ascii_compatible_p = charset->ascii_compatible_p;
int min_char, max_char, nonascii_min_char;
int i;
unsigned char *fast_map = charset->fast_map;
{
if (ascii_compatible_p)
{
- if (! ASCII_BYTE_P (from_c))
+ if (! ASCII_CHAR_P (from_c))
{
if (from_c < nonascii_min_char)
nonascii_min_char = from_c;
}
- else if (! ASCII_BYTE_P (to_c))
+ else if (! ASCII_CHAR_P (to_c))
{
nonascii_min_char = 0x80;
}
/* Read a hexadecimal number (preceded by "0x") from the file FP while
paying attention to comment character '#'. */
-static inline unsigned
-read_hex (FILE *fp, int *eof, int *overflow)
+static unsigned
+read_hex (FILE *fp, bool *eof, bool *overflow)
{
int c;
unsigned n;
`file-name-handler-alist' to avoid running any Lisp code. */
static void
-load_charset_map_from_file (struct charset *charset, Lisp_Object mapfile, int control_flag)
+load_charset_map_from_file (struct charset *charset, Lisp_Object mapfile,
+ int control_flag)
{
unsigned min_code = CHARSET_MIN_CODE (charset);
unsigned max_code = CHARSET_MAX_CODE (charset);
struct charset_map_entries *head, *entries;
int n_entries;
ptrdiff_t count;
- USE_SAFE_ALLOCA;
- suffixes = Fcons (build_string (".map"),
- Fcons (build_string (".TXT"), Qnil));
+ suffixes = list2 (build_string (".map"), build_string (".TXT"));
count = SPECPDL_INDEX ();
+ record_unwind_protect_nothing ();
specbind (Qfile_name_handler_alist, Qnil);
- fd = openp (Vcharset_map_path, mapfile, suffixes, NULL, Qnil);
- unbind_to (count, Qnil);
- if (fd < 0
- || ! (fp = fdopen (fd, "r")))
- error ("Failure in loading charset map: %s", SDATA (mapfile));
+ fd = openp (Vcharset_map_path, mapfile, suffixes, NULL, Qnil, false);
+ fp = fd < 0 ? 0 : fdopen (fd, "r");
+ if (!fp)
+ {
+ int open_errno = errno;
+ emacs_close (fd);
+ report_file_errno ("Loading charset map", mapfile, open_errno);
+ }
+ set_unwind_protect_ptr (count, fclose_unwind, fp);
+ unbind_to (count + 1, Qnil);
- /* Use SAFE_ALLOCA instead of alloca, as `charset_map_entries' is
+ /* Use record_xmalloc, as `charset_map_entries' is
large (larger than MAX_ALLOCA). */
- head = SAFE_ALLOCA (sizeof *head);
+ head = record_xmalloc (sizeof *head);
entries = head;
memset (entries, 0, sizeof (struct charset_map_entries));
{
unsigned from, to, c;
int idx;
- int eof = 0, overflow = 0;
+ bool eof = 0, overflow = 0;
from = read_hex (fp, &eof, &overflow);
if (eof)
if (from < min_code || to > max_code || from > to || c > MAX_CHAR)
continue;
- if (n_entries > 0 && (n_entries % 0x10000) == 0)
+ if (n_entries == 0x10000)
{
- entries->next = SAFE_ALLOCA (sizeof *entries->next);
+ entries->next = record_xmalloc (sizeof *entries->next);
entries = entries->next;
memset (entries, 0, sizeof (struct charset_map_entries));
n_entries = 0;
n_entries++;
}
fclose (fp);
+ clear_unwind_protect (count);
load_charset_map (charset, head, n_entries, control_flag);
- SAFE_FREE ();
+ unbind_to (count, Qnil);
}
static void
else
{
if (! CHARSET_UNIFIED_P (charset))
- abort ();
+ emacs_abort ();
map = CHARSET_UNIFY_MAP (charset);
}
if (STRINGP (map))
{
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;
+ Lisp_Object range = Fcons (Qnil, Qnil);
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
}
c++;
}
- UNGCPRO;
}
void
Lisp_Object arg, struct charset *charset, unsigned from, unsigned to)
{
Lisp_Object range;
- int partial;
-
- partial = (from > CHARSET_MIN_CODE (charset)
- || to < CHARSET_MAX_CODE (charset));
+ bool partial = (from > CHARSET_MIN_CODE (charset)
+ || to < CHARSET_MAX_CODE (charset));
if (CHARSET_METHOD (charset) == CHARSET_METHOD_OFFSET)
{
struct charset charset;
int id;
int dimension;
- int new_definition_p;
+ bool new_definition_p;
int nchars;
if (nargs != charset_arg_max)
CHECK_NATNUM (parent_max_code);
parent_code_offset = Fnth (make_number (3), val);
CHECK_NUMBER (parent_code_offset);
- val = Fmake_vector (make_number (4), Qnil);
+ val = make_uninit_vector (4);
ASET (val, 0, make_number (parent_charset->id));
ASET (val, 1, parent_min_code);
ASET (val, 2, parent_max_code);
{
new_definition_p = 0;
id = XFASTINT (CHARSET_SYMBOL_ID (args[charset_arg_name]));
- set_hash_value (hash_table, charset.hash_index, attrs);
+ set_hash_value_slot (hash_table, charset.hash_index, attrs);
}
else
{
example, the IDs are stuffed into struct
coding_system.charbuf[i] entries, which are 'int'. */
int old_size = charset_table_size;
+ ptrdiff_t new_size = old_size;
struct charset *new_table =
- xpalloc (0, &charset_table_size, 1,
+ xpalloc (0, &new_size, 1,
min (INT_MAX, MOST_POSITIVE_FIXNUM),
sizeof *charset_table);
memcpy (new_table, charset_table, old_size * sizeof *new_table);
charset_table = new_table;
+ charset_table_size = new_size;
/* FIXME: This leaks memory, as the old charset_table becomes
unreachable. If the old charset table is charset_table_init
then this leak is intentional; otherwise, it's unclear.
charset.iso_final) = id;
if (new_definition_p)
Viso_2022_charset_list = nconc2 (Viso_2022_charset_list,
- Fcons (make_number (id), Qnil));
+ list1 (make_number (id)));
if (ISO_CHARSET_TABLE (1, 0, 'J') == id)
charset_jisx0201_roman = id;
else if (ISO_CHARSET_TABLE (2, 0, '@') == id)
emacs_mule_bytes[charset.emacs_mule_id] = charset.dimension + 2;
if (new_definition_p)
Vemacs_mule_charset_list = nconc2 (Vemacs_mule_charset_list,
- Fcons (make_number (id), Qnil));
+ list1 (make_number (id)));
}
if (new_definition_p)
Vcharset_list = Fcons (args[charset_arg_name], Vcharset_list);
if (charset.supplementary_p)
Vcharset_ordered_list = nconc2 (Vcharset_ordered_list,
- Fcons (make_number (id), Qnil));
+ list1 (make_number (id)));
else
{
Lisp_Object tail;
Vcharset_ordered_list);
else if (NILP (tail))
Vcharset_ordered_list = nconc2 (Vcharset_ordered_list,
- Fcons (make_number (id), Qnil));
+ list1 (make_number (id)));
else
{
val = Fcons (XCAR (tail), XCDR (tail));
const char *code_space_chars,
unsigned min_code, unsigned max_code,
int iso_final, int iso_revision, int emacs_mule_id,
- int ascii_compatible, int supplementary,
+ bool ascii_compatible, bool supplementary,
int code_offset)
{
const unsigned char *code_space = (const unsigned char *) code_space_chars;
args[charset_arg_name] = name;
args[charset_arg_dimension] = make_number (dimension);
- val = Fmake_vector (make_number (8), make_number (0));
+ val = make_uninit_vector (8);
for (i = 0; i < 8; i++)
ASET (val, i, make_number (code_space[i]));
args[charset_arg_code_space] = val;
return Qnil;
}
+/* Check that DIMENSION, CHARS, and FINAL_CHAR specify a valid ISO charset.
+ Return true if it's a 96-character set, false if 94. */
+
+static bool
+check_iso_charset_parameter (Lisp_Object dimension, Lisp_Object chars,
+ Lisp_Object final_char)
+{
+ CHECK_NUMBER (dimension);
+ CHECK_NUMBER (chars);
+ CHECK_CHARACTER (final_char);
+
+ if (! (1 <= XINT (dimension) && XINT (dimension) <= 3))
+ error ("Invalid DIMENSION %"pI"d, it should be 1, 2, or 3",
+ XINT (dimension));
+
+ bool chars_flag = XINT (chars) == 96;
+ if (! (chars_flag || XINT (chars) == 94))
+ error ("Invalid CHARS %"pI"d, it should be 94 or 96", XINT (chars));
+
+ int final_ch = XFASTINT (final_char);
+ if (! ('0' <= final_ch && final_ch <= '~'))
+ error ("Invalid FINAL-CHAR '%c', it should be '0'..'~'", final_ch);
+
+ return chars_flag;
+}
+
DEFUN ("get-unused-iso-final-char", Fget_unused_iso_final_char,
Sget_unused_iso_final_char, 2, 2, 0,
doc: /*
return nil. */)
(Lisp_Object dimension, Lisp_Object chars)
{
- int final_char;
-
- CHECK_NUMBER (dimension);
- CHECK_NUMBER (chars);
- if (XINT (dimension) != 1 && XINT (dimension) != 2 && XINT (dimension) != 3)
- args_out_of_range_3 (dimension, make_number (1), make_number (3));
- if (XINT (chars) != 94 && XINT (chars) != 96)
- args_out_of_range_3 (chars, make_number (94), make_number (96));
- for (final_char = '0'; final_char <= '?'; final_char++)
- if (ISO_CHARSET_TABLE (XINT (dimension), XINT (chars), final_char) < 0)
- break;
- return (final_char <= '?' ? make_number (final_char) : Qnil);
-}
-
-static void
-check_iso_charset_parameter (Lisp_Object dimension, Lisp_Object chars, Lisp_Object final_char)
-{
- CHECK_NATNUM (dimension);
- CHECK_NATNUM (chars);
- CHECK_CHARACTER (final_char);
-
- if (XINT (dimension) > 3)
- error ("Invalid DIMENSION %"pI"d, it should be 1, 2, or 3",
- XINT (dimension));
- if (XINT (chars) != 94 && XINT (chars) != 96)
- error ("Invalid CHARS %"pI"d, it should be 94 or 96", XINT (chars));
- if (XINT (final_char) < '0' || XINT (final_char) > '~')
- error ("Invalid FINAL-CHAR %c, it should be `0'..`~'",
- (int)XINT (final_char));
+ bool chars_flag = check_iso_charset_parameter (dimension, chars,
+ make_number ('0'));
+ for (int final_char = '0'; final_char <= '?'; final_char++)
+ if (ISO_CHARSET_TABLE (XINT (dimension), chars_flag, final_char) < 0)
+ return make_number (final_char);
+ return Qnil;
}
(Lisp_Object dimension, Lisp_Object chars, Lisp_Object final_char, Lisp_Object charset)
{
int id;
- int chars_flag;
CHECK_CHARSET_GET_ID (charset, id);
- check_iso_charset_parameter (dimension, chars, final_char);
- chars_flag = XINT (chars) == 96;
- ISO_CHARSET_TABLE (XINT (dimension), chars_flag, XINT (final_char)) = id;
+ bool chars_flag = check_iso_charset_parameter (dimension, chars, final_char);
+ ISO_CHARSET_TABLE (XINT (dimension), chars_flag, XFASTINT (final_char)) = id;
return Qnil;
}
It may lookup a translation table TABLE if supplied. */
static void
-find_charsets_in_text (const unsigned char *ptr, ptrdiff_t nchars, ptrdiff_t nbytes, Lisp_Object charsets, Lisp_Object table, int multibyte)
+find_charsets_in_text (const unsigned char *ptr, ptrdiff_t nchars,
+ ptrdiff_t nbytes, Lisp_Object charsets,
+ Lisp_Object table, bool multibyte)
{
const unsigned char *pend = ptr + nbytes;
if (!NILP (table))
c = translate_char (table, c);
- if (ASCII_BYTE_P (c))
+ if (ASCII_CHAR_P (c))
ASET (charsets, charset_ascii, Qt);
else
ASET (charsets, charset_eight_bit, Qt);
ptrdiff_t from, from_byte, to, stop, stop_byte;
int i;
Lisp_Object val;
- int multibyte = ! NILP (BVAR (current_buffer, enable_multibyte_characters));
+ bool multibyte = ! NILP (BVAR (current_buffer, enable_multibyte_characters));
validate_region (&beg, &end);
from = XFASTINT (beg);
/* 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
+static int
maybe_unify_char (int c, Lisp_Object val)
{
struct charset *charset;
{
c = char_index + CHARSET_CODE_OFFSET (charset);
if (CHARSET_UNIFIED_P (charset)
- && c > MAX_UNICODE_CHAR)
- MAYBE_UNIFY_CHAR (c);
+ && MAX_UNICODE_CHAR < c && c <= MAX_5_BYTE_CHAR)
+ {
+ /* Unify C with a Unicode character if possible. */
+ Lisp_Object val = CHAR_TABLE_REF (Vchar_unify_table, c);
+ c = maybe_unify_char (c, val);
+ }
}
}
/* Variable used temporarily by the macro ENCODE_CHAR. */
Lisp_Object charset_work;
-/* Return a code-point of CHAR in CHARSET. If CHAR doesn't belong to
+/* Return a code-point of C in CHARSET. If C doesn't belong to
CHARSET, return CHARSET_INVALID_CODE (CHARSET). If STRICT is true,
use CHARSET's strict_max_char instead of max_char. */
doc: /* Decode the pair of CHARSET and CODE-POINT into a character.
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 character. Currently not supported and just ignored. */)
+CODE-POINT may be a cons (HIGHER-16-BIT-VALUE . LOWER-16-BIT-VALUE). */)
(Lisp_Object charset, Lisp_Object code_point, Lisp_Object restriction)
{
int c, id;
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 CH to a
-code-point in CCS. Currently not supported and just ignored. */)
+Return nil if CHARSET doesn't include CH. */)
(Lisp_Object ch, Lisp_Object charset, Lisp_Object restriction)
{
int c, id;
struct charset *
char_charset (int c, Lisp_Object charset_list, unsigned int *code_return)
{
- int maybe_null = 0;
+ bool maybe_null = 0;
if (NILP (charset_list))
charset_list = Vcharset_ordered_list;
c = XFASTINT (ch);
charset = CHAR_CHARSET (c);
if (! charset)
- abort ();
+ emacs_abort ();
code = ENCODE_CHAR (charset, c);
if (code == CHARSET_INVALID_CODE (charset))
- abort ();
+ emacs_abort ();
dimension = CHARSET_DIMENSION (charset);
for (val = Qnil; dimension > 0; dimension--)
{
DEFUN ("char-charset", Fchar_charset, Schar_charset, 1, 2, 0,
doc: /* Return the charset of highest priority that contains CH.
+ASCII characters are an exception: for them, this function always
+returns `ascii'.
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. */)
DIMENSION, CHARS, and FINAL-CHAR. */)
(Lisp_Object dimension, Lisp_Object chars, Lisp_Object final_char)
{
- int id;
- int chars_flag;
-
- check_iso_charset_parameter (dimension, chars, final_char);
- chars_flag = XFASTINT (chars) == 96;
- id = ISO_CHARSET_TABLE (XFASTINT (dimension), chars_flag,
- XFASTINT (final_char));
+ bool chars_flag = check_iso_charset_parameter (dimension, chars, final_char);
+ int id = ISO_CHARSET_TABLE (XINT (dimension), chars_flag,
+ XFASTINT (final_char));
return (id >= 0 ? CHARSET_NAME (CHARSET_FROM_ID (id)) : Qnil);
}
{
Lisp_Object tempdir;
tempdir = Fexpand_file_name (build_string ("charsets"), Vdata_directory);
- if (access (SSDATA (tempdir), 0) < 0)
+ if (! file_accessible_directory_p (tempdir))
{
/* This used to be non-fatal (dir_warning), but it should not
happen, and if it does sooner or later it will cause some
obscure problem (eg bug#6401), so better abort. */
fprintf (stderr, "Error: charsets directory not found:\n\
%s\n\
-Emacs will not function correctly without the character map files.\n\
+Emacs will not function correctly without the character map files.\n%s\
Please check your installation!\n",
- SDATA (tempdir));
+ SDATA (tempdir),
+ egetenv("EMACSDATA") ? "The EMACSDATA environment \
+variable is set, maybe it has the wrong value?\n" : "");
exit (1);
}
- Vcharset_map_path = Fcons (tempdir, Qnil);
+ Vcharset_map_path = list1 (tempdir);
}
}
charset_table = charset_table_init;
- charset_table_size = sizeof charset_table_init / sizeof *charset_table_init;
+ charset_table_size = ARRAYELTS (charset_table_init);
charset_table_used = 0;
defsubr (&Scharsetp);
Vcurrent_iso639_language = Qnil;
charset_ascii
- = define_charset_internal (Qascii, 1, "\x00\x7F\x00\x00\x00\x00",
+ = define_charset_internal (Qascii, 1, "\x00\x7F\0\0\0\0\0",
0, 127, 'B', -1, 0, 1, 0, 0);
charset_iso_8859_1
- = define_charset_internal (Qiso_8859_1, 1, "\x00\xFF\x00\x00\x00\x00",
+ = define_charset_internal (Qiso_8859_1, 1, "\x00\xFF\0\0\0\0\0",
0, 255, -1, -1, -1, 1, 0, 0);
charset_unicode
- = define_charset_internal (Qunicode, 3, "\x00\xFF\x00\xFF\x00\x10",
+ = define_charset_internal (Qunicode, 3, "\x00\xFF\x00\xFF\x00\x10\0",
0, MAX_UNICODE_CHAR, -1, 0, -1, 1, 0, 0);
charset_emacs
- = define_charset_internal (Qemacs, 3, "\x00\xFF\x00\xFF\x00\x3F",
+ = define_charset_internal (Qemacs, 3, "\x00\xFF\x00\xFF\x00\x3F\0",
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",
+ = define_charset_internal (Qeight_bit, 1, "\x80\xFF\0\0\0\0\0",
128, 255, -1, 0, -1, 0, 1,
MAX_5_BYTE_CHAR + 1);
charset_unibyte = charset_iso_8859_1;