/* Basic character set support.
- Copyright (C) 2001-2012 Free Software Foundation, Inc.
+ Copyright (C) 2001-2013 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)
#include <config.h>
-#define CHARSET_INLINE EXTERN_INLINE
-
+#include <errno.h>
#include <stdio.h>
#include <unistd.h>
-#include <ctype.h>
#include <limits.h>
#include <sys/types.h>
-#include <setjmp.h>
+#include <c-ctype.h>
#include "lisp.h"
#include "character.h"
#include "charset.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;
/* 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;
return 0;
}
n = 0;
- while (isxdigit (c = getc (fp)))
+ while (c_isxdigit (c = getc (fp)))
{
if (UINT_MAX >> 4 < n)
*overflow = 1;
`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));
+ 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))
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;
(Lisp_Object dimension, Lisp_Object chars, Lisp_Object final_char, Lisp_Object charset)
{
int id;
- int chars_flag;
+ bool chars_flag;
CHECK_CHARSET_GET_ID (charset, id);
check_iso_charset_parameter (dimension, chars, final_char);
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;
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. */)
(Lisp_Object dimension, Lisp_Object chars, Lisp_Object final_char)
{
int id;
- int chars_flag;
+ bool chars_flag;
check_iso_charset_parameter (dimension, chars, final_char);
chars_flag = XFASTINT (chars) == 96;
{
Lisp_Object tempdir;
tempdir = Fexpand_file_name (build_string ("charsets"), Vdata_directory);
- if (access (SSDATA (tempdir), 0) < 0)
+ if (! file_accessible_directory_p (SSDATA (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
exit (1);
}
- Vcharset_map_path = Fcons (tempdir, Qnil);
+ Vcharset_map_path = list1 (tempdir);
}