#include <config.h>
#include "lisp.h"
+
#include "buffer.h"
#include "elhash.h"
#include "insdel.h"
#include "lstream.h"
+#include "opaque.h"
#ifdef MULE
#include "mule-ccl.h"
+#include "chartab.h"
#endif
#include "file-coding.h"
-Lisp_Object Qbuffer_file_coding_system, Qcoding_system_error;
+Lisp_Object Qcoding_system_error;
Lisp_Object Vkeyboard_coding_system;
Lisp_Object Vterminal_coding_system;
/* Table of symbols identifying each coding category. */
Lisp_Object coding_category_symbol[CODING_CATEGORY_LAST + 1];
-/* Coding system currently associated with each coding category. */
-Lisp_Object coding_category_system[CODING_CATEGORY_LAST + 1];
-/* Table of all coding categories in decreasing order of priority.
- This describes a permutation of the possible coding categories. */
-int coding_category_by_priority[CODING_CATEGORY_LAST + 1];
-Lisp_Object Qcoding_system_p;
+struct file_coding_dump {
+ /* Coding system currently associated with each coding category. */
+ Lisp_Object coding_category_system[CODING_CATEGORY_LAST + 1];
+
+ /* Table of all coding categories in decreasing order of priority.
+ This describes a permutation of the possible coding categories. */
+ int coding_category_by_priority[CODING_CATEGORY_LAST + 1];
+
+#ifdef MULE
+ Lisp_Object ucs_to_mule_table[65536];
+#endif
+} *fcd;
+
+static const struct lrecord_description fcd_description_1[] = {
+ { XD_LISP_OBJECT_ARRAY, offsetof (struct file_coding_dump, coding_category_system), CODING_CATEGORY_LAST + 1 },
+#ifdef MULE
+ { XD_LISP_OBJECT_ARRAY, offsetof (struct file_coding_dump, ucs_to_mule_table), 65536 },
+#endif
+ { XD_END }
+};
+
+static const struct struct_description fcd_description = {
+ sizeof (struct file_coding_dump),
+ fcd_description_1
+};
+
+Lisp_Object mule_to_ucs_table;
+
+Lisp_Object Qcoding_systemp;
-Lisp_Object Qno_conversion, Qccl, Qiso2022;
+Lisp_Object Qraw_text, Qno_conversion, Qccl, Qiso2022;
/* Qinternal in general.c */
Lisp_Object Qmnemonic, Qeol_type;
Lisp_Object Qpre_write_conversion;
#ifdef MULE
+Lisp_Object Qucs4, Qutf8;
Lisp_Object Qbig5, Qshift_jis;
Lisp_Object Qcharset_g0, Qcharset_g1, Qcharset_g2, Qcharset_g3;
Lisp_Object Qforce_g0_on_output, Qforce_g1_on_output;
Lisp_Object Qforce_g2_on_output, Qforce_g3_on_output;
Lisp_Object Qno_iso6429;
Lisp_Object Qinput_charset_conversion, Qoutput_charset_conversion;
-Lisp_Object Qctext, Qescape_quoted;
+Lisp_Object Qescape_quoted;
Lisp_Object Qshort, Qno_ascii_eol, Qno_ascii_cntl, Qseven, Qlock_shift;
#endif
Lisp_Object Qencode, Qdecode;
-Lisp_Object Vcoding_system_hashtable;
+Lisp_Object Vcoding_system_hash_table;
int enable_multibyte_characters;
/* Index for next byte to store in ISO escape sequence. */
int esc_bytes_index;
+#ifdef ENABLE_COMPOSITE_CHARS
/* Stuff seen so far when composing a string. */
unsigned_char_dynarr *composite_chars;
+#endif
/* If we saw an invalid designation sequence for a particular
register, we flag it here and switch to ASCII. The next time we
#ifdef MULE
struct detection_state;
static int detect_coding_sjis (struct detection_state *st,
- CONST unsigned char *src,
+ const unsigned char *src,
unsigned int n);
static void decode_coding_sjis (Lstream *decoding,
- CONST unsigned char *src,
+ const unsigned char *src,
unsigned_char_dynarr *dst,
unsigned int n);
static void encode_coding_sjis (Lstream *encoding,
- CONST unsigned char *src,
+ const unsigned char *src,
unsigned_char_dynarr *dst,
unsigned int n);
static int detect_coding_big5 (struct detection_state *st,
- CONST unsigned char *src,
+ const unsigned char *src,
unsigned int n);
static void decode_coding_big5 (Lstream *decoding,
- CONST unsigned char *src,
+ const unsigned char *src,
unsigned_char_dynarr *dst, unsigned int n);
static void encode_coding_big5 (Lstream *encoding,
- CONST unsigned char *src,
+ const unsigned char *src,
+ unsigned_char_dynarr *dst, unsigned int n);
+static int detect_coding_ucs4 (struct detection_state *st,
+ const unsigned char *src,
+ unsigned int n);
+static void decode_coding_ucs4 (Lstream *decoding,
+ const unsigned char *src,
+ unsigned_char_dynarr *dst, unsigned int n);
+static void encode_coding_ucs4 (Lstream *encoding,
+ const unsigned char *src,
+ unsigned_char_dynarr *dst, unsigned int n);
+static int detect_coding_utf8 (struct detection_state *st,
+ const unsigned char *src,
+ unsigned int n);
+static void decode_coding_utf8 (Lstream *decoding,
+ const unsigned char *src,
+ unsigned_char_dynarr *dst, unsigned int n);
+static void encode_coding_utf8 (Lstream *encoding,
+ const unsigned char *src,
unsigned_char_dynarr *dst, unsigned int n);
static int postprocess_iso2022_mask (int mask);
static void reset_iso2022 (Lisp_Object coding_system,
struct iso2022_decoder *iso);
static int detect_coding_iso2022 (struct detection_state *st,
- CONST unsigned char *src,
+ const unsigned char *src,
unsigned int n);
static void decode_coding_iso2022 (Lstream *decoding,
- CONST unsigned char *src,
+ const unsigned char *src,
unsigned_char_dynarr *dst, unsigned int n);
static void encode_coding_iso2022 (Lstream *encoding,
- CONST unsigned char *src,
+ const unsigned char *src,
unsigned_char_dynarr *dst, unsigned int n);
#endif /* MULE */
static void decode_coding_no_conversion (Lstream *decoding,
- CONST unsigned char *src,
+ const unsigned char *src,
unsigned_char_dynarr *dst,
unsigned int n);
static void encode_coding_no_conversion (Lstream *encoding,
- CONST unsigned char *src,
+ const unsigned char *src,
unsigned_char_dynarr *dst,
unsigned int n);
-static void mule_decode (Lstream *decoding, CONST unsigned char *src,
+static void mule_decode (Lstream *decoding, const unsigned char *src,
unsigned_char_dynarr *dst, unsigned int n);
-static void mule_encode (Lstream *encoding, CONST unsigned char *src,
+static void mule_encode (Lstream *encoding, const unsigned char *src,
unsigned_char_dynarr *dst, unsigned int n);
typedef struct codesys_prop codesys_prop;
Dynarr_declare (codesys_prop);
} codesys_prop_dynarr;
+static const struct lrecord_description codesys_prop_description_1[] = {
+ { XD_LISP_OBJECT, offsetof (codesys_prop, sym) },
+ { XD_END }
+};
+
+static const struct struct_description codesys_prop_description = {
+ sizeof (codesys_prop),
+ codesys_prop_description_1
+};
+
+static const struct lrecord_description codesys_prop_dynarr_description_1[] = {
+ XD_DYNARR_DESC (codesys_prop_dynarr, &codesys_prop_description),
+ { XD_END }
+};
+
+static const struct struct_description codesys_prop_dynarr_description = {
+ sizeof (codesys_prop_dynarr),
+ codesys_prop_dynarr_description_1
+};
+
codesys_prop_dynarr *the_codesys_prop_dynarr;
enum codesys_prop_enum
/* Coding system functions */
/************************************************************************/
-static Lisp_Object mark_coding_system (Lisp_Object, void (*) (Lisp_Object));
+static Lisp_Object mark_coding_system (Lisp_Object);
static void print_coding_system (Lisp_Object, Lisp_Object, int);
static void finalize_coding_system (void *header, int for_disksave);
+#ifdef MULE
+static const struct lrecord_description ccs_description_1[] = {
+ { XD_LISP_OBJECT, offsetof (charset_conversion_spec, from_charset) },
+ { XD_LISP_OBJECT, offsetof (charset_conversion_spec, to_charset) },
+ { XD_END }
+};
+
+static const struct struct_description ccs_description = {
+ sizeof (charset_conversion_spec),
+ ccs_description_1
+};
+
+static const struct lrecord_description ccsd_description_1[] = {
+ XD_DYNARR_DESC (charset_conversion_spec_dynarr, &ccs_description),
+ { XD_END }
+};
+
+static const struct struct_description ccsd_description = {
+ sizeof (charset_conversion_spec_dynarr),
+ ccsd_description_1
+};
+#endif
+
+static const struct lrecord_description coding_system_description[] = {
+ { XD_LISP_OBJECT, offsetof (Lisp_Coding_System, name) },
+ { XD_LISP_OBJECT, offsetof (Lisp_Coding_System, doc_string) },
+ { XD_LISP_OBJECT, offsetof (Lisp_Coding_System, mnemonic) },
+ { XD_LISP_OBJECT, offsetof (Lisp_Coding_System, post_read_conversion) },
+ { XD_LISP_OBJECT, offsetof (Lisp_Coding_System, pre_write_conversion) },
+ { XD_LISP_OBJECT, offsetof (Lisp_Coding_System, eol_lf) },
+ { XD_LISP_OBJECT, offsetof (Lisp_Coding_System, eol_crlf) },
+ { XD_LISP_OBJECT, offsetof (Lisp_Coding_System, eol_cr) },
+#ifdef MULE
+ { XD_LISP_OBJECT_ARRAY, offsetof (Lisp_Coding_System, iso2022.initial_charset), 4 },
+ { XD_STRUCT_PTR, offsetof (Lisp_Coding_System, iso2022.input_conv), 1, &ccsd_description },
+ { XD_STRUCT_PTR, offsetof (Lisp_Coding_System, iso2022.output_conv), 1, &ccsd_description },
+ { XD_LISP_OBJECT, offsetof (Lisp_Coding_System, ccl.decode) },
+ { XD_LISP_OBJECT, offsetof (Lisp_Coding_System, ccl.encode) },
+#endif
+ { XD_END }
+};
+
DEFINE_LRECORD_IMPLEMENTATION ("coding-system", coding_system,
mark_coding_system, print_coding_system,
finalize_coding_system,
- 0, 0, struct Lisp_Coding_System);
+ 0, 0, coding_system_description,
+ Lisp_Coding_System);
static Lisp_Object
-mark_coding_system (Lisp_Object obj, void (*markobj) (Lisp_Object))
+mark_coding_system (Lisp_Object obj)
{
- struct Lisp_Coding_System *codesys = XCODING_SYSTEM (obj);
+ Lisp_Coding_System *codesys = XCODING_SYSTEM (obj);
- (markobj) (CODING_SYSTEM_NAME (codesys));
- (markobj) (CODING_SYSTEM_DOC_STRING (codesys));
- (markobj) (CODING_SYSTEM_MNEMONIC (codesys));
- (markobj) (CODING_SYSTEM_EOL_LF (codesys));
- (markobj) (CODING_SYSTEM_EOL_CRLF (codesys));
- (markobj) (CODING_SYSTEM_EOL_CR (codesys));
+ mark_object (CODING_SYSTEM_NAME (codesys));
+ mark_object (CODING_SYSTEM_DOC_STRING (codesys));
+ mark_object (CODING_SYSTEM_MNEMONIC (codesys));
+ mark_object (CODING_SYSTEM_EOL_LF (codesys));
+ mark_object (CODING_SYSTEM_EOL_CRLF (codesys));
+ mark_object (CODING_SYSTEM_EOL_CR (codesys));
switch (CODING_SYSTEM_TYPE (codesys))
{
int i;
case CODESYS_ISO2022:
for (i = 0; i < 4; i++)
- (markobj) (CODING_SYSTEM_ISO2022_INITIAL_CHARSET (codesys, i));
+ mark_object (CODING_SYSTEM_ISO2022_INITIAL_CHARSET (codesys, i));
if (codesys->iso2022.input_conv)
{
for (i = 0; i < Dynarr_length (codesys->iso2022.input_conv); i++)
{
struct charset_conversion_spec *ccs =
Dynarr_atp (codesys->iso2022.input_conv, i);
- (markobj) (ccs->from_charset);
- (markobj) (ccs->to_charset);
+ mark_object (ccs->from_charset);
+ mark_object (ccs->to_charset);
}
}
if (codesys->iso2022.output_conv)
{
struct charset_conversion_spec *ccs =
Dynarr_atp (codesys->iso2022.output_conv, i);
- (markobj) (ccs->from_charset);
- (markobj) (ccs->to_charset);
+ mark_object (ccs->from_charset);
+ mark_object (ccs->to_charset);
}
}
break;
case CODESYS_CCL:
- (markobj) (CODING_SYSTEM_CCL_DECODE (codesys));
- (markobj) (CODING_SYSTEM_CCL_ENCODE (codesys));
+ mark_object (CODING_SYSTEM_CCL_DECODE (codesys));
+ mark_object (CODING_SYSTEM_CCL_ENCODE (codesys));
break;
#endif /* MULE */
default:
break;
}
- (markobj) (CODING_SYSTEM_PRE_WRITE_CONVERSION (codesys));
+ mark_object (CODING_SYSTEM_PRE_WRITE_CONVERSION (codesys));
return CODING_SYSTEM_POST_READ_CONVERSION (codesys);
}
print_coding_system (Lisp_Object obj, Lisp_Object printcharfun,
int escapeflag)
{
- struct Lisp_Coding_System *c = XCODING_SYSTEM (obj);
+ Lisp_Coding_System *c = XCODING_SYSTEM (obj);
if (print_readably)
error ("printing unreadable object #<coding_system 0x%x>",
c->header.uid);
static void
finalize_coding_system (void *header, int for_disksave)
{
- struct Lisp_Coding_System *c = (struct Lisp_Coding_System *) header;
+ Lisp_Coding_System *c = (Lisp_Coding_System *) header;
/* Since coding systems never go away, this function is not
necessary. But it would be necessary if we changed things
so that coding systems could go away. */
}
}
-static enum eol_type
+static eol_type_t
symbol_to_eol_type (Lisp_Object symbol)
{
CHECK_SYMBOL (symbol);
}
static Lisp_Object
-eol_type_to_symbol (enum eol_type type)
+eol_type_to_symbol (eol_type_t type)
{
switch (type)
{
+ default: abort ();
case EOL_LF: return Qlf;
case EOL_CRLF: return Qcrlf;
case EOL_CR: return Qcr;
case EOL_AUTODETECT: return Qnil;
- default: abort (); return Qnil; /* not reached */
}
}
static void
-setup_eol_coding_systems (struct Lisp_Coding_System *codesys)
+setup_eol_coding_systems (Lisp_Coding_System *codesys)
{
Lisp_Object codesys_obj;
int len = string_length (XSYMBOL (CODING_SYSTEM_NAME (codesys))->name);
*/
(coding_system_or_name))
{
- if (CODING_SYSTEMP (coding_system_or_name))
- return coding_system_or_name;
-
if (NILP (coding_system_or_name))
coding_system_or_name = Qbinary;
+ else if (CODING_SYSTEMP (coding_system_or_name))
+ return coding_system_or_name;
else
CHECK_SYMBOL (coding_system_or_name);
- return Fgethash (coding_system_or_name, Vcoding_system_hashtable, Qnil);
+ while (1)
+ {
+ coding_system_or_name =
+ Fgethash (coding_system_or_name, Vcoding_system_hash_table, Qnil);
+
+ if (CODING_SYSTEMP (coding_system_or_name) || NILP (coding_system_or_name))
+ return coding_system_or_name;
+ }
}
DEFUN ("get-coding-system", Fget_coding_system, 1, 1, 0, /*
};
static int
-add_coding_system_to_list_mapper (CONST void *hash_key, void *hash_contents,
+add_coding_system_to_list_mapper (Lisp_Object key, Lisp_Object value,
void *coding_system_list_closure)
{
/* This function can GC */
- Lisp_Object key, contents;
- Lisp_Object *coding_system_list;
struct coding_system_list_closure *cscl =
(struct coding_system_list_closure *) coding_system_list_closure;
- CVOID_TO_LISP (key, hash_key);
- VOID_TO_LISP (contents, hash_contents);
- coding_system_list = cscl->coding_system_list;
+ Lisp_Object *coding_system_list = cscl->coding_system_list;
- *coding_system_list = Fcons (XCODING_SYSTEM (contents)->name,
- *coding_system_list);
+ *coding_system_list = Fcons (key, *coding_system_list);
return 0;
}
GCPRO1 (coding_system_list);
coding_system_list_closure.coding_system_list = &coding_system_list;
- elisp_maphash (add_coding_system_to_list_mapper, Vcoding_system_hashtable,
+ elisp_maphash (add_coding_system_to_list_mapper, Vcoding_system_hash_table,
&coding_system_list_closure);
UNGCPRO;
return XCODING_SYSTEM_NAME (coding_system);
}
-static struct Lisp_Coding_System *
+static Lisp_Coding_System *
allocate_coding_system (enum coding_system_type type, Lisp_Object name)
{
- struct Lisp_Coding_System *codesys =
- alloc_lcrecord_type (struct Lisp_Coding_System, lrecord_coding_system);
+ Lisp_Coding_System *codesys =
+ alloc_lcrecord_type (Lisp_Coding_System, &lrecord_coding_system);
zero_lcrecord (codesys);
CODING_SYSTEM_PRE_WRITE_CONVERSION (codesys) = Qnil;
characters will only be present if you explicitly insert them.)
'shift-jis
Shift-JIS (a Japanese encoding commonly used in PC operating systems).
+'ucs-4
+ ISO 10646 UCS-4 encoding.
+'utf-8
+ ISO 10646 UTF-8 encoding.
'iso2022
Any ISO2022-compliant encoding. Among other things, this includes
JIS (the Japanese encoding commonly used for e-mail), EUC (the
*/
(name, type, doc_string, props))
{
- struct Lisp_Coding_System *codesys;
+ Lisp_Coding_System *codesys;
Lisp_Object rest, key, value;
enum coding_system_type ty;
int need_to_setup_eol_systems = 1;
else if (EQ (type, Qshift_jis)) { ty = CODESYS_SHIFT_JIS; }
else if (EQ (type, Qiso2022)) { ty = CODESYS_ISO2022; }
else if (EQ (type, Qbig5)) { ty = CODESYS_BIG5; }
+ else if (EQ (type, Qucs4)) { ty = CODESYS_UCS4; }
+ else if (EQ (type, Qutf8)) { ty = CODESYS_UTF8; }
else if (EQ (type, Qccl)) { ty = CODESYS_CCL; }
#endif
else if (EQ (type, Qno_conversion)) { ty = CODESYS_NO_CONVERSION; }
{
Lisp_Object codesys_obj;
XSETCODING_SYSTEM (codesys_obj, codesys);
- Fputhash (name, codesys_obj, Vcoding_system_hashtable);
+ Fputhash (name, codesys_obj, Vcoding_system_hash_table);
return codesys_obj;
}
}
allocate_coding_system
(XCODING_SYSTEM_TYPE (old_coding_system),
new_name));
- Fputhash (new_name, new_coding_system, Vcoding_system_hashtable);
+ Fputhash (new_name, new_coding_system, Vcoding_system_hash_table);
}
{
- struct Lisp_Coding_System *to = XCODING_SYSTEM (new_coding_system);
- struct Lisp_Coding_System *from = XCODING_SYSTEM (old_coding_system);
+ Lisp_Coding_System *to = XCODING_SYSTEM (new_coding_system);
+ Lisp_Coding_System *from = XCODING_SYSTEM (old_coding_system);
memcpy (((char *) to ) + sizeof (to->header),
((char *) from) + sizeof (from->header),
sizeof (*from) - sizeof (from->header));
return new_coding_system;
}
+DEFUN ("coding-system-canonical-name-p", Fcoding_system_canonical_name_p, 1, 1, 0, /*
+Return t if OBJECT names a coding system, and is not a coding system alias.
+*/
+ (object))
+{
+ return CODING_SYSTEMP (Fgethash (object, Vcoding_system_hash_table, Qnil))
+ ? Qt : Qnil;
+}
+
+DEFUN ("coding-system-alias-p", Fcoding_system_alias_p, 1, 1, 0, /*
+Return t if OBJECT is a coding system alias.
+All coding system aliases are created by `define-coding-system-alias'.
+*/
+ (object))
+{
+ return SYMBOLP (Fgethash (object, Vcoding_system_hash_table, Qzero))
+ ? Qt : Qnil;
+}
+
+DEFUN ("coding-system-aliasee", Fcoding_system_aliasee, 1, 1, 0, /*
+Return the coding-system symbol for which symbol ALIAS is an alias.
+*/
+ (alias))
+{
+ Lisp_Object aliasee = Fgethash (alias, Vcoding_system_hash_table, Qnil);
+ if (SYMBOLP (aliasee))
+ return aliasee;
+ else
+ signal_simple_error ("Symbol is not a coding system alias", alias);
+}
+
+static Lisp_Object
+append_suffix_to_symbol (Lisp_Object symbol, const char *ascii_string)
+{
+ return Fintern (concat2 (Fsymbol_name (symbol), build_string (ascii_string)),
+ Qnil);
+}
+
+/* A maphash function, for removing dangling coding system aliases. */
+static int
+dangling_coding_system_alias_p (Lisp_Object alias,
+ Lisp_Object aliasee,
+ void *dangling_aliases)
+{
+ if (SYMBOLP (aliasee)
+ && NILP (Fgethash (aliasee, Vcoding_system_hash_table, Qnil)))
+ {
+ (*(int *) dangling_aliases)++;
+ return 1;
+ }
+ else
+ return 0;
+}
+
+DEFUN ("define-coding-system-alias", Fdefine_coding_system_alias, 2, 2, 0, /*
+Define symbol ALIAS as an alias for coding system ALIASEE.
+
+You can use this function to redefine an alias that has already been defined,
+but you cannot redefine a name which is the canonical name for a coding system.
+\(a canonical name of a coding system is what is returned when you call
+`coding-system-name' on a coding system).
+
+ALIASEE itself can be an alias, which allows you to define nested aliases.
+
+You are forbidden, however, from creating alias loops or `dangling' aliases.
+These will be detected, and an error will be signaled if you attempt to do so.
+
+If ALIASEE is nil, then ALIAS will simply be undefined.
+
+See also `coding-system-alias-p', `coding-system-aliasee',
+and `coding-system-canonical-name-p'.
+*/
+ (alias, aliasee))
+{
+ Lisp_Object real_coding_system, probe;
+
+ CHECK_SYMBOL (alias);
+
+ if (!NILP (Fcoding_system_canonical_name_p (alias)))
+ signal_simple_error
+ ("Symbol is the canonical name of a coding system and cannot be redefined",
+ alias);
+
+ if (NILP (aliasee))
+ {
+ Lisp_Object subsidiary_unix = append_suffix_to_symbol (alias, "-unix");
+ Lisp_Object subsidiary_dos = append_suffix_to_symbol (alias, "-dos");
+ Lisp_Object subsidiary_mac = append_suffix_to_symbol (alias, "-mac");
+
+ Fremhash (alias, Vcoding_system_hash_table);
+
+ /* Undefine subsidiary aliases,
+ presumably created by a previous call to this function */
+ if (! NILP (Fcoding_system_alias_p (subsidiary_unix)) &&
+ ! NILP (Fcoding_system_alias_p (subsidiary_dos)) &&
+ ! NILP (Fcoding_system_alias_p (subsidiary_mac)))
+ {
+ Fdefine_coding_system_alias (subsidiary_unix, Qnil);
+ Fdefine_coding_system_alias (subsidiary_dos, Qnil);
+ Fdefine_coding_system_alias (subsidiary_mac, Qnil);
+ }
+
+ /* Undefine dangling coding system aliases. */
+ {
+ int dangling_aliases;
+
+ do {
+ dangling_aliases = 0;
+ elisp_map_remhash (dangling_coding_system_alias_p,
+ Vcoding_system_hash_table,
+ &dangling_aliases);
+ } while (dangling_aliases > 0);
+ }
+
+ return Qnil;
+ }
+
+ if (CODING_SYSTEMP (aliasee))
+ aliasee = XCODING_SYSTEM_NAME (aliasee);
+
+ /* Checks that aliasee names a coding-system */
+ real_coding_system = Fget_coding_system (aliasee);
+
+ /* Check for coding system alias loops */
+ if (EQ (alias, aliasee))
+ alias_loop: signal_simple_error_2
+ ("Attempt to create a coding system alias loop", alias, aliasee);
+
+ for (probe = aliasee;
+ SYMBOLP (probe);
+ probe = Fgethash (probe, Vcoding_system_hash_table, Qzero))
+ {
+ if (EQ (probe, alias))
+ goto alias_loop;
+ }
+
+ Fputhash (alias, aliasee, Vcoding_system_hash_table);
+
+ /* Set up aliases for subsidiaries.
+ #### There must be a better way to handle subsidiary coding systems. */
+ {
+ static const char *suffixes[] = { "-unix", "-dos", "-mac" };
+ int i;
+ for (i = 0; i < countof (suffixes); i++)
+ {
+ Lisp_Object alias_subsidiary =
+ append_suffix_to_symbol (alias, suffixes[i]);
+ Lisp_Object aliasee_subsidiary =
+ append_suffix_to_symbol (aliasee, suffixes[i]);
+
+ if (! NILP (Ffind_coding_system (aliasee_subsidiary)))
+ Fdefine_coding_system_alias (alias_subsidiary, aliasee_subsidiary);
+ }
+ }
+ /* FSF return value is a vector of [ALIAS-unix ALIAS-dos ALIAS-mac],
+ but it doesn't look intentional, so I'd rather return something
+ meaningful or nothing at all. */
+ return Qnil;
+}
+
static Lisp_Object
-subsidiary_coding_system (Lisp_Object coding_system, enum eol_type type)
+subsidiary_coding_system (Lisp_Object coding_system, eol_type_t type)
{
- struct Lisp_Coding_System *cs = XCODING_SYSTEM (coding_system);
+ Lisp_Coding_System *cs = XCODING_SYSTEM (coding_system);
Lisp_Object new_coding_system;
if (CODING_SYSTEM_EOL_TYPE (cs) != EOL_AUTODETECT)
{
switch (XCODING_SYSTEM_TYPE (Fget_coding_system (coding_system)))
{
+ default: abort ();
case CODESYS_AUTODETECT: return Qundecided;
#ifdef MULE
case CODESYS_SHIFT_JIS: return Qshift_jis;
case CODESYS_ISO2022: return Qiso2022;
case CODESYS_BIG5: return Qbig5;
+ case CODESYS_UCS4: return Qucs4;
+ case CODESYS_UTF8: return Qutf8;
case CODESYS_CCL: return Qccl;
#endif
case CODESYS_NO_CONVERSION: return Qno_conversion;
#ifdef DEBUG_XEMACS
case CODESYS_INTERNAL: return Qinternal;
#endif
- default:
- abort ();
}
-
- return Qnil; /* not reached */
}
#ifdef MULE
order. */
for (j = 0; j <= CODING_CATEGORY_LAST; j++)
{
- int cat = coding_category_by_priority[j];
+ int cat = fcd->coding_category_by_priority[j];
if (category_to_priority[cat] < 0)
category_to_priority[cat] = i++;
}
constructed. */
for (i = 0; i <= CODING_CATEGORY_LAST; i++)
- coding_category_by_priority[category_to_priority[i]] = i;
+ fcd->coding_category_by_priority[category_to_priority[i]] = i;
/* Phew! That was confusing. */
return Qnil;
Lisp_Object list = Qnil;
for (i = CODING_CATEGORY_LAST; i >= 0; i--)
- list = Fcons (coding_category_symbol[coding_category_by_priority[i]],
+ list = Fcons (coding_category_symbol[fcd->coding_category_by_priority[i]],
list);
return list;
}
int cat = decode_coding_category (coding_category);
coding_system = Fget_coding_system (coding_system);
- coding_category_system[cat] = coding_system;
+ fcd->coding_category_system[cat] = coding_system;
return Qnil;
}
(coding_category))
{
int cat = decode_coding_category (coding_category);
- Lisp_Object sys = coding_category_system[cat];
+ Lisp_Object sys = fcd->coding_category_system[cat];
if (!NILP (sys))
return XCODING_SYSTEM_NAME (sys);
struct detection_state
{
- enum eol_type eol_type;
+ eol_type_t eol_type;
int seen_non_ascii;
int mask;
#ifdef MULE
struct
{
int mask;
+ int in_byte;
+ }
+ ucs4;
+
+ struct
+ {
+ int mask;
+ int in_byte;
+ }
+ utf8;
+
+ struct
+ {
+ int mask;
int initted;
struct iso2022_decoder iso;
unsigned int flags;
return (mask & (mask - 1)) == 0;
}
-static enum eol_type
-detect_eol_type (struct detection_state *st, CONST unsigned char *src,
+static eol_type_t
+detect_eol_type (struct detection_state *st, const unsigned char *src,
unsigned int n)
{
int c;
while (n--)
{
c = *src++;
- if (c == '\r')
- st->eol.just_saw_cr = 1;
- else
+ if (c == '\n')
{
- if (c == '\n')
- {
- if (st->eol.just_saw_cr)
- return EOL_CRLF;
- else if (st->eol.seen_anything)
- return EOL_LF;
- }
- else if (st->eol.just_saw_cr)
- return EOL_CR;
- st->eol.just_saw_cr = 0;
+ if (st->eol.just_saw_cr)
+ return EOL_CRLF;
+ else if (st->eol.seen_anything)
+ return EOL_LF;
}
+ else if (st->eol.just_saw_cr)
+ return EOL_CR;
+ else if (c == '\r')
+ st->eol.just_saw_cr = 1;
+ else
+ st->eol.just_saw_cr = 0;
st->eol.seen_anything = 1;
}
*/
static int
-detect_coding_type (struct detection_state *st, CONST unsigned char *src,
+detect_coding_type (struct detection_state *st, const Extbyte *src,
unsigned int n, int just_do_eol)
{
int c;
#ifdef MULE
st->shift_jis.mask = ~0;
st->big5.mask = ~0;
+ st->ucs4.mask = ~0;
+ st->utf8.mask = ~0;
st->iso2022.mask = ~0;
#endif
break;
st->shift_jis.mask = detect_coding_sjis (st, src, n);
if (!mask_has_at_most_one_bit_p (st->big5.mask))
st->big5.mask = detect_coding_big5 (st, src, n);
-
- st->mask = st->iso2022.mask | st->shift_jis.mask | st->big5.mask;
+ if (!mask_has_at_most_one_bit_p (st->utf8.mask))
+ st->utf8.mask = detect_coding_utf8 (st, src, n);
+ if (!mask_has_at_most_one_bit_p (st->ucs4.mask))
+ st->ucs4.mask = detect_coding_ucs4 (st, src, n);
+
+ st->mask
+ = st->iso2022.mask | st->shift_jis.mask | st->big5.mask
+ | st->utf8.mask | st->ucs4.mask;
#endif
{
int retval = mask_has_at_most_one_bit_p (st->mask);
}
}
if (NILP (retval))
- retval = Fget_coding_system (Qno_conversion);
+ retval = Fget_coding_system (Qraw_text);
return retval;
}
else
the first one that is allowed. */
for (i = 0; i <= CODING_CATEGORY_LAST; i++)
{
- cat = coding_category_by_priority[i];
+ cat = fcd->coding_category_by_priority[i];
if ((mask & (1 << cat)) &&
- !NILP (coding_category_system[cat]))
+ !NILP (fcd->coding_category_system[cat]))
break;
}
if (cat >= 0)
- return coding_category_system[cat];
+ return fcd->coding_category_system[cat];
else
- return Fget_coding_system (Qno_conversion);
+ return Fget_coding_system (Qraw_text);
}
}
/* Given a seekable read stream and potential coding system and EOL type
as specified, do any autodetection that is called for. If the
- coding system and/or EOL type are not autodetect, they will be left
+ coding system and/or EOL type are not `autodetect', they will be left
alone; but this function will never return an autodetect coding system
or EOL type.
This function does not automatically fetch subsidiary coding systems;
that should be unnecessary with the explicit eol-type argument. */
+#define LENGTH(string_constant) (sizeof (string_constant) - 1)
+
void
determine_real_coding_system (Lstream *stream, Lisp_Object *codesys_in_out,
- enum eol_type *eol_type_in_out)
+ eol_type_t *eol_type_in_out)
{
struct detection_state decst;
decst.mask = ~0;
/* If autodetection is called for, do it now. */
- if (XCODING_SYSTEM_TYPE (*codesys_in_out) == CODESYS_AUTODETECT ||
- *eol_type_in_out == EOL_AUTODETECT)
+ if (XCODING_SYSTEM_TYPE (*codesys_in_out) == CODESYS_AUTODETECT
+ || *eol_type_in_out == EOL_AUTODETECT)
{
+ Extbyte buf[4096];
+ Lisp_Object coding_system = Qnil;
+ Extbyte *p;
+ ssize_t nread = Lstream_read (stream, buf, sizeof (buf));
+ Extbyte *scan_end;
+
+ /* Look for initial "-*-"; mode line prefix */
+ for (p = buf,
+ scan_end = buf + nread - LENGTH ("-*-coding:?-*-");
+ p <= scan_end
+ && *p != '\n'
+ && *p != '\r';
+ p++)
+ if (*p == '-' && *(p+1) == '*' && *(p+2) == '-')
+ {
+ Extbyte *local_vars_beg = p + 3;
+ /* Look for final "-*-"; mode line suffix */
+ for (p = local_vars_beg,
+ scan_end = buf + nread - LENGTH ("-*-");
+ p <= scan_end
+ && *p != '\n'
+ && *p != '\r';
+ p++)
+ if (*p == '-' && *(p+1) == '*' && *(p+2) == '-')
+ {
+ Extbyte *suffix = p;
+ /* Look for "coding:" */
+ for (p = local_vars_beg,
+ scan_end = suffix - LENGTH ("coding:?");
+ p <= scan_end;
+ p++)
+ if (memcmp ("coding:", p, LENGTH ("coding:")) == 0
+ && (p == local_vars_beg
+ || (*(p-1) == ' ' ||
+ *(p-1) == '\t' ||
+ *(p-1) == ';')))
+ {
+ Extbyte save;
+ int n;
+ p += LENGTH ("coding:");
+ while (*p == ' ' || *p == '\t') p++;
+
+ /* Get coding system name */
+ save = *suffix; *suffix = '\0';
+ /* Characters valid in a MIME charset name (rfc 1521),
+ and in a Lisp symbol name. */
+ n = strspn ( (char *) p,
+ "ABCDEFGHIJKLMNOPQRSTUVWXYZ"
+ "abcdefghijklmnopqrstuvwxyz"
+ "0123456789"
+ "!$%&*+-.^_{|}~");
+ *suffix = save;
+ if (n > 0)
+ {
+ save = p[n]; p[n] = '\0';
+ coding_system =
+ Ffind_coding_system (intern ((char *) p));
+ p[n] = save;
+ }
+ break;
+ }
+ break;
+ }
+ break;
+ }
- while (1)
- {
- unsigned char random_buffer[4096];
- int nread;
+ if (NILP (coding_system))
+ do
+ {
+ if (detect_coding_type (&decst, buf, nread,
+ XCODING_SYSTEM_TYPE (*codesys_in_out)
+ != CODESYS_AUTODETECT))
+ break;
+ nread = Lstream_read (stream, buf, sizeof (buf));
+ if (nread == 0)
+ break;
+ }
+ while (1);
- nread = Lstream_read (stream, random_buffer, sizeof (random_buffer));
- if (!nread)
- break;
- if (detect_coding_type (&decst, random_buffer, nread,
- XCODING_SYSTEM_TYPE (*codesys_in_out) !=
- CODESYS_AUTODETECT))
- break;
- }
+ else if (XCODING_SYSTEM_TYPE (*codesys_in_out) == CODESYS_AUTODETECT
+ && XCODING_SYSTEM_EOL_TYPE (coding_system) == EOL_AUTODETECT)
+ do
+ {
+ if (detect_coding_type (&decst, buf, nread, 1))
+ break;
+ nread = Lstream_read (stream, buf, sizeof (buf));
+ if (!nread)
+ break;
+ }
+ while (1);
*eol_type_in_out = decst.eol_type;
if (XCODING_SYSTEM_TYPE (*codesys_in_out) == CODESYS_AUTODETECT)
- *codesys_in_out = coding_system_from_mask (decst.mask);
+ {
+ if (NILP (coding_system))
+ *codesys_in_out = coding_system_from_mask (decst.mask);
+ else
+ *codesys_in_out = coding_system;
+ }
}
/* If we absolutely can't determine the EOL type, just assume LF. */
while (1)
{
unsigned char random_buffer[4096];
- int nread = Lstream_read (istr, random_buffer, sizeof (random_buffer));
+ ssize_t nread = Lstream_read (istr, random_buffer, sizeof (random_buffer));
if (!nread)
break;
#endif
for (i = CODING_CATEGORY_LAST; i >= 0; i--)
{
- int sys = coding_category_by_priority[i];
+ int sys = fcd->coding_category_by_priority[i];
if (decst.mask & (1 << sys))
{
- Lisp_Object codesys = coding_category_system[sys];
+ Lisp_Object codesys = fcd->coding_category_system[sys];
if (!NILP (codesys))
codesys = subsidiary_coding_system (codesys, decst.eol_type);
val = Fcons (codesys, val);
#define DECODE_HANDLE_END_OF_CONVERSION(flags, ch, dst) \
do { \
- DECODE_OUTPUT_PARTIAL_CHAR (ch); \
- if ((flags & CODING_STATE_END) && \
- (flags & CODING_STATE_CR)) \
- Dynarr_add (dst, '\r'); \
+ if (flags & CODING_STATE_END) \
+ { \
+ DECODE_OUTPUT_PARTIAL_CHAR (ch); \
+ if (flags & CODING_STATE_CR) \
+ Dynarr_add (dst, '\r'); \
+ } \
} while (0)
#define DECODING_STREAM_DATA(stream) LSTREAM_TYPE_DATA (stream, decoding)
struct decoding_stream
{
/* Coding system that governs the conversion. */
- struct Lisp_Coding_System *codesys;
+ Lisp_Coding_System *codesys;
/* Stream that we read the encoded data from or
write the decoded data to. */
EOL type stored in CODESYS because the latter might indicate
automatic EOL-type detection while the former will always
indicate a particular EOL type. */
- enum eol_type eol_type;
+ eol_type_t eol_type;
#ifdef MULE
/* Additional ISO2022 information. We define the structure above
because it's also needed by the detection routines. */
/* Additional information (the state of the running CCL program)
used by the CCL decoder. */
struct ccl_program ccl;
+
+ /* counter for UTF-8 or UCS-4 */
+ unsigned char counter;
#endif
struct detection_state decst;
};
-static int decoding_reader (Lstream *stream, unsigned char *data, size_t size);
-static int decoding_writer (Lstream *stream, CONST unsigned char *data, size_t size);
+static ssize_t decoding_reader (Lstream *stream,
+ unsigned char *data, size_t size);
+static ssize_t decoding_writer (Lstream *stream,
+ const unsigned char *data, size_t size);
static int decoding_rewinder (Lstream *stream);
static int decoding_seekable_p (Lstream *stream);
static int decoding_flusher (Lstream *stream);
static int decoding_closer (Lstream *stream);
-static Lisp_Object decoding_marker (Lisp_Object stream,
- void (*markobj) (Lisp_Object));
+static Lisp_Object decoding_marker (Lisp_Object stream);
DEFINE_LSTREAM_IMPLEMENTATION ("decoding", lstream_decoding,
sizeof (struct decoding_stream));
static Lisp_Object
-decoding_marker (Lisp_Object stream, void (*markobj) (Lisp_Object))
+decoding_marker (Lisp_Object stream)
{
Lstream *str = DECODING_STREAM_DATA (XLSTREAM (stream))->other_end;
Lisp_Object str_obj;
and automatically marked. */
XSETLSTREAM (str_obj, str);
- (markobj) (str_obj);
+ mark_object (str_obj);
if (str->imp->marker)
- return (str->imp->marker) (str_obj, markobj);
+ return (str->imp->marker) (str_obj);
else
return Qnil;
}
/* Read SIZE bytes of data and store it into DATA. We are a decoding stream
so we read data from the other end, decode it, and store it into DATA. */
-static int
+static ssize_t
decoding_reader (Lstream *stream, unsigned char *data, size_t size)
{
struct decoding_stream *str = DECODING_STREAM_DATA (stream);
unsigned char *orig_data = data;
- int read_size;
+ ssize_t read_size;
int error_occurred = 0;
/* We need to interface to mule_decode(), which expects to take some
return data - orig_data;
}
-static int
-decoding_writer (Lstream *stream, CONST unsigned char *data, size_t size)
+static ssize_t
+decoding_writer (Lstream *stream, const unsigned char *data, size_t size)
{
struct decoding_stream *str = DECODING_STREAM_DATA (stream);
- int retval;
+ ssize_t retval;
/* Decode all our data into the runoff, and then attempt to write
it all out to the other end. Remove whatever chunk we succeeded
{
setup_ccl_program (&str->ccl, CODING_SYSTEM_CCL_DECODE (str->codesys));
}
+ str->counter = 0;
#endif /* MULE */
str->flags = str->ch = 0;
}
}
Dynarr_free (str->runoff);
#ifdef MULE
+#ifdef ENABLE_COMPOSITE_CHARS
if (str->iso2022.composite_chars)
Dynarr_free (str->iso2022.composite_chars);
#endif
+#endif
return Lstream_close (str->other_end);
}
void
set_decoding_stream_coding_system (Lstream *lstr, Lisp_Object codesys)
{
- struct Lisp_Coding_System *cs = XCODING_SYSTEM (codesys);
+ Lisp_Coding_System *cs = XCODING_SYSTEM (codesys);
struct decoding_stream *str = DECODING_STREAM_DATA (lstr);
str->codesys = cs;
if (CODING_SYSTEM_EOL_TYPE (cs) != EOL_AUTODETECT)
static Lisp_Object
make_decoding_stream_1 (Lstream *stream, Lisp_Object codesys,
- CONST char *mode)
+ const char *mode)
{
Lstream *lstr = Lstream_new (lstream_decoding, mode);
struct decoding_stream *str = DECODING_STREAM_DATA (lstr);
be used for both reading and writing. */
static void
-mule_decode (Lstream *decoding, CONST unsigned char *src,
+mule_decode (Lstream *decoding, const unsigned char *src,
unsigned_char_dynarr *dst, unsigned int n)
{
struct decoding_stream *str = DECODING_STREAM_DATA (decoding);
case CODESYS_BIG5:
decode_coding_big5 (decoding, src, dst, n);
break;
+ case CODESYS_UCS4:
+ decode_coding_ucs4 (decoding, src, dst, n);
+ break;
+ case CODESYS_UTF8:
+ decode_coding_utf8 (decoding, src, dst, n);
+ break;
case CODESYS_CCL:
- ccl_driver (&str->ccl, src, dst, n, 0);
+ str->ccl.last_block = str->flags & CODING_STATE_END;
+ ccl_driver (&str->ccl, src, dst, n, 0, CCL_MODE_DECODING);
break;
case CODESYS_ISO2022:
decode_coding_iso2022 (decoding, src, dst, n);
char tempbuf[1024]; /* some random amount */
Bufpos newpos, even_newer_pos;
Bufpos oldpos = lisp_buffer_stream_startpos (istr);
- int size_in_bytes = Lstream_read (istr, tempbuf, sizeof (tempbuf));
+ ssize_t size_in_bytes = Lstream_read (istr, tempbuf, sizeof (tempbuf));
if (!size_in_bytes)
break;
struct encoding_stream
{
/* Coding system that governs the conversion. */
- struct Lisp_Coding_System *codesys;
+ Lisp_Coding_System *codesys;
/* Stream that we read the encoded data from or
write the decoded data to. */
#endif /* MULE */
};
-static int encoding_reader (Lstream *stream, unsigned char *data, size_t size);
-static int encoding_writer (Lstream *stream, CONST unsigned char *data,
- size_t size);
+static ssize_t encoding_reader (Lstream *stream, unsigned char *data, size_t size);
+static ssize_t encoding_writer (Lstream *stream, const unsigned char *data,
+ size_t size);
static int encoding_rewinder (Lstream *stream);
static int encoding_seekable_p (Lstream *stream);
static int encoding_flusher (Lstream *stream);
static int encoding_closer (Lstream *stream);
-static Lisp_Object encoding_marker (Lisp_Object stream,
- void (*markobj) (Lisp_Object));
+static Lisp_Object encoding_marker (Lisp_Object stream);
DEFINE_LSTREAM_IMPLEMENTATION ("encoding", lstream_encoding,
sizeof (struct encoding_stream));
static Lisp_Object
-encoding_marker (Lisp_Object stream, void (*markobj) (Lisp_Object))
+encoding_marker (Lisp_Object stream)
{
Lstream *str = ENCODING_STREAM_DATA (XLSTREAM (stream))->other_end;
Lisp_Object str_obj;
and automatically marked. */
XSETLSTREAM (str_obj, str);
- (markobj) (str_obj);
+ mark_object (str_obj);
if (str->imp->marker)
- return (str->imp->marker) (str_obj, markobj);
+ return (str->imp->marker) (str_obj);
else
return Qnil;
}
/* Read SIZE bytes of data and store it into DATA. We are a encoding stream
so we read data from the other end, encode it, and store it into DATA. */
-static int
+static ssize_t
encoding_reader (Lstream *stream, unsigned char *data, size_t size)
{
struct encoding_stream *str = ENCODING_STREAM_DATA (stream);
unsigned char *orig_data = data;
- int read_size;
+ ssize_t read_size;
int error_occurred = 0;
/* We need to interface to mule_encode(), which expects to take some
return data - orig_data;
}
-static int
-encoding_writer (Lstream *stream, CONST unsigned char *data, size_t size)
+static ssize_t
+encoding_writer (Lstream *stream, const unsigned char *data, size_t size)
{
struct encoding_stream *str = ENCODING_STREAM_DATA (stream);
- int retval;
+ ssize_t retval;
/* Encode all our data into the runoff, and then attempt to write
it all out to the other end. Remove whatever chunk we succeeded
void
set_encoding_stream_coding_system (Lstream *lstr, Lisp_Object codesys)
{
- struct Lisp_Coding_System *cs = XCODING_SYSTEM (codesys);
+ Lisp_Coding_System *cs = XCODING_SYSTEM (codesys);
struct encoding_stream *str = ENCODING_STREAM_DATA (lstr);
str->codesys = cs;
reset_encoding_stream (str);
static Lisp_Object
make_encoding_stream_1 (Lstream *stream, Lisp_Object codesys,
- CONST char *mode)
+ const char *mode)
{
Lstream *lstr = Lstream_new (lstream_encoding, mode);
struct encoding_stream *str = ENCODING_STREAM_DATA (lstr);
Store the encoded data into DST. */
static void
-mule_encode (Lstream *encoding, CONST unsigned char *src,
+mule_encode (Lstream *encoding, const unsigned char *src,
unsigned_char_dynarr *dst, unsigned int n)
{
struct encoding_stream *str = ENCODING_STREAM_DATA (encoding);
case CODESYS_BIG5:
encode_coding_big5 (encoding, src, dst, n);
break;
+ case CODESYS_UCS4:
+ encode_coding_ucs4 (encoding, src, dst, n);
+ break;
+ case CODESYS_UTF8:
+ encode_coding_utf8 (encoding, src, dst, n);
+ break;
case CODESYS_CCL:
- ccl_driver (&str->ccl, src, dst, n, 0);
+ str->ccl.last_block = str->flags & CODING_STATE_END;
+ ccl_driver (&str->ccl, src, dst, n, 0, CCL_MODE_ENCODING);
break;
case CODESYS_ISO2022:
encode_coding_iso2022 (encoding, src, dst, n);
char tempbuf[1024]; /* some random amount */
Bufpos newpos, even_newer_pos;
Bufpos oldpos = lisp_buffer_stream_startpos (istr);
- int size_in_bytes = Lstream_read (istr, tempbuf, sizeof (tempbuf));
+ ssize_t size_in_bytes = Lstream_read (istr, tempbuf, sizeof (tempbuf));
if (!size_in_bytes)
break;
/* Shift-JIS is a coding system encoding three character sets: ASCII, right
half of JISX0201-Kana, and JISX0208. An ASCII character is encoded
- as is. A character of JISX0201-Kana (TYPE94 character set) is
+ as is. A character of JISX0201-Kana (DIMENSION1_CHARS94 character set) is
encoded by "position-code + 0x80". A character of JISX0208
- (TYPE94x94 character set) is encoded in 2-byte but two
+ (DIMENSION2_CHARS94 character set) is encoded in 2-byte but two
position-codes are divided and shifted so that it fit in the range
below.
((c) >= 0xA1 && (c) <= 0xDF)
static int
-detect_coding_sjis (struct detection_state *st, CONST unsigned char *src,
+detect_coding_sjis (struct detection_state *st, const unsigned char *src,
unsigned int n)
{
int c;
/* Convert Shift-JIS data to internal format. */
static void
-decode_coding_sjis (Lstream *decoding, CONST unsigned char *src,
+decode_coding_sjis (Lstream *decoding, const unsigned char *src,
unsigned_char_dynarr *dst, unsigned int n)
{
unsigned char c;
- unsigned int flags, ch;
- enum eol_type eol_type;
struct decoding_stream *str = DECODING_STREAM_DATA (decoding);
-
- CODING_STREAM_DECOMPOSE (str, flags, ch);
- eol_type = str->eol_type;
+ unsigned int flags = str->flags;
+ unsigned int ch = str->ch;
+ eol_type_t eol_type = str->eol_type;
while (n--)
{
DECODE_HANDLE_END_OF_CONVERSION (flags, ch, dst);
- CODING_STREAM_COMPOSE (str, flags, ch);
+ str->flags = flags;
+ str->ch = ch;
}
/* Convert internally-formatted data to Shift-JIS. */
static void
-encode_coding_sjis (Lstream *encoding, CONST unsigned char *src,
+encode_coding_sjis (Lstream *encoding, const unsigned char *src,
unsigned_char_dynarr *dst, unsigned int n)
{
unsigned char c;
struct encoding_stream *str = ENCODING_STREAM_DATA (encoding);
- unsigned int flags, ch;
- enum eol_type eol_type;
-
- CODING_STREAM_DECOMPOSE (str, flags, ch);
- eol_type = CODING_SYSTEM_EOL_TYPE (str->codesys);
+ unsigned int flags = str->flags;
+ unsigned int ch = str->ch;
+ eol_type_t eol_type = CODING_SYSTEM_EOL_TYPE (str->codesys);
while (n--)
{
}
}
- CODING_STREAM_COMPOSE (str, flags, ch);
+ str->flags = flags;
+ str->ch = ch;
}
DEFUN ("decode-shift-jis-char", Fdecode_shift_jis_char, 1, 1, 0, /*
Since the number of characters in Big5 is larger than maximum
characters in Emacs' charset (96x96), it can't be handled as one
- charset. So, in Emacs, Big5 is devided into two: `charset-big5-1'
- and `charset-big5-2'. Both <type>s are TYPE94x94. The former
+ charset. So, in Emacs, Big5 is divided into two: `charset-big5-1'
+ and `charset-big5-2'. Both <type>s are DIMENSION2_CHARS94. The former
contains frequently used characters and the latter contains less
frequently used characters. */
} while (0)
static int
-detect_coding_big5 (struct detection_state *st, CONST unsigned char *src,
+detect_coding_big5 (struct detection_state *st, const unsigned char *src,
unsigned int n)
{
int c;
/* Convert Big5 data to internal format. */
static void
-decode_coding_big5 (Lstream *decoding, CONST unsigned char *src,
+decode_coding_big5 (Lstream *decoding, const unsigned char *src,
unsigned_char_dynarr *dst, unsigned int n)
{
unsigned char c;
- unsigned int flags, ch;
- enum eol_type eol_type;
struct decoding_stream *str = DECODING_STREAM_DATA (decoding);
-
- CODING_STREAM_DECOMPOSE (str, flags, ch);
- eol_type = str->eol_type;
+ unsigned int flags = str->flags;
+ unsigned int ch = str->ch;
+ eol_type_t eol_type = str->eol_type;
while (n--)
{
DECODE_HANDLE_END_OF_CONVERSION (flags, ch, dst);
- CODING_STREAM_COMPOSE (str, flags, ch);
+ str->flags = flags;
+ str->ch = ch;
}
/* Convert internally-formatted data to Big5. */
static void
-encode_coding_big5 (Lstream *encoding, CONST unsigned char *src,
+encode_coding_big5 (Lstream *encoding, const unsigned char *src,
unsigned_char_dynarr *dst, unsigned int n)
{
unsigned char c;
struct encoding_stream *str = ENCODING_STREAM_DATA (encoding);
- unsigned int flags, ch;
- enum eol_type eol_type;
-
- CODING_STREAM_DECOMPOSE (str, flags, ch);
- eol_type = CODING_SYSTEM_EOL_TYPE (str->codesys);
+ unsigned int flags = str->flags;
+ unsigned int ch = str->ch;
+ eol_type_t eol_type = CODING_SYSTEM_EOL_TYPE (str->codesys);
while (n--)
{
ch = 0;
}
- CODING_STREAM_COMPOSE (str, flags, ch);
+ str->flags = flags;
+ str->ch = ch;
}
\f
/************************************************************************/
-/* ISO2022 methods */
+/* UCS-4 methods */
+/* */
+/* UCS-4 character codes are implemented as nonnegative integers. */
+/* */
/************************************************************************/
-/* The following note describes the coding system ISO2022 briefly.
- Since the intention of this note is to help understanding of the
- programs in this file, some parts are NOT ACCURATE or OVERLY
- SIMPLIFIED. For thorough understanding, please refer to the
- original document of ISO2022.
- ISO2022 provides many mechanisms to encode several character sets
- in 7-bit and 8-bit environments. If one chooses 7-bit environment,
- all text is encoded by codes of less than 128. This may make the
- encoded text a little bit longer, but the text get more stability
- to pass through several gateways (some of them strip off MSB).
+DEFUN ("set-ucs-char", Fset_ucs_char, 2, 2, 0, /*
+Map UCS-4 code CODE to Mule character CHARACTER.
- There are two kind of character sets: control character set and
- graphic character set. The former contains control characters such
- as `newline' and `escape' to provide control functions (control
- functions are provided also by escape sequence). The latter
- contains graphic characters such as 'A' and '-'. Emacs recognizes
- two control character sets and many graphic character sets.
+Return T on success, NIL on failure.
+*/
+ (code, character))
+{
+ unsigned int c;
- Graphic character sets are classified into one of four types,
- according to the dimension and number of characters in the set:
- TYPE94, TYPE96, TYPE94x94, and TYPE96x96. In addition, each
- character set is assigned an identification byte, unique for each
- type, called "final character" (denoted as <F> hereafter). The <F>
- of each character set is decided by ECMA(*) when it is registered
- in ISO. Code range of <F> is 0x30..0x7F (0x30..0x3F are for
- private use only).
+ CHECK_CHAR (character);
+ CHECK_INT (code);
+ c = XINT (code);
- Note (*): ECMA = European Computer Manufacturers Association
+ if (c < sizeof (fcd->ucs_to_mule_table))
+ {
+ fcd->ucs_to_mule_table[c] = character;
+ return Qt;
+ }
+ else
+ return Qnil;
+}
- Here are examples of graphic character set [NAME(<F>)]:
- o TYPE94 -- ASCII('B'), right-half-of-JISX0201('I'), ...
- o TYPE96 -- right-half-of-ISO8859-1('A'), ...
- o TYPE94x94 -- GB2312('A'), JISX0208('B'), ...
- o TYPE96x96 -- none for the moment
+static Lisp_Object
+ucs_to_char (unsigned long code)
+{
+ if (code < sizeof (fcd->ucs_to_mule_table))
+ {
+ return fcd->ucs_to_mule_table[code];
+ }
+ else if ((0xe00000 <= code) && (code <= 0xe00000 + 94 * 94 * 14))
+ {
+ unsigned int c;
+
+ code -= 0xe00000;
+ c = code % (94 * 94);
+ return make_char
+ (MAKE_CHAR (CHARSET_BY_ATTRIBUTES
+ (CHARSET_TYPE_94X94, code / (94 * 94) + '@',
+ CHARSET_LEFT_TO_RIGHT),
+ c / 94 + 33, c % 94 + 33));
+ }
+ else
+ return Qnil;
+}
- A code area (1byte=8bits) is divided into 4 areas, C0, GL, C1, and GR.
- C0 [0x00..0x1F] -- control character plane 0
- GL [0x20..0x7F] -- graphic character plane 0
- C1 [0x80..0x9F] -- control character plane 1
- GR [0xA0..0xFF] -- graphic character plane 1
+DEFUN ("ucs-char", Fucs_char, 1, 1, 0, /*
+Return Mule character corresponding to UCS code CODE (a positive integer).
+*/
+ (code))
+{
+ CHECK_NATNUM (code);
+ return ucs_to_char (XINT (code));
+}
- A control character set is directly designated and invoked to C0 or
- C1 by an escape sequence. The most common case is that:
- - ISO646's control character set is designated/invoked to C0, and
- - ISO6429's control character set is designated/invoked to C1,
- and usually these designations/invocations are omitted in encoded
- text. In a 7-bit environment, only C0 can be used, and a control
- character for C1 is encoded by an appropriate escape sequence to
- fit into the environment. All control characters for C1 are
- defined to have corresponding escape sequences.
+DEFUN ("set-char-ucs", Fset_char_ucs, 2, 2, 0, /*
+Map Mule character CHARACTER to UCS code CODE (a positive integer).
+*/
+ (character, code))
+{
+ /* #### Isn't this gilding the lily? Fput_char_table checks its args.
+ Fset_char_ucs is more restrictive on index arg, but should
+ check code arg in a char_table method. */
+ CHECK_CHAR (character);
+ CHECK_NATNUM (code);
+ return Fput_char_table (character, code, mule_to_ucs_table);
+}
- A graphic character set is at first designated to one of four
- graphic registers (G0 through G3), then these graphic registers are
- invoked to GL or GR. These designations and invocations can be
- done independently. The most common case is that G0 is invoked to
- GL, G1 is invoked to GR, and ASCII is designated to G0. Usually
- these invocations and designations are omitted in encoded text.
- In a 7-bit environment, only GL can be used.
+DEFUN ("char-ucs", Fchar_ucs, 1, 1, 0, /*
+Return the UCS code (a positive integer) corresponding to CHARACTER.
+*/
+ (character))
+{
+ return Fget_char_table (character, mule_to_ucs_table);
+}
- When a graphic character set of TYPE94 or TYPE94x94 is invoked to
- GL, codes 0x20 and 0x7F of the GL area work as control characters
- SPACE and DEL respectively, and code 0xA0 and 0xFF of GR area
- should not be used.
+/* Decode a UCS-4 character into a buffer. If the lookup fails, use
+ <GETA MARK> (U+3013) of JIS X 0208, which means correct character
+ is not found, instead.
+ #### do something more appropriate (use blob?)
+ Danger, Will Robinson! Data loss. Should we signal user? */
+static void
+decode_ucs4 (unsigned long ch, unsigned_char_dynarr *dst)
+{
+ Lisp_Object chr = ucs_to_char (ch);
- There are two ways of invocation: locking-shift and single-shift.
- With locking-shift, the invocation lasts until the next different
- invocation, whereas with single-shift, the invocation works only
- for the following character and doesn't affect locking-shift.
- Invocations are done by the following control characters or escape
- sequences.
+ if (! NILP (chr))
+ {
+ Bufbyte work[MAX_EMCHAR_LEN];
+ int len;
+
+ ch = XCHAR (chr);
+ len = (ch < 128) ?
+ simple_set_charptr_emchar (work, ch) :
+ non_ascii_set_charptr_emchar (work, ch);
+ Dynarr_add_many (dst, work, len);
+ }
+ else
+ {
+ Dynarr_add (dst, LEADING_BYTE_JAPANESE_JISX0208);
+ Dynarr_add (dst, 34 + 128);
+ Dynarr_add (dst, 46 + 128);
+ }
+}
- ----------------------------------------------------------------------
- abbrev function cntrl escape seq description
- ----------------------------------------------------------------------
- SI/LS0 (shift-in) 0x0F none invoke G0 into GL
- SO/LS1 (shift-out) 0x0E none invoke G1 into GL
- LS1R (locking-shift-1 right) none ESC '~' invoke G1 into GR
- LS2 (locking-shift-2) none ESC 'n' invoke G2 into GL
- LS2R (locking-shift-2 right) none ESC '}' invoke G2 into GR
- LS3 (locking-shift-3) none ESC 'o' invoke G3 into GL
- LS3R (locking-shift 3 right) none ESC '|' invoke G3 into GR
- SS2 (single-shift-2) 0x8E ESC 'N' invoke G2 for one char
- SS3 (single-shift-3) 0x8F ESC 'O' invoke G3 for one char
- ----------------------------------------------------------------------
- The first four are for locking-shift. Control characters for these
- functions are defined by macros ISO_CODE_XXX in `coding.h'.
+static unsigned long
+mule_char_to_ucs4 (Lisp_Object charset,
+ unsigned char h, unsigned char l)
+{
+ Lisp_Object code
+ = Fget_char_table (make_char (MAKE_CHAR (charset, h & 127, l & 127)),
+ mule_to_ucs_table);
- Designations are done by the following escape sequences.
- ----------------------------------------------------------------------
- escape sequence description
- ----------------------------------------------------------------------
- ESC '(' <F> designate TYPE94<F> to G0
- ESC ')' <F> designate TYPE94<F> to G1
- ESC '*' <F> designate TYPE94<F> to G2
- ESC '+' <F> designate TYPE94<F> to G3
- ESC ',' <F> designate TYPE96<F> to G0 (*)
- ESC '-' <F> designate TYPE96<F> to G1
- ESC '.' <F> designate TYPE96<F> to G2
- ESC '/' <F> designate TYPE96<F> to G3
- ESC '$' '(' <F> designate TYPE94x94<F> to G0 (**)
- ESC '$' ')' <F> designate TYPE94x94<F> to G1
- ESC '$' '*' <F> designate TYPE94x94<F> to G2
- ESC '$' '+' <F> designate TYPE94x94<F> to G3
- ESC '$' ',' <F> designate TYPE96x96<F> to G0 (*)
- ESC '$' '-' <F> designate TYPE96x96<F> to G1
- ESC '$' '.' <F> designate TYPE96x96<F> to G2
- ESC '$' '/' <F> designate TYPE96x96<F> to G3
+ if (INTP (code))
+ {
+ return XINT (code);
+ }
+ else if ( (XCHARSET_DIMENSION (charset) == 2) &&
+ (XCHARSET_CHARS (charset) == 94) )
+ {
+ unsigned char final = XCHARSET_FINAL (charset);
+
+ if ( ('@' <= final) && (final < 0x7f) )
+ {
+ return 0xe00000 + (final - '@') * 94 * 94
+ + ((h & 127) - 33) * 94 + (l & 127) - 33;
+ }
+ else
+ {
+ return '?';
+ }
+ }
+ else
+ {
+ return '?';
+ }
+}
+
+static void
+encode_ucs4 (Lisp_Object charset,
+ unsigned char h, unsigned char l, unsigned_char_dynarr *dst)
+{
+ unsigned long code = mule_char_to_ucs4 (charset, h, l);
+ Dynarr_add (dst, code >> 24);
+ Dynarr_add (dst, (code >> 16) & 255);
+ Dynarr_add (dst, (code >> 8) & 255);
+ Dynarr_add (dst, code & 255);
+}
+
+static int
+detect_coding_ucs4 (struct detection_state *st, const unsigned char *src,
+ unsigned int n)
+{
+ while (n--)
+ {
+ int c = *src++;
+ switch (st->ucs4.in_byte)
+ {
+ case 0:
+ if (c >= 128)
+ return 0;
+ else
+ st->ucs4.in_byte++;
+ break;
+ case 3:
+ st->ucs4.in_byte = 0;
+ break;
+ default:
+ st->ucs4.in_byte++;
+ }
+ }
+ return CODING_CATEGORY_UCS4_MASK;
+}
+
+static void
+decode_coding_ucs4 (Lstream *decoding, const unsigned char *src,
+ unsigned_char_dynarr *dst, unsigned int n)
+{
+ struct decoding_stream *str = DECODING_STREAM_DATA (decoding);
+ unsigned int flags = str->flags;
+ unsigned int ch = str->ch;
+ unsigned char counter = str->counter;
+
+ while (n--)
+ {
+ unsigned char c = *src++;
+ switch (counter)
+ {
+ case 0:
+ ch = c;
+ counter = 3;
+ break;
+ case 1:
+ decode_ucs4 ( ( ch << 8 ) | c, dst);
+ ch = 0;
+ counter = 0;
+ break;
+ default:
+ ch = ( ch << 8 ) | c;
+ counter--;
+ }
+ }
+ if (counter & CODING_STATE_END)
+ DECODE_OUTPUT_PARTIAL_CHAR (ch);
+
+ str->flags = flags;
+ str->ch = ch;
+ str->counter = counter;
+}
+
+static void
+encode_coding_ucs4 (Lstream *encoding, const unsigned char *src,
+ unsigned_char_dynarr *dst, unsigned int n)
+{
+ struct encoding_stream *str = ENCODING_STREAM_DATA (encoding);
+ unsigned int flags = str->flags;
+ unsigned int ch = str->ch;
+ unsigned char char_boundary = str->iso2022.current_char_boundary;
+ Lisp_Object charset = str->iso2022.current_charset;
+
+#ifdef ENABLE_COMPOSITE_CHARS
+ /* flags for handling composite chars. We do a little switcharoo
+ on the source while we're outputting the composite char. */
+ unsigned int saved_n = 0;
+ const unsigned char *saved_src = NULL;
+ int in_composite = 0;
+
+ back_to_square_n:
+#endif
+
+ while (n--)
+ {
+ unsigned char c = *src++;
+
+ if (BYTE_ASCII_P (c))
+ { /* Processing ASCII character */
+ ch = 0;
+ encode_ucs4 (Vcharset_ascii, c, 0, dst);
+ char_boundary = 1;
+ }
+ else if (BUFBYTE_LEADING_BYTE_P (c) || BUFBYTE_LEADING_BYTE_P (ch))
+ { /* Processing Leading Byte */
+ ch = 0;
+ charset = CHARSET_BY_LEADING_BYTE (c);
+ if (LEADING_BYTE_PREFIX_P(c))
+ ch = c;
+ char_boundary = 0;
+ }
+ else
+ { /* Processing Non-ASCII character */
+ char_boundary = 1;
+ if (EQ (charset, Vcharset_control_1))
+ {
+ encode_ucs4 (Vcharset_control_1, c, 0, dst);
+ }
+ else
+ {
+ switch (XCHARSET_REP_BYTES (charset))
+ {
+ case 2:
+ encode_ucs4 (charset, c, 0, dst);
+ break;
+ case 3:
+ if (XCHARSET_PRIVATE_P (charset))
+ {
+ encode_ucs4 (charset, c, 0, dst);
+ ch = 0;
+ }
+ else if (ch)
+ {
+#ifdef ENABLE_COMPOSITE_CHARS
+ if (EQ (charset, Vcharset_composite))
+ {
+ if (in_composite)
+ {
+ /* #### Bother! We don't know how to
+ handle this yet. */
+ Dynarr_add (dst, 0);
+ Dynarr_add (dst, 0);
+ Dynarr_add (dst, 0);
+ Dynarr_add (dst, '~');
+ }
+ else
+ {
+ Emchar emch = MAKE_CHAR (Vcharset_composite,
+ ch & 0x7F, c & 0x7F);
+ Lisp_Object lstr = composite_char_string (emch);
+ saved_n = n;
+ saved_src = src;
+ in_composite = 1;
+ src = XSTRING_DATA (lstr);
+ n = XSTRING_LENGTH (lstr);
+ }
+ }
+ else
+#endif /* ENABLE_COMPOSITE_CHARS */
+ {
+ encode_ucs4(charset, ch, c, dst);
+ }
+ ch = 0;
+ }
+ else
+ {
+ ch = c;
+ char_boundary = 0;
+ }
+ break;
+ case 4:
+ if (ch)
+ {
+ encode_ucs4 (charset, ch, c, dst);
+ ch = 0;
+ }
+ else
+ {
+ ch = c;
+ char_boundary = 0;
+ }
+ break;
+ default:
+ abort ();
+ }
+ }
+ }
+ }
+
+#ifdef ENABLE_COMPOSITE_CHARS
+ if (in_composite)
+ {
+ n = saved_n;
+ src = saved_src;
+ in_composite = 0;
+ goto back_to_square_n; /* Wheeeeeeeee ..... */
+ }
+#endif /* ENABLE_COMPOSITE_CHARS */
+
+ str->flags = flags;
+ str->ch = ch;
+ str->iso2022.current_char_boundary = char_boundary;
+ str->iso2022.current_charset = charset;
+
+ /* Verbum caro factum est! */
+}
+
+\f
+/************************************************************************/
+/* UTF-8 methods */
+/************************************************************************/
+
+static int
+detect_coding_utf8 (struct detection_state *st, const unsigned char *src,
+ unsigned int n)
+{
+ while (n--)
+ {
+ unsigned char c = *src++;
+ switch (st->utf8.in_byte)
+ {
+ case 0:
+ if (c == ISO_CODE_ESC || c == ISO_CODE_SI || c == ISO_CODE_SO)
+ return 0;
+ else if (c >= 0xfc)
+ st->utf8.in_byte = 5;
+ else if (c >= 0xf8)
+ st->utf8.in_byte = 4;
+ else if (c >= 0xf0)
+ st->utf8.in_byte = 3;
+ else if (c >= 0xe0)
+ st->utf8.in_byte = 2;
+ else if (c >= 0xc0)
+ st->utf8.in_byte = 1;
+ else if (c >= 0x80)
+ return 0;
+ break;
+ default:
+ if ((c & 0xc0) != 0x80)
+ return 0;
+ else
+ st->utf8.in_byte--;
+ }
+ }
+ return CODING_CATEGORY_UTF8_MASK;
+}
+
+static void
+decode_coding_utf8 (Lstream *decoding, const unsigned char *src,
+ unsigned_char_dynarr *dst, unsigned int n)
+{
+ struct decoding_stream *str = DECODING_STREAM_DATA (decoding);
+ unsigned int flags = str->flags;
+ unsigned int ch = str->ch;
+ eol_type_t eol_type = str->eol_type;
+ unsigned char counter = str->counter;
+
+ while (n--)
+ {
+ unsigned char c = *src++;
+ switch (counter)
+ {
+ case 0:
+ if ( c >= 0xfc )
+ {
+ ch = c & 0x01;
+ counter = 5;
+ }
+ else if ( c >= 0xf8 )
+ {
+ ch = c & 0x03;
+ counter = 4;
+ }
+ else if ( c >= 0xf0 )
+ {
+ ch = c & 0x07;
+ counter = 3;
+ }
+ else if ( c >= 0xe0 )
+ {
+ ch = c & 0x0f;
+ counter = 2;
+ }
+ else if ( c >= 0xc0 )
+ {
+ ch = c & 0x1f;
+ counter = 1;
+ }
+ else
+ {
+ DECODE_HANDLE_EOL_TYPE (eol_type, c, flags, dst);
+ decode_ucs4 (c, dst);
+ }
+ break;
+ case 1:
+ ch = ( ch << 6 ) | ( c & 0x3f );
+ decode_ucs4 (ch, dst);
+ ch = 0;
+ counter = 0;
+ break;
+ default:
+ ch = ( ch << 6 ) | ( c & 0x3f );
+ counter--;
+ }
+ label_continue_loop:;
+ }
+
+ if (flags & CODING_STATE_END)
+ DECODE_OUTPUT_PARTIAL_CHAR (ch);
+
+ str->flags = flags;
+ str->ch = ch;
+ str->counter = counter;
+}
+
+static void
+encode_utf8 (Lisp_Object charset,
+ unsigned char h, unsigned char l, unsigned_char_dynarr *dst)
+{
+ unsigned long code = mule_char_to_ucs4 (charset, h, l);
+ if ( code <= 0x7f )
+ {
+ Dynarr_add (dst, code);
+ }
+ else if ( code <= 0x7ff )
+ {
+ Dynarr_add (dst, (code >> 6) | 0xc0);
+ Dynarr_add (dst, (code & 0x3f) | 0x80);
+ }
+ else if ( code <= 0xffff )
+ {
+ Dynarr_add (dst, (code >> 12) | 0xe0);
+ Dynarr_add (dst, ((code >> 6) & 0x3f) | 0x80);
+ Dynarr_add (dst, (code & 0x3f) | 0x80);
+ }
+ else if ( code <= 0x1fffff )
+ {
+ Dynarr_add (dst, (code >> 18) | 0xf0);
+ Dynarr_add (dst, ((code >> 12) & 0x3f) | 0x80);
+ Dynarr_add (dst, ((code >> 6) & 0x3f) | 0x80);
+ Dynarr_add (dst, (code & 0x3f) | 0x80);
+ }
+ else if ( code <= 0x3ffffff )
+ {
+ Dynarr_add (dst, (code >> 24) | 0xf8);
+ Dynarr_add (dst, ((code >> 18) & 0x3f) | 0x80);
+ Dynarr_add (dst, ((code >> 12) & 0x3f) | 0x80);
+ Dynarr_add (dst, ((code >> 6) & 0x3f) | 0x80);
+ Dynarr_add (dst, (code & 0x3f) | 0x80);
+ }
+ else
+ {
+ Dynarr_add (dst, (code >> 30) | 0xfc);
+ Dynarr_add (dst, ((code >> 24) & 0x3f) | 0x80);
+ Dynarr_add (dst, ((code >> 18) & 0x3f) | 0x80);
+ Dynarr_add (dst, ((code >> 12) & 0x3f) | 0x80);
+ Dynarr_add (dst, ((code >> 6) & 0x3f) | 0x80);
+ Dynarr_add (dst, (code & 0x3f) | 0x80);
+ }
+}
+
+static void
+encode_coding_utf8 (Lstream *encoding, const unsigned char *src,
+ unsigned_char_dynarr *dst, unsigned int n)
+{
+ struct encoding_stream *str = ENCODING_STREAM_DATA (encoding);
+ unsigned int flags = str->flags;
+ unsigned int ch = str->ch;
+ eol_type_t eol_type = CODING_SYSTEM_EOL_TYPE (str->codesys);
+ unsigned char char_boundary = str->iso2022.current_char_boundary;
+ Lisp_Object charset = str->iso2022.current_charset;
+
+#ifdef ENABLE_COMPOSITE_CHARS
+ /* flags for handling composite chars. We do a little switcharoo
+ on the source while we're outputting the composite char. */
+ unsigned int saved_n = 0;
+ const unsigned char *saved_src = NULL;
+ int in_composite = 0;
+
+ back_to_square_n:
+#endif /* ENABLE_COMPOSITE_CHARS */
+
+ while (n--)
+ {
+ unsigned char c = *src++;
+
+ if (BYTE_ASCII_P (c))
+ { /* Processing ASCII character */
+ ch = 0;
+ if (c == '\n')
+ {
+ if (eol_type != EOL_LF && eol_type != EOL_AUTODETECT)
+ Dynarr_add (dst, '\r');
+ if (eol_type != EOL_CR)
+ Dynarr_add (dst, c);
+ }
+ else
+ encode_utf8 (Vcharset_ascii, c, 0, dst);
+ char_boundary = 1;
+ }
+ else if (BUFBYTE_LEADING_BYTE_P (c) || BUFBYTE_LEADING_BYTE_P (ch))
+ { /* Processing Leading Byte */
+ ch = 0;
+ charset = CHARSET_BY_LEADING_BYTE (c);
+ if (LEADING_BYTE_PREFIX_P(c))
+ ch = c;
+ char_boundary = 0;
+ }
+ else
+ { /* Processing Non-ASCII character */
+ char_boundary = 1;
+ if (EQ (charset, Vcharset_control_1))
+ {
+ encode_utf8 (Vcharset_control_1, c, 0, dst);
+ }
+ else
+ {
+ switch (XCHARSET_REP_BYTES (charset))
+ {
+ case 2:
+ encode_utf8 (charset, c, 0, dst);
+ break;
+ case 3:
+ if (XCHARSET_PRIVATE_P (charset))
+ {
+ encode_utf8 (charset, c, 0, dst);
+ ch = 0;
+ }
+ else if (ch)
+ {
+#ifdef ENABLE_COMPOSITE_CHARS
+ if (EQ (charset, Vcharset_composite))
+ {
+ if (in_composite)
+ {
+ /* #### Bother! We don't know how to
+ handle this yet. */
+ encode_utf8 (Vcharset_ascii, '~', 0, dst);
+ }
+ else
+ {
+ Emchar emch = MAKE_CHAR (Vcharset_composite,
+ ch & 0x7F, c & 0x7F);
+ Lisp_Object lstr = composite_char_string (emch);
+ saved_n = n;
+ saved_src = src;
+ in_composite = 1;
+ src = XSTRING_DATA (lstr);
+ n = XSTRING_LENGTH (lstr);
+ }
+ }
+ else
+#endif /* ENABLE_COMPOSITE_CHARS */
+ {
+ encode_utf8 (charset, ch, c, dst);
+ }
+ ch = 0;
+ }
+ else
+ {
+ ch = c;
+ char_boundary = 0;
+ }
+ break;
+ case 4:
+ if (ch)
+ {
+ encode_utf8 (charset, ch, c, dst);
+ ch = 0;
+ }
+ else
+ {
+ ch = c;
+ char_boundary = 0;
+ }
+ break;
+ default:
+ abort ();
+ }
+ }
+ }
+ }
+
+#ifdef ENABLE_COMPOSITE_CHARS
+ if (in_composite)
+ {
+ n = saved_n;
+ src = saved_src;
+ in_composite = 0;
+ goto back_to_square_n; /* Wheeeeeeeee ..... */
+ }
+#endif
+
+ str->flags = flags;
+ str->ch = ch;
+ str->iso2022.current_char_boundary = char_boundary;
+ str->iso2022.current_charset = charset;
+
+ /* Verbum caro factum est! */
+}
+
+\f
+/************************************************************************/
+/* ISO2022 methods */
+/************************************************************************/
+
+/* The following note describes the coding system ISO2022 briefly.
+ Since the intention of this note is to help understand the
+ functions in this file, some parts are NOT ACCURATE or OVERLY
+ SIMPLIFIED. For thorough understanding, please refer to the
+ original document of ISO2022.
+
+ ISO2022 provides many mechanisms to encode several character sets
+ in 7-bit and 8-bit environments. For 7-bit environments, all text
+ is encoded using bytes less than 128. This may make the encoded
+ text a little bit longer, but the text passes more easily through
+ several gateways, some of which strip off MSB (Most Signigant Bit).
+
+ There are two kinds of character sets: control character set and
+ graphic character set. The former contains control characters such
+ as `newline' and `escape' to provide control functions (control
+ functions are also provided by escape sequences). The latter
+ contains graphic characters such as 'A' and '-'. Emacs recognizes
+ two control character sets and many graphic character sets.
+
+ Graphic character sets are classified into one of the following
+ four classes, according to the number of bytes (DIMENSION) and
+ number of characters in one dimension (CHARS) of the set:
+ - DIMENSION1_CHARS94
+ - DIMENSION1_CHARS96
+ - DIMENSION2_CHARS94
+ - DIMENSION2_CHARS96
+
+ In addition, each character set is assigned an identification tag,
+ unique for each set, called "final character" (denoted as <F>
+ hereafter). The <F> of each character set is decided by ECMA(*)
+ when it is registered in ISO. The code range of <F> is 0x30..0x7F
+ (0x30..0x3F are for private use only).
+
+ Note (*): ECMA = European Computer Manufacturers Association
+
+ Here are examples of graphic character set [NAME(<F>)]:
+ o DIMENSION1_CHARS94 -- ASCII('B'), right-half-of-JISX0201('I'), ...
+ o DIMENSION1_CHARS96 -- right-half-of-ISO8859-1('A'), ...
+ o DIMENSION2_CHARS94 -- GB2312('A'), JISX0208('B'), ...
+ o DIMENSION2_CHARS96 -- none for the moment
+
+ A code area (1 byte = 8 bits) is divided into 4 areas, C0, GL, C1, and GR.
+ C0 [0x00..0x1F] -- control character plane 0
+ GL [0x20..0x7F] -- graphic character plane 0
+ C1 [0x80..0x9F] -- control character plane 1
+ GR [0xA0..0xFF] -- graphic character plane 1
+
+ A control character set is directly designated and invoked to C0 or
+ C1 by an escape sequence. The most common case is that:
+ - ISO646's control character set is designated/invoked to C0, and
+ - ISO6429's control character set is designated/invoked to C1,
+ and usually these designations/invocations are omitted in encoded
+ text. In a 7-bit environment, only C0 can be used, and a control
+ character for C1 is encoded by an appropriate escape sequence to
+ fit into the environment. All control characters for C1 are
+ defined to have corresponding escape sequences.
+
+ A graphic character set is at first designated to one of four
+ graphic registers (G0 through G3), then these graphic registers are
+ invoked to GL or GR. These designations and invocations can be
+ done independently. The most common case is that G0 is invoked to
+ GL, G1 is invoked to GR, and ASCII is designated to G0. Usually
+ these invocations and designations are omitted in encoded text.
+ In a 7-bit environment, only GL can be used.
+
+ When a graphic character set of CHARS94 is invoked to GL, codes
+ 0x20 and 0x7F of the GL area work as control characters SPACE and
+ DEL respectively, and codes 0xA0 and 0xFF of the GR area should not
+ be used.
+
+ There are two ways of invocation: locking-shift and single-shift.
+ With locking-shift, the invocation lasts until the next different
+ invocation, whereas with single-shift, the invocation affects the
+ following character only and doesn't affect the locking-shift
+ state. Invocations are done by the following control characters or
+ escape sequences:
+
+ ----------------------------------------------------------------------
+ abbrev function cntrl escape seq description
+ ----------------------------------------------------------------------
+ SI/LS0 (shift-in) 0x0F none invoke G0 into GL
+ SO/LS1 (shift-out) 0x0E none invoke G1 into GL
+ LS2 (locking-shift-2) none ESC 'n' invoke G2 into GL
+ LS3 (locking-shift-3) none ESC 'o' invoke G3 into GL
+ LS1R (locking-shift-1 right) none ESC '~' invoke G1 into GR (*)
+ LS2R (locking-shift-2 right) none ESC '}' invoke G2 into GR (*)
+ LS3R (locking-shift 3 right) none ESC '|' invoke G3 into GR (*)
+ SS2 (single-shift-2) 0x8E ESC 'N' invoke G2 for one char
+ SS3 (single-shift-3) 0x8F ESC 'O' invoke G3 for one char
----------------------------------------------------------------------
- In this list, "TYPE94<F>" means a graphic character set of type TYPE94
- and final character <F>, and etc.
+ (*) These are not used by any known coding system.
+
+ Control characters for these functions are defined by macros
+ ISO_CODE_XXX in `coding.h'.
+
+ Designations are done by the following escape sequences:
+ ----------------------------------------------------------------------
+ escape sequence description
+ ----------------------------------------------------------------------
+ ESC '(' <F> designate DIMENSION1_CHARS94<F> to G0
+ ESC ')' <F> designate DIMENSION1_CHARS94<F> to G1
+ ESC '*' <F> designate DIMENSION1_CHARS94<F> to G2
+ ESC '+' <F> designate DIMENSION1_CHARS94<F> to G3
+ ESC ',' <F> designate DIMENSION1_CHARS96<F> to G0 (*)
+ ESC '-' <F> designate DIMENSION1_CHARS96<F> to G1
+ ESC '.' <F> designate DIMENSION1_CHARS96<F> to G2
+ ESC '/' <F> designate DIMENSION1_CHARS96<F> to G3
+ ESC '$' '(' <F> designate DIMENSION2_CHARS94<F> to G0 (**)
+ ESC '$' ')' <F> designate DIMENSION2_CHARS94<F> to G1
+ ESC '$' '*' <F> designate DIMENSION2_CHARS94<F> to G2
+ ESC '$' '+' <F> designate DIMENSION2_CHARS94<F> to G3
+ ESC '$' ',' <F> designate DIMENSION2_CHARS96<F> to G0 (*)
+ ESC '$' '-' <F> designate DIMENSION2_CHARS96<F> to G1
+ ESC '$' '.' <F> designate DIMENSION2_CHARS96<F> to G2
+ ESC '$' '/' <F> designate DIMENSION2_CHARS96<F> to G3
+ ----------------------------------------------------------------------
+
+ In this list, "DIMENSION1_CHARS94<F>" means a graphic character set
+ of dimension 1, chars 94, and final character <F>, etc...
Note (*): Although these designations are not allowed in ISO2022,
Emacs accepts them on decoding, and produces them on encoding
- TYPE96 or TYPE96x96 character set in a coding system which is
- characterized as 7-bit environment, non-locking-shift, and
- non-single-shift.
+ CHARS96 character sets in a coding system which is characterized as
+ 7-bit environment, non-locking-shift, and non-single-shift.
Note (**): If <F> is '@', 'A', or 'B', the intermediate character
- '(' can be omitted. We call this as "short-form" here after.
+ '(' can be omitted. We refer to this as "short-form" hereafter.
Now you may notice that there are a lot of ways for encoding the
same multilingual text in ISO2022. Actually, there exist many
- coding systems such as Compound Text (used in X's inter client
+ coding systems such as Compound Text (used in X11's inter client
communication, ISO-2022-JP (used in Japanese internet), ISO-2022-KR
(used in Korean internet), EUC (Extended UNIX Code, used in Asian
localized platforms), and all of these are variants of ISO2022.
sequences: ISO6429's direction specification and Emacs' private
sequence for specifying character composition.
- ISO6429's direction specification takes the following format:
+ ISO6429's direction specification takes the following form:
o CSI ']' -- end of the current direction
o CSI '0' ']' -- end of the current direction
o CSI '1' ']' -- start of left-to-right text
o CSI '2' ']' -- start of right-to-left text
The control character CSI (0x9B: control sequence introducer) is
- abbreviated to the escape sequence ESC '[' in 7-bit environment.
+ abbreviated to the escape sequence ESC '[' in a 7-bit environment.
- Character composition specification takes the following format:
+ Character composition specification takes the following form:
o ESC '0' -- start character composition
o ESC '1' -- end character composition
- Since these are not standard escape sequences of any ISO, the use
- of them for these meanings is restricted to Emacs only. */
+ Since these are not standard escape sequences of any ISO standard,
+ their use with these meanings is restricted to Emacs only. */
static void
reset_iso2022 (Lisp_Object coding_system, struct iso2022_decoder *iso)
iso->invalid_switch_dir = 0;
iso->output_direction_sequence = 0;
iso->output_literally = 0;
+#ifdef ENABLE_COMPOSITE_CHARS
if (iso->composite_chars)
Dynarr_reset (iso->composite_chars);
+#endif
}
static int
reg = 3; half = 1;
goto locking_shift;
+#ifdef ENABLE_COMPOSITE_CHARS
/**** composite ****/
case '0':
*flags = (*flags & CODING_STATE_ISO2022_LOCK) &
~CODING_STATE_COMPOSITE;
return 1;
+#endif /* ENABLE_COMPOSITE_CHARS */
/**** directionality ****/
}
static int
-detect_coding_iso2022 (struct detection_state *st, CONST unsigned char *src,
+detect_coding_iso2022 (struct detection_state *st, const unsigned char *src,
unsigned int n)
{
- int c;
int mask;
/* #### There are serious deficiencies in the recognition mechanism
- here. This needs to be much smarter if it's going to cut it. */
+ here. This needs to be much smarter if it's going to cut it.
+ The sequence "\xff\x0f" is currently detected as LOCK_SHIFT while
+ it should be detected as Latin-1.
+ All the ISO2022 stuff in this file should be synced up with the
+ code from FSF Emacs-20.4, in which Mule should be more or less stable.
+ Perhaps we should wait till R2L works in FSF Emacs? */
if (!st->iso2022.initted)
{
while (n--)
{
- c = *src++;
+ int c = *src++;
if (c >= 0xA0)
{
mask &= ~CODING_CATEGORY_ISO_7_MASK;
need to handle the CSI differently. */
static void
-restore_left_to_right_direction (struct Lisp_Coding_System *codesys,
+restore_left_to_right_direction (Lisp_Coding_System *codesys,
unsigned_char_dynarr *dst,
unsigned int *flags,
int internal_p)
need to handle the CSI differently. */
static void
-ensure_correct_direction (int direction, struct Lisp_Coding_System *codesys,
+ensure_correct_direction (int direction, Lisp_Coding_System *codesys,
unsigned_char_dynarr *dst, unsigned int *flags,
int internal_p)
{
/* Convert ISO2022-format data to internal format. */
static void
-decode_coding_iso2022 (Lstream *decoding, CONST unsigned char *src,
+decode_coding_iso2022 (Lstream *decoding, const unsigned char *src,
unsigned_char_dynarr *dst, unsigned int n)
{
- unsigned char c;
- unsigned int flags, ch;
- enum eol_type eol_type;
struct decoding_stream *str = DECODING_STREAM_DATA (decoding);
- Lisp_Object coding_system;
+ unsigned int flags = str->flags;
+ unsigned int ch = str->ch;
+ eol_type_t eol_type = str->eol_type;
+#ifdef ENABLE_COMPOSITE_CHARS
unsigned_char_dynarr *real_dst = dst;
+#endif
+ Lisp_Object coding_system;
- CODING_STREAM_DECOMPOSE (str, flags, ch);
- eol_type = str->eol_type;
XSETCODING_SYSTEM (coding_system, str->codesys);
+#ifdef ENABLE_COMPOSITE_CHARS
if (flags & CODING_STATE_COMPOSITE)
dst = str->iso2022.composite_chars;
+#endif /* ENABLE_COMPOSITE_CHARS */
while (n--)
{
- c = *src++;
+ unsigned char c = *src++;
if (flags & CODING_STATE_ESCAPE)
{ /* Within ESC sequence */
int retval = parse_iso2022_esc (coding_system, &str->iso2022,
{
switch (str->iso2022.esc)
{
+#ifdef ENABLE_COMPOSITE_CHARS
case ISO_ESC_START_COMPOSITE:
if (str->iso2022.composite_chars)
Dynarr_reset (str->iso2022.composite_chars);
Dynarr_add_many (dst, comstr, len);
break;
}
+#endif /* ENABLE_COMPOSITE_CHARS */
case ISO_ESC_LITERAL:
DECODE_ADD_BINARY_CHAR (c, dst);
charset = str->iso2022.charset[reg];
/* Error checking: */
- if (NILP (charset) || str->iso2022.invalid_designated[reg]
+ if (! CHARSETP (charset)
+ || str->iso2022.invalid_designated[reg]
|| (((c & 0x7F) == ' ' || (c & 0x7F) == ISO_CODE_DEL)
&& XCHARSET_CHARS (charset) == 94))
/* Mrmph. We are trying to invoke a register that has no
if (flags & CODING_STATE_END)
DECODE_OUTPUT_PARTIAL_CHAR (ch);
- CODING_STREAM_COMPOSE (str, flags, ch);
+ str->flags = flags;
+ str->ch = ch;
}
iso2022_designate (Lisp_Object charset, unsigned char reg,
struct encoding_stream *str, unsigned_char_dynarr *dst)
{
- CONST char *inter94 = "()*+", *inter96= ",-./";
+ static const char inter94[] = "()*+";
+ static const char inter96[] = ",-./";
unsigned int type;
unsigned char final;
Lisp_Object old_charset = str->iso2022.charset[reg];
/* Convert internally-formatted data to ISO2022 format. */
static void
-encode_coding_iso2022 (Lstream *encoding, CONST unsigned char *src,
+encode_coding_iso2022 (Lstream *encoding, const unsigned char *src,
unsigned_char_dynarr *dst, unsigned int n)
{
unsigned char charmask, c;
- unsigned int flags, ch;
- enum eol_type eol_type;
unsigned char char_boundary;
struct encoding_stream *str = ENCODING_STREAM_DATA (encoding);
- struct Lisp_Coding_System *codesys = str->codesys;
+ unsigned int flags = str->flags;
+ unsigned int ch = str->ch;
+ Lisp_Coding_System *codesys = str->codesys;
+ eol_type_t eol_type = CODING_SYSTEM_EOL_TYPE (str->codesys);
int i;
Lisp_Object charset;
int half;
+#ifdef ENABLE_COMPOSITE_CHARS
/* flags for handling composite chars. We do a little switcharoo
on the source while we're outputting the composite char. */
unsigned int saved_n = 0;
- CONST unsigned char *saved_src = NULL;
+ const unsigned char *saved_src = NULL;
int in_composite = 0;
+#endif /* ENABLE_COMPOSITE_CHARS */
- CODING_STREAM_DECOMPOSE (str, flags, ch);
- eol_type = CODING_SYSTEM_EOL_TYPE (str->codesys);
char_boundary = str->iso2022.current_char_boundary;
charset = str->iso2022.current_charset;
half = str->iso2022.current_half;
+#ifdef ENABLE_COMPOSITE_CHARS
back_to_square_n:
+#endif
while (n--)
{
c = *src++;
if (LEADING_BYTE_PREFIX_P(c))
ch = c;
else if (!EQ (charset, Vcharset_control_1)
- && !EQ (charset, Vcharset_composite))
+#ifdef ENABLE_COMPOSITE_CHARS
+ && !EQ (charset, Vcharset_composite)
+#endif
+ )
{
int reg;
}
else if (ch)
{
+#ifdef ENABLE_COMPOSITE_CHARS
if (EQ (charset, Vcharset_composite))
{
if (in_composite)
}
}
else
+#endif /* ENABLE_COMPOSITE_CHARS */
{
Dynarr_add (dst, ch & charmask);
Dynarr_add (dst, c & charmask);
}
}
+#ifdef ENABLE_COMPOSITE_CHARS
if (in_composite)
{
n = saved_n;
Dynarr_add (dst, '1'); /* end composing */
goto back_to_square_n; /* Wheeeeeeeee ..... */
}
+#endif /* ENABLE_COMPOSITE_CHARS */
if (char_boundary && flags & CODING_STATE_END)
{
}
}
- CODING_STREAM_COMPOSE (str, flags, ch);
+ str->flags = flags;
+ str->ch = ch;
str->iso2022.current_char_boundary = char_boundary;
str->iso2022.current_charset = charset;
str->iso2022.current_half = half;
contain all 256 possible byte values and that are not to be
interpreted as being in any particular decoding. */
static void
-decode_coding_no_conversion (Lstream *decoding, CONST unsigned char *src,
+decode_coding_no_conversion (Lstream *decoding, const unsigned char *src,
unsigned_char_dynarr *dst, unsigned int n)
{
unsigned char c;
- unsigned int flags, ch;
- enum eol_type eol_type;
struct decoding_stream *str = DECODING_STREAM_DATA (decoding);
-
- CODING_STREAM_DECOMPOSE (str, flags, ch);
- eol_type = str->eol_type;
+ unsigned int flags = str->flags;
+ unsigned int ch = str->ch;
+ eol_type_t eol_type = str->eol_type;
while (n--)
{
DECODE_HANDLE_END_OF_CONVERSION (flags, ch, dst);
- CODING_STREAM_COMPOSE (str, flags, ch);
+ str->flags = flags;
+ str->ch = ch;
}
static void
-encode_coding_no_conversion (Lstream *encoding, CONST unsigned char *src,
+encode_coding_no_conversion (Lstream *encoding, const unsigned char *src,
unsigned_char_dynarr *dst, unsigned int n)
{
unsigned char c;
struct encoding_stream *str = ENCODING_STREAM_DATA (encoding);
- unsigned int flags, ch;
- enum eol_type eol_type;
-
- CODING_STREAM_DECOMPOSE (str, flags, ch);
- eol_type = CODING_SYSTEM_EOL_TYPE (str->codesys);
+ unsigned int flags = str->flags;
+ unsigned int ch = str->ch;
+ eol_type_t eol_type = CODING_SYSTEM_EOL_TYPE (str->codesys);
while (n--)
{
}
}
- CODING_STREAM_COMPOSE (str, flags, ch);
+ str->flags = flags;
+ str->ch = ch;
}
\f
-/************************************************************************/
-/* Simple internal/external functions */
-/************************************************************************/
-
-static Extbyte_dynarr *conversion_out_dynarr;
-static Bufbyte_dynarr *conversion_in_dynarr;
-
-/* Determine coding system from coding format */
-
-#define FILE_NAME_CODING_SYSTEM \
- ((NILP (Vfile_name_coding_system) || \
- (EQ ((Vfile_name_coding_system), Qbinary))) ? \
- Qnil : Fget_coding_system (Vfile_name_coding_system))
-
-/* #### not correct for all values of `fmt'! */
-#ifdef MULE
-#define FMT_CODING_SYSTEM(fmt) \
- (((fmt) == FORMAT_FILENAME) ? FILE_NAME_CODING_SYSTEM : \
- ((fmt) == FORMAT_CTEXT ) ? Fget_coding_system (Qctext) : \
- ((fmt) == FORMAT_TERMINAL) ? FILE_NAME_CODING_SYSTEM : \
- Qnil)
-#else
-#define FMT_CODING_SYSTEM(fmt) \
- (((fmt) == FORMAT_FILENAME) ? FILE_NAME_CODING_SYSTEM : \
- ((fmt) == FORMAT_TERMINAL) ? FILE_NAME_CODING_SYSTEM : \
- Qnil)
-#endif
-
-Extbyte *
-convert_to_external_format (CONST Bufbyte *ptr,
- Bytecount len,
- Extcount *len_out,
- enum external_data_format fmt)
-{
- Lisp_Object coding_system = FMT_CODING_SYSTEM (fmt);
-
- if (!conversion_out_dynarr)
- conversion_out_dynarr = Dynarr_new (Extbyte);
- else
- Dynarr_reset (conversion_out_dynarr);
-
- if (NILP (coding_system))
- {
- CONST Bufbyte *end = ptr + len;
-
- for (; ptr < end;)
- {
- Bufbyte c =
- (BYTE_ASCII_P (*ptr)) ? *ptr :
- (*ptr == LEADING_BYTE_CONTROL_1) ? (*(ptr+1) - 0x20) :
- (*ptr == LEADING_BYTE_LATIN_ISO8859_1) ? (*(ptr+1)) :
- '~';
-
- Dynarr_add (conversion_out_dynarr, (Extbyte) c);
- INC_CHARPTR (ptr);
- }
-
-#ifdef ERROR_CHECK_BUFPOS
- assert (ptr == end);
-#endif
- }
- else
- {
- Lisp_Object instream, outstream, da_outstream;
- Lstream *istr, *ostr;
- struct gcpro gcpro1, gcpro2, gcpro3;
- char tempbuf[1024]; /* some random amount */
-
- instream = make_fixed_buffer_input_stream ((unsigned char *) ptr, len);
- da_outstream = make_dynarr_output_stream
- ((unsigned_char_dynarr *) conversion_out_dynarr);
- outstream =
- make_encoding_output_stream (XLSTREAM (da_outstream), coding_system);
- istr = XLSTREAM (instream);
- ostr = XLSTREAM (outstream);
- GCPRO3 (instream, outstream, da_outstream);
- while (1)
- {
- int size_in_bytes = Lstream_read (istr, tempbuf, sizeof (tempbuf));
- if (!size_in_bytes)
- break;
- Lstream_write (ostr, tempbuf, size_in_bytes);
- }
- Lstream_close (istr);
- Lstream_close (ostr);
- UNGCPRO;
- Lstream_delete (istr);
- Lstream_delete (ostr);
- Lstream_delete (XLSTREAM (da_outstream));
- }
- *len_out = Dynarr_length (conversion_out_dynarr);
- Dynarr_add (conversion_out_dynarr, 0); /* remember to zero-terminate! */
- return Dynarr_atp (conversion_out_dynarr, 0);
-}
-
-Bufbyte *
-convert_from_external_format (CONST Extbyte *ptr,
- Extcount len,
- Bytecount *len_out,
- enum external_data_format fmt)
-{
- Lisp_Object coding_system = FMT_CODING_SYSTEM (fmt);
-
- if (!conversion_in_dynarr)
- conversion_in_dynarr = Dynarr_new (Bufbyte);
- else
- Dynarr_reset (conversion_in_dynarr);
-
- if (NILP (coding_system))
- {
- CONST Extbyte *end = ptr + len;
- for (; ptr < end; ptr++)
- {
- Extbyte c = *ptr;
- DECODE_ADD_BINARY_CHAR (c, conversion_in_dynarr);
- }
- }
- else
- {
- Lisp_Object instream, outstream, da_outstream;
- Lstream *istr, *ostr;
- struct gcpro gcpro1, gcpro2, gcpro3;
- char tempbuf[1024]; /* some random amount */
-
- instream = make_fixed_buffer_input_stream ((unsigned char *) ptr, len);
- da_outstream = make_dynarr_output_stream
- ((unsigned_char_dynarr *) conversion_in_dynarr);
- outstream =
- make_decoding_output_stream (XLSTREAM (da_outstream), coding_system);
- istr = XLSTREAM (instream);
- ostr = XLSTREAM (outstream);
- GCPRO3 (instream, outstream, da_outstream);
- while (1)
- {
- int size_in_bytes = Lstream_read (istr, tempbuf, sizeof (tempbuf));
- if (!size_in_bytes)
- break;
- Lstream_write (ostr, tempbuf, size_in_bytes);
- }
- Lstream_close (istr);
- Lstream_close (ostr);
- UNGCPRO;
- Lstream_delete (istr);
- Lstream_delete (ostr);
- Lstream_delete (XLSTREAM (da_outstream));
- }
-
- *len_out = Dynarr_length (conversion_in_dynarr);
- Dynarr_add (conversion_in_dynarr, 0); /* remember to zero-terminate! */
- return Dynarr_atp (conversion_in_dynarr, 0);
-}
-
-\f
/************************************************************************/
/* Initialization */
/************************************************************************/
void
-syms_of_mule_coding (void)
+syms_of_file_coding (void)
{
- defsymbol (&Qbuffer_file_coding_system, "buffer-file-coding-system");
deferror (&Qcoding_system_error, "coding-system-error",
"Coding-system error", Qio_error);
DEFSUBR (Fcoding_system_name);
DEFSUBR (Fmake_coding_system);
DEFSUBR (Fcopy_coding_system);
+ DEFSUBR (Fcoding_system_canonical_name_p);
+ DEFSUBR (Fcoding_system_alias_p);
+ DEFSUBR (Fcoding_system_aliasee);
+ DEFSUBR (Fdefine_coding_system_alias);
DEFSUBR (Fsubsidiary_coding_system);
DEFSUBR (Fcoding_system_type);
DEFSUBR (Fencode_shift_jis_char);
DEFSUBR (Fdecode_big5_char);
DEFSUBR (Fencode_big5_char);
+ DEFSUBR (Fset_ucs_char);
+ DEFSUBR (Fucs_char);
+ DEFSUBR (Fset_char_ucs);
+ DEFSUBR (Fchar_ucs);
#endif /* MULE */
- defsymbol (&Qcoding_system_p, "coding-system-p");
+ defsymbol (&Qcoding_systemp, "coding-system-p");
defsymbol (&Qno_conversion, "no-conversion");
+ defsymbol (&Qraw_text, "raw-text");
#ifdef MULE
defsymbol (&Qbig5, "big5");
defsymbol (&Qshift_jis, "shift-jis");
+ defsymbol (&Qucs4, "ucs-4");
+ defsymbol (&Qutf8, "utf-8");
defsymbol (&Qccl, "ccl");
defsymbol (&Qiso2022, "iso2022");
#endif /* MULE */
defsymbol (&Qdecode, "decode");
#ifdef MULE
- defsymbol (&Qctext, "ctext");
defsymbol (&coding_category_symbol[CODING_CATEGORY_SHIFT_JIS],
"shift-jis");
defsymbol (&coding_category_symbol[CODING_CATEGORY_BIG5],
"big5");
+ defsymbol (&coding_category_symbol[CODING_CATEGORY_UCS4],
+ "ucs-4");
+ defsymbol (&coding_category_symbol[CODING_CATEGORY_UTF8],
+ "utf-8");
defsymbol (&coding_category_symbol[CODING_CATEGORY_ISO_7],
"iso-7");
defsymbol (&coding_category_symbol[CODING_CATEGORY_ISO_8_DESIGNATE],
}
void
-lstream_type_create_mule_coding (void)
+lstream_type_create_file_coding (void)
{
LSTREAM_HAS_METHOD (decoding, reader);
LSTREAM_HAS_METHOD (decoding, writer);
}
void
-vars_of_mule_coding (void)
+vars_of_file_coding (void)
{
int i;
+ fcd = xnew (struct file_coding_dump);
+ dumpstruct (&fcd, &fcd_description);
+
/* Initialize to something reasonable ... */
for (i = 0; i <= CODING_CATEGORY_LAST; i++)
{
- coding_category_system[i] = Qnil;
- coding_category_by_priority[i] = i;
+ fcd->coding_category_system[i] = Qnil;
+ fcd->coding_category_by_priority[i] = i;
}
Fprovide (intern ("file-coding"));
Vterminal_coding_system = Qnil;
DEFVAR_LISP ("coding-system-for-read", &Vcoding_system_for_read /*
-Overriding coding system used when writing a file or process.
-You should *bind* this, not set it. If this is non-nil, it specifies
-the coding system that will be used when a file or process is read
-in, and overrides `buffer-file-coding-system-for-read',
+Overriding coding system used when reading from a file or process.
+You should bind this variable with `let', but do not set it globally.
+If this is non-nil, it specifies the coding system that will be used
+to decode input on read operations, such as from a file or process.
+It overrides `buffer-file-coding-system-for-read',
`insert-file-contents-pre-hook', etc. Use those variables instead of
-this one for permanent changes to the environment.
-*/ );
+this one for permanent changes to the environment. */ );
Vcoding_system_for_read = Qnil;
DEFVAR_LISP ("coding-system-for-write",
&Vcoding_system_for_write /*
-Overriding coding system used when writing a file or process.
-You should *bind* this, not set it. If this is non-nil, it specifies
-the coding system that will be used when a file or process is wrote
-in, and overrides `buffer-file-coding-system',
-`write-region-pre-hook', etc. Use those variables instead of this one
-for permanent changes to the environment.
-*/ );
+Overriding coding system used when writing to a file or process.
+You should bind this variable with `let', but do not set it globally.
+If this is non-nil, it specifies the coding system that will be used
+to encode output for write operations, such as to a file or process.
+It overrides `buffer-file-coding-system', `write-region-pre-hook', etc.
+Use those variables instead of this one for permanent changes to the
+environment. */ );
Vcoding_system_for_write = Qnil;
DEFVAR_LISP ("file-name-coding-system", &Vfile_name_coding_system /*
}
void
-complex_vars_of_mule_coding (void)
+complex_vars_of_file_coding (void)
{
- staticpro (&Vcoding_system_hashtable);
- Vcoding_system_hashtable = make_lisp_hashtable (50, HASHTABLE_NONWEAK,
- HASHTABLE_EQ);
+ staticpro (&Vcoding_system_hash_table);
+ Vcoding_system_hash_table =
+ make_lisp_hash_table (50, HASH_TABLE_NON_WEAK, HASH_TABLE_EQ);
the_codesys_prop_dynarr = Dynarr_new (codesys_prop);
+ dumpstruct (&the_codesys_prop_dynarr, &codesys_prop_dynarr_description);
#define DEFINE_CODESYS_PROP(Prop_Type, Sym) do \
{ \
DEFINE_CODESYS_PROP (CODESYS_PROP_CCL, Qdecode);
#endif /* MULE */
/* Need to create this here or we're really screwed. */
- Fmake_coding_system (Qno_conversion, Qno_conversion, build_string ("No conversion"),
- list2 (Qmnemonic, build_string ("Noconv")));
+ Fmake_coding_system
+ (Qraw_text, Qno_conversion,
+ build_string ("Raw text, which means it converts only line-break-codes."),
+ list2 (Qmnemonic, build_string ("Raw")));
+
+ Fmake_coding_system
+ (Qbinary, Qno_conversion,
+ build_string ("Binary, which means it does not convert anything."),
+ list4 (Qeol_type, Qlf,
+ Qmnemonic, build_string ("Binary")));
+
+ Fdefine_coding_system_alias (Qno_conversion, Qraw_text);
- Fcopy_coding_system (Fcoding_system_property (Qno_conversion, Qeol_lf),
- Qbinary);
+ Fdefine_coding_system_alias (Qfile_name, Qbinary);
+
+ Fdefine_coding_system_alias (Qterminal, Qbinary);
+ Fdefine_coding_system_alias (Qkeyboard, Qbinary);
/* Need this for bootstrapping */
- coding_category_system[CODING_CATEGORY_NO_CONVERSION] =
- Fget_coding_system (Qno_conversion);
+ fcd->coding_category_system[CODING_CATEGORY_NO_CONVERSION] =
+ Fget_coding_system (Qraw_text);
+
+#ifdef MULE
+ {
+ unsigned int i;
+
+ for (i = 0; i < 65536; i++)
+ fcd->ucs_to_mule_table[i] = Qnil;
+ }
+ staticpro (&mule_to_ucs_table);
+ mule_to_ucs_table = Fmake_char_table(Qgeneric);
+#endif /* MULE */
}
projects / chise / xemacs-chise.git.1 / blobdiff