/*** TABLE OF CONTENTS ***
+ 0. General comments
1. Preamble
2. Emacs' internal format (emacs-mule) handlers
3. ISO2022 handlers
*/
+/*** 0. General comments ***/
+
+
/*** GENERAL NOTE on CODING SYSTEM ***
Coding system is an encoding mechanism of one or more character
/*** GENERAL NOTES on `decode_coding_XXX ()' functions ***
- These functions decode SRC_BYTES length text at SOURCE encoded in
- CODING to Emacs' internal format (emacs-mule). The resulting text
- goes to a place pointed to by DESTINATION, the length of which
- should not exceed DST_BYTES. These functions set the information of
- original and decoded texts in the members produced, produced_char,
- consumed, and consumed_char of the structure *CODING.
+ These functions decode SRC_BYTES length of unibyte text at SOURCE
+ encoded in CODING to Emacs' internal format. The resulting
+ multibyte text goes to a place pointed to by DESTINATION, the length
+ of which should not exceed DST_BYTES.
- The return value is an integer (CODING_FINISH_XXX) indicating how
- the decoding finished.
+ These functions set the information of original and decoded texts in
+ the members produced, produced_char, consumed, and consumed_char of
+ the structure *CODING. They also set the member result to one of
+ CODING_FINISH_XXX indicating how the decoding finished.
DST_BYTES zero means that source area and destination area are
overlapped, which means that we can produce a decoded text until it
Below is a template of these functions. */
#if 0
+static void
decode_coding_XXX (coding, source, destination, src_bytes, dst_bytes)
struct coding_system *coding;
unsigned char *source, *destination;
/*** GENERAL NOTES on `encode_coding_XXX ()' functions ***
These functions encode SRC_BYTES length text at SOURCE of Emacs'
- internal format (emacs-mule) to CODING. The resulting text goes to
- a place pointed to by DESTINATION, the length of which should not
- exceed DST_BYTES. These functions set the information of
- original and encoded texts in the members produced, produced_char,
- consumed, and consumed_char of the structure *CODING.
+ internal multibyte format to CODING. The resulting unibyte text
+ goes to a place pointed to by DESTINATION, the length of which
+ should not exceed DST_BYTES.
- The return value is an integer (CODING_FINISH_XXX) indicating how
- the encoding finished.
+ These functions set the information of original and encoded texts in
+ the members produced, produced_char, consumed, and consumed_char of
+ the structure *CODING. They also set the member result to one of
+ CODING_FINISH_XXX indicating how the encoding finished.
DST_BYTES zero means that source area and destination area are
- overlapped, which means that we can produce a decoded text until it
- reaches at the head of not-yet-decoded source text.
+ overlapped, which means that we can produce a encoded text until it
+ reaches at the head of not-yet-encoded source text.
Below is a template of these functions. */
#if 0
+static void
encode_coding_XXX (coding, source, destination, src_bytes, dst_bytes)
struct coding_system *coding;
unsigned char *source, *destination;
/*** COMMONLY USED MACROS ***/
-/* The following three macros ONE_MORE_BYTE, TWO_MORE_BYTES, and
- THREE_MORE_BYTES safely get one, two, and three bytes from the
- source text respectively. If there are not enough bytes in the
- source, they jump to `label_end_of_loop'. The caller should set
- variables `src' and `src_end' to appropriate areas in advance. */
-
-#define ONE_MORE_BYTE(c1) \
- do { \
- if (src < src_end) \
- c1 = *src++; \
- else \
- goto label_end_of_loop; \
- } while (0)
+/* The following two macros ONE_MORE_BYTE and TWO_MORE_BYTES safely
+ get one, two, and three bytes from the source text respectively.
+ If there are not enough bytes in the source, they jump to
+ `label_end_of_loop'. The caller should set variables `coding',
+ `src' and `src_end' to appropriate pointer in advance. These
+ macros are called from decoding routines `decode_coding_XXX', thus
+ it is assumed that the source text is unibyte. */
-#define TWO_MORE_BYTES(c1, c2) \
- do { \
- if (src + 1 < src_end) \
- c1 = *src++, c2 = *src++; \
- else \
- goto label_end_of_loop; \
+#define ONE_MORE_BYTE(c1) \
+ do { \
+ if (src >= src_end) \
+ { \
+ coding->result = CODING_FINISH_INSUFFICIENT_SRC; \
+ goto label_end_of_loop; \
+ } \
+ c1 = *src++; \
} while (0)
-#define THREE_MORE_BYTES(c1, c2, c3) \
- do { \
- if (src + 2 < src_end) \
- c1 = *src++, c2 = *src++, c3 = *src++; \
- else \
- goto label_end_of_loop; \
+#define TWO_MORE_BYTES(c1, c2) \
+ do { \
+ if (src + 1 >= src_end) \
+ { \
+ coding->result = CODING_FINISH_INSUFFICIENT_SRC; \
+ goto label_end_of_loop; \
+ } \
+ c1 = *src++; \
+ c2 = *src++; \
} while (0)
-/* The following three macros DECODE_CHARACTER_ASCII,
- DECODE_CHARACTER_DIMENSION1, and DECODE_CHARACTER_DIMENSION2 put
- the multi-byte form of a character of each class at the place
- pointed by `dst'. The caller should set the variable `dst' to
- point to an appropriate area and the variable `coding' to point to
- the coding-system of the currently decoding text in advance. */
-/* Decode one ASCII character C. */
+/* Set C to the next character at the source text pointed by `src'.
+ If there are not enough characters in the source, jump to
+ `label_end_of_loop'. The caller should set variables `coding'
+ `src', `src_end', and `translation_table' to appropriate pointers
+ in advance. This macro is used in encoding routines
+ `encode_coding_XXX', thus it assumes that the source text is in
+ multibyte form except for 8-bit characters. 8-bit characters are
+ in multibyte form if coding->src_multibyte is nonzero, else they
+ are represented by a single byte. */
-#define DECODE_CHARACTER_ASCII(c) \
- do { \
- if (COMPOSING_P (coding->composing)) \
- { \
- *dst++ = 0xA0, *dst++ = (c) | 0x80; \
- coding->composed_chars++; \
- } \
- else \
- { \
- *dst++ = (c); \
- coding->produced_char++; \
- } \
+#define ONE_MORE_CHAR(c) \
+ do { \
+ int len = src_end - src; \
+ int bytes; \
+ if (len <= 0) \
+ { \
+ coding->result = CODING_FINISH_INSUFFICIENT_SRC; \
+ goto label_end_of_loop; \
+ } \
+ if (coding->src_multibyte \
+ || UNIBYTE_STR_AS_MULTIBYTE_P (src, len, bytes)) \
+ c = STRING_CHAR_AND_LENGTH (src, len, bytes); \
+ else \
+ c = *src, bytes = 1; \
+ if (!NILP (translation_table)) \
+ c = translate_char (translation_table, c, 0, 0, 0); \
+ src += bytes; \
} while (0)
-/* Decode one DIMENSION1 character whose charset is CHARSET and whose
- position-code is C. */
-#define DECODE_CHARACTER_DIMENSION1(charset, c) \
+/* Produce a multibyte form of characater C to `dst'. Jump to
+ `label_end_of_loop' if there's not enough space at `dst'.
+
+ If we are now in the middle of composition sequence, the decoded
+ character may be ALTCHAR (for the current composition). In that
+ case, the character goes to coding->cmp_data->data instead of
+ `dst'.
+
+ This macro is used in decoding routines. */
+
+#define EMIT_CHAR(c) \
do { \
- unsigned char leading_code = CHARSET_LEADING_CODE_BASE (charset); \
- if (COMPOSING_P (coding->composing)) \
+ if (! COMPOSING_P (coding) \
+ || coding->composing == COMPOSITION_RELATIVE \
+ || coding->composing == COMPOSITION_WITH_RULE) \
{ \
- *dst++ = leading_code + 0x20; \
- coding->composed_chars++; \
+ int bytes = CHAR_BYTES (c); \
+ if ((dst + bytes) > (dst_bytes ? dst_end : src)) \
+ { \
+ coding->result = CODING_FINISH_INSUFFICIENT_DST; \
+ goto label_end_of_loop; \
+ } \
+ dst += CHAR_STRING (c, dst); \
+ coding->produced_char++; \
} \
- else \
+ \
+ if (COMPOSING_P (coding) \
+ && coding->composing != COMPOSITION_RELATIVE) \
{ \
- *dst++ = leading_code; \
- coding->produced_char++; \
+ CODING_ADD_COMPOSITION_COMPONENT (coding, c); \
+ coding->composition_rule_follows \
+ = coding->composing != COMPOSITION_WITH_ALTCHARS; \
} \
- if (leading_code = CHARSET_LEADING_CODE_EXT (charset)) \
- *dst++ = leading_code; \
- *dst++ = (c) | 0x80; \
} while (0)
-/* Decode one DIMENSION2 character whose charset is CHARSET and whose
- position-codes are C1 and C2. */
-#define DECODE_CHARACTER_DIMENSION2(charset, c1, c2) \
- do { \
- DECODE_CHARACTER_DIMENSION1 (charset, c1); \
- *dst++ = (c2) | 0x80; \
+#define EMIT_ONE_BYTE(c) \
+ do { \
+ if (dst >= (dst_bytes ? dst_end : src)) \
+ { \
+ coding->result = CODING_FINISH_INSUFFICIENT_DST; \
+ goto label_end_of_loop; \
+ } \
+ *dst++ = c; \
+ } while (0)
+
+#define EMIT_TWO_BYTES(c1, c2) \
+ do { \
+ if (dst + 2 > (dst_bytes ? dst_end : src)) \
+ { \
+ coding->result = CODING_FINISH_INSUFFICIENT_DST; \
+ goto label_end_of_loop; \
+ } \
+ *dst++ = c1, *dst++ = c2; \
+ } while (0)
+
+#define EMIT_BYTES(from, to) \
+ do { \
+ if (dst + (to - from) > (dst_bytes ? dst_end : src)) \
+ { \
+ coding->result = CODING_FINISH_INSUFFICIENT_DST; \
+ goto label_end_of_loop; \
+ } \
+ while (from < to) \
+ *dst++ = *from++; \
} while (0)
\f
/*** 1. Preamble ***/
+#ifdef emacs
+#include <config.h>
+#endif
+
#include <stdio.h>
#ifdef emacs
-#include <config.h>
#include "lisp.h"
#include "buffer.h"
#include "charset.h"
+#include "composite.h"
#include "ccl.h"
#include "coding.h"
#include "window.h"
Lisp_Object Qpost_read_conversion, Qpre_write_conversion;
Lisp_Object Qno_conversion, Qundecided;
Lisp_Object Qcoding_system_history;
-Lisp_Object Qsafe_charsets;
+Lisp_Object Qsafe_chars;
Lisp_Object Qvalid_codes;
extern Lisp_Object Qinsert_file_contents, Qwrite_region;
Lisp_Object Vselect_safe_coding_system_function;
-/* Mnemonic character of each format of end-of-line. */
-int eol_mnemonic_unix, eol_mnemonic_dos, eol_mnemonic_mac;
-/* Mnemonic character to indicate format of end-of-line is not yet
+/* Mnemonic string for each format of end-of-line. */
+Lisp_Object eol_mnemonic_unix, eol_mnemonic_dos, eol_mnemonic_mac;
+/* Mnemonic string to indicate format of end-of-line is not yet
decided. */
-int eol_mnemonic_undecided;
+Lisp_Object eol_mnemonic_undecided;
/* Format of end-of-line decided by system. This is CODING_EOL_LF on
Unix, CODING_EOL_CRLF on DOS/Windows, and CODING_EOL_CR on Mac. */
/* Flag to inhibit code conversion of end-of-line format. */
int inhibit_eol_conversion;
+/* Flag to inhibit ISO2022 escape sequence detection. */
+int inhibit_iso_escape_detection;
+
/* Flag to make buffer-file-coding-system inherit from process-coding. */
int inherit_process_coding_system;
Lisp_Object Vprocess_coding_system_alist;
Lisp_Object Vnetwork_coding_system_alist;
+Lisp_Object Vlocale_coding_system;
+
#endif /* emacs */
Lisp_Object Qcoding_category, Qcoding_category_index;
"coding-category-iso-8-else",
"coding-category-ccl",
"coding-category-big5",
+ "coding-category-utf-8",
+ "coding-category-utf-16-be",
+ "coding-category-utf-16-le",
"coding-category-raw-text",
"coding-category-binary"
};
/* Default coding systems used for process I/O. */
Lisp_Object Vdefault_process_coding_system;
+/* Global flag to tell that we can't call post-read-conversion and
+ pre-write-conversion functions. Usually the value is zero, but it
+ is set to 1 temporarily while such functions are running. This is
+ to avoid infinite recursive call. */
+static int inhibit_pre_post_conversion;
+
+/* Char-table containing safe coding systems of each character. */
+Lisp_Object Vchar_coding_system_table;
+Lisp_Object Qchar_coding_system;
+
+/* Return `safe-chars' property of coding system CODING. Don't check
+ validity of CODING. */
+
+Lisp_Object
+coding_safe_chars (coding)
+ struct coding_system *coding;
+{
+ Lisp_Object coding_spec, plist, safe_chars;
+
+ coding_spec = Fget (coding->symbol, Qcoding_system);
+ plist = XVECTOR (coding_spec)->contents[3];
+ safe_chars = Fplist_get (XVECTOR (coding_spec)->contents[3], Qsafe_chars);
+ return (CHAR_TABLE_P (safe_chars) ? safe_chars : Qt);
+}
+
+#define CODING_SAFE_CHAR_P(safe_chars, c) \
+ (EQ (safe_chars, Qt) || !NILP (CHAR_TABLE_REF (safe_chars, c)))
+
\f
/*** 2. Emacs internal format (emacs-mule) handlers ***/
/* Emacs' internal format for encoding multiple character sets is a
kind of multi-byte encoding, i.e. characters are encoded by
- variable-length sequences of one-byte codes. ASCII characters
- and control characters (e.g. `tab', `newline') are represented by
- one-byte sequences which are their ASCII codes, in the range 0x00
- through 0x7F. The other characters are represented by a sequence
- of `base leading-code', optional `extended leading-code', and one
- or two `position-code's. The length of the sequence is determined
- by the base leading-code. Leading-code takes the range 0x80
- through 0x9F, whereas extended leading-code and position-code take
- the range 0xA0 through 0xFF. See `charset.h' for more details
- about leading-code and position-code.
-
- There's one exception to this rule. Special leading-code
- `leading-code-composition' denotes that the following several
- characters should be composed into one character. Leading-codes of
- components (except for ASCII) are added 0x20. An ASCII character
- component is represented by a 2-byte sequence of `0xA0' and
- `ASCII-code + 0x80'. See also the comments in `charset.h' for the
- details of composite character. Hence, we can summarize the code
- range as follows:
+ variable-length sequences of one-byte codes.
+
+ ASCII characters and control characters (e.g. `tab', `newline') are
+ represented by one-byte sequences which are their ASCII codes, in
+ the range 0x00 through 0x7F.
+
+ 8-bit characters of the range 0x80..0x9F are represented by
+ two-byte sequences of LEADING_CODE_8_BIT_CONTROL and (their 8-bit
+ code + 0x20).
+
+ 8-bit characters of the range 0xA0..0xFF are represented by
+ one-byte sequences which are their 8-bit code.
+
+ The other characters are represented by a sequence of `base
+ leading-code', optional `extended leading-code', and one or two
+ `position-code's. The length of the sequence is determined by the
+ base leading-code. Leading-code takes the range 0x80 through 0x9F,
+ whereas extended leading-code and position-code take the range 0xA0
+ through 0xFF. See `charset.h' for more details about leading-code
+ and position-code.
--- CODE RANGE of Emacs' internal format ---
- (character set) (range)
- ASCII 0x00 .. 0x7F
- ELSE (1st byte) 0x80 .. 0x9F
- (rest bytes) 0xA0 .. 0xFF
+ character set range
+ ------------- -----
+ ascii 0x00..0x7F
+ eight-bit-control LEADING_CODE_8_BIT_CONTROL + 0xA0..0xBF
+ eight-bit-graphic 0xA0..0xBF
+ ELSE 0x81..0x9F + [0xA0..0xFF]+
---------------------------------------------
*/
enum emacs_code_class_type emacs_code_class[256];
-/* Go to the next statement only if *SRC is accessible and the code is
- greater than 0xA0. */
-#define CHECK_CODE_RANGE_A0_FF \
- do { \
- if (src >= src_end) \
- goto label_end_of_switch; \
- else if (*src++ < 0xA0) \
- return 0; \
- } while (0)
-
/* See the above "GENERAL NOTES on `detect_coding_XXX ()' functions".
Check if a text is encoded in Emacs' internal format. If it is,
return CODING_CATEGORY_MASK_EMACS_MULE, else return 0. */
int
detect_coding_emacs_mule (src, src_end)
- unsigned char *src, *src_end;
+ unsigned char *src, *src_end;
{
unsigned char c;
int composing = 0;
+ /* Dummy for ONE_MORE_BYTE. */
+ struct coding_system dummy_coding;
+ struct coding_system *coding = &dummy_coding;
- while (src < src_end)
+ while (1)
{
- c = *src++;
+ ONE_MORE_BYTE (c);
if (composing)
{
if (c < 0xA0)
composing = 0;
+ else if (c == 0xA0)
+ {
+ ONE_MORE_BYTE (c);
+ c &= 0x7F;
+ }
else
c -= 0x20;
}
- switch (emacs_code_class[c])
+ if (c < 0x20)
{
- case EMACS_ascii_code:
- case EMACS_linefeed_code:
- break;
-
- case EMACS_control_code:
if (c == ISO_CODE_ESC || c == ISO_CODE_SI || c == ISO_CODE_SO)
return 0;
- break;
+ }
+ else if (c >= 0x80 && c < 0xA0)
+ {
+ if (c == 0x80)
+ /* Old leading code for a composite character. */
+ composing = 1;
+ else
+ {
+ unsigned char *src_base = src - 1;
+ int bytes;
- case EMACS_invalid_code:
- return 0;
+ if (!UNIBYTE_STR_AS_MULTIBYTE_P (src_base, src_end - src_base,
+ bytes))
+ return 0;
+ src = src_base + bytes;
+ }
+ }
+ }
+ label_end_of_loop:
+ return CODING_CATEGORY_MASK_EMACS_MULE;
+}
- case EMACS_leading_code_composition: /* c == 0x80 */
- if (composing)
- CHECK_CODE_RANGE_A0_FF;
- else
- composing = 1;
- break;
- case EMACS_leading_code_4:
- CHECK_CODE_RANGE_A0_FF;
- /* fall down to check it two more times ... */
+/* See the above "GENERAL NOTES on `decode_coding_XXX ()' functions". */
- case EMACS_leading_code_3:
- CHECK_CODE_RANGE_A0_FF;
- /* fall down to check it one more time ... */
+static void
+decode_coding_emacs_mule (coding, source, destination, src_bytes, dst_bytes)
+ struct coding_system *coding;
+ unsigned char *source, *destination;
+ int src_bytes, dst_bytes;
+{
+ unsigned char *src = source;
+ unsigned char *src_end = source + src_bytes;
+ unsigned char *dst = destination;
+ unsigned char *dst_end = destination + dst_bytes;
+ /* SRC_BASE remembers the start position in source in each loop.
+ The loop will be exited when there's not enough source code, or
+ when there's not enough destination area to produce a
+ character. */
+ unsigned char *src_base;
- case EMACS_leading_code_2:
- CHECK_CODE_RANGE_A0_FF;
- break;
+ coding->produced_char = 0;
+ while ((src_base = src) < src_end)
+ {
+ unsigned char tmp[MAX_MULTIBYTE_LENGTH], *p;
+ int bytes;
- default:
- label_end_of_switch:
+ if (UNIBYTE_STR_AS_MULTIBYTE_P (src, src_end - src, bytes))
+ {
+ p = src;
+ src += bytes;
+ }
+ else
+ {
+ bytes = CHAR_STRING (*src, tmp);
+ p = tmp;
+ src++;
+ }
+ if (dst + bytes >= (dst_bytes ? dst_end : src))
+ {
+ coding->result = CODING_FINISH_INSUFFICIENT_DST;
break;
}
+ while (bytes--) *dst++ = *p++;
+ coding->produced_char++;
}
- return CODING_CATEGORY_MASK_EMACS_MULE;
+ coding->consumed = coding->consumed_char = src_base - source;
+ coding->produced = dst - destination;
}
+#define encode_coding_emacs_mule(coding, source, destination, src_bytes, dst_bytes) \
+ encode_eol (coding, source, destination, src_bytes, dst_bytes)
+
+
\f
/*** 3. ISO2022 handlers ***/
/* The following note describes the coding system ISO2022 briefly.
- Since the intention of this note is to help in understanding of
- the programs in this file, some parts are NOT ACCURATE or OVERLY
- SIMPLIFIED. For the thorough understanding, please refer to the
+ 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 environment. If one chooses 7-bite environment,
- all text is encoded by codes of less than 128. This may make the
- encoded text a little bit longer, but the text gets more stability
- to pass through several gateways (some of them strip off the MSB).
+ in 7-bit and 8-bit environments. For 7-bite 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 set: control character set and
+ 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 provided also by escape sequences). The latter
- contains graphic characters such as ' A' and '-'. Emacs recognizes
+ 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, DIMENSION1_CHARS94, DIMENSION1_CHARS96,
- DIMENSION2_CHARS94, DIMENSION2_CHARS96 according to the number of
- bytes (DIMENSION) and the number of characters in one dimension
- (CHARS) of the set. In addition, each character set is assigned an
- identification tag (called "final character" and denoted as <F>
- here after) which is unique in each class. <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).
+ 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
o DIMENSION2_CHARS94 -- GB2312('A'), JISX0208('B'), ...
o DIMENSION2_CHARS96 -- none for the moment
- A code area (1byte=8bits) is divided into 4 areas, C0, GL, C1, and GR.
+ 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 a coded text. With
- 7-bit environment, only C0 can be used, and a control character for
- C1 is encoded by an appropriate escape sequence to fit in the
- environment. All control characters for C1 are defined the
- corresponding escape sequences.
+ 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, and usually
- these invocations and designations are omitted in a coded text.
- With 7-bit environment, only GL can be used.
+ 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, code 0x20
- and 0x7F of GL area work as control characters SPACE and DEL
- respectively, and code 0xA0 and 0xFF of GR area should not 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 works only
- for the following character and doesn't affect locking-shift.
- Invocations are done by the following control characters or escape
- sequences.
+ 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:
----------------------------------------------------------------------
- function control char escape sequence description
+ abbrev function cntrl escape seq description
----------------------------------------------------------------------
- SI (shift-in) 0x0F none invoke G0 to GL
- SO (shift-out) 0x0E none invoke G1 to GL
- LS2 (locking-shift-2) none ESC 'n' invoke G2 into GL
- LS3 (locking-shift-3) none ESC 'o' invoke G3 into GL
- SS2 (single-shift-2) 0x8E ESC 'N' invoke G2 into GL
- SS3 (single-shift-3) 0x8F ESC 'O' invoke G3 into GL
+ 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
----------------------------------------------------------------------
- The first four are for locking-shift. Control characters for these
- functions are defined by macros ISO_CODE_XXX in `coding.h'.
+ (*) 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.
+ Designations are done by the following escape sequences:
----------------------------------------------------------------------
escape sequence description
----------------------------------------------------------------------
----------------------------------------------------------------------
In this list, "DIMENSION1_CHARS94<F>" means a graphic character set
- of dimension 1, chars 94, and final character <F>, and etc.
+ 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
- CHARS96 character set in a coding system which is characterized as
+ 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 exists many
- coding systems such as Compound Text (used in X'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
+ same multilingual text in ISO2022. Actually, there exist many
+ 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.
In addition to the above, Emacs handles two more kinds of escape
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.
-
- Character composition specification takes the following format:
- 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 meaning is restricted to Emacs only. */
+ abbreviated to the escape sequence ESC '[' in a 7-bit environment.
+
+ Character composition specification takes the following form:
+ o ESC '0' -- start relative composition
+ o ESC '1' -- end composition
+ o ESC '2' -- start rule-base composition (*)
+ o ESC '3' -- start relative composition with alternate chars (**)
+ o ESC '4' -- start rule-base composition with alternate chars (**)
+ Since these are not standard escape sequences of any ISO standard,
+ the use of them for these meaning is restricted to Emacs only.
+
+ (*) This form is used only in Emacs 20.5 and the older versions,
+ but the newer versions can safely decode it.
+ (**) This form is used only in Emacs 21.1 and the newer versions,
+ and the older versions can't decode it.
+
+ Here's a list of examples usages of these composition escape
+ sequences (categorized by `enum composition_method').
+
+ COMPOSITION_RELATIVE:
+ ESC 0 CHAR [ CHAR ] ESC 1
+ COMPOSITOIN_WITH_RULE:
+ ESC 2 CHAR [ RULE CHAR ] ESC 1
+ COMPOSITION_WITH_ALTCHARS:
+ ESC 3 ALTCHAR [ ALTCHAR ] ESC 0 CHAR [ CHAR ] ESC 1
+ COMPOSITION_WITH_RULE_ALTCHARS:
+ ESC 4 ALTCHAR [ RULE ALTCHAR ] ESC 0 CHAR [ CHAR ] ESC 1 */
enum iso_code_class_type iso_code_class[256];
-#define CHARSET_OK(idx, charset) \
- (coding_system_table[idx] \
- && (coding_system_table[idx]->safe_charsets[charset] \
- || (CODING_SPEC_ISO_REQUESTED_DESIGNATION \
- (coding_system_table[idx], charset) \
- != CODING_SPEC_ISO_NO_REQUESTED_DESIGNATION)))
+#define CHARSET_OK(idx, charset, c) \
+ (coding_system_table[idx] \
+ && (charset == CHARSET_ASCII \
+ || (safe_chars = coding_safe_chars (coding_system_table[idx]), \
+ CODING_SAFE_CHAR_P (safe_chars, c))) \
+ && (CODING_SPEC_ISO_REQUESTED_DESIGNATION (coding_system_table[idx], \
+ charset) \
+ != CODING_SPEC_ISO_NO_REQUESTED_DESIGNATION))
#define SHIFT_OUT_OK(idx) \
(CODING_SPEC_ISO_INITIAL_DESIGNATION (coding_system_table[idx], 1) >= 0)
int mask_found = 0;
int reg[4], shift_out = 0, single_shifting = 0;
int c, c1, i, charset;
+ /* Dummy for ONE_MORE_BYTE. */
+ struct coding_system dummy_coding;
+ struct coding_system *coding = &dummy_coding;
+ Lisp_Object safe_chars;
reg[0] = CHARSET_ASCII, reg[1] = reg[2] = reg[3] = -1;
while (mask && src < src_end)
{
- c = *src++;
+ ONE_MORE_BYTE (c);
switch (c)
{
case ISO_CODE_ESC:
- single_shifting = 0;
- if (src >= src_end)
+ if (inhibit_iso_escape_detection)
break;
- c = *src++;
+ single_shifting = 0;
+ ONE_MORE_BYTE (c);
if (c >= '(' && c <= '/')
{
/* Designation sequence for a charset of dimension 1. */
- if (src >= src_end)
- break;
- c1 = *src++;
+ ONE_MORE_BYTE (c1);
if (c1 < ' ' || c1 >= 0x80
|| (charset = iso_charset_table[0][c >= ','][c1]) < 0)
/* Invalid designation sequence. Just ignore. */
else if (c == '$')
{
/* Designation sequence for a charset of dimension 2. */
- if (src >= src_end)
- break;
- c = *src++;
+ ONE_MORE_BYTE (c);
if (c >= '@' && c <= 'B')
/* Designation for JISX0208.1978, GB2312, or JISX0208. */
reg[0] = charset = iso_charset_table[1][0][c];
else if (c >= '(' && c <= '/')
{
- if (src >= src_end)
- break;
- c1 = *src++;
+ ONE_MORE_BYTE (c1);
if (c1 < ' ' || c1 >= 0x80
|| (charset = iso_charset_table[1][c >= ','][c1]) < 0)
/* Invalid designation sequence. Just ignore. */
mask &= CODING_CATEGORY_MASK_ISO_7_ELSE;
break;
}
- else if (c == '0' || c == '1' || c == '2')
- /* ESC <Fp> for start/end composition. Just ignore. */
- break;
+ else if (c >= '0' && c <= '4')
+ {
+ /* ESC <Fp> for start/end composition. */
+ mask_found |= CODING_CATEGORY_MASK_ISO;
+ break;
+ }
else
/* Invalid escape sequence. Just ignore. */
break;
/* We found a valid designation sequence for CHARSET. */
mask &= ~CODING_CATEGORY_MASK_ISO_8BIT;
- if (CHARSET_OK (CODING_CATEGORY_IDX_ISO_7, charset))
+ c = MAKE_CHAR (charset, 0, 0);
+ if (CHARSET_OK (CODING_CATEGORY_IDX_ISO_7, charset, c))
mask_found |= CODING_CATEGORY_MASK_ISO_7;
else
mask &= ~CODING_CATEGORY_MASK_ISO_7;
- if (CHARSET_OK (CODING_CATEGORY_IDX_ISO_7_TIGHT, charset))
+ if (CHARSET_OK (CODING_CATEGORY_IDX_ISO_7_TIGHT, charset, c))
mask_found |= CODING_CATEGORY_MASK_ISO_7_TIGHT;
else
mask &= ~CODING_CATEGORY_MASK_ISO_7_TIGHT;
- if (CHARSET_OK (CODING_CATEGORY_IDX_ISO_7_ELSE, charset))
+ if (CHARSET_OK (CODING_CATEGORY_IDX_ISO_7_ELSE, charset, c))
mask_found |= CODING_CATEGORY_MASK_ISO_7_ELSE;
else
mask &= ~CODING_CATEGORY_MASK_ISO_7_ELSE;
- if (CHARSET_OK (CODING_CATEGORY_IDX_ISO_8_ELSE, charset))
+ if (CHARSET_OK (CODING_CATEGORY_IDX_ISO_8_ELSE, charset, c))
mask_found |= CODING_CATEGORY_MASK_ISO_8_ELSE;
else
mask &= ~CODING_CATEGORY_MASK_ISO_8_ELSE;
break;
case ISO_CODE_SO:
+ if (inhibit_iso_escape_detection)
+ break;
single_shifting = 0;
if (shift_out == 0
&& (reg[1] >= 0
break;
case ISO_CODE_SI:
+ if (inhibit_iso_escape_detection)
+ break;
single_shifting = 0;
if (shift_out == 1)
{
{
int newmask = CODING_CATEGORY_MASK_ISO_8_ELSE;
+ if (inhibit_iso_escape_detection)
+ break;
if (c != ISO_CODE_CSI)
{
if (coding_system_table[CODING_CATEGORY_IDX_ISO_8_1]->flags
}
else
{
- unsigned char *src_begin = src;
-
mask &= ~(CODING_CATEGORY_MASK_ISO_7BIT
| CODING_CATEGORY_MASK_ISO_7_ELSE);
mask_found |= CODING_CATEGORY_MASK_ISO_8_1;
0xA0..0FF. If the byte length is odd, we exclude
CODING_CATEGORY_MASK_ISO_8_2. We can check this only
when we are not single shifting. */
- if (!single_shifting)
+ if (!single_shifting
+ && mask & CODING_CATEGORY_MASK_ISO_8_2)
{
- while (src < src_end && *src >= 0xA0)
- src++;
- if ((src - src_begin - 1) & 1 && src < src_end)
+ int i = 1;
+ while (src < src_end)
+ {
+ ONE_MORE_BYTE (c);
+ if (c < 0xA0)
+ break;
+ i++;
+ }
+
+ if (i & 1 && src < src_end)
mask &= ~CODING_CATEGORY_MASK_ISO_8_2;
else
mask_found |= CODING_CATEGORY_MASK_ISO_8_2;
break;
}
}
-
+ label_end_of_loop:
return (mask & mask_found);
}
-/* Decode a character of which charset is CHARSET and the 1st position
- code is C1. If dimension of CHARSET is 2, the 2nd position code is
- fetched from SRC and set to C2. If CHARSET is negative, it means
- that we are decoding ill formed text, and what we can do is just to
- read C1 as is. */
+/* Decode a character of which charset is CHARSET, the 1st position
+ code is C1, the 2nd position code is C2, and return the decoded
+ character code. If the variable `translation_table' is non-nil,
+ returned the translated code. */
-#define DECODE_ISO_CHARACTER(charset, c1) \
- do { \
- int c_alt, charset_alt = (charset); \
- if (COMPOSING_HEAD_P (coding->composing)) \
- { \
- *dst++ = LEADING_CODE_COMPOSITION; \
- if (COMPOSING_WITH_RULE_P (coding->composing)) \
- /* To tell composition rules are embeded. */ \
- *dst++ = 0xFF; \
- coding->composing += 2; \
- } \
- if (charset_alt >= 0) \
- { \
- if (CHARSET_DIMENSION (charset_alt) == 2) \
- { \
- ONE_MORE_BYTE (c2); \
- if (iso_code_class[(c2) & 0x7F] != ISO_0x20_or_0x7F \
- && iso_code_class[(c2) & 0x7F] != ISO_graphic_plane_0) \
- { \
- src--; \
- charset_alt = CHARSET_ASCII; \
- } \
- } \
- if (!NILP (translation_table) \
- && ((c_alt = translate_char (translation_table, \
- -1, charset_alt, c1, c2)) >= 0)) \
- SPLIT_CHAR (c_alt, charset_alt, c1, c2); \
- } \
- if (charset_alt == CHARSET_ASCII || charset_alt < 0) \
- DECODE_CHARACTER_ASCII (c1); \
- else if (CHARSET_DIMENSION (charset_alt) == 1) \
- DECODE_CHARACTER_DIMENSION1 (charset_alt, c1); \
- else \
- DECODE_CHARACTER_DIMENSION2 (charset_alt, c1, c2); \
- if (COMPOSING_WITH_RULE_P (coding->composing)) \
- /* To tell a composition rule follows. */ \
- coding->composing = COMPOSING_WITH_RULE_RULE; \
- } while (0)
+#define DECODE_ISO_CHARACTER(charset, c1, c2) \
+ (NILP (translation_table) \
+ ? MAKE_CHAR (charset, c1, c2) \
+ : translate_char (translation_table, -1, charset, c1, c2))
/* Set designation state into CODING. */
#define DECODE_DESIGNATION(reg, dimension, chars, final_char) \
do { \
- int charset = ISO_CHARSET_TABLE (make_number (dimension), \
- make_number (chars), \
- make_number (final_char)); \
+ int charset, c; \
+ \
+ if (final_char < '0' || final_char >= 128) \
+ goto label_invalid_code; \
+ charset = ISO_CHARSET_TABLE (make_number (dimension), \
+ make_number (chars), \
+ make_number (final_char)); \
+ c = MAKE_CHAR (charset, 0, 0); \
if (charset >= 0 \
&& (CODING_SPEC_ISO_REQUESTED_DESIGNATION (coding, charset) == reg \
- || coding->safe_charsets[charset])) \
+ || CODING_SAFE_CHAR_P (safe_chars, c))) \
{ \
if (coding->spec.iso2022.last_invalid_designation_register == 0 \
&& reg == 0 \
} \
} while (0)
-/* Return 0 if there's a valid composing sequence starting at SRC and
- ending before SRC_END, else return -1. */
+/* Allocate a memory block for storing information about compositions.
+ The block is chained to the already allocated blocks. */
-int
-check_composing_code (coding, src, src_end)
+void
+coding_allocate_composition_data (coding, char_offset)
struct coding_system *coding;
- unsigned char *src, *src_end;
+ int char_offset;
{
- int charset, c, c1, dim;
+ struct composition_data *cmp_data
+ = (struct composition_data *) xmalloc (sizeof *cmp_data);
+
+ cmp_data->char_offset = char_offset;
+ cmp_data->used = 0;
+ cmp_data->prev = coding->cmp_data;
+ cmp_data->next = NULL;
+ if (coding->cmp_data)
+ coding->cmp_data->next = cmp_data;
+ coding->cmp_data = cmp_data;
+ coding->cmp_data_start = 0;
+}
- while (src < src_end)
- {
- c = *src++;
- if (c >= 0x20)
- continue;
- if (c != ISO_CODE_ESC || src >= src_end)
- return -1;
- c = *src++;
- if (c == '1') /* end of compsition */
- return 0;
- if (src + 2 >= src_end
- || !coding->flags & CODING_FLAG_ISO_DESIGNATION)
- return -1;
-
- dim = (c == '$');
- if (dim == 1)
- c = (*src >= '@' && *src <= 'B') ? '(' : *src++;
- if (c >= '(' && c <= '/')
- {
- c1 = *src++;
- if ((c1 < ' ' || c1 >= 0x80)
- || (charset = iso_charset_table[dim][c >= ','][c1]) < 0
- || ! coding->safe_charsets[charset]
- || (CODING_SPEC_ISO_REQUESTED_DESIGNATION (coding, charset)
- == CODING_SPEC_ISO_NO_REQUESTED_DESIGNATION))
- return -1;
- }
- else
- return -1;
- }
+/* Record the starting position START and METHOD of one composition. */
+
+#define CODING_ADD_COMPOSITION_START(coding, start, method) \
+ do { \
+ struct composition_data *cmp_data = coding->cmp_data; \
+ int *data = cmp_data->data + cmp_data->used; \
+ coding->cmp_data_start = cmp_data->used; \
+ data[0] = -1; \
+ data[1] = cmp_data->char_offset + start; \
+ data[3] = (int) method; \
+ cmp_data->used += 4; \
+ } while (0)
+
+/* Record the ending position END of the current composition. */
+
+#define CODING_ADD_COMPOSITION_END(coding, end) \
+ do { \
+ struct composition_data *cmp_data = coding->cmp_data; \
+ int *data = cmp_data->data + coding->cmp_data_start; \
+ data[0] = cmp_data->used - coding->cmp_data_start; \
+ data[2] = cmp_data->char_offset + end; \
+ } while (0)
+
+/* Record one COMPONENT (alternate character or composition rule). */
+
+#define CODING_ADD_COMPOSITION_COMPONENT(coding, component) \
+ (coding->cmp_data->data[coding->cmp_data->used++] = component)
+
+/* Handle compositoin start sequence ESC 0, ESC 2, ESC 3, or ESC 4. */
+
+#define DECODE_COMPOSITION_START(c1) \
+ do { \
+ if (coding->composing == COMPOSITION_DISABLED) \
+ { \
+ *dst++ = ISO_CODE_ESC; \
+ *dst++ = c1 & 0x7f; \
+ coding->produced_char += 2; \
+ } \
+ else if (!COMPOSING_P (coding)) \
+ { \
+ /* This is surely the start of a composition. We must be sure \
+ that coding->cmp_data has enough space to store the \
+ information about the composition. If not, terminate the \
+ current decoding loop, allocate one more memory block for \
+ coding->cmp_data in the calller, then start the decoding \
+ loop again. We can't allocate memory here directly because \
+ it may cause buffer/string relocation. */ \
+ if (!coding->cmp_data \
+ || (coding->cmp_data->used + COMPOSITION_DATA_MAX_BUNCH_LENGTH \
+ >= COMPOSITION_DATA_SIZE)) \
+ { \
+ coding->result = CODING_FINISH_INSUFFICIENT_CMP; \
+ goto label_end_of_loop; \
+ } \
+ coding->composing = (c1 == '0' ? COMPOSITION_RELATIVE \
+ : c1 == '2' ? COMPOSITION_WITH_RULE \
+ : c1 == '3' ? COMPOSITION_WITH_ALTCHARS \
+ : COMPOSITION_WITH_RULE_ALTCHARS); \
+ CODING_ADD_COMPOSITION_START (coding, coding->produced_char, \
+ coding->composing); \
+ coding->composition_rule_follows = 0; \
+ } \
+ else \
+ { \
+ /* We are already handling a composition. If the method is \
+ the following two, the codes following the current escape \
+ sequence are actual characters stored in a buffer. */ \
+ if (coding->composing == COMPOSITION_WITH_ALTCHARS \
+ || coding->composing == COMPOSITION_WITH_RULE_ALTCHARS) \
+ { \
+ coding->composing = COMPOSITION_RELATIVE; \
+ coding->composition_rule_follows = 0; \
+ } \
+ } \
+ } while (0)
+
+/* Handle compositoin end sequence ESC 1. */
+
+#define DECODE_COMPOSITION_END(c1) \
+ do { \
+ if (coding->composing == COMPOSITION_DISABLED) \
+ { \
+ *dst++ = ISO_CODE_ESC; \
+ *dst++ = c1; \
+ coding->produced_char += 2; \
+ } \
+ else \
+ { \
+ CODING_ADD_COMPOSITION_END (coding, coding->produced_char); \
+ coding->composing = COMPOSITION_NO; \
+ } \
+ } while (0)
+
+/* Decode a composition rule from the byte C1 (and maybe one more byte
+ from SRC) and store one encoded composition rule in
+ coding->cmp_data. */
+
+#define DECODE_COMPOSITION_RULE(c1) \
+ do { \
+ int rule = 0; \
+ (c1) -= 32; \
+ if (c1 < 81) /* old format (before ver.21) */ \
+ { \
+ int gref = (c1) / 9; \
+ int nref = (c1) % 9; \
+ if (gref == 4) gref = 10; \
+ if (nref == 4) nref = 10; \
+ rule = COMPOSITION_ENCODE_RULE (gref, nref); \
+ } \
+ else if (c1 < 93) /* new format (after ver.21) */ \
+ { \
+ ONE_MORE_BYTE (c2); \
+ rule = COMPOSITION_ENCODE_RULE (c1 - 81, c2 - 32); \
+ } \
+ CODING_ADD_COMPOSITION_COMPONENT (coding, rule); \
+ coding->composition_rule_follows = 0; \
+ } while (0)
- /* We have not found the sequence "ESC 1". */
- return -1;
-}
/* See the above "GENERAL NOTES on `decode_coding_XXX ()' functions". */
-int
+static void
decode_coding_iso2022 (coding, source, destination, src_bytes, dst_bytes)
struct coding_system *coding;
unsigned char *source, *destination;
unsigned char *src_end = source + src_bytes;
unsigned char *dst = destination;
unsigned char *dst_end = destination + dst_bytes;
- /* Since the maximum bytes produced by each loop is 7, we subtract 6
- from DST_END to assure that overflow checking is necessary only
- at the head of loop. */
- unsigned char *adjusted_dst_end = dst_end - 6;
- int charset;
/* Charsets invoked to graphic plane 0 and 1 respectively. */
int charset0 = CODING_SPEC_ISO_PLANE_CHARSET (coding, 0);
int charset1 = CODING_SPEC_ISO_PLANE_CHARSET (coding, 1);
- Lisp_Object translation_table
- = coding->translation_table_for_decode;
- int result = CODING_FINISH_NORMAL;
+ /* SRC_BASE remembers the start position in source in each loop.
+ The loop will be exited when there's not enough source code
+ (within macro ONE_MORE_BYTE), or when there's not enough
+ destination area to produce a character (within macro
+ EMIT_CHAR). */
+ unsigned char *src_base;
+ int c, charset;
+ Lisp_Object translation_table;
+ Lisp_Object safe_chars;
- if (!NILP (Venable_character_translation) && NILP (translation_table))
- translation_table = Vstandard_translation_table_for_decode;
+ safe_chars = coding_safe_chars (coding);
- coding->produced_char = 0;
- coding->composed_chars = 0;
- coding->fake_multibyte = 0;
- while (src < src_end && (dst_bytes
- ? (dst < adjusted_dst_end)
- : (dst < src - 6)))
+ if (NILP (Venable_character_translation))
+ translation_table = Qnil;
+ else
+ {
+ translation_table = coding->translation_table_for_decode;
+ if (NILP (translation_table))
+ translation_table = Vstandard_translation_table_for_decode;
+ }
+
+ coding->result = CODING_FINISH_NORMAL;
+
+ while (1)
{
- /* SRC_BASE remembers the start position in source in each loop.
- The loop will be exited when there's not enough source text
- to analyze long escape sequence or 2-byte code (within macros
- ONE_MORE_BYTE or TWO_MORE_BYTES). In that case, SRC is reset
- to SRC_BASE before exiting. */
- unsigned char *src_base = src;
- int c1 = *src++, c2;
+ int c1, c2;
+ src_base = src;
+ ONE_MORE_BYTE (c1);
+
+ /* We produce no character or one character. */
switch (iso_code_class [c1])
{
case ISO_0x20_or_0x7F:
- if (!coding->composing
- && (charset0 < 0 || CHARSET_CHARS (charset0) == 94))
+ if (COMPOSING_P (coding) && coding->composition_rule_follows)
+ {
+ DECODE_COMPOSITION_RULE (c1);
+ continue;
+ }
+ if (charset0 < 0 || CHARSET_CHARS (charset0) == 94)
{
/* This is SPACE or DEL. */
- *dst++ = c1;
- coding->produced_char++;
+ charset = CHARSET_ASCII;
break;
}
/* This is a graphic character, we fall down ... */
case ISO_graphic_plane_0:
- if (coding->composing == COMPOSING_WITH_RULE_RULE)
+ if (COMPOSING_P (coding) && coding->composition_rule_follows)
{
- /* This is a composition rule. */
- *dst++ = c1 | 0x80;
- coding->composing = COMPOSING_WITH_RULE_TAIL;
+ DECODE_COMPOSITION_RULE (c1);
+ continue;
}
- else
- DECODE_ISO_CHARACTER (charset0, c1);
+ charset = charset0;
break;
case ISO_0xA0_or_0xFF:
/* This is a graphic character, we fall down ... */
case ISO_graphic_plane_1:
- if (coding->flags & CODING_FLAG_ISO_SEVEN_BITS)
+ if (charset1 < 0)
goto label_invalid_code;
- else
- DECODE_ISO_CHARACTER (charset1, c1);
+ charset = charset1;
break;
- case ISO_control_code:
+ case ISO_control_0:
+ if (COMPOSING_P (coding))
+ DECODE_COMPOSITION_END ('1');
+
/* All ISO2022 control characters in this class have the
same representation in Emacs internal format. */
if (c1 == '\n'
&& (coding->eol_type == CODING_EOL_CR
|| coding->eol_type == CODING_EOL_CRLF))
{
- result = CODING_FINISH_INCONSISTENT_EOL;
- goto label_end_of_loop_2;
+ coding->result = CODING_FINISH_INCONSISTENT_EOL;
+ goto label_end_of_loop;
}
- *dst++ = c1;
- coding->produced_char++;
- if (c1 >= 0x80)
- coding->fake_multibyte = 1;
+ charset = CHARSET_ASCII;
break;
+ case ISO_control_1:
+ if (COMPOSING_P (coding))
+ DECODE_COMPOSITION_END ('1');
+ goto label_invalid_code;
+
case ISO_carriage_return:
+ if (COMPOSING_P (coding))
+ DECODE_COMPOSITION_END ('1');
+
if (coding->eol_type == CODING_EOL_CR)
- *dst++ = '\n';
+ c1 = '\n';
else if (coding->eol_type == CODING_EOL_CRLF)
{
ONE_MORE_BYTE (c1);
- if (c1 == ISO_CODE_LF)
- *dst++ = '\n';
- else
+ if (c1 != ISO_CODE_LF)
{
if (coding->mode & CODING_MODE_INHIBIT_INCONSISTENT_EOL)
{
- result = CODING_FINISH_INCONSISTENT_EOL;
- goto label_end_of_loop_2;
+ coding->result = CODING_FINISH_INCONSISTENT_EOL;
+ goto label_end_of_loop;
}
src--;
- *dst++ = '\r';
+ c1 = '\r';
}
}
- else
- *dst++ = c1;
- coding->produced_char++;
+ charset = CHARSET_ASCII;
break;
case ISO_shift_out:
goto label_invalid_code;
CODING_SPEC_ISO_INVOCATION (coding, 0) = 1;
charset0 = CODING_SPEC_ISO_PLANE_CHARSET (coding, 0);
- break;
+ continue;
case ISO_shift_in:
if (! (coding->flags & CODING_FLAG_ISO_LOCKING_SHIFT))
goto label_invalid_code;
CODING_SPEC_ISO_INVOCATION (coding, 0) = 0;
charset0 = CODING_SPEC_ISO_PLANE_CHARSET (coding, 0);
- break;
+ continue;
case ISO_single_shift_2_7:
case ISO_single_shift_2:
}
else
goto label_invalid_code;
- break;
+ /* We must update these variables now. */
+ charset0 = CODING_SPEC_ISO_PLANE_CHARSET (coding, 0);
+ charset1 = CODING_SPEC_ISO_PLANE_CHARSET (coding, 1);
+ continue;
case 'n': /* invocation of locking-shift-2 */
if (! (coding->flags & CODING_FLAG_ISO_LOCKING_SHIFT)
goto label_invalid_code;
CODING_SPEC_ISO_INVOCATION (coding, 0) = 2;
charset0 = CODING_SPEC_ISO_PLANE_CHARSET (coding, 0);
- break;
+ continue;
case 'o': /* invocation of locking-shift-3 */
if (! (coding->flags & CODING_FLAG_ISO_LOCKING_SHIFT)
goto label_invalid_code;
CODING_SPEC_ISO_INVOCATION (coding, 0) = 3;
charset0 = CODING_SPEC_ISO_PLANE_CHARSET (coding, 0);
- break;
+ continue;
case 'N': /* invocation of single-shift-2 */
if (! (coding->flags & CODING_FLAG_ISO_SINGLE_SHIFT)
|| CODING_SPEC_ISO_DESIGNATION (coding, 2) < 0)
goto label_invalid_code;
- ONE_MORE_BYTE (c1);
charset = CODING_SPEC_ISO_DESIGNATION (coding, 2);
- DECODE_ISO_CHARACTER (charset, c1);
+ ONE_MORE_BYTE (c1);
+ if (c1 < 0x20 || (c1 >= 0x80 && c1 < 0xA0))
+ goto label_invalid_code;
break;
case 'O': /* invocation of single-shift-3 */
if (! (coding->flags & CODING_FLAG_ISO_SINGLE_SHIFT)
|| CODING_SPEC_ISO_DESIGNATION (coding, 3) < 0)
goto label_invalid_code;
- ONE_MORE_BYTE (c1);
charset = CODING_SPEC_ISO_DESIGNATION (coding, 3);
- DECODE_ISO_CHARACTER (charset, c1);
+ ONE_MORE_BYTE (c1);
+ if (c1 < 0x20 || (c1 >= 0x80 && c1 < 0xA0))
+ goto label_invalid_code;
break;
- case '0': case '2': /* start composing */
- /* Before processing composing, we must be sure that all
- characters being composed are supported by CODING.
- If not, we must give up composing. */
- if (check_composing_code (coding, src, src_end) == 0)
- {
- /* We are looking at a valid composition sequence. */
- coding->composing = (c1 == '0'
- ? COMPOSING_NO_RULE_HEAD
- : COMPOSING_WITH_RULE_HEAD);
- coding->composed_chars = 0;
- }
- else
- {
- *dst++ = ISO_CODE_ESC;
- *dst++ = c1;
- coding->produced_char += 2;
- }
- break;
+ case '0': case '2': case '3': case '4': /* start composition */
+ DECODE_COMPOSITION_START (c1);
+ continue;
- case '1': /* end composing */
- if (!coding->composing)
- {
- *dst++ = ISO_CODE_ESC;
- *dst++ = c1;
- coding->produced_char += 2;
- break;
- }
-
- if (coding->composed_chars > 0)
- {
- if (coding->composed_chars == 1)
- {
- unsigned char *this_char_start = dst;
- int this_bytes;
-
- /* Only one character is in the composing
- sequence. Make it a normal character. */
- while (*--this_char_start != LEADING_CODE_COMPOSITION);
- dst = (this_char_start
- + (coding->composing == COMPOSING_NO_RULE_TAIL
- ? 1 : 2));
- *dst -= 0x20;
- if (*dst == 0x80)
- *++dst &= 0x7F;
- this_bytes = BYTES_BY_CHAR_HEAD (*dst);
- while (this_bytes--) *this_char_start++ = *dst++;
- dst = this_char_start;
- }
- coding->produced_char++;
- }
- coding->composing = COMPOSING_NO;
- break;
+ case '1': /* end composition */
+ DECODE_COMPOSITION_END (c1);
+ continue;
case '[': /* specification of direction */
if (coding->flags & CODING_FLAG_ISO_NO_DIRECTION)
default:
goto label_invalid_code;
}
- break;
+ continue;
default:
if (! (coding->flags & CODING_FLAG_ISO_DESIGNATION))
DECODE_DESIGNATION (c1 - 0x2C, 1, 96, c2);
}
else
- {
- goto label_invalid_code;
- }
+ goto label_invalid_code;
+ /* We must update these variables now. */
+ charset0 = CODING_SPEC_ISO_PLANE_CHARSET (coding, 0);
+ charset1 = CODING_SPEC_ISO_PLANE_CHARSET (coding, 1);
+ continue;
}
- /* We must update these variables now. */
- charset0 = CODING_SPEC_ISO_PLANE_CHARSET (coding, 0);
- charset1 = CODING_SPEC_ISO_PLANE_CHARSET (coding, 1);
- break;
+ }
- label_invalid_code:
- while (src_base < src)
- *dst++ = *src_base++;
- coding->fake_multibyte = 1;
+ /* Now we know CHARSET and 1st position code C1 of a character.
+ Produce a multibyte sequence for that character while getting
+ 2nd position code C2 if necessary. */
+ if (CHARSET_DIMENSION (charset) == 2)
+ {
+ ONE_MORE_BYTE (c2);
+ if (c1 < 0x80 ? c2 < 0x20 || c2 >= 0x80 : c2 < 0xA0)
+ /* C2 is not in a valid range. */
+ goto label_invalid_code;
}
+ c = DECODE_ISO_CHARACTER (charset, c1, c2);
+ EMIT_CHAR (c);
continue;
- label_end_of_loop:
- result = CODING_FINISH_INSUFFICIENT_SRC;
- label_end_of_loop_2:
+ label_invalid_code:
+ coding->errors++;
+ if (COMPOSING_P (coding))
+ DECODE_COMPOSITION_END ('1');
src = src_base;
- break;
- }
-
- if (src < src_end)
- {
- if (result == CODING_FINISH_NORMAL)
- result = CODING_FINISH_INSUFFICIENT_DST;
- else if (result != CODING_FINISH_INCONSISTENT_EOL
- && coding->mode & CODING_MODE_LAST_BLOCK)
- {
- /* This is the last block of the text to be decoded. We had
- better just flush out all remaining codes in the text
- although they are not valid characters. */
- src_bytes = src_end - src;
- if (dst_bytes && (dst_end - dst < src_bytes))
- src_bytes = dst_end - dst;
- bcopy (src, dst, src_bytes);
- dst += src_bytes;
- src += src_bytes;
- coding->fake_multibyte = 1;
- }
+ c = *src++;
+ EMIT_CHAR (c);
}
- coding->consumed = coding->consumed_char = src - source;
+ label_end_of_loop:
+ coding->consumed = coding->consumed_char = src_base - source;
coding->produced = dst - destination;
- return result;
+ return;
}
+
/* ISO2022 encoding stuff. */
/*
*/
/* Produce codes (escape sequence) for designating CHARSET to graphic
- register REG. If <final-char> of CHARSET is '@', 'A', or 'B' and
- the coding system CODING allows, produce designation sequence of
- short-form. */
+ register REG at DST, and increment DST. If <final-char> of CHARSET is
+ '@', 'A', or 'B' and the coding system CODING allows, produce
+ designation sequence of short-form. */
#define ENCODE_DESIGNATION(charset, reg, coding) \
do { \
char *intermediate_char_94 = "()*+"; \
char *intermediate_char_96 = ",-./"; \
int revision = CODING_SPEC_ISO_REVISION_NUMBER(coding, charset); \
+ \
if (revision < 255) \
{ \
*dst++ = ISO_CODE_ESC; \
*dst++ = '&'; \
*dst++ = '@' + revision; \
} \
- *dst++ = ISO_CODE_ESC; \
+ *dst++ = ISO_CODE_ESC; \
if (CHARSET_DIMENSION (charset) == 1) \
{ \
if (CHARSET_CHARS (charset) == 94) \
*dst++ = '$'; \
if (CHARSET_CHARS (charset) == 94) \
{ \
- if (! (coding->flags & CODING_FLAG_ISO_SHORT_FORM) \
- || reg != 0 \
- || final_char < '@' || final_char > 'B') \
+ if (! (coding->flags & CODING_FLAG_ISO_SHORT_FORM) \
+ || reg != 0 \
+ || final_char < '@' || final_char > 'B') \
*dst++ = (unsigned char) (intermediate_char_94[reg]); \
} \
else \
- *dst++ = (unsigned char) (intermediate_char_96[reg]); \
+ *dst++ = (unsigned char) (intermediate_char_96[reg]); \
} \
- *dst++ = final_char; \
+ *dst++ = final_char; \
CODING_SPEC_ISO_DESIGNATION (coding, reg) = charset; \
} while (0)
if (coding->flags & CODING_FLAG_ISO_SEVEN_BITS) \
*dst++ = ISO_CODE_ESC, *dst++ = 'N'; \
else \
- { \
- *dst++ = ISO_CODE_SS2; \
- coding->fake_multibyte = 1; \
- } \
+ *dst++ = ISO_CODE_SS2; \
CODING_SPEC_ISO_SINGLE_SHIFTING (coding) = 1; \
} while (0)
if (coding->flags & CODING_FLAG_ISO_SEVEN_BITS) \
*dst++ = ISO_CODE_ESC, *dst++ = 'O'; \
else \
- { \
- *dst++ = ISO_CODE_SS3; \
- coding->fake_multibyte = 1; \
- } \
+ *dst++ = ISO_CODE_SS3; \
CODING_SPEC_ISO_SINGLE_SHIFTING (coding) = 1; \
} while (0)
escape sequence) for ISO2022 locking-shift functions (shift-in,
shift-out, locking-shift-2, and locking-shift-3). */
-#define ENCODE_SHIFT_IN \
- do { \
- *dst++ = ISO_CODE_SI; \
+#define ENCODE_SHIFT_IN \
+ do { \
+ *dst++ = ISO_CODE_SI; \
CODING_SPEC_ISO_INVOCATION (coding, 0) = 0; \
} while (0)
-#define ENCODE_SHIFT_OUT \
- do { \
- *dst++ = ISO_CODE_SO; \
+#define ENCODE_SHIFT_OUT \
+ do { \
+ *dst++ = ISO_CODE_SO; \
CODING_SPEC_ISO_INVOCATION (coding, 0) = 1; \
} while (0)
CODING_SPEC_ISO_INVOCATION (coding, 0) = 2; \
} while (0)
-#define ENCODE_LOCKING_SHIFT_3 \
- do { \
- *dst++ = ISO_CODE_ESC, *dst++ = 'o'; \
+#define ENCODE_LOCKING_SHIFT_3 \
+ do { \
+ *dst++ = ISO_CODE_ESC, *dst++ = 'o'; \
CODING_SPEC_ISO_INVOCATION (coding, 0) = 3; \
} while (0)
CHARSET and whose position-code is C1. Designation and invocation
sequences are also produced in advance if necessary. */
-
#define ENCODE_ISO_CHARACTER_DIMENSION1(charset, c1) \
do { \
if (CODING_SPEC_ISO_SINGLE_SHIFTING (coding)) \
*dst++ = c1 | 0x80; \
break; \
} \
- else if (coding->flags & CODING_FLAG_ISO_SAFE \
- && !coding->safe_charsets[charset]) \
- { \
- /* We should not encode this character, instead produce one or \
- two `?'s. */ \
- *dst++ = CODING_INHIBIT_CHARACTER_SUBSTITUTION; \
- if (CHARSET_WIDTH (charset) == 2) \
- *dst++ = CODING_INHIBIT_CHARACTER_SUBSTITUTION; \
- break; \
- } \
else \
/* Since CHARSET is not yet invoked to any graphic planes, we \
must invoke it, or, at first, designate it to some graphic \
*dst++ = c1 | 0x80, *dst++= c2 | 0x80; \
break; \
} \
- else if (coding->flags & CODING_FLAG_ISO_SAFE \
- && !coding->safe_charsets[charset]) \
- { \
- /* We should not encode this character, instead produce one or \
- two `?'s. */ \
- *dst++ = CODING_INHIBIT_CHARACTER_SUBSTITUTION; \
- if (CHARSET_WIDTH (charset) == 2) \
- *dst++ = CODING_INHIBIT_CHARACTER_SUBSTITUTION; \
- break; \
- } \
else \
/* Since CHARSET is not yet invoked to any graphic planes, we \
must invoke it, or, at first, designate it to some graphic \
dst = encode_invocation_designation (charset, coding, dst); \
} while (1)
-#define ENCODE_ISO_CHARACTER(charset, c1, c2) \
+#define ENCODE_ISO_CHARACTER(c) \
do { \
- int c_alt, charset_alt; \
- if (!NILP (translation_table) \
- && ((c_alt = translate_char (translation_table, -1, \
- charset, c1, c2)) \
- >= 0)) \
- SPLIT_CHAR (c_alt, charset_alt, c1, c2); \
- else \
- charset_alt = charset; \
- if (CHARSET_DIMENSION (charset_alt) == 1) \
+ int charset, c1, c2; \
+ \
+ SPLIT_CHAR (c, charset, c1, c2); \
+ if (CHARSET_DEFINED_P (charset)) \
{ \
- if (charset == CHARSET_ASCII \
- && coding->flags & CODING_FLAG_ISO_USE_ROMAN) \
- charset_alt = charset_latin_jisx0201; \
- ENCODE_ISO_CHARACTER_DIMENSION1 (charset_alt, c1); \
+ if (CHARSET_DIMENSION (charset) == 1) \
+ { \
+ if (charset == CHARSET_ASCII \
+ && coding->flags & CODING_FLAG_ISO_USE_ROMAN) \
+ charset = charset_latin_jisx0201; \
+ ENCODE_ISO_CHARACTER_DIMENSION1 (charset, c1); \
+ } \
+ else \
+ { \
+ if (charset == charset_jisx0208 \
+ && coding->flags & CODING_FLAG_ISO_USE_OLDJIS) \
+ charset = charset_jisx0208_1978; \
+ ENCODE_ISO_CHARACTER_DIMENSION2 (charset, c1, c2); \
+ } \
} \
else \
{ \
- if (charset == charset_jisx0208 \
- && coding->flags & CODING_FLAG_ISO_USE_OLDJIS) \
- charset_alt = charset_jisx0208_1978; \
- ENCODE_ISO_CHARACTER_DIMENSION2 (charset_alt, c1, c2); \
+ *dst++ = c1; \
+ if (c2 >= 0) \
+ *dst++ = c2; \
} \
- if (! COMPOSING_P (coding->composing)) \
- coding->consumed_char++; \
} while (0)
+
+/* Instead of encoding character C, produce one or two `?'s. */
+
+#define ENCODE_UNSAFE_CHARACTER(c) \
+ do { \
+ ENCODE_ISO_CHARACTER (CODING_INHIBIT_CHARACTER_SUBSTITUTION); \
+ if (CHARSET_WIDTH (CHAR_CHARSET (c)) > 1) \
+ ENCODE_ISO_CHARACTER (CODING_INHIBIT_CHARACTER_SUBSTITUTION); \
+ } while (0)
+
+
/* Produce designation and invocation codes at a place pointed by DST
to use CHARSET. The element `spec.iso2022' of *CODING is updated.
Return new DST. */
break;
}
}
+
return dst;
}
-/* The following two macros produce codes for indicating composition. */
-#define ENCODE_COMPOSITION_NO_RULE_START *dst++ = ISO_CODE_ESC, *dst++ = '0'
-#define ENCODE_COMPOSITION_WITH_RULE_START *dst++ = ISO_CODE_ESC, *dst++ = '2'
-#define ENCODE_COMPOSITION_END *dst++ = ISO_CODE_ESC, *dst++ = '1'
+/* Produce 2-byte codes for encoded composition rule RULE. */
+
+#define ENCODE_COMPOSITION_RULE(rule) \
+ do { \
+ int gref, nref; \
+ COMPOSITION_DECODE_RULE (rule, gref, nref); \
+ *dst++ = 32 + 81 + gref; \
+ *dst++ = 32 + nref; \
+ } while (0)
+
+/* Produce codes for indicating the start of a composition sequence
+ (ESC 0, ESC 3, or ESC 4). DATA points to an array of integers
+ which specify information about the composition. See the comment
+ in coding.h for the format of DATA. */
+
+#define ENCODE_COMPOSITION_START(coding, data) \
+ do { \
+ coding->composing = data[3]; \
+ *dst++ = ISO_CODE_ESC; \
+ if (coding->composing == COMPOSITION_RELATIVE) \
+ *dst++ = '0'; \
+ else \
+ { \
+ *dst++ = (coding->composing == COMPOSITION_WITH_ALTCHARS \
+ ? '3' : '4'); \
+ coding->cmp_data_index = coding->cmp_data_start + 4; \
+ coding->composition_rule_follows = 0; \
+ } \
+ } while (0)
+
+/* Produce codes for indicating the end of the current composition. */
+
+#define ENCODE_COMPOSITION_END(coding, data) \
+ do { \
+ *dst++ = ISO_CODE_ESC; \
+ *dst++ = '1'; \
+ coding->cmp_data_start += data[0]; \
+ coding->composing = COMPOSITION_NO; \
+ if (coding->cmp_data_start == coding->cmp_data->used \
+ && coding->cmp_data->next) \
+ { \
+ coding->cmp_data = coding->cmp_data->next; \
+ coding->cmp_data_start = 0; \
+ } \
+ } while (0)
+
+/* Produce composition start sequence ESC 0. Here, this sequence
+ doesn't mean the start of a new composition but means that we have
+ just produced components (alternate chars and composition rules) of
+ the composition and the actual text follows in SRC. */
+
+#define ENCODE_COMPOSITION_FAKE_START(coding) \
+ do { \
+ *dst++ = ISO_CODE_ESC; \
+ *dst++ = '0'; \
+ coding->composing = COMPOSITION_RELATIVE; \
+ } while (0)
/* The following three macros produce codes for indicating direction
of text. */
-#define ENCODE_CONTROL_SEQUENCE_INTRODUCER \
- do { \
+#define ENCODE_CONTROL_SEQUENCE_INTRODUCER \
+ do { \
if (coding->flags == CODING_FLAG_ISO_SEVEN_BITS) \
- *dst++ = ISO_CODE_ESC, *dst++ = '['; \
- else \
- *dst++ = ISO_CODE_CSI; \
+ *dst++ = ISO_CODE_ESC, *dst++ = '['; \
+ else \
+ *dst++ = ISO_CODE_CSI; \
} while (0)
#define ENCODE_DIRECTION_R2L \
- ENCODE_CONTROL_SEQUENCE_INTRODUCER, *dst++ = '2', *dst++ = ']'
+ ENCODE_CONTROL_SEQUENCE_INTRODUCER (dst), *dst++ = '2', *dst++ = ']'
#define ENCODE_DIRECTION_L2R \
- ENCODE_CONTROL_SEQUENCE_INTRODUCER, *dst++ = '0', *dst++ = ']'
+ ENCODE_CONTROL_SEQUENCE_INTRODUCER (dst), *dst++ = '0', *dst++ = ']'
/* Produce codes for designation and invocation to reset the graphic
planes and registers to initial state. */
} while (0)
/* Produce designation sequences of charsets in the line started from
- SRC to a place pointed by *DSTP, and update DSTP.
+ SRC to a place pointed by DST, and return updated DST.
If the current block ends before any end-of-line, we may fail to
find all the necessary designations. */
-void
-encode_designation_at_bol (coding, table, src, src_end, dstp)
+static unsigned char *
+encode_designation_at_bol (coding, translation_table, src, src_end, dst)
struct coding_system *coding;
- Lisp_Object table;
- unsigned char *src, *src_end, **dstp;
+ Lisp_Object translation_table;
+ unsigned char *src, *src_end, *dst;
{
int charset, c, found = 0, reg;
/* Table of charsets to be designated to each graphic register. */
int r[4];
- unsigned char *dst = *dstp;
for (reg = 0; reg < 4; reg++)
r[reg] = -1;
- while (src < src_end && *src != '\n' && found < 4)
+ while (found < 4)
{
- int bytes = BYTES_BY_CHAR_HEAD (*src);
+ ONE_MORE_CHAR (c);
+ if (c == '\n')
+ break;
- if (NILP (table))
- charset = CHARSET_AT (src);
- else
- {
- int c_alt;
- unsigned char c1, c2;
-
- SPLIT_STRING(src, bytes, charset, c1, c2);
- if ((c_alt = translate_char (table, -1, charset, c1, c2)) >= 0)
- charset = CHAR_CHARSET (c_alt);
- }
-
+ charset = CHAR_CHARSET (c);
reg = CODING_SPEC_ISO_REQUESTED_DESIGNATION (coding, charset);
if (reg != CODING_SPEC_ISO_NO_REQUESTED_DESIGNATION && r[reg] < 0)
{
found++;
r[reg] = charset;
}
-
- src += bytes;
}
+ label_end_of_loop:
if (found)
{
for (reg = 0; reg < 4; reg++)
if (r[reg] >= 0
&& CODING_SPEC_ISO_DESIGNATION (coding, reg) != r[reg])
ENCODE_DESIGNATION (r[reg], reg, coding);
- *dstp = dst;
}
+
+ return dst;
}
/* See the above "GENERAL NOTES on `encode_coding_XXX ()' functions". */
-int
+static void
encode_coding_iso2022 (coding, source, destination, src_bytes, dst_bytes)
struct coding_system *coding;
unsigned char *source, *destination;
from DST_END to assure overflow checking is necessary only at the
head of loop. */
unsigned char *adjusted_dst_end = dst_end - 19;
- Lisp_Object translation_table
- = coding->translation_table_for_encode;
- int result = CODING_FINISH_NORMAL;
+ /* SRC_BASE remembers the start position in source in each loop.
+ The loop will be exited when there's not enough source text to
+ analyze multi-byte codes (within macro ONE_MORE_CHAR), or when
+ there's not enough destination area to produce encoded codes
+ (within macro EMIT_BYTES). */
+ unsigned char *src_base;
+ int c;
+ Lisp_Object translation_table;
+ Lisp_Object safe_chars;
- if (!NILP (Venable_character_translation) && NILP (translation_table))
- translation_table = Vstandard_translation_table_for_encode;
+ safe_chars = coding_safe_chars (coding);
+
+ if (NILP (Venable_character_translation))
+ translation_table = Qnil;
+ else
+ {
+ translation_table = coding->translation_table_for_encode;
+ if (NILP (translation_table))
+ translation_table = Vstandard_translation_table_for_encode;
+ }
coding->consumed_char = 0;
- coding->fake_multibyte = 0;
- while (src < src_end && (dst_bytes
- ? (dst < adjusted_dst_end)
- : (dst < src - 19)))
+ coding->errors = 0;
+ while (1)
{
- /* SRC_BASE remembers the start position in source in each loop.
- The loop will be exited when there's not enough source text
- to analyze multi-byte codes (within macros ONE_MORE_BYTE,
- TWO_MORE_BYTES, and THREE_MORE_BYTES). In that case, SRC is
- reset to SRC_BASE before exiting. */
- unsigned char *src_base = src;
- int charset, c1, c2, c3, c4;
+ src_base = src;
+
+ if (dst >= (dst_bytes ? adjusted_dst_end : (src - 19)))
+ {
+ coding->result = CODING_FINISH_INSUFFICIENT_DST;
+ break;
+ }
if (coding->flags & CODING_FLAG_ISO_DESIGNATE_AT_BOL
&& CODING_SPEC_ISO_BOL (coding))
{
/* We have to produce designation sequences if any now. */
- encode_designation_at_bol (coding, translation_table,
- src, src_end, &dst);
+ dst = encode_designation_at_bol (coding, translation_table,
+ src, src_end, dst);
CODING_SPEC_ISO_BOL (coding) = 0;
}
- c1 = *src++;
- /* If we are seeing a component of a composite character, we are
- seeing a leading-code encoded irregularly for composition, or
- a composition rule if composing with rule. We must set C1 to
- a normal leading-code or an ASCII code. If we are not seeing
- a composite character, we must reset composition,
- designation, and invocation states. */
- if (COMPOSING_P (coding->composing))
+ /* Check composition start and end. */
+ if (coding->composing != COMPOSITION_DISABLED
+ && coding->cmp_data_start < coding->cmp_data->used)
{
- if (c1 < 0xA0)
+ struct composition_data *cmp_data = coding->cmp_data;
+ int *data = cmp_data->data + coding->cmp_data_start;
+ int this_pos = cmp_data->char_offset + coding->consumed_char;
+
+ if (coding->composing == COMPOSITION_RELATIVE)
{
- /* We are not in a composite character any longer. */
- coding->composing = COMPOSING_NO;
- ENCODE_RESET_PLANE_AND_REGISTER;
- ENCODE_COMPOSITION_END;
+ if (this_pos == data[2])
+ {
+ ENCODE_COMPOSITION_END (coding, data);
+ cmp_data = coding->cmp_data;
+ data = cmp_data->data + coding->cmp_data_start;
+ }
}
- else
+ else if (COMPOSING_P (coding))
{
- if (coding->composing == COMPOSING_WITH_RULE_RULE)
+ /* COMPOSITION_WITH_ALTCHARS or COMPOSITION_WITH_RULE_ALTCHAR */
+ if (coding->cmp_data_index == coding->cmp_data_start + data[0])
+ /* We have consumed components of the composition.
+ What follows in SRC is the compositions's base
+ text. */
+ ENCODE_COMPOSITION_FAKE_START (coding);
+ else
{
- *dst++ = c1 & 0x7F;
- coding->composing = COMPOSING_WITH_RULE_HEAD;
+ int c = cmp_data->data[coding->cmp_data_index++];
+ if (coding->composition_rule_follows)
+ {
+ ENCODE_COMPOSITION_RULE (c);
+ coding->composition_rule_follows = 0;
+ }
+ else
+ {
+ if (coding->flags & CODING_FLAG_ISO_SAFE
+ && ! CODING_SAFE_CHAR_P (safe_chars, c))
+ ENCODE_UNSAFE_CHARACTER (c);
+ else
+ ENCODE_ISO_CHARACTER (c);
+ if (coding->composing == COMPOSITION_WITH_RULE_ALTCHARS)
+ coding->composition_rule_follows = 1;
+ }
continue;
}
- else if (coding->composing == COMPOSING_WITH_RULE_HEAD)
- coding->composing = COMPOSING_WITH_RULE_RULE;
- if (c1 == 0xA0)
+ }
+ if (!COMPOSING_P (coding))
+ {
+ if (this_pos == data[1])
{
- /* This is an ASCII component. */
- ONE_MORE_BYTE (c1);
- c1 &= 0x7F;
+ ENCODE_COMPOSITION_START (coding, data);
+ continue;
}
- else
- /* This is a leading-code of non ASCII component. */
- c1 -= 0x20;
}
}
-
- /* Now encode one character. C1 is a control character, an
- ASCII character, or a leading-code of multi-byte character. */
- switch (emacs_code_class[c1])
- {
- case EMACS_ascii_code:
- ENCODE_ISO_CHARACTER (CHARSET_ASCII, c1, /* dummy */ c2);
- break;
-
- case EMACS_control_code:
- if (coding->flags & CODING_FLAG_ISO_RESET_AT_CNTL)
- ENCODE_RESET_PLANE_AND_REGISTER;
- *dst++ = c1;
- coding->consumed_char++;
- break;
-
- case EMACS_carriage_return_code:
- if (! (coding->mode & CODING_MODE_SELECTIVE_DISPLAY))
- {
- if (coding->flags & CODING_FLAG_ISO_RESET_AT_CNTL)
- ENCODE_RESET_PLANE_AND_REGISTER;
- *dst++ = c1;
- coding->consumed_char++;
- break;
- }
- /* fall down to treat '\r' as '\n' ... */
-
- case EMACS_linefeed_code:
- if (coding->flags & CODING_FLAG_ISO_RESET_AT_EOL)
- ENCODE_RESET_PLANE_AND_REGISTER;
- if (coding->flags & CODING_FLAG_ISO_INIT_AT_BOL)
- bcopy (coding->spec.iso2022.initial_designation,
- coding->spec.iso2022.current_designation,
- sizeof coding->spec.iso2022.initial_designation);
- if (coding->eol_type == CODING_EOL_LF
- || coding->eol_type == CODING_EOL_UNDECIDED)
- *dst++ = ISO_CODE_LF;
- else if (coding->eol_type == CODING_EOL_CRLF)
- *dst++ = ISO_CODE_CR, *dst++ = ISO_CODE_LF;
- else
- *dst++ = ISO_CODE_CR;
- CODING_SPEC_ISO_BOL (coding) = 1;
- coding->consumed_char++;
- break;
-
- case EMACS_leading_code_2:
- ONE_MORE_BYTE (c2);
- if (c2 < 0xA0)
- {
- /* invalid sequence */
- *dst++ = c1;
- src--;
- coding->consumed_char++;
- }
- else
- ENCODE_ISO_CHARACTER (c1, c2, /* dummy */ c3);
- break;
-
- case EMACS_leading_code_3:
- TWO_MORE_BYTES (c2, c3);
- if (c2 < 0xA0 || c3 < 0xA0)
- {
- /* invalid sequence */
- *dst++ = c1;
- src -= 2;
- coding->consumed_char++;
- }
- else if (c1 < LEADING_CODE_PRIVATE_11)
- ENCODE_ISO_CHARACTER (c1, c2, c3);
- else
- ENCODE_ISO_CHARACTER (c2, c3, /* dummy */ c4);
- break;
- case EMACS_leading_code_4:
- THREE_MORE_BYTES (c2, c3, c4);
- if (c2 < 0xA0 || c3 < 0xA0 || c4 < 0xA0)
- {
- /* invalid sequence */
- *dst++ = c1;
- src -= 3;
- coding->consumed_char++;
- }
- else
- ENCODE_ISO_CHARACTER (c2, c3, c4);
- break;
+ ONE_MORE_CHAR (c);
- case EMACS_leading_code_composition:
- ONE_MORE_BYTE (c2);
- if (c2 < 0xA0)
+ /* Now encode the character C. */
+ if (c < 0x20 || c == 0x7F)
+ {
+ if (c == '\r')
{
- /* invalid sequence */
- *dst++ = c1;
- src--;
- coding->consumed_char++;
+ if (! (coding->mode & CODING_MODE_SELECTIVE_DISPLAY))
+ {
+ if (coding->flags & CODING_FLAG_ISO_RESET_AT_CNTL)
+ ENCODE_RESET_PLANE_AND_REGISTER;
+ *dst++ = c;
+ continue;
+ }
+ /* fall down to treat '\r' as '\n' ... */
+ c = '\n';
}
- else if (c2 == 0xFF)
+ if (c == '\n')
{
- ENCODE_RESET_PLANE_AND_REGISTER;
- coding->composing = COMPOSING_WITH_RULE_HEAD;
- ENCODE_COMPOSITION_WITH_RULE_START;
- coding->consumed_char++;
+ if (coding->flags & CODING_FLAG_ISO_RESET_AT_EOL)
+ ENCODE_RESET_PLANE_AND_REGISTER;
+ if (coding->flags & CODING_FLAG_ISO_INIT_AT_BOL)
+ bcopy (coding->spec.iso2022.initial_designation,
+ coding->spec.iso2022.current_designation,
+ sizeof coding->spec.iso2022.initial_designation);
+ if (coding->eol_type == CODING_EOL_LF
+ || coding->eol_type == CODING_EOL_UNDECIDED)
+ *dst++ = ISO_CODE_LF;
+ else if (coding->eol_type == CODING_EOL_CRLF)
+ *dst++ = ISO_CODE_CR, *dst++ = ISO_CODE_LF;
+ else
+ *dst++ = ISO_CODE_CR;
+ CODING_SPEC_ISO_BOL (coding) = 1;
}
- else
+ else
{
- ENCODE_RESET_PLANE_AND_REGISTER;
- /* Rewind one byte because it is a character code of
- composition elements. */
- src--;
- coding->composing = COMPOSING_NO_RULE_HEAD;
- ENCODE_COMPOSITION_NO_RULE_START;
- coding->consumed_char++;
+ if (coding->flags & CODING_FLAG_ISO_RESET_AT_CNTL)
+ ENCODE_RESET_PLANE_AND_REGISTER;
+ *dst++ = c;
}
- break;
-
- case EMACS_invalid_code:
- *dst++ = c1;
- coding->consumed_char++;
- break;
}
- continue;
- label_end_of_loop:
- result = CODING_FINISH_INSUFFICIENT_SRC;
- src = src_base;
- break;
- }
-
- if (src < src_end && result == CODING_FINISH_NORMAL)
- result = CODING_FINISH_INSUFFICIENT_DST;
-
- /* If this is the last block of the text to be encoded, we must
- reset graphic planes and registers to the initial state, and
- flush out the carryover if any. */
- if (coding->mode & CODING_MODE_LAST_BLOCK)
- {
- ENCODE_RESET_PLANE_AND_REGISTER;
- if (COMPOSING_P (coding->composing))
- ENCODE_COMPOSITION_END;
- if (result == CODING_FINISH_INSUFFICIENT_SRC)
+ else if (ASCII_BYTE_P (c))
+ ENCODE_ISO_CHARACTER (c);
+ else if (SINGLE_BYTE_CHAR_P (c))
{
- while (src < src_end && dst < dst_end)
- *dst++ = *src++;
+ *dst++ = c;
+ coding->errors++;
}
+ else if (coding->flags & CODING_FLAG_ISO_SAFE
+ && ! CODING_SAFE_CHAR_P (safe_chars, c))
+ ENCODE_UNSAFE_CHARACTER (c);
+ else
+ ENCODE_ISO_CHARACTER (c);
+
+ coding->consumed_char++;
}
- coding->consumed = src - source;
+
+ label_end_of_loop:
+ coding->consumed = src_base - source;
coding->produced = coding->produced_char = dst - destination;
- return result;
}
\f
(character set) (range)
ASCII 0x00 .. 0x7F
KATAKANA-JISX0201 0xA0 .. 0xDF
- JISX0208 (1st byte) 0x80 .. 0x9F and 0xE0 .. 0xEF
- (2nd byte) 0x40 .. 0xFF
+ JISX0208 (1st byte) 0x81 .. 0x9F and 0xE0 .. 0xEF
+ (2nd byte) 0x40 .. 0x7E and 0x80 .. 0xFC
-------------------------------
*/
b2 += b2 < 0x3F ? 0x40 : 0x62; \
} while (0)
-#define DECODE_SJIS_BIG5_CHARACTER(charset, c1, c2) \
- do { \
- int c_alt, charset_alt = (charset); \
- if (!NILP (translation_table) \
- && ((c_alt = translate_char (translation_table, \
- -1, (charset), c1, c2)) >= 0)) \
- SPLIT_CHAR (c_alt, charset_alt, c1, c2); \
- if (charset_alt == CHARSET_ASCII || charset_alt < 0) \
- DECODE_CHARACTER_ASCII (c1); \
- else if (CHARSET_DIMENSION (charset_alt) == 1) \
- DECODE_CHARACTER_DIMENSION1 (charset_alt, c1); \
- else \
- DECODE_CHARACTER_DIMENSION2 (charset_alt, c1, c2); \
- } while (0)
-
-#define ENCODE_SJIS_BIG5_CHARACTER(charset, c1, c2) \
- do { \
- int c_alt, charset_alt; \
- if (!NILP (translation_table) \
- && ((c_alt = translate_char (translation_table, -1, \
- charset, c1, c2)) \
- >= 0)) \
- SPLIT_CHAR (c_alt, charset_alt, c1, c2); \
- else \
- charset_alt = charset; \
- if (charset_alt == charset_ascii) \
- *dst++ = c1; \
- else if (CHARSET_DIMENSION (charset_alt) == 1) \
- { \
- if (sjis_p && charset_alt == charset_katakana_jisx0201) \
- *dst++ = c1; \
- else \
- { \
- *dst++ = charset_alt, *dst++ = c1; \
- coding->fake_multibyte = 1; \
- } \
- } \
- else \
- { \
- c1 &= 0x7F, c2 &= 0x7F; \
- if (sjis_p && charset_alt == charset_jisx0208) \
- { \
- unsigned char s1, s2; \
- \
- ENCODE_SJIS (c1, c2, s1, s2); \
- *dst++ = s1, *dst++ = s2; \
- coding->fake_multibyte = 1; \
- } \
- else if (!sjis_p \
- && (charset_alt == charset_big5_1 \
- || charset_alt == charset_big5_2)) \
- { \
- unsigned char b1, b2; \
- \
- ENCODE_BIG5 (charset_alt, c1, c2, b1, b2); \
- *dst++ = b1, *dst++ = b2; \
- } \
- else \
- { \
- *dst++ = charset_alt, *dst++ = c1, *dst++ = c2; \
- coding->fake_multibyte = 1; \
- } \
- } \
- coding->consumed_char++; \
- } while (0);
-
/* See the above "GENERAL NOTES on `detect_coding_XXX ()' functions".
Check if a text is encoded in SJIS. If it is, return
CODING_CATEGORY_MASK_SJIS, else return 0. */
detect_coding_sjis (src, src_end)
unsigned char *src, *src_end;
{
- unsigned char c;
+ int c;
+ /* Dummy for ONE_MORE_BYTE. */
+ struct coding_system dummy_coding;
+ struct coding_system *coding = &dummy_coding;
- while (src < src_end)
+ while (1)
{
- c = *src++;
+ ONE_MORE_BYTE (c);
if ((c >= 0x80 && c < 0xA0) || c >= 0xE0)
{
- if (src < src_end && *src++ < 0x40)
+ ONE_MORE_BYTE (c);
+ if (c < 0x40)
return 0;
}
}
+ label_end_of_loop:
return CODING_CATEGORY_MASK_SJIS;
}
detect_coding_big5 (src, src_end)
unsigned char *src, *src_end;
{
- unsigned char c;
+ int c;
+ /* Dummy for ONE_MORE_BYTE. */
+ struct coding_system dummy_coding;
+ struct coding_system *coding = &dummy_coding;
- while (src < src_end)
+ while (1)
{
- c = *src++;
+ ONE_MORE_BYTE (c);
if (c >= 0xA1)
{
- if (src >= src_end)
- break;
- c = *src++;
+ ONE_MORE_BYTE (c);
if (c < 0x40 || (c >= 0x7F && c <= 0xA0))
return 0;
}
}
+ label_end_of_loop:
return CODING_CATEGORY_MASK_BIG5;
}
+/* See the above "GENERAL NOTES on `detect_coding_XXX ()' functions".
+ Check if a text is encoded in UTF-8. If it is, return
+ CODING_CATEGORY_MASK_UTF_8, else return 0. */
+
+#define UTF_8_1_OCTET_P(c) ((c) < 0x80)
+#define UTF_8_EXTRA_OCTET_P(c) (((c) & 0xC0) == 0x80)
+#define UTF_8_2_OCTET_LEADING_P(c) (((c) & 0xE0) == 0xC0)
+#define UTF_8_3_OCTET_LEADING_P(c) (((c) & 0xF0) == 0xE0)
+#define UTF_8_4_OCTET_LEADING_P(c) (((c) & 0xF8) == 0xF0)
+#define UTF_8_5_OCTET_LEADING_P(c) (((c) & 0xFC) == 0xF8)
+#define UTF_8_6_OCTET_LEADING_P(c) (((c) & 0xFE) == 0xFC)
+
+int
+detect_coding_utf_8 (src, src_end)
+ unsigned char *src, *src_end;
+{
+ unsigned char c;
+ int seq_maybe_bytes;
+ /* Dummy for ONE_MORE_BYTE. */
+ struct coding_system dummy_coding;
+ struct coding_system *coding = &dummy_coding;
+
+ while (1)
+ {
+ ONE_MORE_BYTE (c);
+ if (UTF_8_1_OCTET_P (c))
+ continue;
+ else if (UTF_8_2_OCTET_LEADING_P (c))
+ seq_maybe_bytes = 1;
+ else if (UTF_8_3_OCTET_LEADING_P (c))
+ seq_maybe_bytes = 2;
+ else if (UTF_8_4_OCTET_LEADING_P (c))
+ seq_maybe_bytes = 3;
+ else if (UTF_8_5_OCTET_LEADING_P (c))
+ seq_maybe_bytes = 4;
+ else if (UTF_8_6_OCTET_LEADING_P (c))
+ seq_maybe_bytes = 5;
+ else
+ return 0;
+
+ do
+ {
+ ONE_MORE_BYTE (c);
+ if (!UTF_8_EXTRA_OCTET_P (c))
+ return 0;
+ seq_maybe_bytes--;
+ }
+ while (seq_maybe_bytes > 0);
+ }
+
+ label_end_of_loop:
+ return CODING_CATEGORY_MASK_UTF_8;
+}
+
+/* See the above "GENERAL NOTES on `detect_coding_XXX ()' functions".
+ Check if a text is encoded in UTF-16 Big Endian (endian == 1) or
+ Little Endian (otherwise). If it is, return
+ CODING_CATEGORY_MASK_UTF_16_BE or CODING_CATEGORY_MASK_UTF_16_LE,
+ else return 0. */
+
+#define UTF_16_INVALID_P(val) \
+ (((val) == 0xFFFE) \
+ || ((val) == 0xFFFF))
+
+#define UTF_16_HIGH_SURROGATE_P(val) \
+ (((val) & 0xD800) == 0xD800)
+
+#define UTF_16_LOW_SURROGATE_P(val) \
+ (((val) & 0xDC00) == 0xDC00)
+
+int
+detect_coding_utf_16 (src, src_end)
+ unsigned char *src, *src_end;
+{
+ unsigned char c1, c2;
+ /* Dummy for TWO_MORE_BYTES. */
+ struct coding_system dummy_coding;
+ struct coding_system *coding = &dummy_coding;
+
+ TWO_MORE_BYTES (c1, c2);
+
+ if ((c1 == 0xFF) && (c2 == 0xFE))
+ return CODING_CATEGORY_MASK_UTF_16_LE;
+ else if ((c1 == 0xFE) && (c2 == 0xFF))
+ return CODING_CATEGORY_MASK_UTF_16_BE;
+
+ label_end_of_loop:
+ return 0;
+}
+
/* See the above "GENERAL NOTES on `decode_coding_XXX ()' functions".
If SJIS_P is 1, decode SJIS text, else decode BIG5 test. */
-int
+static void
decode_coding_sjis_big5 (coding, source, destination,
src_bytes, dst_bytes, sjis_p)
struct coding_system *coding;
unsigned char *src_end = source + src_bytes;
unsigned char *dst = destination;
unsigned char *dst_end = destination + dst_bytes;
- /* Since the maximum bytes produced by each loop is 4, we subtract 3
- from DST_END to assure overflow checking is necessary only at the
- head of loop. */
- unsigned char *adjusted_dst_end = dst_end - 3;
- Lisp_Object translation_table
- = coding->translation_table_for_decode;
- int result = CODING_FINISH_NORMAL;
+ /* SRC_BASE remembers the start position in source in each loop.
+ The loop will be exited when there's not enough source code
+ (within macro ONE_MORE_BYTE), or when there's not enough
+ destination area to produce a character (within macro
+ EMIT_CHAR). */
+ unsigned char *src_base;
+ Lisp_Object translation_table;
- if (!NILP (Venable_character_translation) && NILP (translation_table))
- translation_table = Vstandard_translation_table_for_decode;
+ if (NILP (Venable_character_translation))
+ translation_table = Qnil;
+ else
+ {
+ translation_table = coding->translation_table_for_decode;
+ if (NILP (translation_table))
+ translation_table = Vstandard_translation_table_for_decode;
+ }
coding->produced_char = 0;
- coding->fake_multibyte = 0;
- while (src < src_end && (dst_bytes
- ? (dst < adjusted_dst_end)
- : (dst < src - 3)))
+ while (1)
{
- /* SRC_BASE remembers the start position in source in each loop.
- The loop will be exited when there's not enough source text
- to analyze two-byte character (within macro ONE_MORE_BYTE).
- In that case, SRC is reset to SRC_BASE before exiting. */
- unsigned char *src_base = src;
- unsigned char c1 = *src++, c2, c3, c4;
-
- if (c1 < 0x20)
+ int c, charset, c1, c2;
+
+ src_base = src;
+ ONE_MORE_BYTE (c1);
+
+ if (c1 < 0x80)
{
- if (c1 == '\r')
+ charset = CHARSET_ASCII;
+ if (c1 < 0x20)
{
- if (coding->eol_type == CODING_EOL_CRLF)
+ if (c1 == '\r')
{
- ONE_MORE_BYTE (c2);
- if (c2 == '\n')
- *dst++ = c2;
- else if (coding->mode & CODING_MODE_INHIBIT_INCONSISTENT_EOL)
+ if (coding->eol_type == CODING_EOL_CRLF)
{
- result = CODING_FINISH_INCONSISTENT_EOL;
- goto label_end_of_loop_2;
+ ONE_MORE_BYTE (c2);
+ if (c2 == '\n')
+ c1 = c2;
+ else if (coding->mode
+ & CODING_MODE_INHIBIT_INCONSISTENT_EOL)
+ {
+ coding->result = CODING_FINISH_INCONSISTENT_EOL;
+ goto label_end_of_loop;
+ }
+ else
+ /* To process C2 again, SRC is subtracted by 1. */
+ src--;
}
- else
- /* To process C2 again, SRC is subtracted by 1. */
- *dst++ = c1, src--;
+ else if (coding->eol_type == CODING_EOL_CR)
+ c1 = '\n';
+ }
+ else if (c1 == '\n'
+ && (coding->mode & CODING_MODE_INHIBIT_INCONSISTENT_EOL)
+ && (coding->eol_type == CODING_EOL_CR
+ || coding->eol_type == CODING_EOL_CRLF))
+ {
+ coding->result = CODING_FINISH_INCONSISTENT_EOL;
+ goto label_end_of_loop;
}
- else if (coding->eol_type == CODING_EOL_CR)
- *dst++ = '\n';
- else
- *dst++ = c1;
- }
- else if (c1 == '\n'
- && (coding->mode & CODING_MODE_INHIBIT_INCONSISTENT_EOL)
- && (coding->eol_type == CODING_EOL_CR
- || coding->eol_type == CODING_EOL_CRLF))
- {
- result = CODING_FINISH_INCONSISTENT_EOL;
- goto label_end_of_loop_2;
}
- else
- *dst++ = c1;
- coding->produced_char++;
}
- else if (c1 < 0x80)
- DECODE_SJIS_BIG5_CHARACTER (charset_ascii, c1, /* dummy */ c2);
else
- {
+ {
if (sjis_p)
{
- if (c1 < 0xA0 || (c1 >= 0xE0 && c1 < 0xF0))
+ if (c1 >= 0xF0)
+ goto label_invalid_code;
+ if (c1 < 0xA0 || c1 >= 0xE0)
{
/* SJIS -> JISX0208 */
ONE_MORE_BYTE (c2);
- if (c2 >= 0x40)
- {
- DECODE_SJIS (c1, c2, c3, c4);
- DECODE_SJIS_BIG5_CHARACTER (charset_jisx0208, c3, c4);
- }
- else
- goto label_invalid_code_2;
+ if (c2 < 0x40 || c2 == 0x7F || c2 > 0xFC)
+ goto label_invalid_code;
+ DECODE_SJIS (c1, c2, c1, c2);
+ charset = charset_jisx0208;
}
- else if (c1 < 0xE0)
- /* SJIS -> JISX0201-Kana */
- DECODE_SJIS_BIG5_CHARACTER (charset_katakana_jisx0201, c1,
- /* dummy */ c2);
else
- goto label_invalid_code_1;
+ /* SJIS -> JISX0201-Kana */
+ charset = charset_katakana_jisx0201;
}
else
{
/* BIG5 -> Big5 */
- if (c1 >= 0xA1 && c1 <= 0xFE)
- {
- ONE_MORE_BYTE (c2);
- if ((c2 >= 0x40 && c2 <= 0x7E) || (c2 >= 0xA1 && c2 <= 0xFE))
- {
- int charset;
-
- DECODE_BIG5 (c1, c2, charset, c3, c4);
- DECODE_SJIS_BIG5_CHARACTER (charset, c3, c4);
- }
- else
- goto label_invalid_code_2;
- }
- else
- goto label_invalid_code_1;
+ if (c1 < 0xA1 || c1 > 0xFE)
+ goto label_invalid_code;
+ ONE_MORE_BYTE (c2);
+ if (c2 < 0x40 || (c2 > 0x7E && c2 < 0xA1) || c2 > 0xFE)
+ goto label_invalid_code;
+ DECODE_BIG5 (c1, c2, charset, c1, c2);
}
}
- continue;
-
- label_invalid_code_1:
- *dst++ = c1;
- coding->produced_char++;
- coding->fake_multibyte = 1;
- continue;
- label_invalid_code_2:
- *dst++ = c1; *dst++= c2;
- coding->produced_char += 2;
- coding->fake_multibyte = 1;
+ c = DECODE_ISO_CHARACTER (charset, c1, c2);
+ EMIT_CHAR (c);
continue;
- label_end_of_loop:
- result = CODING_FINISH_INSUFFICIENT_SRC;
- label_end_of_loop_2:
+ label_invalid_code:
+ coding->errors++;
src = src_base;
- break;
- }
-
- if (src < src_end)
- {
- if (result == CODING_FINISH_NORMAL)
- result = CODING_FINISH_INSUFFICIENT_DST;
- else if (result != CODING_FINISH_INCONSISTENT_EOL
- && coding->mode & CODING_MODE_LAST_BLOCK)
- {
- src_bytes = src_end - src;
- if (dst_bytes && (dst_end - dst < src_bytes))
- src_bytes = dst_end - dst;
- bcopy (dst, src, src_bytes);
- src += src_bytes;
- dst += src_bytes;
- coding->fake_multibyte = 1;
- }
+ c = *src++;
+ EMIT_CHAR (c);
}
- coding->consumed = coding->consumed_char = src - source;
+ label_end_of_loop:
+ coding->consumed = coding->consumed_char = src_base - source;
coding->produced = dst - destination;
- return result;
+ return;
}
/* See the above "GENERAL NOTES on `encode_coding_XXX ()' functions".
- This function can encode `charset_ascii', `charset_katakana_jisx0201',
- `charset_jisx0208', `charset_big5_1', and `charset_big5-2'. We are
- sure that all these charsets are registered as official charset
+ This function can encode charsets `ascii', `katakana-jisx0201',
+ `japanese-jisx0208', `chinese-big5-1', and `chinese-big5-2'. We
+ are sure that all these charsets are registered as official charset
(i.e. do not have extended leading-codes). Characters of other
charsets are produced without any encoding. If SJIS_P is 1, encode
SJIS text, else encode BIG5 text. */
-int
+static void
encode_coding_sjis_big5 (coding, source, destination,
src_bytes, dst_bytes, sjis_p)
struct coding_system *coding;
unsigned char *src_end = source + src_bytes;
unsigned char *dst = destination;
unsigned char *dst_end = destination + dst_bytes;
- /* Since the maximum bytes produced by each loop is 2, we subtract 1
- from DST_END to assure overflow checking is necessary only at the
- head of loop. */
- unsigned char *adjusted_dst_end = dst_end - 1;
- Lisp_Object translation_table
- = coding->translation_table_for_encode;
- int result = CODING_FINISH_NORMAL;
+ /* SRC_BASE remembers the start position in source in each loop.
+ The loop will be exited when there's not enough source text to
+ analyze multi-byte codes (within macro ONE_MORE_CHAR), or when
+ there's not enough destination area to produce encoded codes
+ (within macro EMIT_BYTES). */
+ unsigned char *src_base;
+ Lisp_Object translation_table;
- if (!NILP (Venable_character_translation) && NILP (translation_table))
- translation_table = Vstandard_translation_table_for_encode;
+ if (NILP (Venable_character_translation))
+ translation_table = Qnil;
+ else
+ {
+ translation_table = coding->translation_table_for_decode;
+ if (NILP (translation_table))
+ translation_table = Vstandard_translation_table_for_decode;
+ }
- coding->consumed_char = 0;
- coding->fake_multibyte = 0;
- while (src < src_end && (dst_bytes
- ? (dst < adjusted_dst_end)
- : (dst < src - 1)))
+ while (1)
{
- /* SRC_BASE remembers the start position in source in each loop.
- The loop will be exited when there's not enough source text
- to analyze multi-byte codes (within macros ONE_MORE_BYTE and
- TWO_MORE_BYTES). In that case, SRC is reset to SRC_BASE
- before exiting. */
- unsigned char *src_base = src;
- unsigned char c1 = *src++, c2, c3, c4;
-
- if (coding->composing)
+ int c, charset, c1, c2;
+
+ src_base = src;
+ ONE_MORE_CHAR (c);
+
+ /* Now encode the character C. */
+ if (SINGLE_BYTE_CHAR_P (c))
{
- if (c1 == 0xA0)
+ switch (c)
{
- ONE_MORE_BYTE (c1);
- c1 &= 0x7F;
+ case '\r':
+ if (!coding->mode & CODING_MODE_SELECTIVE_DISPLAY)
+ {
+ EMIT_ONE_BYTE (c);
+ break;
+ }
+ c = '\n';
+ case '\n':
+ if (coding->eol_type == CODING_EOL_CRLF)
+ {
+ EMIT_TWO_BYTES ('\r', c);
+ break;
+ }
+ else if (coding->eol_type == CODING_EOL_CR)
+ c = '\r';
+ default:
+ EMIT_ONE_BYTE (c);
}
- else if (c1 >= 0xA0)
- c1 -= 0x20;
- else
- coding->composing = 0;
}
-
- switch (emacs_code_class[c1])
+ else
{
- case EMACS_ascii_code:
- ENCODE_SJIS_BIG5_CHARACTER (charset_ascii, c1, /* dummy */ c2);
- break;
-
- case EMACS_control_code:
- *dst++ = c1;
- coding->consumed_char++;
- break;
-
- case EMACS_carriage_return_code:
- if (! (coding->mode & CODING_MODE_SELECTIVE_DISPLAY))
+ SPLIT_CHAR (c, charset, c1, c2);
+ if (sjis_p)
{
- *dst++ = c1;
- coding->consumed_char++;
- break;
+ if (charset == charset_jisx0208
+ || charset == charset_jisx0208_1978)
+ {
+ ENCODE_SJIS (c1, c2, c1, c2);
+ EMIT_TWO_BYTES (c1, c2);
+ }
+ else if (charset == charset_latin_jisx0201)
+ EMIT_ONE_BYTE (c1);
+ else
+ /* There's no way other than producing the internal
+ codes as is. */
+ EMIT_BYTES (src_base, src);
}
- /* fall down to treat '\r' as '\n' ... */
-
- case EMACS_linefeed_code:
- if (coding->eol_type == CODING_EOL_LF
- || coding->eol_type == CODING_EOL_UNDECIDED)
- *dst++ = '\n';
- else if (coding->eol_type == CODING_EOL_CRLF)
- *dst++ = '\r', *dst++ = '\n';
else
- *dst++ = '\r';
- coding->consumed_char++;
- break;
-
- case EMACS_leading_code_2:
- ONE_MORE_BYTE (c2);
- ENCODE_SJIS_BIG5_CHARACTER (c1, c2, /* dummy */ c3);
- break;
-
- case EMACS_leading_code_3:
- TWO_MORE_BYTES (c2, c3);
- ENCODE_SJIS_BIG5_CHARACTER (c1, c2, c3);
- break;
-
- case EMACS_leading_code_4:
- THREE_MORE_BYTES (c2, c3, c4);
- ENCODE_SJIS_BIG5_CHARACTER (c2, c3, c4);
- break;
-
- case EMACS_leading_code_composition:
- coding->composing = 1;
- break;
-
- default: /* i.e. case EMACS_invalid_code: */
- *dst++ = c1;
- coding->consumed_char++;
+ {
+ if (charset == charset_big5_1 || charset == charset_big5_2)
+ {
+ ENCODE_BIG5 (charset, c1, c2, c1, c2);
+ EMIT_TWO_BYTES (c1, c2);
+ }
+ else
+ /* There's no way other than producing the internal
+ codes as is. */
+ EMIT_BYTES (src_base, src);
+ }
}
- continue;
-
- label_end_of_loop:
- result = CODING_FINISH_INSUFFICIENT_SRC;
- src = src_base;
- break;
+ coding->consumed_char++;
}
- if (result == CODING_FINISH_NORMAL
- && src < src_end)
- result = CODING_FINISH_INSUFFICIENT_DST;
- coding->consumed = src - source;
+ label_end_of_loop:
+ coding->consumed = src_base - source;
coding->produced = coding->produced_char = dst - destination;
- return result;
}
\f
unsigned char *src, *src_end;
{
unsigned char *valid;
+ int c;
+ /* Dummy for ONE_MORE_BYTE. */
+ struct coding_system dummy_coding;
+ struct coding_system *coding = &dummy_coding;
/* No coding system is assigned to coding-category-ccl. */
if (!coding_system_table[CODING_CATEGORY_IDX_CCL])
return 0;
valid = coding_system_table[CODING_CATEGORY_IDX_CCL]->spec.ccl.valid_codes;
- while (src < src_end)
+ while (1)
{
- if (! valid[*src]) return 0;
- src++;
+ ONE_MORE_BYTE (c);
+ if (! valid[c])
+ return 0;
}
+ label_end_of_loop:
return CODING_CATEGORY_MASK_CCL;
}
\f
/*** 6. End-of-line handlers ***/
-/* See the above "GENERAL NOTES on `decode_coding_XXX ()' functions".
- This function is called only when `coding->eol_type' is
- CODING_EOL_CRLF or CODING_EOL_CR. */
+/* See the above "GENERAL NOTES on `decode_coding_XXX ()' functions". */
-int
+static void
decode_eol (coding, source, destination, src_bytes, dst_bytes)
struct coding_system *coding;
unsigned char *source, *destination;
int src_bytes, dst_bytes;
{
unsigned char *src = source;
- unsigned char *src_end = source + src_bytes;
unsigned char *dst = destination;
- unsigned char *dst_end = destination + dst_bytes;
- unsigned char c;
- int result = CODING_FINISH_NORMAL;
-
- coding->fake_multibyte = 0;
-
- if (src_bytes <= 0)
- return result;
-
+ unsigned char *src_end = src + src_bytes;
+ unsigned char *dst_end = dst + dst_bytes;
+ Lisp_Object translation_table;
+ /* SRC_BASE remembers the start position in source in each loop.
+ The loop will be exited when there's not enough source code
+ (within macro ONE_MORE_BYTE), or when there's not enough
+ destination area to produce a character (within macro
+ EMIT_CHAR). */
+ unsigned char *src_base;
+ int c;
+
+ translation_table = Qnil;
switch (coding->eol_type)
{
case CODING_EOL_CRLF:
- {
- /* Since the maximum bytes produced by each loop is 2, we
- subtract 1 from DST_END to assure overflow checking is
- necessary only at the head of loop. */
- unsigned char *adjusted_dst_end = dst_end - 1;
-
- while (src < src_end && (dst_bytes
- ? (dst < adjusted_dst_end)
- : (dst < src - 1)))
- {
- unsigned char *src_base = src;
-
- c = *src++;
- if (c == '\r')
- {
- ONE_MORE_BYTE (c);
- if (c == '\n')
- *dst++ = c;
- else
- {
- if (coding->mode & CODING_MODE_INHIBIT_INCONSISTENT_EOL)
- {
- result = CODING_FINISH_INCONSISTENT_EOL;
- goto label_end_of_loop_2;
- }
- src--;
- *dst++ = '\r';
- if (BASE_LEADING_CODE_P (c))
- coding->fake_multibyte = 1;
- }
- }
- else if (c == '\n'
- && (coding->mode & CODING_MODE_INHIBIT_INCONSISTENT_EOL))
- {
- result = CODING_FINISH_INCONSISTENT_EOL;
- goto label_end_of_loop_2;
- }
- else
- {
- *dst++ = c;
- if (BASE_LEADING_CODE_P (c))
- coding->fake_multibyte = 1;
- }
- continue;
-
- label_end_of_loop:
- result = CODING_FINISH_INSUFFICIENT_SRC;
- label_end_of_loop_2:
- src = src_base;
- break;
- }
- if (src < src_end)
- {
- if (result == CODING_FINISH_NORMAL)
- result = CODING_FINISH_INSUFFICIENT_DST;
- else if (result != CODING_FINISH_INCONSISTENT_EOL
- && coding->mode & CODING_MODE_LAST_BLOCK)
- {
- /* This is the last block of the text to be decoded.
- We flush out all remaining codes. */
- src_bytes = src_end - src;
- if (dst_bytes && (dst_end - dst < src_bytes))
- src_bytes = dst_end - dst;
- bcopy (src, dst, src_bytes);
- dst += src_bytes;
- src += src_bytes;
- }
- }
- }
- break;
-
- case CODING_EOL_CR:
- if (coding->mode & CODING_MODE_INHIBIT_INCONSISTENT_EOL)
+ while (1)
{
- while (src < src_end)
+ src_base = src;
+ ONE_MORE_BYTE (c);
+ if (c == '\r')
{
- if ((c = *src++) == '\n')
- break;
- if (BASE_LEADING_CODE_P (c))
- coding->fake_multibyte = 1;
+ ONE_MORE_BYTE (c);
+ if (c != '\n')
+ {
+ if (coding->mode & CODING_MODE_INHIBIT_INCONSISTENT_EOL)
+ {
+ coding->result = CODING_FINISH_INCONSISTENT_EOL;
+ goto label_end_of_loop;
+ }
+ src--;
+ c = '\r';
+ }
}
- if (*--src == '\n')
+ else if (c == '\n'
+ && (coding->mode & CODING_MODE_INHIBIT_INCONSISTENT_EOL))
{
- src_bytes = src - source;
- result = CODING_FINISH_INCONSISTENT_EOL;
+ coding->result = CODING_FINISH_INCONSISTENT_EOL;
+ goto label_end_of_loop;
}
+ EMIT_CHAR (c);
}
- if (dst_bytes && src_bytes > dst_bytes)
+ break;
+
+ case CODING_EOL_CR:
+ while (1)
{
- result = CODING_FINISH_INSUFFICIENT_DST;
- src_bytes = dst_bytes;
+ src_base = src;
+ ONE_MORE_BYTE (c);
+ if (c == '\n')
+ {
+ if (coding->mode & CODING_MODE_INHIBIT_INCONSISTENT_EOL)
+ {
+ coding->result = CODING_FINISH_INCONSISTENT_EOL;
+ goto label_end_of_loop;
+ }
+ }
+ else if (c == '\r')
+ c = '\n';
+ EMIT_CHAR (c);
}
- if (dst_bytes)
- bcopy (source, destination, src_bytes);
- else
- safe_bcopy (source, destination, src_bytes);
- src = source + src_bytes;
- while (src_bytes--) if (*dst++ == '\r') dst[-1] = '\n';
break;
- default: /* i.e. case: CODING_EOL_LF */
- if (dst_bytes && src_bytes > dst_bytes)
+ default: /* no need for EOL handling */
+ while (1)
{
- result = CODING_FINISH_INSUFFICIENT_DST;
- src_bytes = dst_bytes;
+ src_base = src;
+ ONE_MORE_BYTE (c);
+ EMIT_CHAR (c);
}
- if (dst_bytes)
- bcopy (source, destination, src_bytes);
- else
- safe_bcopy (source, destination, src_bytes);
- src += src_bytes;
- dst += src_bytes;
- coding->fake_multibyte = 1;
- break;
}
- coding->consumed = coding->consumed_char = src - source;
- coding->produced = coding->produced_char = dst - destination;
- return result;
+ label_end_of_loop:
+ coding->consumed = coding->consumed_char = src_base - source;
+ coding->produced = dst - destination;
+ return;
}
/* See "GENERAL NOTES about `encode_coding_XXX ()' functions". Encode
- format of end-of-line according to `coding->eol_type'. If
- `coding->mode & CODING_MODE_SELECTIVE_DISPLAY' is nonzero, code
- '\r' in source text also means end-of-line. */
+ format of end-of-line according to `coding->eol_type'. It also
+ convert multibyte form 8-bit characers to unibyte if
+ CODING->src_multibyte is nonzero. If `coding->mode &
+ CODING_MODE_SELECTIVE_DISPLAY' is nonzero, code '\r' in source text
+ also means end-of-line. */
-int
+static void
encode_eol (coding, source, destination, src_bytes, dst_bytes)
struct coding_system *coding;
unsigned char *source, *destination;
{
unsigned char *src = source;
unsigned char *dst = destination;
- int result = CODING_FINISH_NORMAL;
-
- coding->fake_multibyte = 0;
+ unsigned char *src_end = src + src_bytes;
+ unsigned char *dst_end = dst + dst_bytes;
+ Lisp_Object translation_table;
+ /* SRC_BASE remembers the start position in source in each loop.
+ The loop will be exited when there's not enough source text to
+ analyze multi-byte codes (within macro ONE_MORE_CHAR), or when
+ there's not enough destination area to produce encoded codes
+ (within macro EMIT_BYTES). */
+ unsigned char *src_base;
+ int c;
+ int selective_display = coding->mode & CODING_MODE_SELECTIVE_DISPLAY;
+
+ translation_table = Qnil;
+ if (coding->src_multibyte
+ && *(src_end - 1) == LEADING_CODE_8_BIT_CONTROL)
+ {
+ src_end--;
+ src_bytes--;
+ coding->result = CODING_FINISH_INSUFFICIENT_SRC;
+ }
if (coding->eol_type == CODING_EOL_CRLF)
{
- unsigned char c;
- unsigned char *src_end = source + src_bytes;
- unsigned char *dst_end = destination + dst_bytes;
- /* Since the maximum bytes produced by each loop is 2, we
- subtract 1 from DST_END to assure overflow checking is
- necessary only at the head of loop. */
- unsigned char *adjusted_dst_end = dst_end - 1;
-
- while (src < src_end && (dst_bytes
- ? (dst < adjusted_dst_end)
- : (dst < src - 1)))
+ while (src < src_end)
{
+ src_base = src;
c = *src++;
- if (c == '\n'
- || (c == '\r' && (coding->mode & CODING_MODE_SELECTIVE_DISPLAY)))
- *dst++ = '\r', *dst++ = '\n';
+ if (c >= 0x20)
+ EMIT_ONE_BYTE (c);
+ else if (c == '\n' || (c == '\r' && selective_display))
+ EMIT_TWO_BYTES ('\r', '\n');
else
- {
- *dst++ = c;
- if (BASE_LEADING_CODE_P (c))
- coding->fake_multibyte = 1;
- }
+ EMIT_ONE_BYTE (c);
}
- if (src < src_end)
- result = CODING_FINISH_INSUFFICIENT_DST;
+ src_base = src;
+ label_end_of_loop:
+ ;
}
else
{
- unsigned char c;
-
- if (dst_bytes && src_bytes > dst_bytes)
+ if (src_bytes <= dst_bytes)
{
- src_bytes = dst_bytes;
- result = CODING_FINISH_INSUFFICIENT_DST;
+ safe_bcopy (src, dst, src_bytes);
+ src_base = src_end;
+ dst += src_bytes;
}
- if (dst_bytes)
- bcopy (source, destination, src_bytes);
else
- safe_bcopy (source, destination, src_bytes);
- dst_bytes = src_bytes;
+ {
+ if (coding->src_multibyte
+ && *(src + dst_bytes - 1) == LEADING_CODE_8_BIT_CONTROL)
+ dst_bytes--;
+ safe_bcopy (src, dst, dst_bytes);
+ src_base = src + dst_bytes;
+ dst = destination + dst_bytes;
+ coding->result = CODING_FINISH_INSUFFICIENT_DST;
+ }
if (coding->eol_type == CODING_EOL_CR)
{
- while (src_bytes--)
- {
- if ((c = *dst++) == '\n')
- dst[-1] = '\r';
- else if (BASE_LEADING_CODE_P (c))
- coding->fake_multibyte = 1;
- }
+ for (src = destination; src < dst; src++)
+ if (*src == '\n') *src = '\r';
}
- else
+ else if (selective_display)
{
- if (coding->mode & CODING_MODE_SELECTIVE_DISPLAY)
- {
- while (src_bytes--)
- if (*dst++ == '\r') dst[-1] = '\n';
- }
- coding->fake_multibyte = 1;
+ for (src = destination; src < dst; src++)
+ if (*src == '\r') *src = '\n';
}
- src = source + dst_bytes;
- dst = destination + dst_bytes;
}
+ if (coding->src_multibyte)
+ dst = destination + str_as_unibyte (destination, dst - destination);
- coding->consumed = coding->consumed_char = src - source;
- coding->produced = coding->produced_char = dst - destination;
- return result;
+ coding->consumed = src_base - source;
+ coding->produced = dst - destination;
}
\f
coding->mode = 0;
coding->heading_ascii = -1;
coding->post_read_conversion = coding->pre_write_conversion = Qnil;
+ coding->composing = COMPOSITION_DISABLED;
+ coding->cmp_data = NULL;
+
+ if (NILP (coding_system))
+ goto label_invalid_coding_system;
+
coding_spec = Fget (coding_system, Qcoding_system);
+
if (!VECTORP (coding_spec)
|| XVECTOR (coding_spec)->size != 5
|| !CONSP (XVECTOR (coding_spec)->contents[3]))
return 0;
}
- /* Initialize remaining fields. */
- coding->composing = 0;
-
/* Get values of coding system properties:
`post-read-conversion', `pre-write-conversion',
`translation-table-for-decode', `translation-table-for-encode'. */
plist = XVECTOR (coding_spec)->contents[3];
- coding->post_read_conversion = Fplist_get (plist, Qpost_read_conversion);
- coding->pre_write_conversion = Fplist_get (plist, Qpre_write_conversion);
+ /* Pre & post conversion functions should be disabled if
+ inhibit_eol_conversion is nozero. This is the case that a code
+ conversion function is called while those functions are running. */
+ if (! inhibit_pre_post_conversion)
+ {
+ coding->post_read_conversion = Fplist_get (plist, Qpost_read_conversion);
+ coding->pre_write_conversion = Fplist_get (plist, Qpre_write_conversion);
+ }
val = Fplist_get (plist, Qtranslation_table_for_decode);
if (SYMBOLP (val))
val = Fget (val, Qtranslation_table_for_decode);
else
goto label_invalid_coding_system;
- val = Fplist_get (plist, Qsafe_charsets);
- if (EQ (val, Qt))
- {
- for (i = 0; i <= MAX_CHARSET; i++)
- coding->safe_charsets[i] = 1;
- }
- else
- {
- bzero (coding->safe_charsets, MAX_CHARSET + 1);
- while (CONSP (val))
- {
- if ((i = get_charset_id (XCONS (val)->car)) >= 0)
- coding->safe_charsets[i] = 1;
- val = XCONS (val)->cdr;
- }
- }
+ /* If the coding system has non-nil `composition' property, enable
+ composition handling. */
+ val = Fplist_get (plist, Qcomposition);
+ if (!NILP (val))
+ coding->composing = COMPOSITION_NO;
switch (XFASTINT (coding_type))
{
val = Vcharset_revision_alist;
while (CONSP (val))
{
- charset = get_charset_id (Fcar_safe (XCONS (val)->car));
+ charset = get_charset_id (Fcar_safe (XCAR (val)));
if (charset >= 0
- && (temp = Fcdr_safe (XCONS (val)->car), INTEGERP (temp))
+ && (temp = Fcdr_safe (XCAR (val)), INTEGERP (temp))
&& (i = XINT (temp), (i >= 0 && (i + '@') < 128)))
CODING_SPEC_ISO_REVISION_NUMBER (coding, charset) = i;
- val = XCONS (val)->cdr;
+ val = XCDR (val);
}
/* Checks FLAGS[REG] (REG = 0, 1, 2 3) and decide designations.
tail = flags[i];
coding->flags |= CODING_FLAG_ISO_DESIGNATION;
- if (INTEGERP (XCONS (tail)->car)
- && (charset = XINT (XCONS (tail)->car),
+ if (INTEGERP (XCAR (tail))
+ && (charset = XINT (XCAR (tail)),
CHARSET_VALID_P (charset))
- || (charset = get_charset_id (XCONS (tail)->car)) >= 0)
+ || (charset = get_charset_id (XCAR (tail))) >= 0)
{
CODING_SPEC_ISO_INITIAL_DESIGNATION (coding, i) = charset;
CODING_SPEC_ISO_REQUESTED_DESIGNATION (coding, charset) =i;
}
else
CODING_SPEC_ISO_INITIAL_DESIGNATION (coding, i) = -1;
- tail = XCONS (tail)->cdr;
+ tail = XCDR (tail);
while (CONSP (tail))
{
- if (INTEGERP (XCONS (tail)->car)
- && (charset = XINT (XCONS (tail)->car),
+ if (INTEGERP (XCAR (tail))
+ && (charset = XINT (XCAR (tail)),
CHARSET_VALID_P (charset))
- || (charset = get_charset_id (XCONS (tail)->car)) >= 0)
+ || (charset = get_charset_id (XCAR (tail))) >= 0)
CODING_SPEC_ISO_REQUESTED_DESIGNATION (coding, charset)
= i;
- else if (EQ (XCONS (tail)->car, Qt))
+ else if (EQ (XCAR (tail), Qt))
reg_bits |= 1 << i;
- tail = XCONS (tail)->cdr;
+ tail = XCDR (tail);
}
}
else
if (reg_bits)
for (charset = 0; charset <= MAX_CHARSET; charset++)
{
- if (CHARSET_VALID_P (charset))
+ if (CHARSET_VALID_P (charset)
+ && (CODING_SPEC_ISO_REQUESTED_DESIGNATION (coding, charset)
+ == CODING_SPEC_ISO_NO_REQUESTED_DESIGNATION))
{
/* There exist some default graphic registers to be
- used CHARSET. */
+ used by CHARSET. */
/* We had better avoid designating a charset of
CHARS96 to REG 0 as far as possible. */
coding->common_flags
|= CODING_REQUIRE_DECODING_MASK | CODING_REQUIRE_ENCODING_MASK;
{
- Lisp_Object val;
- Lisp_Object decoder, encoder;
-
val = XVECTOR (coding_spec)->contents[4];
- if (CONSP (val)
- && SYMBOLP (XCONS (val)->car)
- && !NILP (decoder = Fget (XCONS (val)->car, Qccl_program_idx))
- && !NILP (decoder = Fcdr (Faref (Vccl_program_table, decoder)))
- && SYMBOLP (XCONS (val)->cdr)
- && !NILP (encoder = Fget (XCONS (val)->cdr, Qccl_program_idx))
- && !NILP (encoder = Fcdr (Faref (Vccl_program_table, encoder))))
- {
- setup_ccl_program (&(coding->spec.ccl.decoder), decoder);
- setup_ccl_program (&(coding->spec.ccl.encoder), encoder);
- }
- else
+ if (! CONSP (val)
+ || setup_ccl_program (&(coding->spec.ccl.decoder),
+ XCAR (val)) < 0
+ || setup_ccl_program (&(coding->spec.ccl.encoder),
+ XCDR (val)) < 0)
goto label_invalid_coding_system;
bzero (coding->spec.ccl.valid_codes, 256);
{
Lisp_Object this;
- for (; CONSP (val); val = XCONS (val)->cdr)
+ for (; CONSP (val); val = XCDR (val))
{
- this = XCONS (val)->car;
+ this = XCAR (val);
if (INTEGERP (this)
&& XINT (this) >= 0 && XINT (this) < 256)
coding->spec.ccl.valid_codes[XINT (this)] = 1;
else if (CONSP (this)
- && INTEGERP (XCONS (this)->car)
- && INTEGERP (XCONS (this)->cdr))
+ && INTEGERP (XCAR (this))
+ && INTEGERP (XCDR (this)))
{
- int start = XINT (XCONS (this)->car);
- int end = XINT (XCONS (this)->cdr);
+ int start = XINT (XCAR (this));
+ int end = XINT (XCDR (this));
if (start >= 0 && start <= end && end < 256)
while (start <= end)
}
}
coding->common_flags |= CODING_REQUIRE_FLUSHING_MASK;
+ coding->spec.ccl.cr_carryover = 0;
break;
case 5:
return -1;
}
+/* Free memory blocks allocated for storing composition information. */
+
+void
+coding_free_composition_data (coding)
+ struct coding_system *coding;
+{
+ struct composition_data *cmp_data = coding->cmp_data, *next;
+
+ if (!cmp_data)
+ return;
+ /* Memory blocks are chained. At first, rewind to the first, then,
+ free blocks one by one. */
+ while (cmp_data->prev)
+ cmp_data = cmp_data->prev;
+ while (cmp_data)
+ {
+ next = cmp_data->next;
+ xfree (cmp_data);
+ cmp_data = next;
+ }
+ coding->cmp_data = NULL;
+}
+
+/* Set `char_offset' member of all memory blocks pointed by
+ coding->cmp_data to POS. */
+
+void
+coding_adjust_composition_offset (coding, pos)
+ struct coding_system *coding;
+ int pos;
+{
+ struct composition_data *cmp_data;
+
+ for (cmp_data = coding->cmp_data; cmp_data; cmp_data = cmp_data->next)
+ cmp_data->char_offset = pos;
+}
+
/* Setup raw-text or one of its subsidiaries in the structure
coding_system CODING according to the already setup value eol_type
in CODING. CODING should be setup for some coding system in
coding->symbol
= XVECTOR (subsidiaries)->contents[coding->eol_type];
}
+ setup_coding_system (coding->symbol, coding);
}
return;
}
as BIG5. Assigned the coding-system (Lisp symbol)
`cn-big5' by default.
+ o coding-category-utf-8
+
+ The category for a coding system which has the same code range
+ as UTF-8 (cf. RFC2279). Assigned the coding-system (Lisp
+ symbol) `utf-8' by default.
+
+ o coding-category-utf-16-be
+
+ The category for a coding system in which a text has an
+ Unicode signature (cf. Unicode Standard) in the order of BIG
+ endian at the head. Assigned the coding-system (Lisp symbol)
+ `utf-16-be' by default.
+
+ o coding-category-utf-16-le
+
+ The category for a coding system in which a text has an
+ Unicode signature (cf. Unicode Standard) in the order of
+ LITTLE endian at the head. Assigned the coding-system (Lisp
+ symbol) `utf-16-le' by default.
+
o coding-category-ccl
The category for a coding system of which encoder/decoder is
/* Detect how a text of length SRC_BYTES pointed by SOURCE is encoded.
If it detects possible coding systems, return an integer in which
appropriate flag bits are set. Flag bits are defined by macros
- CODING_CATEGORY_MASK_XXX in `coding.h'.
+ CODING_CATEGORY_MASK_XXX in `coding.h'. If PRIORITIES is non-NULL,
+ it should point the table `coding_priorities'. In that case, only
+ the flag bit for a coding system of the highest priority is set in
+ the returned value.
How many ASCII characters are at the head is returned as *SKIP. */
{
register unsigned char c;
unsigned char *src = source, *src_end = source + src_bytes;
- unsigned int mask;
- int i;
+ unsigned int mask, utf16_examined_p, iso2022_examined_p;
+ int i, idx;
/* At first, skip all ASCII characters and control characters except
for three ISO2022 specific control characters. */
goto label_loop_detect_coding;
}
if (priorities)
- goto label_return_highest_only;
+ {
+ for (i = 0; i < CODING_CATEGORY_IDX_MAX; i++)
+ {
+ if (mask & priorities[i])
+ return priorities[i];
+ }
+ return CODING_CATEGORY_MASK_RAW_TEXT;
+ }
}
else
{
if (c < 0xA0)
{
/* C is the first byte of SJIS character code,
- or a leading-code of Emacs' internal format (emacs-mule). */
- try = CODING_CATEGORY_MASK_SJIS | CODING_CATEGORY_MASK_EMACS_MULE;
+ or a leading-code of Emacs' internal format (emacs-mule),
+ or the first byte of UTF-16. */
+ try = (CODING_CATEGORY_MASK_SJIS
+ | CODING_CATEGORY_MASK_EMACS_MULE
+ | CODING_CATEGORY_MASK_UTF_16_BE
+ | CODING_CATEGORY_MASK_UTF_16_LE);
/* Or, if C is a special latin extra code,
or is an ISO2022 specific control code of C1 (SS2 or SS3),
else
/* C is a character of ISO2022 in graphic plane right,
or a SJIS's 1-byte character code (i.e. JISX0201),
- or the first byte of BIG5's 2-byte code. */
+ or the first byte of BIG5's 2-byte code,
+ or the first byte of UTF-8/16. */
try = (CODING_CATEGORY_MASK_ISO_8_ELSE
| CODING_CATEGORY_MASK_ISO_8BIT
| CODING_CATEGORY_MASK_SJIS
- | CODING_CATEGORY_MASK_BIG5);
+ | CODING_CATEGORY_MASK_BIG5
+ | CODING_CATEGORY_MASK_UTF_8
+ | CODING_CATEGORY_MASK_UTF_16_BE
+ | CODING_CATEGORY_MASK_UTF_16_LE);
/* Or, we may have to consider the possibility of CCL. */
if (coding_system_table[CODING_CATEGORY_IDX_CCL]
try |= CODING_CATEGORY_MASK_CCL;
mask = 0;
+ utf16_examined_p = iso2022_examined_p = 0;
if (priorities)
{
for (i = 0; i < CODING_CATEGORY_IDX_MAX; i++)
{
- if (priorities[i] & try & CODING_CATEGORY_MASK_ISO)
- mask = detect_coding_iso2022 (src, src_end);
+ if (!iso2022_examined_p
+ && (priorities[i] & try & CODING_CATEGORY_MASK_ISO))
+ {
+ mask |= detect_coding_iso2022 (src, src_end);
+ iso2022_examined_p = 1;
+ }
else if (priorities[i] & try & CODING_CATEGORY_MASK_SJIS)
- mask = detect_coding_sjis (src, src_end);
+ mask |= detect_coding_sjis (src, src_end);
+ else if (priorities[i] & try & CODING_CATEGORY_MASK_UTF_8)
+ mask |= detect_coding_utf_8 (src, src_end);
+ else if (!utf16_examined_p
+ && (priorities[i] & try &
+ CODING_CATEGORY_MASK_UTF_16_BE_LE))
+ {
+ mask |= detect_coding_utf_16 (src, src_end);
+ utf16_examined_p = 1;
+ }
else if (priorities[i] & try & CODING_CATEGORY_MASK_BIG5)
- mask = detect_coding_big5 (src, src_end);
+ mask |= detect_coding_big5 (src, src_end);
else if (priorities[i] & try & CODING_CATEGORY_MASK_EMACS_MULE)
- mask = detect_coding_emacs_mule (src, src_end);
+ mask |= detect_coding_emacs_mule (src, src_end);
else if (priorities[i] & try & CODING_CATEGORY_MASK_CCL)
- mask = detect_coding_ccl (src, src_end);
+ mask |= detect_coding_ccl (src, src_end);
else if (priorities[i] & CODING_CATEGORY_MASK_RAW_TEXT)
- mask = CODING_CATEGORY_MASK_RAW_TEXT;
+ mask |= CODING_CATEGORY_MASK_RAW_TEXT;
else if (priorities[i] & CODING_CATEGORY_MASK_BINARY)
- mask = CODING_CATEGORY_MASK_BINARY;
- if (mask)
- goto label_return_highest_only;
+ mask |= CODING_CATEGORY_MASK_BINARY;
+ if (mask & priorities[i])
+ return priorities[i];
}
return CODING_CATEGORY_MASK_RAW_TEXT;
}
mask |= detect_coding_sjis (src, src_end);
if (try & CODING_CATEGORY_MASK_BIG5)
mask |= detect_coding_big5 (src, src_end);
+ if (try & CODING_CATEGORY_MASK_UTF_8)
+ mask |= detect_coding_utf_8 (src, src_end);
+ if (try & CODING_CATEGORY_MASK_UTF_16_BE_LE)
+ mask |= detect_coding_utf_16 (src, src_end);
if (try & CODING_CATEGORY_MASK_EMACS_MULE)
mask |= detect_coding_emacs_mule (src, src_end);
if (try & CODING_CATEGORY_MASK_CCL)
mask |= detect_coding_ccl (src, src_end);
}
return (mask | CODING_CATEGORY_MASK_RAW_TEXT | CODING_CATEGORY_MASK_BINARY);
-
- label_return_highest_only:
- for (i = 0; i < CODING_CATEGORY_IDX_MAX; i++)
- {
- if (mask & priorities[i])
- return priorities[i];
- }
- return CODING_CATEGORY_MASK_RAW_TEXT;
}
/* Detect how a text of length SRC_BYTES pointed by SRC is encoded.
if (VECTORP (tmp))
val = XVECTOR (tmp)->contents[coding->eol_type];
}
- setup_coding_system (val, coding);
- /* Set this again because setup_coding_system reset this member. */
- coding->heading_ascii = skip;
+
+ /* Setup this new coding system while preserving some slots. */
+ {
+ int src_multibyte = coding->src_multibyte;
+ int dst_multibyte = coding->dst_multibyte;
+
+ setup_coding_system (val, coding);
+ coding->src_multibyte = src_multibyte;
+ coding->dst_multibyte = dst_multibyte;
+ coding->heading_ascii = skip;
+ }
}
/* Detect how end-of-line of a text of length SRC_BYTES pointed by
return eol_type;
}
+/* Like detect_eol_type, but detect EOL type in 2-octet
+ big-endian/little-endian format for coding systems utf-16-be and
+ utf-16-le. */
+
+static int
+detect_eol_type_in_2_octet_form (source, src_bytes, skip, big_endian_p)
+ unsigned char *source;
+ int src_bytes, *skip;
+{
+ unsigned char *src = source, *src_end = src + src_bytes;
+ unsigned int c1, c2;
+ int total = 0; /* How many end-of-lines are found so far. */
+ int eol_type = CODING_EOL_UNDECIDED;
+ int this_eol_type;
+ int msb, lsb;
+
+ if (big_endian_p)
+ msb = 0, lsb = 1;
+ else
+ msb = 1, lsb = 0;
+
+ *skip = 0;
+
+ while ((src + 1) < src_end && total < MAX_EOL_CHECK_COUNT)
+ {
+ c1 = (src[msb] << 8) | (src[lsb]);
+ src += 2;
+
+ if (c1 == '\n' || c1 == '\r')
+ {
+ if (*skip == 0)
+ *skip = src - 2 - source;
+ total++;
+ if (c1 == '\n')
+ {
+ this_eol_type = CODING_EOL_LF;
+ }
+ else
+ {
+ if ((src + 1) >= src_end)
+ {
+ this_eol_type = CODING_EOL_CR;
+ }
+ else
+ {
+ c2 = (src[msb] << 8) | (src[lsb]);
+ if (c2 == '\n')
+ this_eol_type = CODING_EOL_CRLF, src += 2;
+ else
+ this_eol_type = CODING_EOL_CR;
+ }
+ }
+
+ if (eol_type == CODING_EOL_UNDECIDED)
+ /* This is the first end-of-line. */
+ eol_type = this_eol_type;
+ else if (eol_type != this_eol_type)
+ {
+ /* The found type is different from what found before. */
+ eol_type = CODING_EOL_INCONSISTENT;
+ break;
+ }
+ }
+ }
+
+ if (*skip == 0)
+ *skip = src_end - source;
+ return eol_type;
+}
+
/* Detect how end-of-line of a text of length SRC_BYTES pointed by SRC
is encoded. If it detects an appropriate format of end-of-line, it
sets the information in *CODING. */
{
Lisp_Object val;
int skip;
- int eol_type = detect_eol_type (src, src_bytes, &skip);
+ int eol_type;
+
+ switch (coding->category_idx)
+ {
+ case CODING_CATEGORY_IDX_UTF_16_BE:
+ eol_type = detect_eol_type_in_2_octet_form (src, src_bytes, &skip, 1);
+ break;
+ case CODING_CATEGORY_IDX_UTF_16_LE:
+ eol_type = detect_eol_type_in_2_octet_form (src, src_bytes, &skip, 0);
+ break;
+ default:
+ eol_type = detect_eol_type (src, src_bytes, &skip);
+ break;
+ }
if (coding->heading_ascii > skip)
coding->heading_ascii = skip;
val = Fget (coding->symbol, Qeol_type);
if (VECTORP (val) && XVECTOR (val)->size == 3)
{
+ int src_multibyte = coding->src_multibyte;
+ int dst_multibyte = coding->dst_multibyte;
+
setup_coding_system (XVECTOR (val)->contents[eol_type], coding);
+ coding->src_multibyte = src_multibyte;
+ coding->dst_multibyte = dst_multibyte;
coding->heading_ascii = skip;
}
}
#define CONVERSION_BUFFER_EXTRA_ROOM 256
-#define DECODING_BUFFER_MAG(coding) \
- (coding->type == coding_type_iso2022 \
- ? 3 \
- : ((coding->type == coding_type_sjis || coding->type == coding_type_big5) \
- ? 2 \
- : (coding->type == coding_type_raw_text \
- ? 1 \
- : (coding->type == coding_type_ccl \
- ? coding->spec.ccl.decoder.buf_magnification \
- : 2))))
+#define DECODING_BUFFER_MAG(coding) \
+ (coding->type == coding_type_iso2022 \
+ ? 3 \
+ : (coding->type == coding_type_ccl \
+ ? coding->spec.ccl.decoder.buf_magnification \
+ : 2))
/* Return maximum size (bytes) of a buffer enough for decoding
SRC_BYTES of text encoded in CODING. */
if (coding->type == coding_type_ccl)
magnification = coding->spec.ccl.encoder.buf_magnification;
- else
+ else if (CODING_REQUIRE_ENCODING (coding))
magnification = 3;
+ else
+ magnification = 1;
return (src_bytes * magnification + CONVERSION_BUFFER_EXTRA_ROOM);
}
int result;
ccl->last_block = coding->mode & CODING_MODE_LAST_BLOCK;
-
+ if (encodep)
+ ccl->eol_type = coding->eol_type;
coding->produced = ccl_driver (ccl, source, destination,
src_bytes, dst_bytes, &(coding->consumed));
- coding->produced_char
- = multibyte_chars_in_text (destination, coding->produced);
- coding->consumed_char
- = multibyte_chars_in_text (source, coding->consumed);
+ if (encodep)
+ coding->produced_char = coding->produced;
+ else
+ {
+ int bytes
+ = dst_bytes ? dst_bytes : source + coding->consumed - destination;
+ coding->produced = str_as_multibyte (destination, bytes,
+ coding->produced,
+ &(coding->produced_char));
+ }
switch (ccl->status)
{
return result;
}
+/* Decode EOL format of the text at PTR of BYTES length destructively
+ according to CODING->eol_type. This is called after the CCL
+ program produced a decoded text at PTR. If we do CRLF->LF
+ conversion, update CODING->produced and CODING->produced_char. */
+
+static void
+decode_eol_post_ccl (coding, ptr, bytes)
+ struct coding_system *coding;
+ unsigned char *ptr;
+ int bytes;
+{
+ Lisp_Object val, saved_coding_symbol;
+ unsigned char *pend = ptr + bytes;
+ int dummy;
+
+ /* Remember the current coding system symbol. We set it back when
+ an inconsistent EOL is found so that `last-coding-system-used' is
+ set to the coding system that doesn't specify EOL conversion. */
+ saved_coding_symbol = coding->symbol;
+
+ coding->spec.ccl.cr_carryover = 0;
+ if (coding->eol_type == CODING_EOL_UNDECIDED)
+ {
+ /* Here, to avoid the call of setup_coding_system, we directly
+ call detect_eol_type. */
+ coding->eol_type = detect_eol_type (ptr, bytes, &dummy);
+ if (coding->eol_type == CODING_EOL_INCONSISTENT)
+ coding->eol_type = CODING_EOL_LF;
+ if (coding->eol_type != CODING_EOL_UNDECIDED)
+ {
+ val = Fget (coding->symbol, Qeol_type);
+ if (VECTORP (val) && XVECTOR (val)->size == 3)
+ coding->symbol = XVECTOR (val)->contents[coding->eol_type];
+ }
+ coding->mode |= CODING_MODE_INHIBIT_INCONSISTENT_EOL;
+ }
+
+ if (coding->eol_type == CODING_EOL_LF
+ || coding->eol_type == CODING_EOL_UNDECIDED)
+ {
+ /* We have nothing to do. */
+ ptr = pend;
+ }
+ else if (coding->eol_type == CODING_EOL_CRLF)
+ {
+ unsigned char *pstart = ptr, *p = ptr;
+
+ if (! (coding->mode & CODING_MODE_LAST_BLOCK)
+ && *(pend - 1) == '\r')
+ {
+ /* If the last character is CR, we can't handle it here
+ because LF will be in the not-yet-decoded source text.
+ Recorded that the CR is not yet processed. */
+ coding->spec.ccl.cr_carryover = 1;
+ coding->produced--;
+ coding->produced_char--;
+ pend--;
+ }
+ while (ptr < pend)
+ {
+ if (*ptr == '\r')
+ {
+ if (ptr + 1 < pend && *(ptr + 1) == '\n')
+ {
+ *p++ = '\n';
+ ptr += 2;
+ }
+ else
+ {
+ if (coding->mode & CODING_MODE_INHIBIT_INCONSISTENT_EOL)
+ goto undo_eol_conversion;
+ *p++ = *ptr++;
+ }
+ }
+ else if (*ptr == '\n'
+ && coding->mode & CODING_MODE_INHIBIT_INCONSISTENT_EOL)
+ goto undo_eol_conversion;
+ else
+ *p++ = *ptr++;
+ continue;
+
+ undo_eol_conversion:
+ /* We have faced with inconsistent EOL format at PTR.
+ Convert all LFs before PTR back to CRLFs. */
+ for (p--, ptr--; p >= pstart; p--)
+ {
+ if (*p == '\n')
+ *ptr-- = '\n', *ptr-- = '\r';
+ else
+ *ptr-- = *p;
+ }
+ /* If carryover is recorded, cancel it because we don't
+ convert CRLF anymore. */
+ if (coding->spec.ccl.cr_carryover)
+ {
+ coding->spec.ccl.cr_carryover = 0;
+ coding->produced++;
+ coding->produced_char++;
+ pend++;
+ }
+ p = ptr = pend;
+ coding->eol_type = CODING_EOL_LF;
+ coding->symbol = saved_coding_symbol;
+ }
+ if (p < pend)
+ {
+ /* As each two-byte sequence CRLF was converted to LF, (PEND
+ - P) is the number of deleted characters. */
+ coding->produced -= pend - p;
+ coding->produced_char -= pend - p;
+ }
+ }
+ else /* i.e. coding->eol_type == CODING_EOL_CR */
+ {
+ unsigned char *p = ptr;
+
+ for (; ptr < pend; ptr++)
+ {
+ if (*ptr == '\r')
+ *ptr = '\n';
+ else if (*ptr == '\n'
+ && coding->mode & CODING_MODE_INHIBIT_INCONSISTENT_EOL)
+ {
+ for (; p < ptr; p++)
+ {
+ if (*p == '\n')
+ *p = '\r';
+ }
+ ptr = pend;
+ coding->eol_type = CODING_EOL_LF;
+ coding->symbol = saved_coding_symbol;
+ }
+ }
+ }
+}
+
/* See "GENERAL NOTES about `decode_coding_XXX ()' functions". Before
decoding, it may detect coding system and format of end-of-line if
- those are not yet decided. */
+ those are not yet decided. The source should be unibyte, the
+ result is multibyte if CODING->dst_multibyte is nonzero, else
+ unibyte. */
int
decode_coding (coding, source, destination, src_bytes, dst_bytes)
unsigned char *source, *destination;
int src_bytes, dst_bytes;
{
- int result;
-
- if (src_bytes <= 0
- && ! (coding->mode & CODING_MODE_LAST_BLOCK
- && CODING_REQUIRE_FLUSHING (coding)))
- {
- coding->produced = coding->produced_char = 0;
- coding->consumed = coding->consumed_char = 0;
- coding->fake_multibyte = 0;
- return CODING_FINISH_NORMAL;
- }
-
if (coding->type == coding_type_undecided)
detect_coding (coding, source, src_bytes);
- if (coding->eol_type == CODING_EOL_UNDECIDED)
+ if (coding->eol_type == CODING_EOL_UNDECIDED
+ && coding->type != coding_type_ccl)
detect_eol (coding, source, src_bytes);
+ coding->produced = coding->produced_char = 0;
+ coding->consumed = coding->consumed_char = 0;
+ coding->errors = 0;
+ coding->result = CODING_FINISH_NORMAL;
+
switch (coding->type)
{
- case coding_type_emacs_mule:
- case coding_type_undecided:
- case coding_type_raw_text:
- if (coding->eol_type == CODING_EOL_LF
- || coding->eol_type == CODING_EOL_UNDECIDED)
- goto label_no_conversion;
- result = decode_eol (coding, source, destination, src_bytes, dst_bytes);
- break;
-
case coding_type_sjis:
- result = decode_coding_sjis_big5 (coding, source, destination,
- src_bytes, dst_bytes, 1);
+ decode_coding_sjis_big5 (coding, source, destination,
+ src_bytes, dst_bytes, 1);
break;
case coding_type_iso2022:
- result = decode_coding_iso2022 (coding, source, destination,
- src_bytes, dst_bytes);
+ decode_coding_iso2022 (coding, source, destination,
+ src_bytes, dst_bytes);
break;
case coding_type_big5:
- result = decode_coding_sjis_big5 (coding, source, destination,
- src_bytes, dst_bytes, 0);
+ decode_coding_sjis_big5 (coding, source, destination,
+ src_bytes, dst_bytes, 0);
break;
- case coding_type_ccl:
- result = ccl_coding_driver (coding, source, destination,
- src_bytes, dst_bytes, 0);
+ case coding_type_emacs_mule:
+ decode_coding_emacs_mule (coding, source, destination,
+ src_bytes, dst_bytes);
break;
- default: /* i.e. case coding_type_no_conversion: */
- label_no_conversion:
- if (dst_bytes && src_bytes > dst_bytes)
+ case coding_type_ccl:
+ if (coding->spec.ccl.cr_carryover)
{
- coding->produced = dst_bytes;
- result = CODING_FINISH_INSUFFICIENT_DST;
+ /* Set the CR which is not processed by the previous call of
+ decode_eol_post_ccl in DESTINATION. */
+ *destination = '\r';
+ coding->produced++;
+ coding->produced_char++;
+ dst_bytes--;
}
- else
+ ccl_coding_driver (coding, source,
+ destination + coding->spec.ccl.cr_carryover,
+ src_bytes, dst_bytes, 0);
+ if (coding->eol_type != CODING_EOL_LF)
+ decode_eol_post_ccl (coding, destination, coding->produced);
+ break;
+
+ default:
+ decode_eol (coding, source, destination, src_bytes, dst_bytes);
+ }
+
+ if (coding->result == CODING_FINISH_INSUFFICIENT_SRC
+ && coding->consumed == src_bytes)
+ coding->result = CODING_FINISH_NORMAL;
+
+ if (coding->mode & CODING_MODE_LAST_BLOCK
+ && coding->result == CODING_FINISH_INSUFFICIENT_SRC)
+ {
+ unsigned char *src = source + coding->consumed;
+ unsigned char *dst = destination + coding->produced;
+
+ src_bytes -= coding->consumed;
+ coding->errors++;
+ if (COMPOSING_P (coding))
+ DECODE_COMPOSITION_END ('1');
+ while (src_bytes--)
{
- coding->produced = src_bytes;
- result = CODING_FINISH_NORMAL;
+ int c = *src++;
+ dst += CHAR_STRING (c, dst);
+ coding->produced_char++;
}
- if (dst_bytes)
- bcopy (source, destination, coding->produced);
- else
- safe_bcopy (source, destination, coding->produced);
- coding->fake_multibyte = 1;
- coding->consumed
- = coding->consumed_char = coding->produced_char = coding->produced;
- break;
+ coding->consumed = coding->consumed_char = src - source;
+ coding->produced = dst - destination;
}
- return result;
+ if (!coding->dst_multibyte)
+ {
+ coding->produced = str_as_unibyte (destination, coding->produced);
+ coding->produced_char = coding->produced;
+ }
+
+ return coding->result;
}
-/* See "GENERAL NOTES about `encode_coding_XXX ()' functions". */
+/* See "GENERAL NOTES about `encode_coding_XXX ()' functions". The
+ multibyteness of the source is CODING->src_multibyte, the
+ multibyteness of the result is always unibyte. */
int
encode_coding (coding, source, destination, src_bytes, dst_bytes)
unsigned char *source, *destination;
int src_bytes, dst_bytes;
{
- int result;
-
- if (src_bytes <= 0
- && ! (coding->mode & CODING_MODE_LAST_BLOCK
- && CODING_REQUIRE_FLUSHING (coding)))
- {
- coding->produced = coding->produced_char = 0;
- coding->consumed = coding->consumed_char = 0;
- coding->fake_multibyte = 0;
- return CODING_FINISH_NORMAL;
- }
+ coding->produced = coding->produced_char = 0;
+ coding->consumed = coding->consumed_char = 0;
+ coding->errors = 0;
+ coding->result = CODING_FINISH_NORMAL;
switch (coding->type)
{
- case coding_type_emacs_mule:
- case coding_type_undecided:
- case coding_type_raw_text:
- if (coding->eol_type == CODING_EOL_LF
- || coding->eol_type == CODING_EOL_UNDECIDED)
- goto label_no_conversion;
- result = encode_eol (coding, source, destination, src_bytes, dst_bytes);
- break;
-
case coding_type_sjis:
- result = encode_coding_sjis_big5 (coding, source, destination,
- src_bytes, dst_bytes, 1);
+ encode_coding_sjis_big5 (coding, source, destination,
+ src_bytes, dst_bytes, 1);
break;
case coding_type_iso2022:
- result = encode_coding_iso2022 (coding, source, destination,
- src_bytes, dst_bytes);
+ encode_coding_iso2022 (coding, source, destination,
+ src_bytes, dst_bytes);
break;
case coding_type_big5:
- result = encode_coding_sjis_big5 (coding, source, destination,
- src_bytes, dst_bytes, 0);
+ encode_coding_sjis_big5 (coding, source, destination,
+ src_bytes, dst_bytes, 0);
+ break;
+
+ case coding_type_emacs_mule:
+ encode_coding_emacs_mule (coding, source, destination,
+ src_bytes, dst_bytes);
break;
case coding_type_ccl:
- result = ccl_coding_driver (coding, source, destination,
- src_bytes, dst_bytes, 1);
+ ccl_coding_driver (coding, source, destination,
+ src_bytes, dst_bytes, 1);
break;
- default: /* i.e. case coding_type_no_conversion: */
- label_no_conversion:
- if (dst_bytes && src_bytes > dst_bytes)
- {
- coding->produced = dst_bytes;
- result = CODING_FINISH_INSUFFICIENT_DST;
- }
- else
- {
- coding->produced = src_bytes;
- result = CODING_FINISH_NORMAL;
- }
- if (dst_bytes)
- bcopy (source, destination, coding->produced);
- else
- safe_bcopy (source, destination, coding->produced);
- if (coding->mode & CODING_MODE_SELECTIVE_DISPLAY)
+ default:
+ encode_eol (coding, source, destination, src_bytes, dst_bytes);
+ }
+
+ if (coding->result == CODING_FINISH_INSUFFICIENT_SRC
+ && coding->consumed == src_bytes)
+ coding->result = CODING_FINISH_NORMAL;
+
+ if (coding->mode & CODING_MODE_LAST_BLOCK)
+ {
+ unsigned char *src = source + coding->consumed;
+ unsigned char *src_end = src + src_bytes;
+ unsigned char *dst = destination + coding->produced;
+
+ if (coding->type == coding_type_iso2022)
+ ENCODE_RESET_PLANE_AND_REGISTER;
+ if (COMPOSING_P (coding))
+ *dst++ = ISO_CODE_ESC, *dst++ = '1';
+ if (coding->consumed < src_bytes)
{
- unsigned char *p = destination, *pend = p + coding->produced;
- while (p < pend)
- if (*p++ == '\015') p[-1] = '\n';
+ int len = src_bytes - coding->consumed;
+
+ BCOPY_SHORT (source + coding->consumed, dst, len);
+ if (coding->src_multibyte)
+ len = str_as_unibyte (dst, len);
+ dst += len;
+ coding->consumed = src_bytes;
}
- coding->fake_multibyte = 1;
- coding->consumed
- = coding->consumed_char = coding->produced_char = coding->produced;
- break;
+ coding->produced = coding->produced_char = dst - destination;
}
- return result;
+ return coding->result;
}
/* Scan text in the region between *BEG and *END (byte positions),
skip characters which we don't have to decode by coding system
CODING at the head and tail, then set *BEG and *END to the region
of the text we actually have to convert. The caller should move
- the gap out of the region in advance.
+ the gap out of the region in advance if the region is from a
+ buffer.
If STR is not NULL, *BEG and *END are indices into STR. */
if (coding->type == coding_type_ccl
|| coding->type == coding_type_undecided
- || !NILP (coding->post_read_conversion))
+ || coding->eol_type != CODING_EOL_LF
+ || !NILP (coding->post_read_conversion)
+ || coding->composing != COMPOSITION_DISABLED)
{
/* We can't skip any data. */
return;
}
- else if (coding->type == coding_type_no_conversion)
+ if (coding->type == coding_type_no_conversion
+ || coding->type == coding_type_raw_text
+ || coding->type == coding_type_emacs_mule)
{
/* We need no conversion, but don't have to skip any data here.
Decoding routine handles them effectively anyway. */
if (!NILP (CHAR_TABLE_REF (translation_table, i)))
break;
if (i < 128)
- /* Some ASCII character should be tranlsated. We give up
+ /* Some ASCII character should be translated. We give up
shrinking. */
return;
}
- eol_conversion = (coding->eol_type != CODING_EOL_LF);
-
- if ((! eol_conversion) && (coding->heading_ascii >= 0))
+ if (coding->heading_ascii >= 0)
/* Detection routine has already found how much we can skip at the
head. */
*beg += coding->heading_ascii;
endp_orig = endp = begp + *end - *beg;
}
+ eol_conversion = (coding->eol_type == CODING_EOL_CR
+ || coding->eol_type == CODING_EOL_CRLF);
+
switch (coding->type)
{
- case coding_type_emacs_mule:
- case coding_type_raw_text:
- if (eol_conversion)
- {
- if (coding->heading_ascii < 0)
- while (begp < endp && *begp != '\r' && *begp < 0x80) begp++;
- while (begp < endp && endp[-1] != '\r' && endp[-1] < 0x80)
- endp--;
- /* Do not consider LF as ascii if preceded by CR, since that
- confuses eol decoding. */
- if (begp < endp && endp < endp_orig && endp[-1] == '\r' && endp[0] == '\n')
- endp++;
- }
- else
- begp = endp;
- break;
-
case coding_type_sjis:
case coding_type_big5:
/* We can skip all ASCII characters at the head. */
endp++;
break;
- default: /* i.e. case coding_type_iso2022: */
+ case coding_type_iso2022:
if (CODING_SPEC_ISO_INITIAL_DESIGNATION (coding, 0) != CHARSET_ASCII)
/* We can't skip any data. */
break;
endp = eight_bit;
}
}
+ break;
+
+ default:
+ abort ();
}
*beg += begp - begp_orig;
*end += endp - endp_orig;
int eol_conversion;
Lisp_Object translation_table;
- if (coding->type == coding_type_ccl)
- /* We can't skip any data. */
- return;
- else if (coding->type == coding_type_no_conversion)
+ if (coding->type == coding_type_ccl
+ || coding->eol_type == CODING_EOL_CRLF
+ || coding->eol_type == CODING_EOL_CR
+ || coding->cmp_data && coding->cmp_data->used > 0)
{
- /* We need no conversion. */
- *beg = *end;
+ /* We can't skip any data. */
+ return;
+ }
+ if (coding->type == coding_type_no_conversion
+ || coding->type == coding_type_raw_text
+ || coding->type == coding_type_emacs_mule
+ || coding->type == coding_type_undecided)
+ {
+ /* We need no conversion, but don't have to skip any data here.
+ Encoding routine handles them effectively anyway. */
return;
}
the caller is expected to have handled it already. */
switch (coding->type)
{
- case coding_type_undecided:
- case coding_type_emacs_mule:
- case coding_type_raw_text:
- if (eol_conversion)
- {
- while (begp < endp && *begp != '\n') begp++;
- while (begp < endp && endp[-1] != '\n') endp--;
- }
- else
- begp = endp;
- break;
-
case coding_type_iso2022:
if (CODING_SPEC_ISO_INITIAL_DESIGNATION (coding, 0) != CHARSET_ASCII)
/* We can't skip any data. */
}
/* fall down ... */
- default:
+ case coding_type_sjis:
+ case coding_type_big5:
/* We can skip all ASCII characters at the head and tail. */
if (eol_conversion)
while (begp < endp && *begp < 0x80 && *begp != '\n') begp++;
else
while (begp < endp && *(endp - 1) < 0x80) endp--;
break;
+
+ default:
+ abort ();
}
*beg += begp - begp_orig;
} \
} while (0)
+static Lisp_Object
+code_convert_region_unwind (dummy)
+ Lisp_Object dummy;
+{
+ inhibit_pre_post_conversion = 0;
+ return Qnil;
+}
+
+/* Store information about all compositions in the range FROM and TO
+ of OBJ in memory blocks pointed by CODING->cmp_data. OBJ is a
+ buffer or a string, defaults to the current buffer. */
+
+void
+coding_save_composition (coding, from, to, obj)
+ struct coding_system *coding;
+ int from, to;
+ Lisp_Object obj;
+{
+ Lisp_Object prop;
+ int start, end;
+
+ if (coding->composing == COMPOSITION_DISABLED)
+ return;
+ if (!coding->cmp_data)
+ coding_allocate_composition_data (coding, from);
+ if (!find_composition (from, to, &start, &end, &prop, obj)
+ || end > to)
+ return;
+ if (start < from
+ && (!find_composition (end, to, &start, &end, &prop, obj)
+ || end > to))
+ return;
+ coding->composing = COMPOSITION_NO;
+ do
+ {
+ if (COMPOSITION_VALID_P (start, end, prop))
+ {
+ enum composition_method method = COMPOSITION_METHOD (prop);
+ if (coding->cmp_data->used + COMPOSITION_DATA_MAX_BUNCH_LENGTH
+ >= COMPOSITION_DATA_SIZE)
+ coding_allocate_composition_data (coding, from);
+ /* For relative composition, we remember start and end
+ positions, for the other compositions, we also remember
+ components. */
+ CODING_ADD_COMPOSITION_START (coding, start - from, method);
+ if (method != COMPOSITION_RELATIVE)
+ {
+ /* We must store a*/
+ Lisp_Object val, ch;
+
+ val = COMPOSITION_COMPONENTS (prop);
+ if (CONSP (val))
+ while (CONSP (val))
+ {
+ ch = XCAR (val), val = XCDR (val);
+ CODING_ADD_COMPOSITION_COMPONENT (coding, XINT (ch));
+ }
+ else if (VECTORP (val) || STRINGP (val))
+ {
+ int len = (VECTORP (val)
+ ? XVECTOR (val)->size : XSTRING (val)->size);
+ int i;
+ for (i = 0; i < len; i++)
+ {
+ ch = (STRINGP (val)
+ ? Faref (val, make_number (i))
+ : XVECTOR (val)->contents[i]);
+ CODING_ADD_COMPOSITION_COMPONENT (coding, XINT (ch));
+ }
+ }
+ else /* INTEGERP (val) */
+ CODING_ADD_COMPOSITION_COMPONENT (coding, XINT (val));
+ }
+ CODING_ADD_COMPOSITION_END (coding, end - from);
+ }
+ start = end;
+ }
+ while (start < to
+ && find_composition (start, to, &start, &end, &prop, obj)
+ && end <= to);
+
+ /* Make coding->cmp_data point to the first memory block. */
+ while (coding->cmp_data->prev)
+ coding->cmp_data = coding->cmp_data->prev;
+ coding->cmp_data_start = 0;
+}
+
+/* Reflect the saved information about compositions to OBJ.
+ CODING->cmp_data points to a memory block for the informaiton. OBJ
+ is a buffer or a string, defaults to the current buffer. */
+
+void
+coding_restore_composition (coding, obj)
+ struct coding_system *coding;
+ Lisp_Object obj;
+{
+ struct composition_data *cmp_data = coding->cmp_data;
+
+ if (!cmp_data)
+ return;
+
+ while (cmp_data->prev)
+ cmp_data = cmp_data->prev;
+
+ while (cmp_data)
+ {
+ int i;
+
+ for (i = 0; i < cmp_data->used; i += cmp_data->data[i])
+ {
+ int *data = cmp_data->data + i;
+ enum composition_method method = (enum composition_method) data[3];
+ Lisp_Object components;
+
+ if (method == COMPOSITION_RELATIVE)
+ components = Qnil;
+ else
+ {
+ int len = data[0] - 4, j;
+ Lisp_Object args[MAX_COMPOSITION_COMPONENTS * 2 - 1];
+
+ for (j = 0; j < len; j++)
+ args[j] = make_number (data[4 + j]);
+ components = (method == COMPOSITION_WITH_ALTCHARS
+ ? Fstring (len, args) : Fvector (len, args));
+ }
+ compose_text (data[1], data[2], components, Qnil, obj);
+ }
+ cmp_data = cmp_data->next;
+ }
+}
+
/* Decode (if ENCODEP is zero) or encode (if ENCODEP is nonzero) the
text from FROM to TO (byte positions are FROM_BYTE and TO_BYTE) by
coding system CODING, and return the status code of code conversion
If REPLACE is nonzero, we do various things as if the original text
is deleted and a new text is inserted. See the comments in
- replace_range (insdel.c) to know what we are doing. */
+ replace_range (insdel.c) to know what we are doing.
+
+ If REPLACE is zero, it is assumed that the source text is unibyte.
+ Otherwize, it is assumed that the source text is multibyte. */
int
code_convert_region (from, from_byte, to, to_byte, coding, encodep, replace)
{
int len = to - from, len_byte = to_byte - from_byte;
int require, inserted, inserted_byte;
- int head_skip, tail_skip, total_skip;
+ int head_skip, tail_skip, total_skip = 0;
Lisp_Object saved_coding_symbol;
- int multibyte = !NILP (current_buffer->enable_multibyte_characters);
int first = 1;
- int fake_multibyte = 0;
unsigned char *src, *dst;
Lisp_Object deletion;
int orig_point = PT, orig_len = len;
int prev_Z;
+ int multibyte_p = !NILP (current_buffer->enable_multibyte_characters);
+
+ coding->src_multibyte = replace && multibyte_p;
+ coding->dst_multibyte = multibyte_p;
deletion = Qnil;
saved_coding_symbol = Qnil;
if (replace)
{
int saved_from = from;
+ int saved_inhibit_modification_hooks;
prepare_to_modify_buffer (from, to, &from);
if (saved_from != from)
{
to = from + len;
- if (multibyte)
- from_byte = CHAR_TO_BYTE (from), to_byte = CHAR_TO_BYTE (to);
- else
- from_byte = from, to_byte = to;
+ from_byte = CHAR_TO_BYTE (from), to_byte = CHAR_TO_BYTE (to);
len_byte = to_byte - from_byte;
}
+
+ /* The code conversion routine can not preserve text properties
+ for now. So, we must remove all text properties in the
+ region. Here, we must suppress all modification hooks. */
+ saved_inhibit_modification_hooks = inhibit_modification_hooks;
+ inhibit_modification_hooks = 1;
+ Fset_text_properties (make_number (from), make_number (to), Qnil, Qnil);
+ inhibit_modification_hooks = saved_inhibit_modification_hooks;
}
if (! encodep && CODING_REQUIRE_DETECTION (coding))
encodings again in vain. */
coding->type = coding_type_emacs_mule;
}
- if (coding->eol_type == CODING_EOL_UNDECIDED)
+ if (coding->eol_type == CODING_EOL_UNDECIDED
+ && coding->type != coding_type_ccl)
{
saved_coding_symbol = coding->symbol;
detect_eol (coding, BYTE_POS_ADDR (from_byte), len_byte);
}
}
- coding->consumed_char = len, coding->consumed = len_byte;
-
- if (encodep
- ? ! CODING_REQUIRE_ENCODING (coding)
- : ! CODING_REQUIRE_DECODING (coding))
- {
- coding->produced = len_byte;
- if (multibyte
- && ! replace
- /* See the comment of the member heading_ascii in coding.h. */
- && coding->heading_ascii < len_byte)
- {
- /* We still may have to combine byte at the head and the
- tail of the text in the region. */
- if (from < GPT && GPT < to)
- move_gap_both (to, to_byte);
- len = multibyte_chars_in_text (BYTE_POS_ADDR (from_byte), len_byte);
- adjust_after_insert (from, from_byte, to, to_byte, len);
- coding->produced_char = len;
- }
- else
- {
- if (!replace)
- adjust_after_insert (from, from_byte, to, to_byte, len_byte);
- coding->produced_char = len_byte;
- }
- return 0;
- }
-
/* Now we convert the text. */
/* For encoding, we must process pre-write-conversion in advance. */
- if (encodep
- && ! NILP (coding->pre_write_conversion)
+ if (! inhibit_pre_post_conversion
+ && encodep
&& SYMBOLP (coding->pre_write_conversion)
&& ! NILP (Ffboundp (coding->pre_write_conversion)))
{
/* The function in pre-write-conversion may put a new text in a
new buffer. */
- struct buffer *prev = current_buffer, *new;
+ struct buffer *prev = current_buffer;
+ Lisp_Object new;
+ int count = specpdl_ptr - specpdl;
+ record_unwind_protect (code_convert_region_unwind, Qnil);
+ /* We should not call any more pre-write/post-read-conversion
+ functions while this pre-write-conversion is running. */
+ inhibit_pre_post_conversion = 1;
call2 (coding->pre_write_conversion,
make_number (from), make_number (to));
+ inhibit_pre_post_conversion = 0;
+ /* Discard the unwind protect. */
+ specpdl_ptr--;
+
if (current_buffer != prev)
{
len = ZV - BEGV;
- new = current_buffer;
+ new = Fcurrent_buffer ();
set_buffer_internal_1 (prev);
- del_range_2 (from, from_byte, to, to_byte);
+ del_range_2 (from, from_byte, to, to_byte, 0);
TEMP_SET_PT_BOTH (from, from_byte);
- insert_from_buffer (new, 1, len, 0);
+ insert_from_buffer (XBUFFER (new), 1, len, 0);
+ Fkill_buffer (new);
if (orig_point >= to)
orig_point += len - orig_len;
else if (orig_point > from)
orig_point = from;
orig_len = len;
to = from + len;
- from_byte = multibyte ? CHAR_TO_BYTE (from) : from_byte;
- to_byte = multibyte ? CHAR_TO_BYTE (to) : to;
+ from_byte = CHAR_TO_BYTE (from);
+ to_byte = CHAR_TO_BYTE (to);
len_byte = to_byte - from_byte;
TEMP_SET_PT_BOTH (from, from_byte);
}
if (replace)
deletion = make_buffer_string_both (from, from_byte, to, to_byte, 1);
+ if (coding->composing != COMPOSITION_DISABLED)
+ {
+ if (encodep)
+ coding_save_composition (coding, from, to, Fcurrent_buffer ());
+ else
+ coding_allocate_composition_data (coding, from);
+ }
+
/* Try to skip the heading and tailing ASCIIs. */
- {
- int from_byte_orig = from_byte, to_byte_orig = to_byte;
-
- if (from < GPT && GPT < to)
- move_gap_both (from, from_byte);
- SHRINK_CONVERSION_REGION (&from_byte, &to_byte, coding, NULL, encodep);
- if (from_byte == to_byte
- && ! (coding->mode & CODING_MODE_LAST_BLOCK
- && CODING_REQUIRE_FLUSHING (coding)))
- {
- coding->produced = len_byte;
- coding->produced_char = multibyte ? len : len_byte;
- if (!replace)
- /* We must record and adjust for this new text now. */
- adjust_after_insert (from, from_byte_orig, to, to_byte_orig, len);
- return 0;
- }
+ if (coding->type != coding_type_ccl)
+ {
+ int from_byte_orig = from_byte, to_byte_orig = to_byte;
- head_skip = from_byte - from_byte_orig;
- tail_skip = to_byte_orig - to_byte;
- total_skip = head_skip + tail_skip;
- from += head_skip;
- to -= tail_skip;
- len -= total_skip; len_byte -= total_skip;
- }
+ if (from < GPT && GPT < to)
+ move_gap_both (from, from_byte);
+ SHRINK_CONVERSION_REGION (&from_byte, &to_byte, coding, NULL, encodep);
+ if (from_byte == to_byte
+ && (encodep || NILP (coding->post_read_conversion))
+ && ! CODING_REQUIRE_FLUSHING (coding))
+ {
+ coding->produced = len_byte;
+ coding->produced_char = len;
+ if (!replace)
+ /* We must record and adjust for this new text now. */
+ adjust_after_insert (from, from_byte_orig, to, to_byte_orig, len);
+ return 0;
+ }
- /* The code conversion routine can not preserve text properties for
- now. So, we must remove all text properties in the region.
- Here, we must suppress all modification hooks. */
- if (replace)
- {
- int saved_inhibit_modification_hooks = inhibit_modification_hooks;
- inhibit_modification_hooks = 1;
- Fset_text_properties (make_number (from), make_number (to), Qnil, Qnil);
- inhibit_modification_hooks = saved_inhibit_modification_hooks;
+ head_skip = from_byte - from_byte_orig;
+ tail_skip = to_byte_orig - to_byte;
+ total_skip = head_skip + tail_skip;
+ from += head_skip;
+ to -= tail_skip;
+ len -= total_skip; len_byte -= total_skip;
}
/* For converion, we must put the gap before the text in addition to
move_gap_both (from, from_byte);
inserted = inserted_byte = 0;
- src = GAP_END_ADDR, dst = GPT_ADDR;
GAP_SIZE += len_byte;
ZV -= len;
ZV_BYTE -= len_byte;
Z_BYTE -= len_byte;
- if (GPT - BEG < beg_unchanged)
- beg_unchanged = GPT - BEG;
- if (Z - GPT < end_unchanged)
- end_unchanged = Z - GPT;
+ if (GPT - BEG < BEG_UNCHANGED)
+ BEG_UNCHANGED = GPT - BEG;
+ if (Z - GPT < END_UNCHANGED)
+ END_UNCHANGED = Z - GPT;
+
+ if (!encodep && coding->src_multibyte)
+ {
+ /* Decoding routines expects that the source text is unibyte.
+ We must convert 8-bit characters of multibyte form to
+ unibyte. */
+ int len_byte_orig = len_byte;
+ len_byte = str_as_unibyte (GAP_END_ADDR - len_byte, len_byte);
+ if (len_byte < len_byte_orig)
+ safe_bcopy (GAP_END_ADDR - len_byte_orig, GAP_END_ADDR - len_byte,
+ len_byte);
+ coding->src_multibyte = 0;
+ }
for (;;)
{
int result;
- /* The buffer memory is changed from:
- +--------+converted-text+---------+-------original-text------+---+
- |<-from->|<--inserted-->|---------|<-----------len---------->|---|
- |<------------------- GAP_SIZE -------------------->| */
+ /* The buffer memory is now:
+ +--------+converted-text+---------+-------original-text-------+---+
+ |<-from->|<--inserted-->|---------|<--------len_byte--------->|---|
+ |<---------------------- GAP ----------------------->| */
+ src = GAP_END_ADDR - len_byte;
+ dst = GPT_ADDR + inserted_byte;
+
if (encodep)
result = encode_coding (coding, src, dst, len_byte, 0);
else
result = decode_coding (coding, src, dst, len_byte, 0);
- /* to:
- +--------+-------converted-text--------+--+---original-text--+---+
- |<-from->|<--inserted-->|<--produced-->|--|<-(len-consumed)->|---|
- |<------------------- GAP_SIZE -------------------->| */
- if (coding->fake_multibyte)
- fake_multibyte = 1;
-
- if (!encodep && !multibyte)
- coding->produced_char = coding->produced;
+
+ /* The buffer memory is now:
+ +--------+-------converted-text----+--+------original-text----+---+
+ |<-from->|<-inserted->|<-produced->|--|<-(len_byte-consumed)->|---|
+ |<---------------------- GAP ----------------------->| */
+
inserted += coding->produced_char;
inserted_byte += coding->produced;
len_byte -= coding->consumed;
+
+ if (result == CODING_FINISH_INSUFFICIENT_CMP)
+ {
+ coding_allocate_composition_data (coding, from + inserted);
+ continue;
+ }
+
src += coding->consumed;
- dst += inserted_byte;
+ dst += coding->produced;
if (result == CODING_FINISH_NORMAL)
{
if (! encodep && result == CODING_FINISH_INCONSISTENT_EOL)
{
unsigned char *pend = dst, *p = pend - inserted_byte;
+ Lisp_Object eol_type;
/* Encode LFs back to the original eol format (CR or CRLF). */
if (coding->eol_type == CODING_EOL_CR)
while (p < pend) if (*p++ == '\n') count++;
if (src - dst < count)
{
- /* We don't have sufficient room for putting LFs
+ /* We don't have sufficient room for encoding LFs
back to CRLF. We must record converted and
not-yet-converted text back to the buffer
content, enlarge the gap, then record them out of
/* Suppress eol-format conversion in the further conversion. */
coding->eol_type = CODING_EOL_LF;
- /* Restore the original symbol. */
- coding->symbol = saved_coding_symbol;
+ /* Set the coding system symbol to that for Unix-like EOL. */
+ eol_type = Fget (saved_coding_symbol, Qeol_type);
+ if (VECTORP (eol_type)
+ && XVECTOR (eol_type)->size == 3
+ && SYMBOLP (XVECTOR (eol_type)->contents[CODING_EOL_LF]))
+ coding->symbol = XVECTOR (eol_type)->contents[CODING_EOL_LF];
+ else
+ coding->symbol = saved_coding_symbol;
continue;
}
if (len_byte <= 0)
- break;
+ {
+ if (coding->type != coding_type_ccl
+ || coding->mode & CODING_MODE_LAST_BLOCK)
+ break;
+ coding->mode |= CODING_MODE_LAST_BLOCK;
+ continue;
+ }
if (result == CODING_FINISH_INSUFFICIENT_SRC)
{
/* The source text ends in invalid codes. Let's just
inserted_byte += len_byte;
while (len_byte--)
*dst++ = *src++;
- fake_multibyte = 1;
break;
}
if (result == CODING_FINISH_INTERRUPT)
{
/* The conversion procedure was interrupted by a user. */
- fake_multibyte = 1;
break;
}
/* Now RESULT == CODING_FINISH_INSUFFICIENT_DST */
{
/* It's quite strange to require more memory without
consuming any bytes. Perhaps CCL program bug. */
- fake_multibyte = 1;
break;
}
if (first)
GAP_SIZE += add;
ZV -= add; Z -= add; ZV_BYTE -= add; Z_BYTE -= add;
GPT -= inserted_byte; GPT_BYTE -= inserted_byte;
- /* Don't forget to update SRC, DST. */
- src = GAP_END_ADDR - len_byte;
- dst = GPT_ADDR + inserted_byte;
}
}
if (src - dst > 0) *dst = 0; /* Put an anchor. */
- if (multibyte
- && (encodep
- || fake_multibyte
- || (to - from) != (to_byte - from_byte)))
- inserted = multibyte_chars_in_text (GPT_ADDR, inserted_byte);
+ if (encodep && coding->dst_multibyte)
+ {
+ /* The output is unibyte. We must convert 8-bit characters to
+ multibyte form. */
+ if (inserted_byte * 2 > GAP_SIZE)
+ {
+ GAP_SIZE -= inserted_byte;
+ ZV += inserted_byte; Z += inserted_byte;
+ ZV_BYTE += inserted_byte; Z_BYTE += inserted_byte;
+ GPT += inserted_byte; GPT_BYTE += inserted_byte;
+ make_gap (inserted_byte - GAP_SIZE);
+ GAP_SIZE += inserted_byte;
+ ZV -= inserted_byte; Z -= inserted_byte;
+ ZV_BYTE -= inserted_byte; Z_BYTE -= inserted_byte;
+ GPT -= inserted_byte; GPT_BYTE -= inserted_byte;
+ }
+ inserted_byte = str_to_multibyte (GPT_ADDR, GAP_SIZE, inserted_byte);
+ }
/* If we have shrinked the conversion area, adjust it now. */
if (total_skip > 0)
adjust_after_replace (from, from_byte, deletion, inserted, inserted_byte);
inserted = Z - prev_Z;
- if (! encodep && ! NILP (coding->post_read_conversion))
+ if (!encodep && coding->cmp_data && coding->cmp_data->used)
+ coding_restore_composition (coding, Fcurrent_buffer ());
+ coding_free_composition_data (coding);
+
+ if (! inhibit_pre_post_conversion
+ && ! encodep && ! NILP (coding->post_read_conversion))
{
Lisp_Object val;
+ int count = specpdl_ptr - specpdl;
if (from != PT)
TEMP_SET_PT_BOTH (from, from_byte);
prev_Z = Z;
+ record_unwind_protect (code_convert_region_unwind, Qnil);
+ /* We should not call any more pre-write/post-read-conversion
+ functions while this post-read-conversion is running. */
+ inhibit_pre_post_conversion = 1;
val = call1 (coding->post_read_conversion, make_number (inserted));
+ inhibit_pre_post_conversion = 0;
+ /* Discard the unwind protect. */
+ specpdl_ptr--;
CHECK_NUMBER (val, 0);
- inserted = Z - prev_Z;
+ inserted += Z - prev_Z;
}
if (orig_point >= from)
TEMP_SET_PT (orig_point);
}
- signal_after_change (from, to - from, inserted);
+ if (replace)
+ {
+ signal_after_change (from, to - from, inserted);
+ update_compositions (from, from + inserted, CHECK_BORDER);
+ }
{
coding->consumed = to_byte - from_byte;
}
Lisp_Object
-code_convert_string (str, coding, encodep, nocopy)
+run_pre_post_conversion_on_str (str, coding, encodep)
+ Lisp_Object str;
+ struct coding_system *coding;
+ int encodep;
+{
+ int count = specpdl_ptr - specpdl;
+ struct gcpro gcpro1;
+ struct buffer *prev = current_buffer;
+ int multibyte = STRING_MULTIBYTE (str);
+
+ record_unwind_protect (Fset_buffer, Fcurrent_buffer ());
+ record_unwind_protect (code_convert_region_unwind, Qnil);
+ GCPRO1 (str);
+ temp_output_buffer_setup (" *code-converting-work*");
+ set_buffer_internal (XBUFFER (Vstandard_output));
+ /* We must insert the contents of STR as is without
+ unibyte<->multibyte conversion. For that, we adjust the
+ multibyteness of the working buffer to that of STR. */
+ Ferase_buffer ();
+ current_buffer->enable_multibyte_characters = multibyte ? Qt : Qnil;
+ insert_from_string (str, 0, 0,
+ XSTRING (str)->size, STRING_BYTES (XSTRING (str)), 0);
+ UNGCPRO;
+ inhibit_pre_post_conversion = 1;
+ if (encodep)
+ call2 (coding->pre_write_conversion, make_number (BEG), make_number (Z));
+ else
+ {
+ TEMP_SET_PT_BOTH (BEG, BEG_BYTE);
+ call1 (coding->post_read_conversion, make_number (Z - BEG));
+ }
+ inhibit_pre_post_conversion = 0;
+ str = make_buffer_string (BEG, Z, 0);
+ return unbind_to (count, str);
+}
+
+Lisp_Object
+decode_coding_string (str, coding, nocopy)
Lisp_Object str;
struct coding_system *coding;
- int encodep, nocopy;
+ int nocopy;
{
int len;
char *buf;
- int from = 0, to = XSTRING (str)->size;
- int to_byte = STRING_BYTES (XSTRING (str));
+ int from, to, to_byte;
struct gcpro gcpro1;
Lisp_Object saved_coding_symbol;
int result;
- saved_coding_symbol = Qnil;
- if (encodep && !NILP (coding->pre_write_conversion)
- || !encodep && !NILP (coding->post_read_conversion))
- {
- /* Since we have to call Lisp functions which assume target text
- is in a buffer, after setting a temporary buffer, call
- code_convert_region. */
- int count = specpdl_ptr - specpdl;
- struct buffer *prev = current_buffer;
-
- record_unwind_protect (Fset_buffer, Fcurrent_buffer ());
- temp_output_buffer_setup (" *code-converting-work*");
- set_buffer_internal (XBUFFER (Vstandard_output));
- if (encodep)
- insert_from_string (str, 0, 0, to, to_byte, 0);
- else
- {
- /* We must insert the contents of STR as is without
- unibyte<->multibyte conversion. */
- current_buffer->enable_multibyte_characters = Qnil;
- insert_from_string (str, 0, 0, to_byte, to_byte, 0);
- current_buffer->enable_multibyte_characters = Qt;
- }
- code_convert_region (BEGV, BEGV_BYTE, ZV, ZV_BYTE, coding, encodep, 1);
- if (encodep)
- /* We must return the buffer contents as unibyte string. */
- current_buffer->enable_multibyte_characters = Qnil;
- str = make_buffer_string (BEGV, ZV, 0);
- set_buffer_internal (prev);
- return unbind_to (count, str);
- }
+ from = 0;
+ to = XSTRING (str)->size;
+ to_byte = STRING_BYTES (XSTRING (str));
- if (! encodep && CODING_REQUIRE_DETECTION (coding))
+ saved_coding_symbol = Qnil;
+ if (CODING_REQUIRE_DETECTION (coding))
{
/* See the comments in code_convert_region. */
if (coding->type == coding_type_undecided)
if (coding->type == coding_type_undecided)
coding->type = coding_type_emacs_mule;
}
- if (coding->eol_type == CODING_EOL_UNDECIDED)
+ if (coding->eol_type == CODING_EOL_UNDECIDED
+ && coding->type != coding_type_ccl)
{
saved_coding_symbol = coding->symbol;
detect_eol (coding, XSTRING (str)->data, to_byte);
}
}
- if (encodep
- ? ! CODING_REQUIRE_ENCODING (coding)
- : ! CODING_REQUIRE_DECODING (coding))
- from = to_byte;
- else
+ if (! CODING_REQUIRE_DECODING (coding))
+ {
+ if (!STRING_MULTIBYTE (str))
+ {
+ str = Fstring_as_multibyte (str);
+ nocopy = 1;
+ }
+ return (nocopy ? str : Fcopy_sequence (str));
+ }
+
+ if (STRING_MULTIBYTE (str))
{
- /* Try to skip the heading and tailing ASCIIs. */
+ /* Decoding routines expect the source text to be unibyte. */
+ str = Fstring_as_unibyte (str);
+ to_byte = STRING_BYTES (XSTRING (str));
+ nocopy = 1;
+ coding->src_multibyte = 0;
+ }
+ coding->dst_multibyte = 1;
+
+ if (coding->composing != COMPOSITION_DISABLED)
+ coding_allocate_composition_data (coding, from);
+
+ /* Try to skip the heading and tailing ASCIIs. */
+ if (coding->type != coding_type_ccl)
+ {
+ int from_orig = from;
+
SHRINK_CONVERSION_REGION (&from, &to_byte, coding, XSTRING (str)->data,
- encodep);
+ 0);
+ if (from == to_byte)
+ return (nocopy ? str : Fcopy_sequence (str));
}
- if (from == to_byte
- && coding->type != coding_type_ccl)
- return (nocopy ? str : Fcopy_sequence (str));
- if (encodep)
- len = encoding_buffer_size (coding, to_byte - from);
- else
- len = decoding_buffer_size (coding, to_byte - from);
+ len = decoding_buffer_size (coding, to_byte - from);
len += from + STRING_BYTES (XSTRING (str)) - to_byte;
GCPRO1 (str);
buf = get_conversion_buffer (len);
if (from > 0)
bcopy (XSTRING (str)->data, buf, from);
- result = (encodep
- ? encode_coding (coding, XSTRING (str)->data + from,
- buf + from, to_byte - from, len)
- : decode_coding (coding, XSTRING (str)->data + from,
- buf + from, to_byte - from, len));
- if (! encodep && result == CODING_FINISH_INCONSISTENT_EOL)
+ result = decode_coding (coding, XSTRING (str)->data + from,
+ buf + from, to_byte - from, len);
+ if (result == CODING_FINISH_INCONSISTENT_EOL)
{
- /* We simple try to decode the whole string again but without
+ /* We simply try to decode the whole string again but without
eol-conversion this time. */
coding->eol_type = CODING_EOL_LF;
coding->symbol = saved_coding_symbol;
- return code_convert_string (str, coding, encodep, nocopy);
+ coding_free_composition_data (coding);
+ return decode_coding_string (str, coding, nocopy);
}
bcopy (XSTRING (str)->data + to_byte, buf + from + coding->produced,
STRING_BYTES (XSTRING (str)) - to_byte);
len = from + STRING_BYTES (XSTRING (str)) - to_byte;
- if (encodep)
- str = make_unibyte_string (buf, len + coding->produced);
- else
+ str = make_multibyte_string (buf, len + coding->produced_char,
+ len + coding->produced);
+
+ if (coding->cmp_data && coding->cmp_data->used)
+ coding_restore_composition (coding, str);
+ coding_free_composition_data (coding);
+
+ if (SYMBOLP (coding->post_read_conversion)
+ && !NILP (Ffboundp (coding->post_read_conversion)))
+ str = run_pre_post_conversion_on_str (str, coding, 0);
+
+ return str;
+}
+
+Lisp_Object
+encode_coding_string (str, coding, nocopy)
+ Lisp_Object str;
+ struct coding_system *coding;
+ int nocopy;
+{
+ int len;
+ char *buf;
+ int from, to, to_byte;
+ struct gcpro gcpro1;
+ Lisp_Object saved_coding_symbol;
+ int result;
+
+ if (SYMBOLP (coding->pre_write_conversion)
+ && !NILP (Ffboundp (coding->pre_write_conversion)))
+ str = run_pre_post_conversion_on_str (str, coding, 1);
+
+ from = 0;
+ to = XSTRING (str)->size;
+ to_byte = STRING_BYTES (XSTRING (str));
+
+ saved_coding_symbol = Qnil;
+ if (! CODING_REQUIRE_ENCODING (coding))
+ {
+ if (STRING_MULTIBYTE (str))
+ {
+ str = Fstring_as_unibyte (str);
+ nocopy = 1;
+ }
+ return (nocopy ? str : Fcopy_sequence (str));
+ }
+
+ /* Encoding routines determine the multibyteness of the source text
+ by coding->src_multibyte. */
+ coding->src_multibyte = STRING_MULTIBYTE (str);
+ coding->dst_multibyte = 0;
+
+ if (coding->composing != COMPOSITION_DISABLED)
+ coding_save_composition (coding, from, to, str);
+
+ /* Try to skip the heading and tailing ASCIIs. */
+ if (coding->type != coding_type_ccl)
{
- int chars= (coding->fake_multibyte
- ? multibyte_chars_in_text (buf + from, coding->produced)
- : coding->produced_char);
- str = make_multibyte_string (buf, len + chars, len + coding->produced);
+ int from_orig = from;
+
+ SHRINK_CONVERSION_REGION (&from, &to_byte, coding, XSTRING (str)->data,
+ 1);
+ if (from == to_byte)
+ return (nocopy ? str : Fcopy_sequence (str));
}
+ len = encoding_buffer_size (coding, to_byte - from);
+ len += from + STRING_BYTES (XSTRING (str)) - to_byte;
+ GCPRO1 (str);
+ buf = get_conversion_buffer (len);
+ UNGCPRO;
+
+ if (from > 0)
+ bcopy (XSTRING (str)->data, buf, from);
+ result = encode_coding (coding, XSTRING (str)->data + from,
+ buf + from, to_byte - from, len);
+ bcopy (XSTRING (str)->data + to_byte, buf + from + coding->produced,
+ STRING_BYTES (XSTRING (str)) - to_byte);
+
+ len = from + STRING_BYTES (XSTRING (str)) - to_byte;
+ str = make_unibyte_string (buf, len + coding->produced);
+ coding_free_composition_data (coding);
+
return str;
}
/* At first, gather possible coding systems in VAL. */
val = Qnil;
- for (tmp = Vcoding_category_list; !NILP (tmp); tmp = XCONS (tmp)->cdr)
+ for (tmp = Vcoding_category_list; CONSP (tmp); tmp = XCDR (tmp))
{
- int idx
- = XFASTINT (Fget (XCONS (tmp)->car, Qcoding_category_index));
- if (coding_mask & (1 << idx))
+ Lisp_Object category_val, category_index;
+
+ category_index = Fget (XCAR (tmp), Qcoding_category_index);
+ category_val = Fsymbol_value (XCAR (tmp));
+ if (!NILP (category_val)
+ && NATNUMP (category_index)
+ && (coding_mask & (1 << XFASTINT (category_index))))
{
- val = Fcons (Fsymbol_value (XCONS (tmp)->car), val);
+ val = Fcons (category_val, val);
if (highest)
break;
}
val = Fnreverse (val);
/* Then, replace the elements with subsidiary coding systems. */
- for (tmp = val; !NILP (tmp); tmp = XCONS (tmp)->cdr)
+ for (tmp = val; CONSP (tmp); tmp = XCDR (tmp))
{
if (eol_type != CODING_EOL_UNDECIDED
&& eol_type != CODING_EOL_INCONSISTENT)
{
Lisp_Object eol;
- eol = Fget (XCONS (tmp)->car, Qeol_type);
+ eol = Fget (XCAR (tmp), Qeol_type);
if (VECTORP (eol))
- XCONS (tmp)->car = XVECTOR (eol)->contents[eol_type];
+ XCAR (tmp) = XVECTOR (eol)->contents[eol_type];
}
}
- return (highest ? XCONS (val)->car : val);
+ return (highest ? XCAR (val) : val);
}
DEFUN ("detect-coding-region", Fdetect_coding_region, Sdetect_coding_region,
!NILP (highest));
}
+/* Return an intersection of lists L1 and L2. */
+
+static Lisp_Object
+intersection (l1, l2)
+ Lisp_Object l1, l2;
+{
+ Lisp_Object val;
+
+ for (val = Qnil; CONSP (l1); l1 = XCDR (l1))
+ {
+ if (!NILP (Fmemq (XCAR (l1), l2)))
+ val = Fcons (XCAR (l1), val);
+ }
+ return val;
+}
+
+
+/* Subroutine for Fsafe_coding_systems_region_internal.
+
+ Return a list of coding systems that safely encode the multibyte
+ text between P and PEND. SAFE_CODINGS, if non-nil, is a list of
+ possible coding systems. If it is nil, it means that we have not
+ yet found any coding systems.
+
+ WORK_TABLE is a copy of the char-table Vchar_coding_system_table. An
+ element of WORK_TABLE is set to t once the element is looked up.
+
+ If a non-ASCII single byte char is found, set
+ *single_byte_char_found to 1. */
+
+static Lisp_Object
+find_safe_codings (p, pend, safe_codings, work_table, single_byte_char_found)
+ unsigned char *p, *pend;
+ Lisp_Object safe_codings, work_table;
+ int *single_byte_char_found;
+{
+ int c, len, idx;
+ Lisp_Object val;
+
+ while (p < pend)
+ {
+ c = STRING_CHAR_AND_LENGTH (p, pend - p, len);
+ p += len;
+ if (ASCII_BYTE_P (c))
+ /* We can ignore ASCII characters here. */
+ continue;
+ if (SINGLE_BYTE_CHAR_P (c))
+ *single_byte_char_found = 1;
+ if (NILP (safe_codings))
+ continue;
+ /* Check the safe coding systems for C. */
+ val = char_table_ref_and_index (work_table, c, &idx);
+ if (EQ (val, Qt))
+ /* This element was already checked. Ignore it. */
+ continue;
+ /* Remember that we checked this element. */
+ CHAR_TABLE_SET (work_table, idx, Qt);
+
+ /* If there are some safe coding systems for C and we have
+ already found the other set of coding systems for the
+ different characters, get the intersection of them. */
+ if (!EQ (safe_codings, Qt) && !NILP (val))
+ val = intersection (safe_codings, val);
+ safe_codings = val;
+ }
+ return safe_codings;
+}
+
+
+/* Return a list of coding systems that safely encode the text between
+ START and END. If the text contains only ASCII or is unibyte,
+ return t. */
+
+DEFUN ("find-coding-systems-region-internal",
+ Ffind_coding_systems_region_internal,
+ Sfind_coding_systems_region_internal, 2, 2, 0,
+ "Internal use only.")
+ (start, end)
+ Lisp_Object start, end;
+{
+ Lisp_Object work_table, safe_codings;
+ int non_ascii_p = 0;
+ int single_byte_char_found = 0;
+ unsigned char *p1, *p1end, *p2, *p2end, *p;
+ Lisp_Object args[2];
+
+ if (STRINGP (start))
+ {
+ if (!STRING_MULTIBYTE (start))
+ return Qt;
+ p1 = XSTRING (start)->data, p1end = p1 + STRING_BYTES (XSTRING (start));
+ p2 = p2end = p1end;
+ if (XSTRING (start)->size != STRING_BYTES (XSTRING (start)))
+ non_ascii_p = 1;
+ }
+ else
+ {
+ int from, to, stop;
+
+ CHECK_NUMBER_COERCE_MARKER (start, 0);
+ CHECK_NUMBER_COERCE_MARKER (end, 1);
+ if (XINT (start) < BEG || XINT (end) > Z || XINT (start) > XINT (end))
+ args_out_of_range (start, end);
+ if (NILP (current_buffer->enable_multibyte_characters))
+ return Qt;
+ from = CHAR_TO_BYTE (XINT (start));
+ to = CHAR_TO_BYTE (XINT (end));
+ stop = from < GPT_BYTE && GPT_BYTE < to ? GPT_BYTE : to;
+ p1 = BYTE_POS_ADDR (from), p1end = p1 + (stop - from);
+ if (stop == to)
+ p2 = p2end = p1end;
+ else
+ p2 = BYTE_POS_ADDR (stop), p2end = p2 + (to - stop);
+ if (XINT (end) - XINT (start) != to - from)
+ non_ascii_p = 1;
+ }
+
+ if (!non_ascii_p)
+ {
+ /* We are sure that the text contains no multibyte character.
+ Check if it contains eight-bit-graphic. */
+ p = p1;
+ for (p = p1; p < p1end && ASCII_BYTE_P (*p); p++);
+ if (p == p1end)
+ {
+ for (p = p2; p < p2end && ASCII_BYTE_P (*p); p++);
+ if (p == p2end)
+ return Qt;
+ }
+ }
+
+ /* The text contains non-ASCII characters. */
+ work_table = Fcopy_sequence (Vchar_coding_system_table);
+ safe_codings = find_safe_codings (p1, p1end, Qt, work_table,
+ &single_byte_char_found);
+ if (p2 < p2end)
+ safe_codings = find_safe_codings (p2, p2end, safe_codings, work_table,
+ &single_byte_char_found);
+
+ if (!single_byte_char_found)
+ {
+ /* Append generic coding systems. */
+ Lisp_Object args[2];
+ args[0] = safe_codings;
+ args[1] = Fchar_table_extra_slot (Vchar_coding_system_table,
+ make_number (0));
+ safe_codings = Fappend (make_number (2), args);
+ }
+ else
+ safe_codings = Fcons (Qraw_text, Fcons (Qemacs_mule, safe_codings));
+ return safe_codings;
+}
+
+
Lisp_Object
code_convert_region1 (start, end, coding_system, encodep)
Lisp_Object start, end, coding_system;
error ("Invalid coding system: %s", XSYMBOL (coding_system)->name->data);
coding.mode |= CODING_MODE_LAST_BLOCK;
+ coding.src_multibyte = coding.dst_multibyte
+ = !NILP (current_buffer->enable_multibyte_characters);
code_convert_region (from, CHAR_TO_BYTE (from), to, CHAR_TO_BYTE (to),
&coding, encodep, 1);
Vlast_coding_system_used = coding.symbol;
error ("Invalid coding system: %s", XSYMBOL (coding_system)->name->data);
coding.mode |= CODING_MODE_LAST_BLOCK;
+ string = (encodep
+ ? encode_coding_string (string, &coding, !NILP (nocopy))
+ : decode_coding_string (string, &coding, !NILP (nocopy)));
Vlast_coding_system_used = coding.symbol;
- return code_convert_string (string, &coding, encodep, !NILP (nocopy));
+
+ return string;
}
DEFUN ("decode-coding-string", Fdecode_coding_string, Sdecode_coding_string,
}
/* Encode or decode STRING according to CODING_SYSTEM.
- Do not set Vlast_coding_system_used. */
+ Do not set Vlast_coding_system_used.
+
+ This function is called only from macros DECODE_FILE and
+ ENCODE_FILE, thus we ignore character composition. */
Lisp_Object
code_convert_string_norecord (string, coding_system, encodep)
if (setup_coding_system (Fcheck_coding_system (coding_system), &coding) < 0)
error ("Invalid coding system: %s", XSYMBOL (coding_system)->name->data);
+ coding.composing = COMPOSITION_DISABLED;
coding.mode |= CODING_MODE_LAST_BLOCK;
- return code_convert_string (string, &coding, encodep, Qt);
+ return (encodep
+ ? encode_coding_string (string, &coding, 1)
+ : decode_coding_string (string, &coding, 1));
}
\f
DEFUN ("decode-sjis-char", Fdecode_sjis_char, Sdecode_sjis_char, 1, 1, 0,
- "Decode a JISX0208 character of shift-jis encoding.\n\
-CODE is the character code in SJIS.\n\
+ "Decode a Japanese character which has CODE in shift_jis encoding.\n\
Return the corresponding character.")
(code)
Lisp_Object code;
CHECK_NUMBER (code, 0);
s1 = (XFASTINT (code)) >> 8, s2 = (XFASTINT (code)) & 0xFF;
- DECODE_SJIS (s1, s2, c1, c2);
- XSETFASTINT (val, MAKE_NON_ASCII_CHAR (charset_jisx0208, c1, c2));
+ if (s1 == 0)
+ {
+ if (s2 < 0x80)
+ XSETFASTINT (val, s2);
+ else if (s2 >= 0xA0 || s2 <= 0xDF)
+ XSETFASTINT (val, MAKE_CHAR (charset_katakana_jisx0201, s2, 0));
+ else
+ error ("Invalid Shift JIS code: %x", XFASTINT (code));
+ }
+ else
+ {
+ if ((s1 < 0x80 || s1 > 0x9F && s1 < 0xE0 || s1 > 0xEF)
+ || (s2 < 0x40 || s2 == 0x7F || s2 > 0xFC))
+ error ("Invalid Shift JIS code: %x", XFASTINT (code));
+ DECODE_SJIS (s1, s2, c1, c2);
+ XSETFASTINT (val, MAKE_CHAR (charset_jisx0208, c1, c2));
+ }
return val;
}
DEFUN ("encode-sjis-char", Fencode_sjis_char, Sencode_sjis_char, 1, 1, 0,
- "Encode a JISX0208 character CHAR to SJIS coding system.\n\
-Return the corresponding character code in SJIS.")
+ "Encode a Japanese character CHAR to shift_jis encoding.\n\
+Return the corresponding code in SJIS.")
(ch)
Lisp_Object ch;
{
CHECK_NUMBER (ch, 0);
SPLIT_CHAR (XFASTINT (ch), charset, c1, c2);
- if (charset == charset_jisx0208)
+ if (charset == CHARSET_ASCII)
+ {
+ val = ch;
+ }
+ else if (charset == charset_jisx0208
+ && c1 > 0x20 && c1 < 0x7F && c2 > 0x20 && c2 < 0x7F)
{
ENCODE_SJIS (c1, c2, s1, s2);
XSETFASTINT (val, (s1 << 8) | s2);
}
+ else if (charset == charset_katakana_jisx0201
+ && c1 > 0x20 && c2 < 0xE0)
+ {
+ XSETFASTINT (val, c1 | 0x80);
+ }
else
- XSETFASTINT (val, 0);
+ error ("Can't encode to shift_jis: %d", XFASTINT (ch));
return val;
}
DEFUN ("decode-big5-char", Fdecode_big5_char, Sdecode_big5_char, 1, 1, 0,
- "Decode a Big5 character CODE of BIG5 coding system.\n\
-CODE is the character code in BIG5.\n\
+ "Decode a Big5 character which has CODE in BIG5 coding system.\n\
Return the corresponding character.")
(code)
Lisp_Object code;
CHECK_NUMBER (code, 0);
b1 = (XFASTINT (code)) >> 8, b2 = (XFASTINT (code)) & 0xFF;
- DECODE_BIG5 (b1, b2, charset, c1, c2);
- XSETFASTINT (val, MAKE_NON_ASCII_CHAR (charset, c1, c2));
+ if (b1 == 0)
+ {
+ if (b2 >= 0x80)
+ error ("Invalid BIG5 code: %x", XFASTINT (code));
+ val = code;
+ }
+ else
+ {
+ if ((b1 < 0xA1 || b1 > 0xFE)
+ || (b2 < 0x40 || (b2 > 0x7E && b2 < 0xA1) || b2 > 0xFE))
+ error ("Invalid BIG5 code: %x", XFASTINT (code));
+ DECODE_BIG5 (b1, b2, charset, c1, c2);
+ XSETFASTINT (val, MAKE_CHAR (charset, c1, c2));
+ }
return val;
}
CHECK_NUMBER (ch, 0);
SPLIT_CHAR (XFASTINT (ch), charset, c1, c2);
- if (charset == charset_big5_1 || charset == charset_big5_2)
+ if (charset == CHARSET_ASCII)
+ {
+ val = ch;
+ }
+ else if ((charset == charset_big5_1
+ && (XFASTINT (ch) >= 0x250a1 && XFASTINT (ch) <= 0x271ec))
+ || (charset == charset_big5_2
+ && XFASTINT (ch) >= 0x290a1 && XFASTINT (ch) <= 0x2bdb2))
{
ENCODE_BIG5 (charset, c1, c2, b1, b2);
XSETFASTINT (val, (b1 << 8) | b2);
}
else
- XSETFASTINT (val, 0);
+ error ("Can't encode to Big5: %d", XFASTINT (ch));
return val;
}
\f
setup_coding_system (Fcheck_coding_system (coding_system), &terminal_coding);
/* We had better not send unsafe characters to terminal. */
terminal_coding.flags |= CODING_FLAG_ISO_SAFE;
-
+ /* Characer composition should be disabled. */
+ terminal_coding.composing = COMPOSITION_DISABLED;
+ terminal_coding.src_multibyte = 1;
+ terminal_coding.dst_multibyte = 0;
return Qnil;
}
CHECK_SYMBOL (coding_system, 0);
setup_coding_system (Fcheck_coding_system (coding_system),
&safe_terminal_coding);
+ /* Characer composition should be disabled. */
+ safe_terminal_coding.composing = COMPOSITION_DISABLED;
+ safe_terminal_coding.src_multibyte = 1;
+ safe_terminal_coding.dst_multibyte = 0;
return Qnil;
}
{
CHECK_SYMBOL (coding_system, 0);
setup_coding_system (Fcheck_coding_system (coding_system), &keyboard_coding);
+ /* Characer composition should be disabled. */
+ keyboard_coding.composing = COMPOSITION_DISABLED;
return Qnil;
}
if (NILP (chain))
return Qnil;
- for (; CONSP (chain); chain = XCONS (chain)->cdr)
+ for (; CONSP (chain); chain = XCDR (chain))
{
Lisp_Object elt;
- elt = XCONS (chain)->car;
+ elt = XCAR (chain);
if (CONSP (elt)
&& ((STRINGP (target)
- && STRINGP (XCONS (elt)->car)
- && fast_string_match (XCONS (elt)->car, target) >= 0)
- || (INTEGERP (target) && EQ (target, XCONS (elt)->car))))
+ && STRINGP (XCAR (elt))
+ && fast_string_match (XCAR (elt), target) >= 0)
+ || (INTEGERP (target) && EQ (target, XCAR (elt)))))
{
- val = XCONS (elt)->cdr;
+ val = XCDR (elt);
/* Here, if VAL is both a valid coding system and a valid
function symbol, we return VAL as a coding system. */
if (CONSP (val))
DEFUN ("update-coding-systems-internal", Fupdate_coding_systems_internal,
Supdate_coding_systems_internal, 0, 0, 0,
"Update internal database for ISO2022 and CCL based coding systems.\n\
-When values of the following coding categories are changed, you must\n\
-call this function:\n\
- coding-category-iso-7, coding-category-iso-7-tight,\n\
- coding-category-iso-8-1, coding-category-iso-8-2,\n\
- coding-category-iso-7-else, coding-category-iso-8-else,\n\
- coding-category-ccl")
+When values of any coding categories are changed, you must\n\
+call this function")
()
{
int i;
- for (i = CODING_CATEGORY_IDX_ISO_7; i <= CODING_CATEGORY_IDX_CCL; i++)
+ for (i = CODING_CATEGORY_IDX_EMACS_MULE; i < CODING_CATEGORY_IDX_MAX; i++)
{
Lisp_Object val;
while (CONSP (val) && i < CODING_CATEGORY_IDX_MAX)
{
- if (! SYMBOLP (XCONS (val)->car))
+ if (! SYMBOLP (XCAR (val)))
break;
- idx = XFASTINT (Fget (XCONS (val)->car, Qcoding_category_index));
+ idx = XFASTINT (Fget (XCAR (val), Qcoding_category_index));
if (idx >= CODING_CATEGORY_IDX_MAX)
break;
coding_priorities[i++] = (1 << idx);
- val = XCONS (val)->cdr;
+ val = XCDR (val);
}
/* If coding-category-list is valid and contains all coding
categories, `i' should be CODING_CATEGORY_IDX_MAX now. If not,
- the following code saves Emacs from craching. */
+ the following code saves Emacs from crashing. */
while (i < CODING_CATEGORY_IDX_MAX)
coding_priorities[i++] = CODING_CATEGORY_MASK_RAW_TEXT;
for (i = 0x21 ; i < 0x7F; i++)
emacs_code_class[i] = EMACS_ascii_code;
emacs_code_class[0x7F] = EMACS_control_code;
- emacs_code_class[0x80] = EMACS_leading_code_composition;
- for (i = 0x81; i < 0xFF; i++)
+ for (i = 0x80; i < 0xFF; i++)
emacs_code_class[i] = EMACS_invalid_code;
emacs_code_class[LEADING_CODE_PRIVATE_11] = EMACS_leading_code_3;
emacs_code_class[LEADING_CODE_PRIVATE_12] = EMACS_leading_code_3;
/* ISO2022 specific initialize routine. */
for (i = 0; i < 0x20; i++)
- iso_code_class[i] = ISO_control_code;
+ iso_code_class[i] = ISO_control_0;
for (i = 0x21; i < 0x7F; i++)
iso_code_class[i] = ISO_graphic_plane_0;
for (i = 0x80; i < 0xA0; i++)
- iso_code_class[i] = ISO_control_code;
+ iso_code_class[i] = ISO_control_1;
for (i = 0xA1; i < 0xFF; i++)
iso_code_class[i] = ISO_graphic_plane_1;
iso_code_class[0x20] = iso_code_class[0x7F] = ISO_0x20_or_0x7F;
#else
system_eol_type = CODING_EOL_LF;
#endif
+
+ inhibit_pre_post_conversion = 0;
}
#ifdef emacs
Qtranslation_table_for_encode = intern ("translation-table-for-encode");
staticpro (&Qtranslation_table_for_encode);
- Qsafe_charsets = intern ("safe-charsets");
- staticpro (&Qsafe_charsets);
+ Qsafe_chars = intern ("safe-chars");
+ staticpro (&Qsafe_chars);
+
+ Qchar_coding_system = intern ("char-coding-system");
+ staticpro (&Qchar_coding_system);
+
+ /* Intern this now in case it isn't already done.
+ Setting this variable twice is harmless.
+ But don't staticpro it here--that is done in alloc.c. */
+ Qchar_table_extra_slots = intern ("char-table-extra-slots");
+ Fput (Qsafe_chars, Qchar_table_extra_slots, make_number (0));
+ Fput (Qchar_coding_system, Qchar_table_extra_slots, make_number (1));
Qvalid_codes = intern ("valid-codes");
staticpro (&Qvalid_codes);
defsubr (&Scheck_coding_system);
defsubr (&Sdetect_coding_region);
defsubr (&Sdetect_coding_string);
+ defsubr (&Sfind_coding_systems_region_internal);
defsubr (&Sdecode_coding_region);
defsubr (&Sencode_coding_region);
defsubr (&Sdecode_coding_string);
DEFVAR_LISP ("coding-system-for-write", &Vcoding_system_for_write,
"Specify the coding system for write operations.\n\
-It is useful to bind this variable with `let', but do not set it globally.\n\
-If the value is a coding system, it is used for encoding on write operation.\n\
-If not, an appropriate element is used from one of the coding system alists:\n\
+Programs bind this variable with `let', but you should not set it globally.\n\
+If the value is a coding system, it is used for encoding of output,\n\
+when writing it to a file and when sending it to a file or subprocess.\n\
+\n\
+If this does not specify a coding system, an appropriate element\n\
+is used from one of the coding system alists:\n\
There are three such tables, `file-coding-system-alist',\n\
-`process-coding-system-alist', and `network-coding-system-alist'.");
+`process-coding-system-alist', and `network-coding-system-alist'.\n\
+For output to files, if the above procedure does not specify a coding system,\n\
+the value of `buffer-file-coding-system' is used.");
Vcoding_system_for_write = Qnil;
DEFVAR_LISP ("last-coding-system-used", &Vlast_coding_system_used,
Vlast_coding_system_used = Qnil;
DEFVAR_BOOL ("inhibit-eol-conversion", &inhibit_eol_conversion,
- "*Non-nil inhibit code conversion of end-of-line format in any cases.");
+ "*Non-nil means always inhibit code conversion of end-of-line format.\n\
+See info node `Coding Systems' and info node `Text and Binary' concerning\n\
+such conversion.");
inhibit_eol_conversion = 0;
DEFVAR_BOOL ("inherit-process-coding-system", &inherit_process_coding_system,
See also the function `find-operation-coding-system'.");
Vnetwork_coding_system_alist = Qnil;
- DEFVAR_INT ("eol-mnemonic-unix", &eol_mnemonic_unix,
- "Mnemonic character indicating UNIX-like end-of-line format (i.e. LF) .");
- eol_mnemonic_unix = ':';
+ DEFVAR_LISP ("locale-coding-system", &Vlocale_coding_system,
+ "Coding system to use with system messages.");
+ Vlocale_coding_system = Qnil;
+
+ /* The eol mnemonics are reset in startup.el system-dependently. */
+ DEFVAR_LISP ("eol-mnemonic-unix", &eol_mnemonic_unix,
+ "*String displayed in mode line for UNIX-like (LF) end-of-line format.");
+ eol_mnemonic_unix = build_string (":");
- DEFVAR_INT ("eol-mnemonic-dos", &eol_mnemonic_dos,
- "Mnemonic character indicating DOS-like end-of-line format (i.e. CRLF).");
- eol_mnemonic_dos = '\\';
+ DEFVAR_LISP ("eol-mnemonic-dos", &eol_mnemonic_dos,
+ "*String displayed in mode line for DOS-like (CRLF) end-of-line format.");
+ eol_mnemonic_dos = build_string ("\\");
- DEFVAR_INT ("eol-mnemonic-mac", &eol_mnemonic_mac,
- "Mnemonic character indicating MAC-like end-of-line format (i.e. CR).");
- eol_mnemonic_mac = '/';
+ DEFVAR_LISP ("eol-mnemonic-mac", &eol_mnemonic_mac,
+ "*String displayed in mode line for MAC-like (CR) end-of-line format.");
+ eol_mnemonic_mac = build_string ("/");
- DEFVAR_INT ("eol-mnemonic-undecided", &eol_mnemonic_undecided,
- "Mnemonic character indicating end-of-line format is not yet decided.");
- eol_mnemonic_undecided = ':';
+ DEFVAR_LISP ("eol-mnemonic-undecided", &eol_mnemonic_undecided,
+ "*String displayed in mode line when end-of-line format is not yet determined.");
+ eol_mnemonic_undecided = build_string (":");
DEFVAR_LISP ("enable-character-translation", &Venable_character_translation,
"*Non-nil enables character translation while encoding and decoding.");
The default value is `select-safe-coding-system' (which see).");
Vselect_safe_coding_system_function = Qnil;
+ DEFVAR_LISP ("char-coding-system-table", &Vchar_coding_system_table,
+ "Char-table containing safe coding systems of each characters.\n\
+Each element doesn't include such generic coding systems that can\n\
+encode any characters. They are in the first extra slot.");
+ Vchar_coding_system_table = Fmake_char_table (Qchar_coding_system, Qnil);
+
+ DEFVAR_BOOL ("inhibit-iso-escape-detection",
+ &inhibit_iso_escape_detection,
+ "If non-nil, Emacs ignores ISO2022's escape sequence on code detection.\n\
+\n\
+By default, on reading a file, Emacs tries to detect how the text is\n\
+encoded. This code detection is sensitive to escape sequences. If\n\
+the sequence is valid as ISO2022, the code is determined as one of\n\
+the ISO2022 encodings, and the file is decoded by the corresponding\n\
+coding system (e.g. `iso-2022-7bit').\n\
+\n\
+However, there may be a case that you want to read escape sequences in\n\
+a file as is. In such a case, you can set this variable to non-nil.\n\
+Then, as the code detection ignores any escape sequences, no file is\n\
+detected as encoded in some ISO2022 encoding. The result is that all\n\
+escape sequences become visible in a buffer.\n\
+\n\
+The default value is nil, and it is strongly recommended not to change\n\
+it. That is because many Emacs Lisp source files that contain\n\
+non-ASCII characters are encoded by the coding system `iso-2022-7bit'\n\
+in Emacs's distribution, and they won't be decoded correctly on\n\
+reading if you suppress escape sequence detection.\n\
+\n\
+The other way to read escape sequences in a file without decoding is\n\
+to explicitly specify some coding system that doesn't use ISO2022's\n\
+escape sequence (e.g `latin-1') on reading by \\[universal-coding-system-argument].");
+ inhibit_iso_escape_detection = 0;
+}
+
+char *
+emacs_strerror (error_number)
+ int error_number;
+{
+ char *str;
+
+ synchronize_system_messages_locale ();
+ str = strerror (error_number);
+
+ if (! NILP (Vlocale_coding_system))
+ {
+ Lisp_Object dec = code_convert_string_norecord (build_string (str),
+ Vlocale_coding_system,
+ 0);
+ str = (char *) XSTRING (dec)->data;
+ }
+
+ return str;
}
#endif /* emacs */
+
HCoop / bpt / emacs.git / blobdiff