/* String search routines for GNU Emacs.
- Copyright (C) 1985, 86, 87, 93, 94, 97, 1998 Free Software Foundation, Inc.
+ Copyright (C) 1985, 86,87,93,94,97,98, 1999 Free Software Foundation, Inc.
This file is part of GNU Emacs.
error ("Stack overflow in regexp matcher");
}
-#ifdef __STDC__
-#define CONST const
-#else
-#define CONST
-#endif
-
/* Compile a regexp and signal a Lisp error if anything goes wrong.
PATTERN is the pattern to compile.
CP is the place to put the result.
int posix;
int multibyte;
{
- char *raw_pattern;
+ unsigned char *raw_pattern;
int raw_pattern_size;
char *val;
reg_syntax_t old;
if (multibyte == STRING_MULTIBYTE (pattern))
{
- raw_pattern = (char *) XSTRING (pattern)->data;
- raw_pattern_size = XSTRING (pattern)->size_byte;
+ raw_pattern = (unsigned char *) SDATA (pattern);
+ raw_pattern_size = SBYTES (pattern);
}
else if (multibyte)
{
- raw_pattern_size = count_size_as_multibyte (XSTRING (pattern)->data,
- XSTRING (pattern)->size);
- raw_pattern = (char *) alloca (raw_pattern_size + 1);
- copy_text (XSTRING (pattern)->data, raw_pattern,
- XSTRING (pattern)->size, 0, 1);
+ raw_pattern_size = count_size_as_multibyte (SDATA (pattern),
+ SCHARS (pattern));
+ raw_pattern = (unsigned char *) alloca (raw_pattern_size + 1);
+ copy_text (SDATA (pattern), raw_pattern,
+ SCHARS (pattern), 0, 1);
}
else
{
by subtracting nonascii-insert-offset from each non-ASCII char,
so that only the multibyte chars which really correspond to
the chosen single-byte character set can possibly match. */
- raw_pattern_size = XSTRING (pattern)->size;
- raw_pattern = (char *) alloca (raw_pattern_size + 1);
- copy_text (XSTRING (pattern)->data, raw_pattern,
- XSTRING (pattern)->size_byte, 1, 0);
+ raw_pattern_size = SCHARS (pattern);
+ raw_pattern = (unsigned char *) alloca (raw_pattern_size + 1);
+ copy_text (SDATA (pattern), raw_pattern,
+ SBYTES (pattern), 1, 0);
}
cp->regexp = Qnil;
- cp->buf.translate = (! NILP (translate) ? translate : 0);
+ cp->buf.translate = (! NILP (translate) ? translate : make_number (0));
cp->posix = posix;
cp->buf.multibyte = multibyte;
BLOCK_INPUT;
old = re_set_syntax (RE_SYNTAX_EMACS
| (posix ? 0 : RE_NO_POSIX_BACKTRACKING));
- val = (char *) re_compile_pattern (raw_pattern, raw_pattern_size, &cp->buf);
+ val = (char *) re_compile_pattern ((char *)raw_pattern,
+ raw_pattern_size, &cp->buf);
re_set_syntax (old);
UNBLOCK_INPUT;
if (val)
cp->regexp = Fcopy_sequence (pattern);
}
+/* Shrink each compiled regexp buffer in the cache
+ to the size actually used right now.
+ This is called from garbage collection. */
+
+void
+shrink_regexp_cache ()
+{
+ struct regexp_cache *cp;
+
+ for (cp = searchbuf_head; cp != 0; cp = cp->next)
+ {
+ cp->buf.allocated = cp->buf.used;
+ cp->buf.buffer
+ = (unsigned char *) xrealloc (cp->buf.buffer, cp->buf.used);
+ }
+}
+
/* Compile a regexp if necessary, but first check to see if there's one in
the cache.
PATTERN is the pattern to compile.
for (cpp = &searchbuf_head; ; cpp = &cp->next)
{
cp = *cpp;
- if (XSTRING (cp->regexp)->size == XSTRING (pattern)->size
+ /* Entries are initialized to nil, and may be set to nil by
+ compile_pattern_1 if the pattern isn't valid. Don't apply
+ string accessors in those cases. However, compile_pattern_1
+ is only applied to the cache entry we pick here to reuse. So
+ nil should never appear before a non-nil entry. */
+ if (NILP (cp->regexp))
+ goto compile_it;
+ if (SCHARS (cp->regexp) == SCHARS (pattern)
+ && STRING_MULTIBYTE (cp->regexp) == STRING_MULTIBYTE (pattern)
&& !NILP (Fstring_equal (cp->regexp, pattern))
- && cp->buf.translate == (! NILP (translate) ? translate : 0)
+ && EQ (cp->buf.translate, (! NILP (translate) ? translate : make_number (0)))
&& cp->posix == posix
&& cp->buf.multibyte == multibyte)
break;
- /* If we're at the end of the cache, compile into the last cell. */
+ /* If we're at the end of the cache, compile into the nil cell
+ we found, or the last (least recently used) cell with a
+ string value. */
if (cp->next == 0)
{
+ compile_it:
compile_pattern_1 (cp, pattern, translate, regp, posix, multibyte);
break;
}
if (running_asynch_code)
save_search_regs ();
- CHECK_STRING (string, 0);
+ CHECK_STRING (string);
bufp = compile_pattern (string, &search_regs,
(!NILP (current_buffer->case_fold_search)
? DOWNCASE_TABLE : Qnil),
}
re_match_object = Qnil;
-
+
i = re_match_2 (bufp, (char *) p1, s1, (char *) p2, s2,
PT_BYTE - BEGV_BYTE, &search_regs,
ZV_BYTE - BEGV_BYTE);
+ immediate_quit = 0;
+
if (i == -2)
matcher_overflow ();
= BYTE_TO_CHAR (search_regs.end[i] + BEGV_BYTE);
}
XSETBUFFER (last_thing_searched, current_buffer);
- immediate_quit = 0;
return val;
}
DEFUN ("looking-at", Flooking_at, Slooking_at, 1, 1, 0,
- "Return t if text after point matches regular expression REGEXP.\n\
-This function modifies the match data that `match-beginning',\n\
-`match-end' and `match-data' access; save and restore the match\n\
-data if you want to preserve them.")
- (regexp)
+ doc: /* Return t if text after point matches regular expression REGEXP.
+This function modifies the match data that `match-beginning',
+`match-end' and `match-data' access; save and restore the match
+data if you want to preserve them. */)
+ (regexp)
Lisp_Object regexp;
{
return looking_at_1 (regexp, 0);
}
DEFUN ("posix-looking-at", Fposix_looking_at, Sposix_looking_at, 1, 1, 0,
- "Return t if text after point matches regular expression REGEXP.\n\
-Find the longest match, in accord with Posix regular expression rules.\n\
-This function modifies the match data that `match-beginning',\n\
-`match-end' and `match-data' access; save and restore the match\n\
-data if you want to preserve them.")
- (regexp)
+ doc: /* Return t if text after point matches regular expression REGEXP.
+Find the longest match, in accord with Posix regular expression rules.
+This function modifies the match data that `match-beginning',
+`match-end' and `match-data' access; save and restore the match
+data if you want to preserve them. */)
+ (regexp)
Lisp_Object regexp;
{
return looking_at_1 (regexp, 1);
if (running_asynch_code)
save_search_regs ();
- CHECK_STRING (regexp, 0);
- CHECK_STRING (string, 1);
+ CHECK_STRING (regexp);
+ CHECK_STRING (string);
if (NILP (start))
pos = 0, pos_byte = 0;
else
{
- int len = XSTRING (string)->size;
+ int len = SCHARS (string);
- CHECK_NUMBER (start, 2);
+ CHECK_NUMBER (start);
pos = XINT (start);
if (pos < 0 && -pos <= len)
pos = len + pos;
STRING_MULTIBYTE (string));
immediate_quit = 1;
re_match_object = string;
-
- val = re_search (bufp, (char *) XSTRING (string)->data,
- XSTRING (string)->size_byte, pos_byte,
- XSTRING (string)->size_byte - pos_byte,
+
+ val = re_search (bufp, (char *) SDATA (string),
+ SBYTES (string), pos_byte,
+ SBYTES (string) - pos_byte,
&search_regs);
immediate_quit = 0;
last_thing_searched = Qt;
}
DEFUN ("string-match", Fstring_match, Sstring_match, 2, 3, 0,
- "Return index of start of first match for REGEXP in STRING, or nil.\n\
-If third arg START is non-nil, start search at that index in STRING.\n\
-For index of first char beyond the match, do (match-end 0).\n\
-`match-end' and `match-beginning' also give indices of substrings\n\
-matched by parenthesis constructs in the pattern.")
- (regexp, string, start)
+ doc: /* Return index of start of first match for REGEXP in STRING, or nil.
+Case is ignored if `case-fold-search' is non-nil in the current buffer.
+If third arg START is non-nil, start search at that index in STRING.
+For index of first char beyond the match, do (match-end 0).
+`match-end' and `match-beginning' also give indices of substrings
+matched by parenthesis constructs in the pattern.
+
+You can use the function `match-string' to extract the substrings
+matched by the parenthesis constructions in REGEXP. */)
+ (regexp, string, start)
Lisp_Object regexp, string, start;
{
return string_match_1 (regexp, string, start, 0);
}
DEFUN ("posix-string-match", Fposix_string_match, Sposix_string_match, 2, 3, 0,
- "Return index of start of first match for REGEXP in STRING, or nil.\n\
-Find the longest match, in accord with Posix regular expression rules.\n\
-If third arg START is non-nil, start search at that index in STRING.\n\
-For index of first char beyond the match, do (match-end 0).\n\
-`match-end' and `match-beginning' also give indices of substrings\n\
-matched by parenthesis constructs in the pattern.")
- (regexp, string, start)
+ doc: /* Return index of start of first match for REGEXP in STRING, or nil.
+Find the longest match, in accord with Posix regular expression rules.
+Case is ignored if `case-fold-search' is non-nil in the current buffer.
+If third arg START is non-nil, start search at that index in STRING.
+For index of first char beyond the match, do (match-end 0).
+`match-end' and `match-beginning' also give indices of substrings
+matched by parenthesis constructs in the pattern. */)
+ (regexp, string, start)
Lisp_Object regexp, string, start;
{
return string_match_1 (regexp, string, start, 1);
0, STRING_MULTIBYTE (string));
immediate_quit = 1;
re_match_object = string;
-
- val = re_search (bufp, (char *) XSTRING (string)->data,
- XSTRING (string)->size_byte, 0, XSTRING (string)->size_byte,
- 0);
+
+ val = re_search (bufp, (char *) SDATA (string),
+ SBYTES (string), 0,
+ SBYTES (string), 0);
immediate_quit = 0;
return val;
}
int
fast_c_string_match_ignore_case (regexp, string)
Lisp_Object regexp;
- char *string;
+ const char *string;
{
int val;
struct re_pattern_buffer *bufp;
return val;
}
\f
-/* max and min. */
-
-static int
-max (a, b)
- int a, b;
-{
- return ((a > b) ? a : b);
-}
-
-static int
-min (a, b)
- int a, b;
-{
- return ((a < b) ? a : b);
-}
-
-\f
/* The newline cache: remembering which sections of text have no newlines. */
/* If the user has requested newline caching, make sure it's on.
If ALLOW_QUIT is non-zero, set immediate_quit. That's good to do
except when inside redisplay. */
+int
scan_buffer (target, start, end, count, shortage, allow_quit)
register int target;
int start, end;
int allow_quit;
{
struct region_cache *newline_cache;
- int direction;
+ int direction;
if (count > 0)
{
examine. */
int ceiling_byte = CHAR_TO_BYTE (end) - 1;
int start_byte = CHAR_TO_BYTE (start);
+ int tem;
/* If we're looking for a newline, consult the newline cache
to see where we can avoid some scanning. */
bytes. BUFFER_CEILING_OF returns the last character
position that is contiguous, so the ceiling is the
position after that. */
- ceiling_byte = min (BUFFER_CEILING_OF (start_byte), ceiling_byte);
+ tem = BUFFER_CEILING_OF (start_byte);
+ ceiling_byte = min (tem, ceiling_byte);
{
- /* The termination address of the dumb loop. */
+ /* The termination address of the dumb loop. */
register unsigned char *ceiling_addr
= BYTE_POS_ADDR (ceiling_byte) + 1;
register unsigned char *cursor
/* The last character to check before the next obstacle. */
int ceiling_byte = CHAR_TO_BYTE (end);
int start_byte = CHAR_TO_BYTE (start);
+ int tem;
/* Consult the newline cache, if appropriate. */
if (target == '\n' && newline_cache)
}
/* Stop scanning before the gap. */
- ceiling_byte = max (BUFFER_FLOOR_OF (start_byte - 1), ceiling_byte);
+ tem = BUFFER_FLOOR_OF (start_byte - 1);
+ ceiling_byte = max (tem, ceiling_byte);
{
/* The termination address of the dumb loop. */
int old_immediate_quit = immediate_quit;
- /* If we are not in selective display mode,
- check only for newlines. */
- int selective_display = (!NILP (current_buffer->selective_display)
- && !INTEGERP (current_buffer->selective_display));
-
/* The code that follows is like scan_buffer
but checks for either newline or carriage return. */
}
else
{
- int start_byte = CHAR_TO_BYTE (start);
while (start_byte > limit_byte)
{
ceiling = BUFFER_FLOOR_OF (start_byte - 1);
if (shortage == 0)
pos--;
-
+
return pos;
}
\f
if (!NILP (count))
{
- CHECK_NUMBER (count, 3);
+ CHECK_NUMBER (count);
n *= XINT (count);
}
- CHECK_STRING (string, 0);
+ CHECK_STRING (string);
if (NILP (bound))
{
if (n > 0)
}
else
{
- CHECK_NUMBER_COERCE_MARKER (bound, 1);
+ CHECK_NUMBER_COERCE_MARKER (bound);
lim = XINT (bound);
- lim_byte = CHAR_TO_BYTE (lim);
if (n > 0 ? lim < PT : lim > PT)
error ("Invalid search bound (wrong side of point)");
if (lim > ZV)
lim = ZV, lim_byte = ZV_BYTE;
- if (lim < BEGV)
+ else if (lim < BEGV)
lim = BEGV, lim_byte = BEGV_BYTE;
+ else
+ lim_byte = CHAR_TO_BYTE (lim);
}
np = search_buffer (string, PT, PT_BYTE, lim, lim_byte, n, RE,
(!NILP (current_buffer->case_fold_search)
? current_buffer->case_canon_table
- : make_number (0)),
+ : Qnil),
(!NILP (current_buffer->case_fold_search)
? current_buffer->case_eqv_table
- : make_number (0)),
+ : Qnil),
posix);
if (np <= 0)
{
trivial_regexp_p (regexp)
Lisp_Object regexp;
{
- int len = XSTRING (regexp)->size_byte;
- unsigned char *s = XSTRING (regexp)->data;
- unsigned char c;
+ int len = SBYTES (regexp);
+ unsigned char *s = SDATA (regexp);
while (--len >= 0)
{
switch (*s++)
{
case '|': case '(': case ')': case '`': case '\'': case 'b':
case 'B': case '<': case '>': case 'w': case 'W': case 's':
- case 'S': case '=':
+ case 'S': case '=': case '{': case '}':
case 'c': case 'C': /* for categoryspec and notcategoryspec */
case '1': case '2': case '3': case '4': case '5':
case '6': case '7': case '8': case '9':
POSIX is nonzero if we want full backtracking (POSIX style)
for this pattern. 0 means backtrack only enough to get a valid match. */
-#define TRANSLATE(trt, d) \
- (! NILP (trt) ? XINT (Faref (trt, make_number (d))) : (d))
+#define TRANSLATE(out, trt, d) \
+do \
+ { \
+ if (! NILP (trt)) \
+ { \
+ Lisp_Object temp; \
+ temp = Faref (trt, make_number (d)); \
+ if (INTEGERP (temp)) \
+ out = XINT (temp); \
+ else \
+ out = d; \
+ } \
+ else \
+ out = d; \
+ } \
+while (0)
static int
search_buffer (string, pos, pos_byte, lim, lim_byte, n,
Lisp_Object inverse_trt;
int posix;
{
- int len = XSTRING (string)->size;
- int len_byte = XSTRING (string)->size_byte;
+ int len = SCHARS (string);
+ int len_byte = SBYTES (string);
register int i;
if (running_asynch_code)
save_search_regs ();
+ /* Searching 0 times means don't move. */
/* Null string is found at starting position. */
- if (len == 0)
+ if (len == 0 || n == 0)
{
- set_search_regs (pos, 0);
+ set_search_regs (pos_byte, 0);
return pos;
}
- /* Searching 0 times means don't move. */
- if (n == 0)
- return pos;
-
if (RE && !trivial_regexp_p (string))
{
unsigned char *p1, *p2;
s2 = 0;
}
re_match_object = Qnil;
-
+
while (n < 0)
{
int val;
}
if (val >= 0)
{
+ pos_byte = search_regs.start[0] + BEGV_BYTE;
for (i = 0; i < search_regs.num_regs; i++)
if (search_regs.start[i] >= 0)
{
}
if (val >= 0)
{
+ pos_byte = search_regs.end[0] + BEGV_BYTE;
for (i = 0; i < search_regs.num_regs; i++)
if (search_regs.start[i] >= 0)
{
int raw_pattern_size_byte;
unsigned char *patbuf;
int multibyte = !NILP (current_buffer->enable_multibyte_characters);
- unsigned char *base_pat = XSTRING (string)->data;
+ unsigned char *base_pat = SDATA (string);
int charset_base = -1;
- int simple = 1;
+ int boyer_moore_ok = 1;
/* MULTIBYTE says whether the text to be searched is multibyte.
We must convert PATTERN to match that, or we will not really
if (multibyte == STRING_MULTIBYTE (string))
{
- raw_pattern = (char *) XSTRING (string)->data;
- raw_pattern_size = XSTRING (string)->size;
- raw_pattern_size_byte = XSTRING (string)->size_byte;
+ raw_pattern = (unsigned char *) SDATA (string);
+ raw_pattern_size = SCHARS (string);
+ raw_pattern_size_byte = SBYTES (string);
}
else if (multibyte)
{
- raw_pattern_size = XSTRING (string)->size;
+ raw_pattern_size = SCHARS (string);
raw_pattern_size_byte
- = count_size_as_multibyte (XSTRING (string)->data,
+ = count_size_as_multibyte (SDATA (string),
raw_pattern_size);
- raw_pattern = (char *) alloca (raw_pattern_size_byte + 1);
- copy_text (XSTRING (string)->data, raw_pattern,
- XSTRING (string)->size, 0, 1);
+ raw_pattern = (unsigned char *) alloca (raw_pattern_size_byte + 1);
+ copy_text (SDATA (string), raw_pattern,
+ SCHARS (string), 0, 1);
}
else
{
by subtracting nonascii-insert-offset from each non-ASCII char,
so that only the multibyte chars which really correspond to
the chosen single-byte character set can possibly match. */
- raw_pattern_size = XSTRING (string)->size;
- raw_pattern_size_byte = XSTRING (string)->size;
- raw_pattern = (char *) alloca (raw_pattern_size + 1);
- copy_text (XSTRING (string)->data, raw_pattern,
- XSTRING (string)->size_byte, 1, 0);
+ raw_pattern_size = SCHARS (string);
+ raw_pattern_size_byte = SCHARS (string);
+ raw_pattern = (unsigned char *) alloca (raw_pattern_size + 1);
+ copy_text (SDATA (string), raw_pattern,
+ SBYTES (string), 1, 0);
}
/* Copy and optionally translate the pattern. */
{
while (--len >= 0)
{
- unsigned char workbuf[4], *str;
- int c, translated;
+ unsigned char str[MAX_MULTIBYTE_LENGTH];
+ int c, translated, inverse;
int in_charlen, charlen;
/* If we got here and the RE flag is set, it's because we're
}
c = STRING_CHAR_AND_LENGTH (base_pat, len_byte, in_charlen);
+
/* Translate the character, if requested. */
- translated = TRANSLATE (trt, c);
+ TRANSLATE (translated, trt, c);
/* If translation changed the byte-length, go back
to the original character. */
- charlen = CHAR_STRING (translated, workbuf, str);
+ charlen = CHAR_STRING (translated, str);
if (in_charlen != charlen)
{
translated = c;
- charlen = CHAR_STRING (c, workbuf, str);
+ charlen = CHAR_STRING (c, str);
}
+ /* If we are searching for something strange,
+ an invalid multibyte code, don't use boyer-moore. */
+ if (! ASCII_BYTE_P (translated)
+ && (charlen == 1 /* 8bit code */
+ || charlen != in_charlen /* invalid multibyte code */
+ ))
+ boyer_moore_ok = 0;
+
+ TRANSLATE (inverse, inverse_trt, c);
+
/* Did this char actually get translated?
Would any other char get translated into it? */
- if (translated != c
- || TRANSLATE (inverse_trt, c) != c)
+ if (translated != c || inverse != c)
{
/* Keep track of which character set row
contains the characters that need translation. */
- int charset_base_code = c & ~0xff;
- if (charset_base == -1)
+ int charset_base_code = c & ~CHAR_FIELD3_MASK;
+ int inverse_charset_base = inverse & ~CHAR_FIELD3_MASK;
+
+ if (charset_base_code != inverse_charset_base)
+ boyer_moore_ok = 0;
+ else if (charset_base == -1)
charset_base = charset_base_code;
else if (charset_base != charset_base_code)
/* If two different rows appear, needing translation,
then we cannot use boyer_moore search. */
- simple = 0;
- /* ??? Handa: this must do simple = 0
- if c is a composite character. */
- }
+ boyer_moore_ok = 0;
+ }
/* Store this character into the translated pattern. */
bcopy (str, pat, charlen);
}
else
{
+ /* Unibyte buffer. */
+ charset_base = 0;
while (--len >= 0)
{
int c, translated;
base_pat++;
}
c = *base_pat++;
- translated = TRANSLATE (trt, c);
-
- /* Did this char actually get translated?
- Would any other char get translated into it? */
- if (translated != c
- || TRANSLATE (inverse_trt, c) != c)
- {
- /* Keep track of which character set row
- contains the characters that need translation. */
- int charset_base_code = c & ~0xff;
- if (charset_base == -1)
- charset_base = charset_base_code;
- else if (charset_base != charset_base_code)
- /* If two different rows appear, needing translation,
- then we cannot use boyer_moore search. */
- simple = 0;
- }
+ TRANSLATE (translated, trt, c);
*pat++ = translated;
}
}
len = raw_pattern_size;
pat = base_pat = patbuf;
- if (simple)
+ if (boyer_moore_ok)
return boyer_moore (n, pat, len, len_byte, trt, inverse_trt,
- pos, pos_byte, lim, lim_byte);
+ pos, pos_byte, lim, lim_byte,
+ charset_base);
else
return simple_search (n, pat, len, len_byte, trt,
pos, pos_byte, lim, lim_byte);
int lim, lim_byte;
{
int multibyte = ! NILP (current_buffer->enable_multibyte_characters);
+ int forward = n > 0;
if (lim > pos && multibyte)
while (n > 0)
while (this_len > 0)
{
int charlen, buf_charlen;
- int pat_ch = STRING_CHAR_AND_LENGTH (p, this_len_byte, charlen);
- int buf_ch;
-
- this_len_byte -= charlen;
- this_len--;
- p += charlen;
+ int pat_ch, buf_ch;
+ pat_ch = STRING_CHAR_AND_LENGTH (p, this_len_byte, charlen);
buf_ch = STRING_CHAR_AND_LENGTH (BYTE_POS_ADDR (this_pos_byte),
ZV_BYTE - this_pos_byte,
buf_charlen);
- this_pos_byte += buf_charlen;
- this_pos++;
- buf_ch = TRANSLATE (trt, buf_ch);
+ TRANSLATE (buf_ch, trt, buf_ch);
if (buf_ch != pat_ch)
break;
+
+ this_len_byte -= charlen;
+ this_len--;
+ p += charlen;
+
+ this_pos_byte += buf_charlen;
+ this_pos++;
}
if (this_len == 0)
{
int pat_ch = *p++;
int buf_ch = FETCH_BYTE (this_pos);
- this_len--;
- this_pos++;
- buf_ch = TRANSLATE (trt, buf_ch);
+ TRANSLATE (buf_ch, trt, buf_ch);
if (buf_ch != pat_ch)
break;
+
+ this_len--;
+ this_pos++;
}
if (this_len == 0)
while (this_len > 0)
{
int charlen, buf_charlen;
- int pat_ch = STRING_CHAR_AND_LENGTH (p, this_len_byte, charlen);
- int buf_ch;
-
- this_len_byte -= charlen;
- this_len--;
- p += charlen;
+ int pat_ch, buf_ch;
+ pat_ch = STRING_CHAR_AND_LENGTH (p, this_len_byte, charlen);
buf_ch = STRING_CHAR_AND_LENGTH (BYTE_POS_ADDR (this_pos_byte),
ZV_BYTE - this_pos_byte,
buf_charlen);
- this_pos_byte += buf_charlen;
- this_pos++;
- buf_ch = TRANSLATE (trt, buf_ch);
+ TRANSLATE (buf_ch, trt, buf_ch);
if (buf_ch != pat_ch)
break;
+
+ this_len_byte -= charlen;
+ this_len--;
+ p += charlen;
+ this_pos_byte += buf_charlen;
+ this_pos++;
}
if (this_len == 0)
{
int pat_ch = *p++;
int buf_ch = FETCH_BYTE (this_pos);
- this_len--;
- this_pos++;
- buf_ch = TRANSLATE (trt, buf_ch);
+ TRANSLATE (buf_ch, trt, buf_ch);
if (buf_ch != pat_ch)
break;
+ this_len--;
+ this_pos++;
}
if (this_len == 0)
stop:
if (n == 0)
- return pos;
+ {
+ if (forward)
+ set_search_regs ((multibyte ? pos_byte : pos) - len_byte, len_byte);
+ else
+ set_search_regs (multibyte ? pos_byte : pos, len_byte);
+
+ return pos;
+ }
else if (n > 0)
return -n;
else
static int
boyer_moore (n, base_pat, len, len_byte, trt, inverse_trt,
- pos, pos_byte, lim, lim_byte)
+ pos, pos_byte, lim, lim_byte, charset_base)
int n;
unsigned char *base_pat;
int len, len_byte;
Lisp_Object inverse_trt;
int pos, pos_byte;
int lim, lim_byte;
+ int charset_base;
{
int direction = ((n > 0) ? 1 : -1);
register int dirlen;
- int infinity, limit, k, stride_for_teases;
+ int infinity, limit, stride_for_teases = 0;
register int *BM_tab;
int *BM_tab_base;
- register unsigned char *cursor, *p_limit;
+ register unsigned char *cursor, *p_limit;
register int i, j;
- unsigned char *pat;
+ unsigned char *pat, *pat_end;
int multibyte = ! NILP (current_buffer->enable_multibyte_characters);
unsigned char simple_translate[0400];
- int translate_prev_byte;
- int translate_anteprev_byte;
+ int translate_prev_byte = 0;
+ int translate_anteprev_byte = 0;
#ifdef C_ALLOCA
int BM_tab_space[0400];
/* a single test, a test for having gone past the end of the */
/* permissible match region, to test for both possible matches (when */
/* the stride goes past the end immediately) and failure to */
- /* match (where you get nudged past the end one stride at a time). */
+ /* match (where you get nudged past the end one stride at a time). */
/* Here we make a "mickey mouse" BM table. The stride of the search */
/* is determined only by the last character of the putative match. */
/* If that character does not match, we will stride the proper */
/* distance to propose a match that superimposes it on the last */
/* instance of a character that matches it (per trt), or misses */
- /* it entirely if there is none. */
+ /* it entirely if there is none. */
dirlen = len_byte * direction;
infinity = dirlen - (lim_byte + pos_byte + len_byte + len_byte) * direction;
+
+ /* Record position after the end of the pattern. */
+ pat_end = base_pat + len_byte;
+ /* BASE_PAT points to a character that we start scanning from.
+ It is the first character in a forward search,
+ the last character in a backward search. */
if (direction < 0)
- pat = (base_pat += len_byte - 1);
- else
- pat = base_pat;
+ base_pat = pat_end - 1;
+
BM_tab_base = BM_tab;
BM_tab += 0400;
j = dirlen; /* to get it in a register */
i = 0;
while (i != infinity)
{
- unsigned char *ptr = pat + i;
+ unsigned char *ptr = base_pat + i;
i += direction;
if (i == dirlen)
i = infinity;
if (! NILP (trt))
{
int ch;
+ int untranslated;
int this_translated = 1;
if (multibyte
- && (ptr + 1 == pat + len_byte || CHAR_HEAD_P (ptr[1])))
+ /* Is *PTR the last byte of a character? */
+ && (pat_end - ptr == 1 || CHAR_HEAD_P (ptr[1])))
{
unsigned char *charstart = ptr;
while (! CHAR_HEAD_P (*charstart))
charstart--;
- if (! CHAR_HEAD_P (*ptr))
+ untranslated = STRING_CHAR (charstart, ptr - charstart + 1);
+ if (charset_base == (untranslated & ~CHAR_FIELD3_MASK))
{
- translate_prev_byte = ptr[-1];
- if (! CHAR_HEAD_P (translate_prev_byte))
- translate_anteprev_byte = ptr[-2];
+ TRANSLATE (ch, trt, untranslated);
+ if (! CHAR_HEAD_P (*ptr))
+ {
+ translate_prev_byte = ptr[-1];
+ if (! CHAR_HEAD_P (translate_prev_byte))
+ translate_anteprev_byte = ptr[-2];
+ }
+ }
+ else
+ {
+ this_translated = 0;
+ ch = *ptr;
}
- ch = STRING_CHAR (charstart, ptr - charstart + 1);
- ch = TRANSLATE (trt, ch);
}
else if (!multibyte)
- ch = TRANSLATE (trt, *ptr);
+ TRANSLATE (ch, trt, *ptr);
else
{
ch = *ptr;
this_translated = 0;
}
- k = j = (unsigned char) ch;
+ if (ch > 0400)
+ j = ((unsigned char) ch) | 0200;
+ else
+ j = (unsigned char) ch;
+
if (i == infinity)
stride_for_teases = BM_tab[j];
+
BM_tab[j] = dirlen - i;
/* A translation table is accompanied by its inverse -- see */
- /* comment following downcase_table for details */
+ /* comment following downcase_table for details */
if (this_translated)
- while (1)
- {
- ch = TRANSLATE (inverse_trt, ch);
- /* For all the characters that map into K,
- set up simple_translate to map them into K. */
- simple_translate[(unsigned char) ch] = k;
- if ((unsigned char) ch == k)
- break;
- BM_tab[(unsigned char) ch] = dirlen - i;
- }
+ {
+ int starting_ch = ch;
+ int starting_j = j;
+ while (1)
+ {
+ TRANSLATE (ch, inverse_trt, ch);
+ if (ch > 0400)
+ j = ((unsigned char) ch) | 0200;
+ else
+ j = (unsigned char) ch;
+
+ /* For all the characters that map into CH,
+ set up simple_translate to map the last byte
+ into STARTING_J. */
+ simple_translate[j] = starting_j;
+ if (ch == starting_ch)
+ break;
+ BM_tab[j] = dirlen - i;
+ }
+ }
}
else
{
QUIT;
pat = base_pat;
limit = pos_byte - dirlen + direction;
- limit = ((direction > 0)
- ? BUFFER_CEILING_OF (limit)
- : BUFFER_FLOOR_OF (limit));
- /* LIMIT is now the last (not beyond-last!) value POS_BYTE
- can take on without hitting edge of buffer or the gap. */
- limit = ((direction > 0)
- ? min (lim_byte - 1, min (limit, pos_byte + 20000))
- : max (lim_byte, max (limit, pos_byte - 20000)));
+ if (direction > 0)
+ {
+ limit = BUFFER_CEILING_OF (limit);
+ /* LIMIT is now the last (not beyond-last!) value POS_BYTE
+ can take on without hitting edge of buffer or the gap. */
+ limit = min (limit, pos_byte + 20000);
+ limit = min (limit, lim_byte - 1);
+ }
+ else
+ {
+ limit = BUFFER_FLOOR_OF (limit);
+ /* LIMIT is now the last (not beyond-last!) value POS_BYTE
+ can take on without hitting edge of buffer or the gap. */
+ limit = max (limit, pos_byte - 20000);
+ limit = max (limit, lim_byte);
+ }
tail_end = BUFFER_CEILING_OF (pos_byte) + 1;
tail_end_ptr = BYTE_POS_ADDR (tail_end);
/* This loop can be coded for space rather than */
/* speed because it will usually run only once. */
/* (the reach is at most len + 21, and typically */
- /* does not exceed len) */
+ /* does not exceed len) */
while ((limit - pos_byte) * direction >= 0)
pos_byte += BM_tab[FETCH_BYTE (pos_byte)];
/* now run the same tests to distinguish going off the */
}
/* Record beginning BEG_BYTE and end BEG_BYTE + NBYTES
- for a match just found in the current buffer. */
+ for the overall match just found in the current buffer.
+ Also clear out the match data for registers 1 and up. */
static void
set_search_regs (beg_byte, nbytes)
int beg_byte, nbytes;
{
+ int i;
+
/* Make sure we have registers in which to store
the match position. */
if (search_regs.num_regs == 0)
search_regs.num_regs = 2;
}
+ /* Clear out the other registers. */
+ for (i = 1; i < search_regs.num_regs; i++)
+ {
+ search_regs.start[i] = -1;
+ search_regs.end[i] = -1;
+ }
+
search_regs.start[0] = BYTE_TO_CHAR (beg_byte);
search_regs.end[0] = BYTE_TO_CHAR (beg_byte + nbytes);
XSETBUFFER (last_thing_searched, current_buffer);
int prev_c = 0;
int adjust;
- CHECK_STRING (string, 0);
- p = XSTRING (string)->data;
- len = XSTRING (string)->size;
+ CHECK_STRING (string);
+ p = SDATA (string);
+ len = SCHARS (string);
for (i = 0, i_byte = 0; i < len; )
{
int c;
-
- if (STRING_MULTIBYTE (string))
- FETCH_STRING_CHAR_ADVANCE (c, string, i, i_byte);
- else
- c = XSTRING (string)->data[i++];
+
+ FETCH_STRING_CHAR_ADVANCE (c, string, i, i_byte);
if (SYNTAX (c) != Sword)
{
if (SYNTAX (prev_c) == Sword)
word_count++;
if (!word_count)
- return build_string ("");
+ return empty_string;
adjust = - punct_count + 5 * (word_count - 1) + 4;
- val = make_uninit_multibyte_string (len + adjust,
- XSTRING (string)->size_byte + adjust);
+ if (STRING_MULTIBYTE (string))
+ val = make_uninit_multibyte_string (len + adjust,
+ SBYTES (string)
+ + adjust);
+ else
+ val = make_uninit_string (len + adjust);
- o = XSTRING (val)->data;
+ o = SDATA (val);
*o++ = '\\';
*o++ = 'b';
+ prev_c = 0;
- for (i = 0; i < XSTRING (val)->size_byte; i++)
- if (SYNTAX (p[i]) == Sword)
- *o++ = p[i];
- else if (i > 0 && SYNTAX (p[i-1]) == Sword && --word_count)
- {
- *o++ = '\\';
- *o++ = 'W';
- *o++ = '\\';
- *o++ = 'W';
- *o++ = '*';
- }
+ for (i = 0, i_byte = 0; i < len; )
+ {
+ int c;
+ int i_byte_orig = i_byte;
+
+ FETCH_STRING_CHAR_ADVANCE (c, string, i, i_byte);
+
+ if (SYNTAX (c) == Sword)
+ {
+ bcopy (SDATA (string) + i_byte_orig, o,
+ i_byte - i_byte_orig);
+ o += i_byte - i_byte_orig;
+ }
+ else if (i > 0 && SYNTAX (prev_c) == Sword && --word_count)
+ {
+ *o++ = '\\';
+ *o++ = 'W';
+ *o++ = '\\';
+ *o++ = 'W';
+ *o++ = '*';
+ }
+
+ prev_c = c;
+ }
*o++ = '\\';
*o++ = 'b';
}
\f
DEFUN ("search-backward", Fsearch_backward, Ssearch_backward, 1, 4,
- "MSearch backward: ",
- "Search backward from point for STRING.\n\
-Set point to the beginning of the occurrence found, and return point.\n\
-An optional second argument bounds the search; it is a buffer position.\n\
-The match found must not extend before that position.\n\
-Optional third argument, if t, means if fail just return nil (no error).\n\
- If not nil and not t, position at limit of search and return nil.\n\
-Optional fourth argument is repeat count--search for successive occurrences.\n\
-See also the functions `match-beginning', `match-end' and `replace-match'.")
- (string, bound, noerror, count)
+ "MSearch backward: ",
+ doc: /* Search backward from point for STRING.
+Set point to the beginning of the occurrence found, and return point.
+An optional second argument bounds the search; it is a buffer position.
+The match found must not extend before that position.
+Optional third argument, if t, means if fail just return nil (no error).
+ If not nil and not t, position at limit of search and return nil.
+Optional fourth argument is repeat count--search for successive occurrences.
+
+Search case-sensitivity is determined by the value of the variable
+`case-fold-search', which see.
+
+See also the functions `match-beginning', `match-end' and `replace-match'. */)
+ (string, bound, noerror, count)
Lisp_Object string, bound, noerror, count;
{
return search_command (string, bound, noerror, count, -1, 0, 0);
}
DEFUN ("search-forward", Fsearch_forward, Ssearch_forward, 1, 4, "MSearch: ",
- "Search forward from point for STRING.\n\
-Set point to the end of the occurrence found, and return point.\n\
-An optional second argument bounds the search; it is a buffer position.\n\
-The match found must not extend after that position. nil is equivalent\n\
- to (point-max).\n\
-Optional third argument, if t, means if fail just return nil (no error).\n\
- If not nil and not t, move to limit of search and return nil.\n\
-Optional fourth argument is repeat count--search for successive occurrences.\n\
-See also the functions `match-beginning', `match-end' and `replace-match'.")
- (string, bound, noerror, count)
+ doc: /* Search forward from point for STRING.
+Set point to the end of the occurrence found, and return point.
+An optional second argument bounds the search; it is a buffer position.
+The match found must not extend after that position. nil is equivalent
+ to (point-max).
+Optional third argument, if t, means if fail just return nil (no error).
+ If not nil and not t, move to limit of search and return nil.
+Optional fourth argument is repeat count--search for successive occurrences.
+
+Search case-sensitivity is determined by the value of the variable
+`case-fold-search', which see.
+
+See also the functions `match-beginning', `match-end' and `replace-match'. */)
+ (string, bound, noerror, count)
Lisp_Object string, bound, noerror, count;
{
return search_command (string, bound, noerror, count, 1, 0, 0);
}
DEFUN ("word-search-backward", Fword_search_backward, Sword_search_backward, 1, 4,
- "sWord search backward: ",
- "Search backward from point for STRING, ignoring differences in punctuation.\n\
-Set point to the beginning of the occurrence found, and return point.\n\
-An optional second argument bounds the search; it is a buffer position.\n\
-The match found must not extend before that position.\n\
-Optional third argument, if t, means if fail just return nil (no error).\n\
- If not nil and not t, move to limit of search and return nil.\n\
-Optional fourth argument is repeat count--search for successive occurrences.")
- (string, bound, noerror, count)
+ "sWord search backward: ",
+ doc: /* Search backward from point for STRING, ignoring differences in punctuation.
+Set point to the beginning of the occurrence found, and return point.
+An optional second argument bounds the search; it is a buffer position.
+The match found must not extend before that position.
+Optional third argument, if t, means if fail just return nil (no error).
+ If not nil and not t, move to limit of search and return nil.
+Optional fourth argument is repeat count--search for successive occurrences. */)
+ (string, bound, noerror, count)
Lisp_Object string, bound, noerror, count;
{
return search_command (wordify (string), bound, noerror, count, -1, 1, 0);
}
DEFUN ("word-search-forward", Fword_search_forward, Sword_search_forward, 1, 4,
- "sWord search: ",
- "Search forward from point for STRING, ignoring differences in punctuation.\n\
-Set point to the end of the occurrence found, and return point.\n\
-An optional second argument bounds the search; it is a buffer position.\n\
-The match found must not extend after that position.\n\
-Optional third argument, if t, means if fail just return nil (no error).\n\
- If not nil and not t, move to limit of search and return nil.\n\
-Optional fourth argument is repeat count--search for successive occurrences.")
- (string, bound, noerror, count)
+ "sWord search: ",
+ doc: /* Search forward from point for STRING, ignoring differences in punctuation.
+Set point to the end of the occurrence found, and return point.
+An optional second argument bounds the search; it is a buffer position.
+The match found must not extend after that position.
+Optional third argument, if t, means if fail just return nil (no error).
+ If not nil and not t, move to limit of search and return nil.
+Optional fourth argument is repeat count--search for successive occurrences. */)
+ (string, bound, noerror, count)
Lisp_Object string, bound, noerror, count;
{
return search_command (wordify (string), bound, noerror, count, 1, 1, 0);
}
DEFUN ("re-search-backward", Fre_search_backward, Sre_search_backward, 1, 4,
- "sRE search backward: ",
- "Search backward from point for match for regular expression REGEXP.\n\
-Set point to the beginning of the match, and return point.\n\
-The match found is the one starting last in the buffer\n\
-and yet ending before the origin of the search.\n\
-An optional second argument bounds the search; it is a buffer position.\n\
-The match found must start at or after that position.\n\
-Optional third argument, if t, means if fail just return nil (no error).\n\
- If not nil and not t, move to limit of search and return nil.\n\
-Optional fourth argument is repeat count--search for successive occurrences.\n\
-See also the functions `match-beginning', `match-end' and `replace-match'.")
- (regexp, bound, noerror, count)
+ "sRE search backward: ",
+ doc: /* Search backward from point for match for regular expression REGEXP.
+Set point to the beginning of the match, and return point.
+The match found is the one starting last in the buffer
+and yet ending before the origin of the search.
+An optional second argument bounds the search; it is a buffer position.
+The match found must start at or after that position.
+Optional third argument, if t, means if fail just return nil (no error).
+ If not nil and not t, move to limit of search and return nil.
+Optional fourth argument is repeat count--search for successive occurrences.
+See also the functions `match-beginning', `match-end', `match-string',
+and `replace-match'. */)
+ (regexp, bound, noerror, count)
Lisp_Object regexp, bound, noerror, count;
{
return search_command (regexp, bound, noerror, count, -1, 1, 0);
}
DEFUN ("re-search-forward", Fre_search_forward, Sre_search_forward, 1, 4,
- "sRE search: ",
- "Search forward from point for regular expression REGEXP.\n\
-Set point to the end of the occurrence found, and return point.\n\
-An optional second argument bounds the search; it is a buffer position.\n\
-The match found must not extend after that position.\n\
-Optional third argument, if t, means if fail just return nil (no error).\n\
- If not nil and not t, move to limit of search and return nil.\n\
-Optional fourth argument is repeat count--search for successive occurrences.\n\
-See also the functions `match-beginning', `match-end' and `replace-match'.")
- (regexp, bound, noerror, count)
+ "sRE search: ",
+ doc: /* Search forward from point for regular expression REGEXP.
+Set point to the end of the occurrence found, and return point.
+An optional second argument bounds the search; it is a buffer position.
+The match found must not extend after that position.
+Optional third argument, if t, means if fail just return nil (no error).
+ If not nil and not t, move to limit of search and return nil.
+Optional fourth argument is repeat count--search for successive occurrences.
+See also the functions `match-beginning', `match-end', `match-string',
+and `replace-match'. */)
+ (regexp, bound, noerror, count)
Lisp_Object regexp, bound, noerror, count;
{
return search_command (regexp, bound, noerror, count, 1, 1, 0);
}
DEFUN ("posix-search-backward", Fposix_search_backward, Sposix_search_backward, 1, 4,
- "sPosix search backward: ",
- "Search backward from point for match for regular expression REGEXP.\n\
-Find the longest match in accord with Posix regular expression rules.\n\
-Set point to the beginning of the match, and return point.\n\
-The match found is the one starting last in the buffer\n\
-and yet ending before the origin of the search.\n\
-An optional second argument bounds the search; it is a buffer position.\n\
-The match found must start at or after that position.\n\
-Optional third argument, if t, means if fail just return nil (no error).\n\
- If not nil and not t, move to limit of search and return nil.\n\
-Optional fourth argument is repeat count--search for successive occurrences.\n\
-See also the functions `match-beginning', `match-end' and `replace-match'.")
- (regexp, bound, noerror, count)
+ "sPosix search backward: ",
+ doc: /* Search backward from point for match for regular expression REGEXP.
+Find the longest match in accord with Posix regular expression rules.
+Set point to the beginning of the match, and return point.
+The match found is the one starting last in the buffer
+and yet ending before the origin of the search.
+An optional second argument bounds the search; it is a buffer position.
+The match found must start at or after that position.
+Optional third argument, if t, means if fail just return nil (no error).
+ If not nil and not t, move to limit of search and return nil.
+Optional fourth argument is repeat count--search for successive occurrences.
+See also the functions `match-beginning', `match-end', `match-string',
+and `replace-match'. */)
+ (regexp, bound, noerror, count)
Lisp_Object regexp, bound, noerror, count;
{
return search_command (regexp, bound, noerror, count, -1, 1, 1);
}
DEFUN ("posix-search-forward", Fposix_search_forward, Sposix_search_forward, 1, 4,
- "sPosix search: ",
- "Search forward from point for regular expression REGEXP.\n\
-Find the longest match in accord with Posix regular expression rules.\n\
-Set point to the end of the occurrence found, and return point.\n\
-An optional second argument bounds the search; it is a buffer position.\n\
-The match found must not extend after that position.\n\
-Optional third argument, if t, means if fail just return nil (no error).\n\
- If not nil and not t, move to limit of search and return nil.\n\
-Optional fourth argument is repeat count--search for successive occurrences.\n\
-See also the functions `match-beginning', `match-end' and `replace-match'.")
- (regexp, bound, noerror, count)
+ "sPosix search: ",
+ doc: /* Search forward from point for regular expression REGEXP.
+Find the longest match in accord with Posix regular expression rules.
+Set point to the end of the occurrence found, and return point.
+An optional second argument bounds the search; it is a buffer position.
+The match found must not extend after that position.
+Optional third argument, if t, means if fail just return nil (no error).
+ If not nil and not t, move to limit of search and return nil.
+Optional fourth argument is repeat count--search for successive occurrences.
+See also the functions `match-beginning', `match-end', `match-string',
+and `replace-match'. */)
+ (regexp, bound, noerror, count)
Lisp_Object regexp, bound, noerror, count;
{
return search_command (regexp, bound, noerror, count, 1, 1, 1);
}
\f
DEFUN ("replace-match", Freplace_match, Sreplace_match, 1, 5, 0,
- "Replace text matched by last search with NEWTEXT.\n\
-If second arg FIXEDCASE is non-nil, do not alter case of replacement text.\n\
-Otherwise maybe capitalize the whole text, or maybe just word initials,\n\
-based on the replaced text.\n\
-If the replaced text has only capital letters\n\
-and has at least one multiletter word, convert NEWTEXT to all caps.\n\
-If the replaced text has at least one word starting with a capital letter,\n\
-then capitalize each word in NEWTEXT.\n\n\
-If third arg LITERAL is non-nil, insert NEWTEXT literally.\n\
-Otherwise treat `\\' as special:\n\
- `\\&' in NEWTEXT means substitute original matched text.\n\
- `\\N' means substitute what matched the Nth `\\(...\\)'.\n\
- If Nth parens didn't match, substitute nothing.\n\
- `\\\\' means insert one `\\'.\n\
-FIXEDCASE and LITERAL are optional arguments.\n\
-Leaves point at end of replacement text.\n\
-\n\
-The optional fourth argument STRING can be a string to modify.\n\
-In that case, this function creates and returns a new string\n\
-which is made by replacing the part of STRING that was matched.\n\
-\n\
-The optional fifth argument SUBEXP specifies a subexpression of the match.\n\
-It says to replace just that subexpression instead of the whole match.\n\
-This is useful only after a regular expression search or match\n\
-since only regular expressions have distinguished subexpressions.")
- (newtext, fixedcase, literal, string, subexp)
+ doc: /* Replace text matched by last search with NEWTEXT.
+Leave point at the end of the replacement text.
+
+If second arg FIXEDCASE is non-nil, do not alter case of replacement text.
+Otherwise maybe capitalize the whole text, or maybe just word initials,
+based on the replaced text.
+If the replaced text has only capital letters
+and has at least one multiletter word, convert NEWTEXT to all caps.
+Otherwise if all words are capitalized in the replaced text,
+capitalize each word in NEWTEXT.
+
+If third arg LITERAL is non-nil, insert NEWTEXT literally.
+Otherwise treat `\\' as special:
+ `\\&' in NEWTEXT means substitute original matched text.
+ `\\N' means substitute what matched the Nth `\\(...\\)'.
+ If Nth parens didn't match, substitute nothing.
+ `\\\\' means insert one `\\'.
+Case conversion does not apply to these substitutions.
+
+FIXEDCASE and LITERAL are optional arguments.
+
+The optional fourth argument STRING can be a string to modify.
+This is meaningful when the previous match was done against STRING,
+using `string-match'. When used this way, `replace-match'
+creates and returns a new string made by copying STRING and replacing
+the part of STRING that was matched.
+
+The optional fifth argument SUBEXP specifies a subexpression;
+it says to replace just that subexpression with NEWTEXT,
+rather than replacing the entire matched text.
+This is, in a vague sense, the inverse of using `\\N' in NEWTEXT;
+`\\N' copies subexp N into NEWTEXT, but using N as SUBEXP puts
+NEWTEXT in place of subexp N.
+This is useful only after a regular expression search or match,
+since only regular expressions have distinguished subexpressions. */)
+ (newtext, fixedcase, literal, string, subexp)
Lisp_Object newtext, fixedcase, literal, string, subexp;
{
enum { nochange, all_caps, cap_initial } case_action;
- register int pos, last;
+ register int pos, pos_byte;
int some_multiletter_word;
int some_lowercase;
int some_uppercase;
int some_nonuppercase_initial;
register int c, prevc;
- int inslen;
int sub;
int opoint, newpoint;
- CHECK_STRING (newtext, 0);
+ CHECK_STRING (newtext);
if (! NILP (string))
- CHECK_STRING (string, 4);
+ CHECK_STRING (string);
case_action = nochange; /* We tried an initialization */
/* but some C compilers blew it */
sub = 0;
else
{
- CHECK_NUMBER (subexp, 3);
+ CHECK_NUMBER (subexp);
sub = XINT (subexp);
if (sub < 0 || sub >= search_regs.num_regs)
args_out_of_range (subexp, make_number (search_regs.num_regs));
{
if (search_regs.start[sub] < 0
|| search_regs.start[sub] > search_regs.end[sub]
- || search_regs.end[sub] > XSTRING (string)->size)
+ || search_regs.end[sub] > SCHARS (string))
args_out_of_range (make_number (search_regs.start[sub]),
make_number (search_regs.end[sub]));
}
if (NILP (fixedcase))
{
- int beg;
/* Decide how to casify by examining the matched text. */
+ int last;
- if (NILP (string))
- last = CHAR_TO_BYTE (search_regs.end[sub]);
- else
- last = search_regs.end[sub];
+ pos = search_regs.start[sub];
+ last = search_regs.end[sub];
if (NILP (string))
- beg = CHAR_TO_BYTE (search_regs.start[sub]);
+ pos_byte = CHAR_TO_BYTE (pos);
else
- beg = search_regs.start[sub];
+ pos_byte = string_char_to_byte (string, pos);
prevc = '\n';
case_action = all_caps;
some_nonuppercase_initial = 0;
some_uppercase = 0;
- for (pos = beg; pos < last; pos++)
+ while (pos < last)
{
if (NILP (string))
- c = FETCH_BYTE (pos);
+ {
+ c = FETCH_CHAR (pos_byte);
+ INC_BOTH (pos, pos_byte);
+ }
else
- c = XSTRING (string)->data[pos];
+ FETCH_STRING_CHAR_ADVANCE (c, string, pos, pos_byte);
if (LOWERCASEP (c))
{
if desired. */
if (NILP (literal))
{
- int lastpos = -1;
- int lastpos_byte = -1;
+ int lastpos = 0;
+ int lastpos_byte = 0;
/* We build up the substituted string in ACCUM. */
Lisp_Object accum;
Lisp_Object middle;
- int pos_byte;
+ int length = SBYTES (newtext);
accum = Qnil;
- for (pos_byte = 0, pos = 0; pos_byte < XSTRING (newtext)->size_byte;)
+ for (pos_byte = 0, pos = 0; pos_byte < length;)
{
int substart = -1;
- int subend;
+ int subend = 0;
int delbackslash = 0;
FETCH_STRING_CHAR_ADVANCE (c, newtext, pos, pos_byte);
if (c == '\\')
{
FETCH_STRING_CHAR_ADVANCE (c, newtext, pos, pos_byte);
+
if (c == '&')
{
substart = search_regs.start[sub];
subend = search_regs.end[sub];
}
- else if (c >= '1' && c <= '9' && c <= search_regs.num_regs + '0')
+ else if (c >= '1' && c <= '9')
{
- if (search_regs.start[c - '0'] >= 0)
+ if (search_regs.start[c - '0'] >= 0
+ && c <= search_regs.num_regs + '0')
{
substart = search_regs.start[c - '0'];
subend = search_regs.end[c - '0'];
}
+ else
+ {
+ /* If that subexp did not match,
+ replace \\N with nothing. */
+ substart = 0;
+ subend = 0;
+ }
}
else if (c == '\\')
delbackslash = 1;
}
if (substart >= 0)
{
- if (pos - 1 != lastpos + 1)
- middle = substring_both (newtext, lastpos + 1,
- lastpos_byte + 1,
- pos - 1, pos_byte - 1);
+ if (pos - 2 != lastpos)
+ middle = substring_both (newtext, lastpos,
+ lastpos_byte,
+ pos - 2, pos_byte - 2);
else
middle = Qnil;
accum = concat3 (accum, middle,
}
else if (delbackslash)
{
- middle = substring_both (newtext, lastpos + 1,
- lastpos_byte + 1,
- pos, pos_byte);
+ middle = substring_both (newtext, lastpos,
+ lastpos_byte,
+ pos - 1, pos_byte - 1);
accum = concat2 (accum, middle);
lastpos = pos;
}
}
- if (pos != lastpos + 1)
- middle = substring_both (newtext, lastpos + 1,
- lastpos_byte + 1,
+ if (pos != lastpos)
+ middle = substring_both (newtext, lastpos,
+ lastpos_byte,
pos, pos_byte);
else
middle = Qnil;
return concat3 (before, newtext, after);
}
- /* Record point, the move (quietly) to the start of the match. */
- if (PT > search_regs.start[sub])
+ /* Record point, then move (quietly) to the start of the match. */
+ if (PT >= search_regs.end[sub])
opoint = PT - ZV;
+ else if (PT > search_regs.start[sub])
+ opoint = search_regs.end[sub] - ZV;
else
opoint = PT;
- TEMP_SET_PT (search_regs.start[sub]);
-
- /* We insert the replacement text before the old text, and then
- delete the original text. This means that markers at the
- beginning or end of the original will float to the corresponding
- position in the replacement. */
- if (!NILP (literal))
- Finsert_and_inherit (1, &newtext);
- else
+ /* If we want non-literal replacement,
+ perform substitution on the replacement string. */
+ if (NILP (literal))
{
- struct gcpro gcpro1;
- GCPRO1 (newtext);
-
- for (pos = 0; pos < XSTRING (newtext)->size; pos++)
+ int length = SBYTES (newtext);
+ unsigned char *substed;
+ int substed_alloc_size, substed_len;
+ int buf_multibyte = !NILP (current_buffer->enable_multibyte_characters);
+ int str_multibyte = STRING_MULTIBYTE (newtext);
+ Lisp_Object rev_tbl;
+ int really_changed = 0;
+
+ rev_tbl= (!buf_multibyte && CHAR_TABLE_P (Vnonascii_translation_table)
+ ? Fchar_table_extra_slot (Vnonascii_translation_table,
+ make_number (0))
+ : Qnil);
+
+ substed_alloc_size = length * 2 + 100;
+ substed = (unsigned char *) xmalloc (substed_alloc_size + 1);
+ substed_len = 0;
+
+ /* Go thru NEWTEXT, producing the actual text to insert in
+ SUBSTED while adjusting multibyteness to that of the current
+ buffer. */
+
+ for (pos_byte = 0, pos = 0; pos_byte < length;)
{
- int offset = PT - search_regs.start[sub];
+ unsigned char str[MAX_MULTIBYTE_LENGTH];
+ unsigned char *add_stuff = NULL;
+ int add_len = 0;
+ int idx = -1;
+
+ if (str_multibyte)
+ {
+ FETCH_STRING_CHAR_ADVANCE_NO_CHECK (c, newtext, pos, pos_byte);
+ if (!buf_multibyte)
+ c = multibyte_char_to_unibyte (c, rev_tbl);
+ }
+ else
+ {
+ /* Note that we don't have to increment POS. */
+ c = SREF (newtext, pos_byte++);
+ if (buf_multibyte)
+ c = unibyte_char_to_multibyte (c);
+ }
+
+ /* Either set ADD_STUFF and ADD_LEN to the text to put in SUBSTED,
+ or set IDX to a match index, which means put that part
+ of the buffer text into SUBSTED. */
- c = XSTRING (newtext)->data[pos];
if (c == '\\')
{
- c = XSTRING (newtext)->data[++pos];
+ really_changed = 1;
+
+ if (str_multibyte)
+ {
+ FETCH_STRING_CHAR_ADVANCE_NO_CHECK (c, newtext,
+ pos, pos_byte);
+ if (!buf_multibyte && !SINGLE_BYTE_CHAR_P (c))
+ c = multibyte_char_to_unibyte (c, rev_tbl);
+ }
+ else
+ {
+ c = SREF (newtext, pos_byte++);
+ if (buf_multibyte)
+ c = unibyte_char_to_multibyte (c);
+ }
+
if (c == '&')
- Finsert_buffer_substring
- (Fcurrent_buffer (),
- make_number (search_regs.start[sub] + offset),
- make_number (search_regs.end[sub] + offset));
+ idx = sub;
else if (c >= '1' && c <= '9' && c <= search_regs.num_regs + '0')
{
if (search_regs.start[c - '0'] >= 1)
- Finsert_buffer_substring
- (Fcurrent_buffer (),
- make_number (search_regs.start[c - '0'] + offset),
- make_number (search_regs.end[c - '0'] + offset));
+ idx = c - '0';
}
else if (c == '\\')
- insert_char (c);
+ add_len = 1, add_stuff = "\\";
else
- error ("Invalid use of `\\' in replacement text");
+ {
+ xfree (substed);
+ error ("Invalid use of `\\' in replacement text");
+ }
+ }
+ else
+ {
+ add_len = CHAR_STRING (c, str);
+ add_stuff = str;
+ }
+
+ /* If we want to copy part of a previous match,
+ set up ADD_STUFF and ADD_LEN to point to it. */
+ if (idx >= 0)
+ {
+ int begbyte = CHAR_TO_BYTE (search_regs.start[idx]);
+ add_len = CHAR_TO_BYTE (search_regs.end[idx]) - begbyte;
+ if (search_regs.start[idx] < GPT && GPT < search_regs.end[idx])
+ move_gap (search_regs.start[idx]);
+ add_stuff = BYTE_POS_ADDR (begbyte);
+ }
+
+ /* Now the stuff we want to add to SUBSTED
+ is invariably ADD_LEN bytes starting at ADD_STUFF. */
+
+ /* Make sure SUBSTED is big enough. */
+ if (substed_len + add_len >= substed_alloc_size)
+ {
+ substed_alloc_size = substed_len + add_len + 500;
+ substed = (unsigned char *) xrealloc (substed,
+ substed_alloc_size + 1);
+ }
+
+ /* Now add to the end of SUBSTED. */
+ if (add_stuff)
+ {
+ bcopy (add_stuff, substed + substed_len, add_len);
+ substed_len += add_len;
+ }
+ }
+
+ if (really_changed)
+ {
+ if (buf_multibyte)
+ {
+ int nchars = multibyte_chars_in_text (substed, substed_len);
+
+ newtext = make_multibyte_string (substed, nchars, substed_len);
}
else
- insert_char (c);
+ newtext = make_unibyte_string (substed, substed_len);
}
- UNGCPRO;
+ xfree (substed);
}
- inslen = PT - (search_regs.start[sub]);
- del_range (search_regs.start[sub] + inslen, search_regs.end[sub] + inslen);
+ /* Replace the old text with the new in the cleanest possible way. */
+ replace_range (search_regs.start[sub], search_regs.end[sub],
+ newtext, 1, 0, 1);
+ newpoint = search_regs.start[sub] + SCHARS (newtext);
if (case_action == all_caps)
- Fupcase_region (make_number (PT - inslen), make_number (PT));
+ Fupcase_region (make_number (search_regs.start[sub]),
+ make_number (newpoint));
else if (case_action == cap_initial)
- Fupcase_initials_region (make_number (PT - inslen), make_number (PT));
+ Fupcase_initials_region (make_number (search_regs.start[sub]),
+ make_number (newpoint));
- newpoint = PT;
+ /* Adjust search data for this change. */
+ {
+ int oldend = search_regs.end[sub];
+ int change = newpoint - search_regs.end[sub];
+ int i;
+
+ for (i = 0; i < search_regs.num_regs; i++)
+ {
+ if (search_regs.start[i] > oldend)
+ search_regs.start[i] += change;
+ if (search_regs.end[i] > oldend)
+ search_regs.end[i] += change;
+ }
+ }
/* Put point back where it was in the text. */
if (opoint <= 0)
/* Now move point "officially" to the start of the inserted replacement. */
move_if_not_intangible (newpoint);
-
+
return Qnil;
}
\f
{
register int n;
- CHECK_NUMBER (num, 0);
+ CHECK_NUMBER (num);
n = XINT (num);
if (n < 0 || n >= search_regs.num_regs)
args_out_of_range (num, make_number (search_regs.num_regs));
}
DEFUN ("match-beginning", Fmatch_beginning, Smatch_beginning, 1, 1, 0,
- "Return position of start of text matched by last search.\n\
-SUBEXP, a number, specifies which parenthesized expression in the last\n\
- regexp.\n\
-Value is nil if SUBEXPth pair didn't match, or there were less than\n\
- SUBEXP pairs.\n\
-Zero means the entire text matched by the whole regexp or whole string.")
- (subexp)
+ doc: /* Return position of start of text matched by last search.
+SUBEXP, a number, specifies which parenthesized expression in the last
+ regexp.
+Value is nil if SUBEXPth pair didn't match, or there were less than
+ SUBEXP pairs.
+Zero means the entire text matched by the whole regexp or whole string. */)
+ (subexp)
Lisp_Object subexp;
{
return match_limit (subexp, 1);
}
DEFUN ("match-end", Fmatch_end, Smatch_end, 1, 1, 0,
- "Return position of end of text matched by last search.\n\
-SUBEXP, a number, specifies which parenthesized expression in the last\n\
- regexp.\n\
-Value is nil if SUBEXPth pair didn't match, or there were less than\n\
- SUBEXP pairs.\n\
-Zero means the entire text matched by the whole regexp or whole string.")
- (subexp)
+ doc: /* Return position of end of text matched by last search.
+SUBEXP, a number, specifies which parenthesized expression in the last
+ regexp.
+Value is nil if SUBEXPth pair didn't match, or there were less than
+ SUBEXP pairs.
+Zero means the entire text matched by the whole regexp or whole string. */)
+ (subexp)
Lisp_Object subexp;
{
return match_limit (subexp, 0);
-}
+}
DEFUN ("match-data", Fmatch_data, Smatch_data, 0, 2, 0,
- "Return a list containing all info on what the last search matched.\n\
-Element 2N is `(match-beginning N)'; element 2N + 1 is `(match-end N)'.\n\
-All the elements are markers or nil (nil if the Nth pair didn't match)\n\
-if the last match was on a buffer; integers or nil if a string was matched.\n\
-Use `store-match-data' to reinstate the data in this list.\n\
-\n\
-If INTEGERS (the optional first argument) is non-nil, always use integers\n\
-\(rather than markers) to represent buffer positions.\n\
-If REUSE is a list, reuse it as part of the value. If REUSE is long enough\n\
-to hold all the values, and if INTEGERS is non-nil, no consing is done.")
- (integers, reuse)
+ doc: /* Return a list containing all info on what the last search matched.
+Element 2N is `(match-beginning N)'; element 2N + 1 is `(match-end N)'.
+All the elements are markers or nil (nil if the Nth pair didn't match)
+if the last match was on a buffer; integers or nil if a string was matched.
+Use `store-match-data' to reinstate the data in this list.
+
+If INTEGERS (the optional first argument) is non-nil, always use integers
+\(rather than markers) to represent buffer positions.
+If REUSE is a list, reuse it as part of the value. If REUSE is long enough
+to hold all the values, and if INTEGERS is non-nil, no consing is done.
+
+Return value is undefined if the last search failed. */)
+ (integers, reuse)
Lisp_Object integers, reuse;
{
Lisp_Object tail, prev;
if (NILP (last_thing_searched))
return Qnil;
+ prev = Qnil;
+
data = (Lisp_Object *) alloca ((2 * search_regs.num_regs)
* sizeof (Lisp_Object));
last_thing_searched);
data[2 * i + 1] = Fmake_marker ();
Fset_marker (data[2 * i + 1],
- make_number (search_regs.end[i]),
+ make_number (search_regs.end[i]),
last_thing_searched);
}
else
/* If REUSE is a list, store as many value elements as will fit
into the elements of REUSE. */
for (i = 0, tail = reuse; CONSP (tail);
- i++, tail = XCONS (tail)->cdr)
+ i++, tail = XCDR (tail))
{
if (i < 2 * len + 2)
- XCONS (tail)->car = data[i];
+ XSETCAR (tail, data[i]);
else
- XCONS (tail)->car = Qnil;
+ XSETCAR (tail, Qnil);
prev = tail;
}
/* If we couldn't fit all value elements into REUSE,
cons up the rest of them and add them to the end of REUSE. */
if (i < 2 * len + 2)
- XCONS (prev)->cdr = Flist (2 * len + 2 - i, data + i);
+ XSETCDR (prev, Flist (2 * len + 2 - i, data + i));
return reuse;
}
-DEFUN ("store-match-data", Fstore_match_data, Sstore_match_data, 1, 1, 0,
- "Set internal data on last search match from elements of LIST.\n\
-LIST should have been created by calling `match-data' previously.")
- (list)
+DEFUN ("set-match-data", Fset_match_data, Sset_match_data, 1, 1, 0,
+ doc: /* Set internal data on last search match from elements of LIST.
+LIST should have been created by calling `match-data' previously. */)
+ (list)
register Lisp_Object list;
{
register int i;
if (!CONSP (list) && !NILP (list))
list = wrong_type_argument (Qconsp, list);
- /* Unless we find a marker with a buffer in LIST, assume that this
+ /* Unless we find a marker with a buffer in LIST, assume that this
match data came from a string. */
last_thing_searched = Qt;
length * sizeof (regoff_t));
}
+ for (i = search_regs.num_regs; i < length; i++)
+ search_regs.start[i] = -1;
+
search_regs.num_regs = length;
}
}
}
else
{
+ int from;
+
if (MARKERP (marker))
{
if (XMARKER (marker)->buffer == 0)
XSETBUFFER (last_thing_searched, XMARKER (marker)->buffer);
}
- CHECK_NUMBER_COERCE_MARKER (marker, 0);
- search_regs.start[i] = XINT (marker);
+ CHECK_NUMBER_COERCE_MARKER (marker);
+ from = XINT (marker);
list = Fcdr (list);
marker = Fcar (list);
if (MARKERP (marker) && XMARKER (marker)->buffer == 0)
XSETFASTINT (marker, 0);
- CHECK_NUMBER_COERCE_MARKER (marker, 0);
+ CHECK_NUMBER_COERCE_MARKER (marker);
+ search_regs.start[i] = from;
search_regs.end[i] = XINT (marker);
}
list = Fcdr (list);
}
- return Qnil;
+ return Qnil;
}
/* If non-zero the match data have been saved in saved_search_regs
/* Quote a string to inactivate reg-expr chars */
DEFUN ("regexp-quote", Fregexp_quote, Sregexp_quote, 1, 1, 0,
- "Return a regexp string which matches exactly STRING and nothing else.")
- (string)
+ doc: /* Return a regexp string which matches exactly STRING and nothing else. */)
+ (string)
Lisp_Object string;
{
register unsigned char *in, *out, *end;
register unsigned char *temp;
int backslashes_added = 0;
- CHECK_STRING (string, 0);
+ CHECK_STRING (string);
- temp = (unsigned char *) alloca (XSTRING (string)->size_byte * 2);
+ temp = (unsigned char *) alloca (SBYTES (string) * 2);
/* Now copy the data into the new string, inserting escapes. */
- in = XSTRING (string)->data;
- end = in + XSTRING (string)->size_byte;
- out = temp;
+ in = SDATA (string);
+ end = in + SBYTES (string);
+ out = temp;
for (; in != end; in++)
{
*out++ = *in;
}
- return make_multibyte_string (temp,
- XSTRING (string)->size + backslashes_added,
- out - temp);
+ return make_specified_string (temp,
+ SCHARS (string) + backslashes_added,
+ out - temp,
+ STRING_MULTIBYTE (string));
}
-\f
+\f
+void
syms_of_search ()
{
register int i;
for (i = 0; i < REGEXP_CACHE_SIZE; ++i)
{
searchbufs[i].buf.allocated = 100;
- searchbufs[i].buf.buffer = (unsigned char *) malloc (100);
+ searchbufs[i].buf.buffer = (unsigned char *) xmalloc (100);
searchbufs[i].buf.fastmap = searchbufs[i].fastmap;
searchbufs[i].regexp = Qnil;
staticpro (&searchbufs[i].regexp);
defsubr (&Smatch_beginning);
defsubr (&Smatch_end);
defsubr (&Smatch_data);
- defsubr (&Sstore_match_data);
+ defsubr (&Sset_match_data);
defsubr (&Sregexp_quote);
}
+
+/* arch-tag: a6059d79-0552-4f14-a2cb-d379a4e3c78f
+ (do not change this comment) */
HCoop / bpt / emacs.git / blobdiff