#include <config.h>
#include "lisp.h"
#include "syntax.h"
+#include "category.h"
#include "buffer.h"
+#include "charset.h"
#include "region-cache.h"
#include "commands.h"
#include "blockinput.h"
If it is 0, we should compile the pattern not to record any
subexpression bounds.
POSIX is nonzero if we want full backtracking (POSIX style)
- for this pattern. 0 means backtrack only enough to get a valid match. */
+ for this pattern. 0 means backtrack only enough to get a valid match.
+ MULTIBYTE is nonzero if we want to handle multibyte characters in
+ PATTERN. 0 means all multibyte characters are recognized just as
+ sequences of binary data. */
static void
-compile_pattern_1 (cp, pattern, translate, regp, posix)
+compile_pattern_1 (cp, pattern, translate, regp, posix, multibyte)
struct regexp_cache *cp;
Lisp_Object pattern;
Lisp_Object *translate;
struct re_registers *regp;
int posix;
+ int multibyte;
{
CONST char *val;
reg_syntax_t old;
cp->regexp = Qnil;
cp->buf.translate = translate;
cp->posix = posix;
+ cp->buf.multibyte = multibyte;
BLOCK_INPUT;
old = re_set_syntax (RE_SYNTAX_EMACS
| (posix ? 0 : RE_NO_POSIX_BACKTRACKING));
int posix;
{
struct regexp_cache *cp, **cpp;
+ /* Should we check it here, or add an argument `multibyte' to this
+ function? */
+ int multibyte = !NILP (current_buffer->enable_multibyte_characters);
for (cpp = &searchbuf_head; ; cpp = &cp->next)
{
if (XSTRING (cp->regexp)->size == XSTRING (pattern)->size
&& !NILP (Fstring_equal (cp->regexp, pattern))
&& cp->buf.translate == translate
- && cp->posix == posix)
+ && cp->posix == posix
+ && cp->buf.multibyte == multibyte)
break;
/* If we're at the end of the cache, compile into the last cell. */
if (cp->next == 0)
{
- compile_pattern_1 (cp, pattern, translate, regp, posix);
+ compile_pattern_1 (cp, pattern, translate, regp, posix, multibyte);
break;
}
}
s1 = ZV - BEGV;
s2 = 0;
}
+
+ re_match_object = Qnil;
i = re_match_2 (bufp, (char *) p1, s1, (char *) p2, s2,
PT - BEGV, &search_regs,
? DOWNCASE_TABLE : 0),
posix);
immediate_quit = 1;
+ re_match_object = string;
+
val = re_search (bufp, (char *) XSTRING (string)->data,
XSTRING (string)->size, s, XSTRING (string)->size - s,
&search_regs);
bufp = compile_pattern (regexp, 0, 0, 0);
immediate_quit = 1;
+ re_match_object = string;
+
val = re_search (bufp, (char *) XSTRING (string)->data,
XSTRING (string)->size, 0, XSTRING (string)->size,
0);
immediate_quit = 0;
return val;
}
+
+/* Match REGEXP against STRING, searching all of STRING ignoring case,
+ and return the index of the match, or negative on failure.
+ This does not clobber the match data. */
+
+extern Lisp_Object Vascii_downcase_table;
+
+int
+fast_string_match_ignore_case (regexp, string)
+ Lisp_Object regexp;
+ char *string;
+{
+ int val;
+ struct re_pattern_buffer *bufp;
+ int len = strlen (string);
+
+ bufp = compile_pattern (regexp, 0,
+ XCHAR_TABLE (Vascii_downcase_table)->contents, 0);
+ immediate_quit = 1;
+ val = re_search (bufp, string, len, 0, len, 0);
+ immediate_quit = 0;
+ return val;
+}
\f
/* max and min. */
{
/* The termination address of the dumb loop. */
- register unsigned char *ceiling_addr = &FETCH_CHAR (ceiling) + 1;
- register unsigned char *cursor = &FETCH_CHAR (start);
+ register unsigned char *ceiling_addr = POS_ADDR (ceiling) + 1;
+ register unsigned char *cursor = POS_ADDR (start);
unsigned char *base = cursor;
while (cursor < ceiling_addr)
{
/* The termination address of the dumb loop. */
- register unsigned char *ceiling_addr = &FETCH_CHAR (ceiling);
- register unsigned char *cursor = &FETCH_CHAR (start - 1);
+ register unsigned char *ceiling_addr = POS_ADDR (ceiling);
+ register unsigned char *cursor = POS_ADDR (start - 1);
unsigned char *base = cursor;
while (cursor >= ceiling_addr)
return pos;
}
\f
-Lisp_Object skip_chars ();
-
-DEFUN ("skip-chars-forward", Fskip_chars_forward, Sskip_chars_forward, 1, 2, 0,
- "Move point forward, stopping before a char not in STRING, or at pos LIM.\n\
-STRING is like the inside of a `[...]' in a regular expression\n\
-except that `]' is never special and `\\' quotes `^', `-' or `\\'.\n\
-Thus, with arg \"a-zA-Z\", this skips letters stopping before first nonletter.\n\
-With arg \"^a-zA-Z\", skips nonletters stopping before first letter.\n\
-Returns the distance traveled, either zero or positive.")
- (string, lim)
- Lisp_Object string, lim;
-{
- return skip_chars (1, 0, string, lim);
-}
-
-DEFUN ("skip-chars-backward", Fskip_chars_backward, Sskip_chars_backward, 1, 2, 0,
- "Move point backward, stopping after a char not in STRING, or at pos LIM.\n\
-See `skip-chars-forward' for details.\n\
-Returns the distance traveled, either zero or negative.")
- (string, lim)
- Lisp_Object string, lim;
-{
- return skip_chars (0, 0, string, lim);
-}
-
-DEFUN ("skip-syntax-forward", Fskip_syntax_forward, Sskip_syntax_forward, 1, 2, 0,
- "Move point forward across chars in specified syntax classes.\n\
-SYNTAX is a string of syntax code characters.\n\
-Stop before a char whose syntax is not in SYNTAX, or at position LIM.\n\
-If SYNTAX starts with ^, skip characters whose syntax is NOT in SYNTAX.\n\
-This function returns the distance traveled, either zero or positive.")
- (syntax, lim)
- Lisp_Object syntax, lim;
-{
- return skip_chars (1, 1, syntax, lim);
-}
-
-DEFUN ("skip-syntax-backward", Fskip_syntax_backward, Sskip_syntax_backward, 1, 2, 0,
- "Move point backward across chars in specified syntax classes.\n\
-SYNTAX is a string of syntax code characters.\n\
-Stop on reaching a char whose syntax is not in SYNTAX, or at position LIM.\n\
-If SYNTAX starts with ^, skip characters whose syntax is NOT in SYNTAX.\n\
-This function returns the distance traveled, either zero or negative.")
- (syntax, lim)
- Lisp_Object syntax, lim;
-{
- return skip_chars (0, 1, syntax, lim);
-}
-
-Lisp_Object
-skip_chars (forwardp, syntaxp, string, lim)
- int forwardp, syntaxp;
- Lisp_Object string, lim;
-{
- register unsigned char *p, *pend;
- register unsigned char c;
- unsigned char fastmap[0400];
- int negate = 0;
- register int i;
-
- CHECK_STRING (string, 0);
-
- if (NILP (lim))
- XSETINT (lim, forwardp ? ZV : BEGV);
- else
- CHECK_NUMBER_COERCE_MARKER (lim, 1);
-
- /* In any case, don't allow scan outside bounds of buffer. */
- /* jla turned this off, for no known reason.
- bfox turned the ZV part on, and rms turned the
- BEGV part back on. */
- if (XINT (lim) > ZV)
- XSETFASTINT (lim, ZV);
- if (XINT (lim) < BEGV)
- XSETFASTINT (lim, BEGV);
-
- p = XSTRING (string)->data;
- pend = p + XSTRING (string)->size;
- bzero (fastmap, sizeof fastmap);
-
- if (p != pend && *p == '^')
- {
- negate = 1; p++;
- }
-
- /* Find the characters specified and set their elements of fastmap.
- If syntaxp, each character counts as itself.
- Otherwise, handle backslashes and ranges specially */
-
- while (p != pend)
- {
- c = *p++;
- if (syntaxp)
- fastmap[c] = 1;
- else
- {
- if (c == '\\')
- {
- if (p == pend) break;
- c = *p++;
- }
- if (p != pend && *p == '-')
- {
- p++;
- if (p == pend) break;
- while (c <= *p)
- {
- fastmap[c] = 1;
- c++;
- }
- p++;
- }
- else
- fastmap[c] = 1;
- }
- }
-
- if (syntaxp && fastmap['-'] != 0)
- fastmap[' '] = 1;
-
- /* If ^ was the first character, complement the fastmap. */
-
- if (negate)
- for (i = 0; i < sizeof fastmap; i++)
- fastmap[i] ^= 1;
-
- {
- int start_point = PT;
-
- immediate_quit = 1;
- if (syntaxp)
- {
-
- if (forwardp)
- {
- while (PT < XINT (lim)
- && fastmap[(unsigned char) syntax_code_spec[(int) SYNTAX (FETCH_CHAR (PT))]])
- SET_PT (PT + 1);
- }
- else
- {
- while (PT > XINT (lim)
- && fastmap[(unsigned char) syntax_code_spec[(int) SYNTAX (FETCH_CHAR (PT - 1))]])
- SET_PT (PT - 1);
- }
- }
- else
- {
- if (forwardp)
- {
- while (PT < XINT (lim) && fastmap[FETCH_CHAR (PT)])
- SET_PT (PT + 1);
- }
- else
- {
- while (PT > XINT (lim) && fastmap[FETCH_CHAR (PT - 1)])
- SET_PT (PT - 1);
- }
- }
- immediate_quit = 0;
-
- return make_number (PT - start_point);
- }
-}
-\f
/* Subroutines of Lisp buffer search functions. */
static Lisp_Object
case '|': case '(': case ')': case '`': case '\'': case 'b':
case 'B': case '<': case '>': case 'w': case 'W': case 's':
case 'S': 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':
return 0;
s1 = ZV - BEGV;
s2 = 0;
}
+ re_match_object = Qnil;
+
while (n < 0)
{
int val;
: max (lim, max (limit, pos - 20000)));
if ((limit - pos) * direction > 20)
{
- p_limit = &FETCH_CHAR (limit);
- p2 = (cursor = &FETCH_CHAR (pos));
+ p_limit = POS_ADDR (limit);
+ p2 = (cursor = POS_ADDR (pos));
/* In this loop, pos + cursor - p2 is the surrogate for pos */
while (1) /* use one cursor setting as long as i can */
{
/* (the reach is at most len + 21, and typically */
/* does not exceed len) */
while ((limit - pos) * direction >= 0)
- pos += BM_tab[FETCH_CHAR(pos)];
+ pos += BM_tab[FETCH_BYTE (pos)];
/* now run the same tests to distinguish going off the */
/* end, a match or a phony match. */
if ((pos - limit) * direction <= len)
{
pos -= direction;
if (pat[i] != (trt != 0
- ? trt[FETCH_CHAR(pos)]
- : FETCH_CHAR (pos)))
+ ? trt[FETCH_BYTE (pos)]
+ : FETCH_BYTE (pos)))
break;
}
/* Above loop has moved POS part or all the way
for (pos = search_regs.start[sub]; pos < last; pos++)
{
if (NILP (string))
- c = FETCH_CHAR (pos);
+ c = FETCH_BYTE (pos);
else
c = XSTRING (string)->data[pos];
make_number (search_regs.start[sub]));
after = Fsubstring (string, make_number (search_regs.end[sub]), Qnil);
- /* Do case substitution into NEWTEXT if desired. */
+ /* Substitute parts of the match into NEWTEXT
+ if desired. */
if (NILP (literal))
{
int lastpos = -1;
}
else if (c == '\\')
delbackslash = 1;
+ else
+ error ("Invalid use of `\\' in replacement text");
}
if (substart >= 0)
{
newtext = concat2 (accum, middle);
}
+ /* Do case substitution in NEWTEXT if desired. */
if (case_action == all_caps)
newtext = Fupcase (newtext);
else if (case_action == cap_initial)
make_number (search_regs.start[c - '0'] + offset),
make_number (search_regs.end[c - '0'] + offset));
}
- else
+ else if (c == '\\')
insert_char (c);
+ else
+ error ("Invalid use of `\\' in replacement text");
}
else
insert_char (c);
defsubr (&Sposix_looking_at);
defsubr (&Sstring_match);
defsubr (&Sposix_string_match);
- defsubr (&Sskip_chars_forward);
- defsubr (&Sskip_chars_backward);
- defsubr (&Sskip_syntax_forward);
- defsubr (&Sskip_syntax_backward);
defsubr (&Ssearch_forward);
defsubr (&Ssearch_backward);
defsubr (&Sword_search_forward);
HCoop / bpt / emacs.git / blobdiff