/* GNU Emacs routines to deal with syntax tables; also word and list parsing.
- Copyright (C) 1985, 87, 93, 94, 95, 97, 1998 Free Software Foundation, Inc.
+ Copyright (C) 1985, 87, 93, 94, 95, 97, 1998, 1999 Free Software Foundation, Inc.
This file is part of GNU Emacs.
int words_include_escapes;
int parse_sexp_lookup_properties;
+/* Nonzero means `scan-sexps' treat all multibyte characters as symbol. */
+int multibyte_syntax_as_symbol;
+
/* Used as a temporary in SYNTAX_ENTRY and other macros in syntax.h,
if not compiled with GCC. No need to mark it, since it is used
only very temporarily. */
Lisp_Object syntax_temp;
+/* Non-zero means an open parenthesis in column 0 is always considered
+ to be the start of a defun. Zero means an open parenthesis in
+ column 0 has no special meaning. */
+
+int open_paren_in_column_0_is_defun_start;
+
/* This is the internal form of the parse state used in parse-partial-sexp. */
struct lisp_parse_state
{
int depth; /* Depth at end of parsing. */
int instring; /* -1 if not within string, else desired terminator. */
- int incomment; /* Nonzero if within a comment at end of parsing. */
+ int incomment; /* -1 if in unnestable comment else comment nesting */
int comstyle; /* comment style a=0, or b=1, or ST_COMMENT_STYLE. */
int quoted; /* Nonzero if just after an escape char at end of parsing */
int thislevelstart; /* Char number of most recent start-of-expression at current level */
static int find_defun_start P_ ((int, int));
-static int back_comment P_ ((int, int, int, int, int *, int *));
+static int back_comment P_ ((int, int, int, int, int, int *, int *));
static int char_quoted P_ ((int, int));
static Lisp_Object skip_chars P_ ((int, int, Lisp_Object, Lisp_Object));
static Lisp_Object scan_lists P_ ((int, int, int, int));
Lisp_Object object;
{
Lisp_Object tmp_table;
- int cnt = 0, doing_extra = 0, invalidate = 1;
+ int cnt = 0, invalidate = 1;
INTERVAL i, oldi;
if (init)
while (!NULL_PARENT (i))
{
if (AM_RIGHT_CHILD (i))
- i->parent->position = i->position
+ INTERVAL_PARENT (i)->position = i->position
- LEFT_TOTAL_LENGTH (i) + TOTAL_LENGTH (i) /* right end */
- - TOTAL_LENGTH (i->parent)
- + LEFT_TOTAL_LENGTH (i->parent);
+ - TOTAL_LENGTH (INTERVAL_PARENT (i))
+ + LEFT_TOTAL_LENGTH (INTERVAL_PARENT (i));
else
- i->parent->position = i->position - LEFT_TOTAL_LENGTH (i)
+ INTERVAL_PARENT (i)->position = i->position - LEFT_TOTAL_LENGTH (i)
+ TOTAL_LENGTH (i);
- i = i->parent;
+ i = INTERVAL_PARENT (i);
}
i = gl_state.forward_i;
gl_state.b_property = i->position - 1 - gl_state.offset;
find_defun_start (pos, pos_byte)
int pos, pos_byte;
{
- int tem;
- int shortage;
int opoint = PT, opoint_byte = PT_BYTE;
/* Use previous finding, if it's valid and applies to this inquiry. */
gl_state.use_global = 0;
while (PT > BEGV)
{
- /* Open-paren at start of line means we found our defun-start. */
+ /* Open-paren at start of line means we may have found our
+ defun-start. */
if (SYNTAX (FETCH_CHAR (PT_BYTE)) == Sopen)
{
SETUP_SYNTAX_TABLE (PT + 1, -1); /* Try again... */
- if (SYNTAX (FETCH_CHAR (PT_BYTE)) == Sopen)
+ if (SYNTAX (FETCH_CHAR (PT_BYTE)) == Sopen
+ && open_paren_in_column_0_is_defun_start)
break;
/* Now fallback to the default value. */
gl_state.current_syntax_table = current_buffer->syntax_table;
the returned value (or at FROM, if the search was not successful). */
static int
-back_comment (from, from_byte, stop, comstyle, charpos_ptr, bytepos_ptr)
+back_comment (from, from_byte, stop, comnested, comstyle, charpos_ptr, bytepos_ptr)
int from, from_byte, stop;
- int comstyle;
+ int comnested, comstyle;
int *charpos_ptr, *bytepos_ptr;
{
/* Look back, counting the parity of string-quotes,
int defun_start = 0;
int defun_start_byte = 0;
register enum syntaxcode code;
+ int nesting = 1; /* current comment nesting */
int c;
/* At beginning of range to scan, we're outside of strings;
that determines quote parity to the comment-end. */
while (from != stop)
{
- int temp_byte, prev_comend_second;
+ int temp_byte;
/* Move back and examine a character. */
DEC_BOTH (from, from_byte);
code = Scomment;
UPDATE_SYNTAX_TABLE_BACKWARD (from);
}
+ else if (code == Scomment && comstyle != SYNTAX_COMMENT_STYLE (c))
+ /* Ignore comment starters of a different style. */
+ continue;
/* Ignore escaped characters, except comment-enders. */
if (code != Sendcomment && char_quoted (from, from_byte))
string_lossage = 1;
}
- /* Record comment-starters according to that
- quote-parity to the comment-end. */
if (code == Scomment)
+ /* We've already checked that it is the relevant comstyle. */
{
+ if (comnested && --nesting <= 0 && parity == 0 && !string_lossage)
+ /* nested comments have to be balanced, so we don't need to
+ keep looking for earlier ones. We use here the same (slightly
+ incorrect) reasoning as below: since it is followed by uniform
+ paired string quotes, this comment-start has to be outside of
+ strings, else the comment-end itself would be inside a string. */
+ goto done;
+
+ /* Record comment-starters according to that
+ quote-parity to the comment-end. */
comstart_parity = parity;
comstart_pos = from;
comstart_byte = from_byte;
(because they go with the earlier comment-ender). */
if (code == Sendcomment
&& SYNTAX_COMMENT_STYLE (FETCH_CHAR (from_byte)) == comstyle)
- break;
+ {
+ if (comnested)
+ nesting++;
+ else
+ break;
+ }
/* Assume a defun-start point is outside of strings. */
if (code == Sopen
we know it can't be inside a string
since if it were then the comment ender would be inside one.
So it does start a comment. Skip back to it. */
- else if (comstart_parity == 0 && !string_lossage)
+ else if (!comnested && comstart_parity == 0 && !string_lossage)
{
from = comstart_pos;
from_byte = comstart_byte;
UPDATE_SYNTAX_TABLE_FORWARD (from - 1);
}
+ done:
*charpos_ptr = from;
*bytepos_ptr = from_byte;
- return from;
+ return (from == comment_end) ? -1 : from;
}
\f
DEFUN ("syntax-table-p", Fsyntax_table_p, Ssyntax_table_p, 1, 1, 0,
(table)
Lisp_Object table;
{
+ int idx;
check_syntax_table (table);
current_buffer->syntax_table = table;
/* Indicate that this buffer now has a specified syntax table. */
- current_buffer->local_var_flags
- |= XFASTINT (buffer_local_flags.syntax_table);
+ idx = PER_BUFFER_VAR_IDX (syntax_table);
+ SET_PER_BUFFER_VALUE_P (current_buffer, idx, 1);
return table;
}
\f
The second character of S is the matching parenthesis,\n\
used only if the first character is `(' or `)'.\n\
Any additional characters are flags.\n\
-Defined flags are the characters 1, 2, 3, 4, b, and p.\n\
+Defined flags are the characters 1, 2, 3, 4, b, p, and n.\n\
1 means CHAR is the start of a two-char comment start sequence.\n\
2 means CHAR is the second character of such a sequence.\n\
3 means CHAR is the start of a two-char comment end sequence.\n\
of a comment-start, or the first character of a comment-end sequence) using\n\
this flag:\n\
b means CHAR is part of comment sequence b.\n\
+ n means CHAR is part of a nestable comment sequence.\n\
\n\
p means CHAR is a prefix character for `backward-prefix-chars';\n\
such characters are treated as whitespace when they occur\n\
case 'b':
val |= 1 << 21;
break;
+
+ case 'n':
+ val |= 1 << 22;
+ break;
}
if (val < XVECTOR (Vsyntax_code_object)->size && NILP (match))
Lisp_Object value;
{
register enum syntaxcode code;
- char desc, match, start1, start2, end1, end2, prefix, comstyle;
+ char desc, start1, start2, end1, end2, prefix, comstyle;
char str[2];
Lisp_Object first, match_lisp;
return;
}
- first = XCONS (value)->car;
- match_lisp = XCONS (value)->cdr;
+ first = XCAR (value);
+ match_lisp = XCDR (value);
if (!INTEGERP (first) || !(NILP (match_lisp) || INTEGERP (match_lisp)))
{
DEFUN ("forward-word", Fforward_word, Sforward_word, 1, 1, "p",
"Move point forward ARG words (backward if ARG is negative).\n\
Normally returns t.\n\
-If an edge of the buffer is reached, point is left there\n\
-and nil is returned.")
+If an edge of the buffer or a field boundary is reached, point is left there\n\
+and the function returns nil. Field boundaries are not noticed if\n\
+`inhibit-field-text-motion' is non-nil.")
(count)
Lisp_Object count;
{
- int val;
+ int orig_val, val;
CHECK_NUMBER (count, 0);
- if (!(val = scan_words (PT, XINT (count))))
- {
- SET_PT (XINT (count) > 0 ? ZV : BEGV);
- return Qnil;
- }
+ val = orig_val = scan_words (PT, XINT (count));
+ if (! orig_val)
+ val = XINT (count) > 0 ? ZV : BEGV;
+
+ /* Avoid jumping out of an input field. */
+ val = XFASTINT (Fconstrain_to_field (make_number (val), make_number (PT),
+ Qt, Qnil));
+
SET_PT (val);
- return Qt;
+ return val == orig_val ? Qt : Qnil;
}
\f
Lisp_Object skip_chars ();
int forwardp, syntaxp;
Lisp_Object string, lim;
{
- register unsigned char *p, *pend;
register unsigned int c;
register int ch;
unsigned char fastmap[0400];
}
}
\f
+/* Jump over a comment, assuming we are at the beginning of one.
+ FROM is the current position.
+ FROM_BYTE is the bytepos corresponding to FROM.
+ Do not move past STOP (a charpos).
+ The comment over which we have to jump is of style STYLE
+ (either SYNTAX_COMMENT_STYLE(foo) or ST_COMMENT_STYLE).
+ NESTING should be positive to indicate the nesting at the beginning
+ for nested comments and should be zero or negative else.
+ ST_COMMENT_STYLE cannot be nested.
+ PREV_SYNTAX is the SYNTAX_WITH_FLAGS of the previous character
+ (or 0 If the search cannot start in the middle of a two-character).
+
+ If successful, return 1 and store the charpos of the comment's end
+ into *CHARPOS_PTR and the corresponding bytepos into *BYTEPOS_PTR.
+ Else, return 0 and store the charpos STOP into *CHARPOS_PTR, the
+ corresponding bytepos into *BYTEPOS_PTR and the current nesting
+ (as defined for state.incomment) in *INCOMMENT_PTR.
+
+ The comment end is the last character of the comment rather than the
+ character just after the comment.
+
+ Global syntax data is assumed to initially be valid for FROM and
+ remains valid for forward search starting at the returned position. */
+
+static int
+forw_comment (from, from_byte, stop, nesting, style, prev_syntax,
+ charpos_ptr, bytepos_ptr, incomment_ptr)
+ int from, from_byte, stop;
+ int nesting, style, prev_syntax;
+ int *charpos_ptr, *bytepos_ptr, *incomment_ptr;
+{
+ register int c, c1;
+ register enum syntaxcode code;
+ register int syntax;
+
+ if (nesting <= 0) nesting = -1;
+
+ /* Enter the loop in the middle so that we find
+ a 2-char comment ender if we start in the middle of it. */
+ syntax = prev_syntax;
+ if (syntax != 0) goto forw_incomment;
+
+ while (1)
+ {
+ if (from == stop)
+ {
+ *incomment_ptr = nesting;
+ *charpos_ptr = from;
+ *bytepos_ptr = from_byte;
+ return 0;
+ }
+ c = FETCH_CHAR (from_byte);
+ syntax = SYNTAX_WITH_FLAGS (c);
+ code = syntax & 0xff;
+ if (code == Sendcomment
+ && SYNTAX_FLAGS_COMMENT_STYLE (syntax) == style
+ && --nesting <= 0)
+ /* we have encountered a comment end of the same style
+ as the comment sequence which began this comment
+ section */
+ break;
+ if (code == Scomment_fence
+ && style == ST_COMMENT_STYLE)
+ /* we have encountered a comment end of the same style
+ as the comment sequence which began this comment
+ section. */
+ break;
+ if (nesting > 0
+ && code == Scomment
+ && SYNTAX_FLAGS_COMMENT_STYLE (syntax) == style)
+ /* we have encountered a nested comment of the same style
+ as the comment sequence which began this comment section */
+ nesting++;
+ INC_BOTH (from, from_byte);
+ UPDATE_SYNTAX_TABLE_FORWARD (from);
+
+ forw_incomment:
+ if (from < stop && SYNTAX_FLAGS_COMEND_FIRST (syntax)
+ && SYNTAX_FLAGS_COMMENT_STYLE (syntax) == style
+ && (c1 = FETCH_CHAR (from_byte),
+ SYNTAX_COMEND_SECOND (c1)))
+ {
+ if (--nesting <= 0)
+ /* we have encountered a comment end of the same style
+ as the comment sequence which began this comment
+ section */
+ break;
+ else
+ {
+ INC_BOTH (from, from_byte);
+ UPDATE_SYNTAX_TABLE_FORWARD (from);
+ }
+ }
+ if (nesting > 0
+ && from < stop
+ && SYNTAX_FLAGS_COMSTART_FIRST (syntax)
+ && (c1 = FETCH_CHAR (from_byte),
+ SYNTAX_COMMENT_STYLE (c1) == style
+ && SYNTAX_COMSTART_SECOND (c1)))
+ /* we have encountered a nested comment of the same style
+ as the comment sequence which began this comment
+ section */
+ {
+ INC_BOTH (from, from_byte);
+ UPDATE_SYNTAX_TABLE_FORWARD (from);
+ nesting++;
+ }
+ }
+ *charpos_ptr = from;
+ *bytepos_ptr = from_byte;
+ return 1;
+}
+
DEFUN ("forward-comment", Fforward_comment, Sforward_comment, 1, 1, 0,
"Move forward across up to N comments. If N is negative, move backward.\n\
Stop scanning if we find something other than a comment or whitespace.\n\
register int c, c1;
register enum syntaxcode code;
int comstyle = 0; /* style of comment encountered */
+ int comnested = 0; /* whether the comment is nestable or not */
int found;
int count1;
int out_charpos, out_bytepos;
+ int dummy;
CHECK_NUMBER (count, 0);
count1 = XINT (count);
immediate_quit = 0;
return Qnil;
}
- UPDATE_SYNTAX_TABLE_FORWARD (from);
c = FETCH_CHAR (from_byte);
code = SYNTAX (c);
comstart_first = SYNTAX_COMSTART_FIRST (c);
+ comnested = SYNTAX_COMMENT_NESTED (c);
INC_BOTH (from, from_byte);
UPDATE_SYNTAX_TABLE_FORWARD (from);
comstyle = 0;
the comment section. */
code = Scomment;
comstyle = SYNTAX_COMMENT_STYLE (c1);
+ comnested = comnested || SYNTAX_COMMENT_NESTED (c1);
INC_BOTH (from, from_byte);
+ UPDATE_SYNTAX_TABLE_FORWARD (from);
}
+ /* FIXME: here we ignore 2-char endcomments while we don't
+ when going backwards. */
}
while (code == Swhitespace || code == Sendcomment);
return Qnil;
}
/* We're at the start of a comment. */
- while (1)
+ found = forw_comment (from, from_byte, stop, comnested, comstyle, 0,
+ &out_charpos, &out_bytepos, &dummy);
+ from = out_charpos; from_byte = out_bytepos;
+ if (!found)
{
- if (from == stop)
- {
- immediate_quit = 0;
- SET_PT_BOTH (from, from_byte);
- return Qnil;
- }
- UPDATE_SYNTAX_TABLE_FORWARD (from);
- c = FETCH_CHAR (from_byte);
- INC_BOTH (from, from_byte);
- if (SYNTAX (c) == Sendcomment
- && SYNTAX_COMMENT_STYLE (c) == comstyle)
- /* we have encountered a comment end of the same style
- as the comment sequence which began this comment
- section */
- break;
- if (SYNTAX (c) == Scomment_fence
- && comstyle == ST_COMMENT_STYLE)
- /* we have encountered a comment end of the same style
- as the comment sequence which began this comment
- section. */
- break;
- if (from < stop && SYNTAX_COMEND_FIRST (c)
- && SYNTAX_COMMENT_STYLE (c) == comstyle
- && (c1 = FETCH_CHAR (from_byte),
- UPDATE_SYNTAX_TABLE_FORWARD (from),
- SYNTAX_COMEND_SECOND (c1)))
- /* we have encountered a comment end of the same style
- as the comment sequence which began this comment
- section */
- {
- INC_BOTH (from, from_byte);
- break;
- }
+ immediate_quit = 0;
+ SET_PT_BOTH (from, from_byte);
+ return Qnil;
}
+ INC_BOTH (from, from_byte);
+ UPDATE_SYNTAX_TABLE_FORWARD (from);
/* We have skipped one comment. */
count1--;
}
c = FETCH_CHAR (from_byte);
code = SYNTAX (c);
comstyle = 0;
+ comnested = SYNTAX_COMMENT_NESTED (c);
if (code == Sendcomment)
comstyle = SYNTAX_COMMENT_STYLE (c);
comstart_second = SYNTAX_COMSTART_SECOND (c);
code = Sendcomment;
/* Calling char_quoted, above, set up global syntax position
at the new value of FROM. */
- comstyle = SYNTAX_COMMENT_STYLE (FETCH_CHAR (from_byte));
+ c1 = FETCH_CHAR (from_byte);
+ comstyle = SYNTAX_COMMENT_STYLE (c1);
+ comnested = comnested || SYNTAX_COMMENT_NESTED (c1);
}
if (from > stop && comstart_second
&& prev_char_comstart_first (from, from_byte)
&& !char_quoted (from - 1, dec_bytepos (from_byte)))
{
- /* We must record the comment style encountered so that
- later, we can match only the proper comment begin
- sequence of the same style. */
code = Scomment;
DEC_BOTH (from, from_byte);
}
}
else if (code == Sendcomment)
{
- found = back_comment (from, from_byte, stop, comstyle,
+ found = back_comment (from, from_byte, stop, comnested, comstyle,
&out_charpos, &out_bytepos);
- if (found != -1)
+ if (found == -1)
+ {
+#if 0 /* cc-mode (and maybe others) relies on the bogus behavior. */
+ /* Failure: we should go back to the end of this
+ not-quite-endcomment. */
+ if (SYNTAX(c) != code)
+ /* It was a two-char Sendcomment. */
+ INC_BOTH (from, from_byte);
+ goto leave;
+#endif
+ }
+ else
+ /* We have skipped one comment. */
from = out_charpos, from_byte = out_bytepos;
- /* We have skipped one comment. */
break;
}
else if (code != Swhitespace && code != Scomment)
return Qt;
}
\f
+/* Return syntax code of character C if C is a single byte character
+ or `multibyte_symbol_p' is zero. Otherwise, retrun Ssymbol. */
+
+#define SYNTAX_WITH_MULTIBYTE_CHECK(c) \
+ ((SINGLE_BYTE_CHAR_P (c) || !multibyte_symbol_p) \
+ ? SYNTAX (c) : Ssymbol)
+
static Lisp_Object
scan_lists (from, count, depth, sexpflag)
register int from;
register enum syntaxcode code, temp_code;
int min_depth = depth; /* Err out if depth gets less than this. */
int comstyle = 0; /* style of comment encountered */
+ int comnested = 0; /* whether the comment is nestable or not */
int temp_pos;
int last_good = from;
int found;
int from_byte;
int out_bytepos, out_charpos;
- int temp;
+ int temp, dummy;
+ int multibyte_symbol_p = sexpflag && multibyte_syntax_as_symbol;
if (depth > 0) min_depth = 0;
int comstart_first, prefix;
UPDATE_SYNTAX_TABLE_FORWARD (from);
c = FETCH_CHAR (from_byte);
- code = SYNTAX (c);
+ code = SYNTAX_WITH_MULTIBYTE_CHECK (c);
comstart_first = SYNTAX_COMSTART_FIRST (c);
+ comnested = SYNTAX_COMMENT_NESTED (c);
prefix = SYNTAX_PREFIX (c);
if (depth == min_depth)
last_good = from;
only a comment end of the same style actually ends
the comment section */
code = Scomment;
- comstyle = SYNTAX_COMMENT_STYLE (FETCH_CHAR (from_byte));
+ c1 = FETCH_CHAR (from_byte);
+ comstyle = SYNTAX_COMMENT_STYLE (c1);
+ comnested = comnested || SYNTAX_COMMENT_NESTED (c1);
INC_BOTH (from, from_byte);
UPDATE_SYNTAX_TABLE_FORWARD (from);
}
UPDATE_SYNTAX_TABLE_FORWARD (from);
/* Some compilers can't handle this inside the switch. */
- temp = SYNTAX (FETCH_CHAR (from_byte));
+ c = FETCH_CHAR (from_byte);
+ temp = SYNTAX_WITH_MULTIBYTE_CHECK (c);
switch (temp)
{
case Scharquote:
}
goto done;
- case Scomment:
case Scomment_fence:
+ comstyle = ST_COMMENT_STYLE;
+ /* FALLTHROUGH */
+ case Scomment:
if (!parse_sexp_ignore_comments) break;
- while (1)
+ UPDATE_SYNTAX_TABLE_FORWARD (from);
+ found = forw_comment (from, from_byte, stop,
+ comnested, comstyle, 0,
+ &out_charpos, &out_bytepos, &dummy);
+ from = out_charpos, from_byte = out_bytepos;
+ if (!found)
{
- if (from == stop)
- {
- if (depth == 0)
- goto done;
- goto lose;
- }
- UPDATE_SYNTAX_TABLE_FORWARD (from);
- c = FETCH_CHAR (from_byte);
- INC_BOTH (from, from_byte);
- if (code == Scomment
- ? (SYNTAX (c) == Sendcomment
- && SYNTAX_COMMENT_STYLE (c) == comstyle)
- : (SYNTAX (c) == Scomment_fence))
- /* we have encountered a comment end of the same style
- as the comment sequence which began this comment
- section */
- break;
- if (from < stop && SYNTAX_COMEND_FIRST (c)
- && SYNTAX_COMMENT_STYLE (c) == comstyle
- && (UPDATE_SYNTAX_TABLE_FORWARD (from),
- SYNTAX_COMEND_SECOND (FETCH_CHAR (from_byte)))
- && code == Scomment)
- /* we have encountered a comment end of the same style
- as the comment sequence which began this comment
- section */
- {
- INC_BOTH (from, from_byte);
- break;
- }
+ if (depth == 0)
+ goto done;
+ goto lose;
}
+ INC_BOTH (from, from_byte);
+ UPDATE_SYNTAX_TABLE_FORWARD (from);
break;
case Smath:
{
if (from >= stop) goto lose;
UPDATE_SYNTAX_TABLE_FORWARD (from);
+ c = FETCH_CHAR (from_byte);
if (code == Sstring
- ? (FETCH_CHAR (from_byte) == stringterm)
- : SYNTAX (FETCH_CHAR (from_byte)) == Sstring_fence)
+ ? c == stringterm
+ : SYNTAX_WITH_MULTIBYTE_CHECK (c) == Sstring_fence)
break;
/* Some compilers can't handle this inside the switch. */
- temp = SYNTAX (FETCH_CHAR (from_byte));
+ temp = SYNTAX_WITH_MULTIBYTE_CHECK (c);
switch (temp)
{
case Scharquote:
DEC_BOTH (from, from_byte);
UPDATE_SYNTAX_TABLE_BACKWARD (from);
c = FETCH_CHAR (from_byte);
- code = SYNTAX (c);
+ code = SYNTAX_WITH_MULTIBYTE_CHECK (c);
if (depth == min_depth)
last_good = from;
comstyle = 0;
+ comnested = SYNTAX_COMMENT_NESTED (c);
if (code == Sendcomment)
comstyle = SYNTAX_COMMENT_STYLE (c);
if (from > stop && SYNTAX_COMEND_SECOND (c)
DEC_BOTH (from, from_byte);
UPDATE_SYNTAX_TABLE_BACKWARD (from);
code = Sendcomment;
- comstyle = SYNTAX_COMMENT_STYLE (FETCH_CHAR (from_byte));
+ c1 = FETCH_CHAR (from_byte);
+ comstyle = SYNTAX_COMMENT_STYLE (c1);
+ comnested = comnested || SYNTAX_COMMENT_NESTED (c1);
}
/* Quoting turns anything except a comment-ender
temp_pos--;
UPDATE_SYNTAX_TABLE_BACKWARD (from - 1);
c1 = FETCH_CHAR (temp_pos);
- temp_code = SYNTAX (c1);
+ temp_code = SYNTAX_WITH_MULTIBYTE_CHECK (c1);
/* Don't allow comment-end to be quoted. */
if (temp_code == Sendcomment)
goto done2;
UPDATE_SYNTAX_TABLE_BACKWARD (from - 1);
}
c1 = FETCH_CHAR (temp_pos);
- temp_code = SYNTAX (c1);
+ temp_code = SYNTAX_WITH_MULTIBYTE_CHECK (c1);
if (! (quoted || temp_code == Sword
|| temp_code == Ssymbol
|| temp_code == Squote))
case Sendcomment:
if (!parse_sexp_ignore_comments)
break;
- found = back_comment (from, from_byte, stop, comstyle,
+ found = back_comment (from, from_byte, stop, comnested, comstyle,
&out_charpos, &out_bytepos);
+ /* FIXME: if found == -1, then it really wasn't a comment-end.
+ For single-char Sendcomment, we can't do much about it apart
+ from skipping the char.
+ For 2-char endcomments, we could try again, taking both
+ chars as separate entities, but it's a lot of trouble
+ for very little gain, so we don't bother either. -sm */
if (found != -1)
from = out_charpos, from_byte = out_bytepos;
break;
if (from == stop) goto lose;
UPDATE_SYNTAX_TABLE_BACKWARD (from);
if (!char_quoted (from, from_byte)
- && SYNTAX (FETCH_CHAR (from_byte)) == code)
+ && (c = FETCH_CHAR (from_byte),
+ SYNTAX_WITH_MULTIBYTE_CHECK (c) == code))
break;
}
if (code == Sstring_fence && !depth && sexpflag) goto done2;
struct lisp_parse_state state;
register enum syntaxcode code;
+ int c1;
+ int comnested;
struct level { int last, prev; };
struct level levelstart[100];
register struct level *curlevel = levelstart;
struct level *endlevel = levelstart + 100;
- int prev;
register int depth; /* Paren depth of current scanning location.
level - levelstart equals this except
when the depth becomes negative. */
int prev_from_syntax;
int boundary_stop = commentstop == -1;
int nofence;
+ int found;
+ int out_bytepos, out_charpos;
int temp;
prev_from = from;
oldstate = Fcdr (oldstate);
tem = Fcar (oldstate);
/* Check whether we are inside string_fence-style string: */
- state.instring = ( !NILP (tem)
- ? ( INTEGERP (tem) ? XINT (tem) : ST_STRING_STYLE)
- : -1);
+ state.instring = (!NILP (tem)
+ ? (INTEGERP (tem) ? XINT (tem) : ST_STRING_STYLE)
+ : -1);
oldstate = Fcdr (oldstate);
tem = Fcar (oldstate);
- state.incomment = !NILP (tem);
+ state.incomment = (!NILP (tem)
+ ? (INTEGERP (tem) ? XINT (tem) : -1)
+ : 0);
oldstate = Fcdr (oldstate);
tem = Fcar (oldstate);
start_quoted = !NILP (tem);
- /* if the eight element of the list is nil, we are in comment
+ /* if the eighth element of the list is nil, we are in comment
style a. If it is non-nil, we are in comment style b */
oldstate = Fcdr (oldstate);
oldstate = Fcdr (oldstate);
tem = Fcar (oldstate);
- state.comstyle = NILP (tem) ? 0 : ( EQ (tem, Qsyntax_table)
- ? ST_COMMENT_STYLE : 1 );
+ state.comstyle = NILP (tem) ? 0 : (EQ (tem, Qsyntax_table)
+ ? ST_COMMENT_STYLE : 1);
oldstate = Fcdr (oldstate);
tem = Fcar (oldstate);
code = prev_from_syntax & 0xff;
if (code == Scomment)
- state.comstr_start = prev_from;
+ {
+ state.incomment = (SYNTAX_FLAGS_COMMENT_NESTED (prev_from_syntax) ?
+ 1 : -1);
+ state.comstr_start = prev_from;
+ }
else if (code == Scomment_fence)
{
/* Record the comment style we have entered so that only
the comment-end sequence of the same style actually
terminates the comment section. */
- state.comstyle = ( code == Scomment_fence
- ? ST_COMMENT_STYLE
- : SYNTAX_COMMENT_STYLE (FETCH_CHAR (from_byte)));
+ state.comstyle = ST_COMMENT_STYLE;
+ state.incomment = -1;
state.comstr_start = prev_from;
- if (code != Scomment_fence)
- INC_FROM;
code = Scomment;
}
else if (from < end)
if (SYNTAX_FLAGS_COMSTART_FIRST (prev_from_syntax))
- if (SYNTAX_COMSTART_SECOND (FETCH_CHAR (from_byte)))
+ if (c1 = FETCH_CHAR (from_byte),
+ SYNTAX_COMSTART_SECOND (c1))
/* Duplicate code to avoid a complex if-expression
which causes trouble for the SGI compiler. */
{
/* Record the comment style we have entered so that only
the comment-end sequence of the same style actually
terminates the comment section. */
- state.comstyle = ( code == Scomment_fence
- ? ST_COMMENT_STYLE
- : SYNTAX_COMMENT_STYLE (FETCH_CHAR (from_byte)));
+ state.comstyle = SYNTAX_COMMENT_STYLE (FETCH_CHAR (from_byte));
+ comnested = SYNTAX_FLAGS_COMMENT_NESTED (prev_from_syntax);
+ comnested = comnested || SYNTAX_COMMENT_NESTED (c1);
+ state.incomment = comnested ? 1 : -1;
state.comstr_start = prev_from;
- if (code != Scomment_fence)
- INC_FROM;
+ INC_FROM;
code = Scomment;
}
startincomment:
if (commentstop == 1)
goto done;
- if (from != BEGV)
- {
- /* Enter the loop in the middle so that we find
- a 2-char comment ender if we start in the middle of it. */
- goto startincomment_1;
- }
- /* At beginning of buffer, enter the loop the ordinary way. */
- state.incomment = 1;
goto commentloop;
case Scomment:
- state.incomment = 1;
+ if (! state.incomment)
+ abort ();
if (commentstop || boundary_stop) goto done;
commentloop:
- while (1)
- {
- if (from == end) goto done;
- prev = FETCH_CHAR (from_byte);
- if (SYNTAX (prev) == Sendcomment
- && SYNTAX_COMMENT_STYLE (prev) == state.comstyle)
- /* Only terminate the comment section if the endcomment
- of the same style as the start sequence has been
- encountered. */
- break;
- if (state.comstyle == ST_COMMENT_STYLE
- && SYNTAX (prev) == Scomment_fence)
- break;
- INC_FROM;
- startincomment_1:
- if (from < end && SYNTAX_FLAGS_COMEND_FIRST (prev_from_syntax)
- && SYNTAX_COMEND_SECOND (FETCH_CHAR (from_byte))
- && (SYNTAX_FLAGS_COMMENT_STYLE (prev_from_syntax)
- == state.comstyle))
- /* Only terminate the comment section if the end-comment
- sequence of the same style as the start sequence has
- been encountered. */
- break;
- }
+ /* The (from == BEGV) test is to enter the loop in the middle so
+ that we find a 2-char comment ender even if we start in the
+ middle of it. */
+ found = forw_comment (from, from_byte, end,
+ state.incomment, state.comstyle,
+ (from == BEGV) ? 0 : prev_from_syntax,
+ &out_charpos, &out_bytepos, &state.incomment);
+ from = out_charpos; from_byte = out_bytepos;
+ /* Beware! prev_from and friends are invalid now.
+ Luckily, the `done' doesn't use them and the INC_FROM
+ sets them to a sane value without looking at them. */
+ if (!found) goto done;
INC_FROM;
state.incomment = 0;
state.comstyle = 0; /* reset the comment style */
3. non-nil if inside a string.\n\
(it is the character that will terminate the string,\n\
or t if the string should be terminated by a generic string delimiter.)\n\
- 4. t if inside a comment.\n\
+ 4. nil if outside a comment, t if inside a non-nestable comment, \n\
+ else an integer (the current comment nesting).\n\
5. t if following a quote character.\n\
6. the minimum paren-depth encountered during this scan.\n\
7. t if in a comment of style b; `syntax-table' if the comment\n\
Fcons (state.instring >= 0
? (state.instring == ST_STRING_STYLE
? Qt : make_number (state.instring)) : Qnil,
- Fcons (state.incomment ? Qt : Qnil,
+ Fcons (state.incomment < 0 ? Qt :
+ (state.incomment == 0 ? Qnil :
+ make_number (state.incomment)),
Fcons (state.quoted ? Qt : Qnil,
Fcons (make_number (state.mindepth),
Fcons ((state.comstyle
c = ".,;:?!#@~^'`"[i];
SET_RAW_SYNTAX_ENTRY (Vstandard_syntax_table, c, temp);
}
+
+ /* All multibyte characters have syntax `word' by default. */
+ temp = XVECTOR (Vsyntax_code_object)->contents[(int) Sword];
+ for (i = CHAR_TABLE_SINGLE_BYTE_SLOTS; i < CHAR_TABLE_ORDINARY_SLOTS; i++)
+ XCHAR_TABLE (Vstandard_syntax_table)->contents[i] = temp;
}
void
DEFVAR_BOOL ("words-include-escapes", &words_include_escapes,
"Non-nil means `forward-word', etc., should treat escape chars part of words.");
+ DEFVAR_BOOL ("multibyte-syntax-as-symbol", &multibyte_syntax_as_symbol,
+ "Non-nil means `scan-sexps' treats all multibyte characters as symbol.");
+ multibyte_syntax_as_symbol = 0;
+
+ DEFVAR_BOOL ("open-paren-in-column-0-is-defun-start",
+ &open_paren_in_column_0_is_defun_start,
+ "Non-nil means an open paren in column 0 denotes the start of a defun.");
+ open_paren_in_column_0_is_defun_start = 1;
+
defsubr (&Ssyntax_table_p);
defsubr (&Ssyntax_table);
defsubr (&Sstandard_syntax_table);