X-Git-Url: https://code.delx.au/gnu-emacs/blobdiff_plain/ab5796a9f97180707734a81320e3eb81937281fe..a5e700caf2a2f93ed6d39af407d67d8244a6b2a7:/src/syntax.c diff --git a/src/syntax.c b/src/syntax.c index 79df112d26..fa34c2433c 100644 --- a/src/syntax.c +++ b/src/syntax.c @@ -1,5 +1,5 @@ /* GNU Emacs routines to deal with syntax tables; also word and list parsing. - Copyright (C) 1985, 87, 93, 94, 95, 97, 1998, 1999 Free Software Foundation, Inc. + Copyright (C) 1985, 87, 93, 94, 95, 97, 1998, 1999, 2004 Free Software Foundation, Inc. This file is part of GNU Emacs. @@ -26,6 +26,7 @@ Boston, MA 02111-1307, USA. */ #include "buffer.h" #include "charset.h" #include "keymap.h" +#include "regex.h" /* Make syntax table lookup grant data in gl_state. */ #define SYNTAX_ENTRY_VIA_PROPERTY @@ -97,11 +98,12 @@ static int find_start_modiff; static int find_defun_start P_ ((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 skip_chars P_ ((int, int, Lisp_Object, Lisp_Object, int)); static Lisp_Object scan_lists P_ ((int, int, int, int)); static void scan_sexps_forward P_ ((struct lisp_parse_state *, int, int, int, int, int, Lisp_Object, int)); +static int in_classes P_ ((int, Lisp_Object)); struct gl_state_s gl_state; /* Global state of syntax parser. */ @@ -290,10 +292,13 @@ char_quoted (charpos, bytepos) DEC_BOTH (charpos, bytepos); - while (bytepos >= beg) + while (charpos >= beg) { + int c; + UPDATE_SYNTAX_TABLE_BACKWARD (charpos); - code = SYNTAX (FETCH_CHAR (bytepos)); + c = FETCH_CHAR (bytepos); + code = SYNTAX (c); if (! (code == Scharquote || code == Sescape)) break; @@ -380,12 +385,16 @@ find_defun_start (pos, pos_byte) gl_state.use_global = 0; while (PT > BEGV) { + int c; + /* Open-paren at start of line means we may have found our defun-start. */ - if (SYNTAX (FETCH_CHAR (PT_BYTE)) == Sopen) + c = FETCH_CHAR (PT_BYTE); + if (SYNTAX (c) == Sopen) { SETUP_SYNTAX_TABLE (PT + 1, -1); /* Try again... */ - if (SYNTAX (FETCH_CHAR (PT_BYTE)) == Sopen) + c = FETCH_CHAR (PT_BYTE); + if (SYNTAX (c) == Sopen) break; /* Now fallback to the default value. */ gl_state.current_syntax_table = current_buffer->syntax_table; @@ -664,7 +673,7 @@ back_comment (from, from_byte, stop, comnested, comstyle, charpos_ptr, bytepos_p { from = comstart_pos; from_byte = comstart_byte; - /* Globals are correct now. */ + UPDATE_SYNTAX_TABLE_FORWARD (from - 1); } else { @@ -977,7 +986,7 @@ text property. */) DEFUN ("modify-syntax-entry", Fmodify_syntax_entry, Smodify_syntax_entry, 2, 3, "cSet syntax for character: \nsSet syntax for %s to: ", doc: /* Set syntax for character CHAR according to string NEWENTRY. -The syntax is changed only for table SYNTAX_TABLE, which defaults to +The syntax is changed only for table SYNTAX-TABLE, which defaults to the current buffer's syntax table. The first character of NEWENTRY should be one of the following: Space or - whitespace syntax. w word constituent. @@ -1286,6 +1295,7 @@ and the function returns nil. Field boundaries are not noticed if (arg) Lisp_Object arg; { + Lisp_Object tmp; int orig_val, val; if (NILP (arg)) @@ -1298,8 +1308,9 @@ and the function returns nil. Field boundaries are not noticed if val = XINT (arg) > 0 ? ZV : BEGV; /* Avoid jumping out of an input field. */ - val = XFASTINT (Fconstrain_to_field (make_number (val), make_number (PT), - Qt, Qnil, Qnil)); + tmp = Fconstrain_to_field (make_number (val), make_number (PT), + Qt, Qnil, Qnil); + val = XFASTINT (tmp); SET_PT (val); return val == orig_val ? Qt : Qnil; @@ -1314,13 +1325,13 @@ except that `]' is never special and `\\' quotes `^', `-' or `\\' (but not as the end of a range; quoting is never needed there). Thus, with arg "a-zA-Z", this skips letters stopping before first nonletter. With arg "^a-zA-Z", skips nonletters stopping before first letter. -Returns the distance traveled, either zero or positive. -Note that char classes, e.g. `[:alpha:]', are not currently supported; -they will be treated as literals. */) +Char classes, e.g. `[:alpha:]', are supported. + +Returns the distance traveled, either zero or positive. */) (string, lim) Lisp_Object string, lim; { - return skip_chars (1, 0, string, lim); + return skip_chars (1, 0, string, lim, 1); } DEFUN ("skip-chars-backward", Fskip_chars_backward, Sskip_chars_backward, 1, 2, 0, @@ -1330,7 +1341,7 @@ Returns the distance traveled, either zero or negative. */) (string, lim) Lisp_Object string, lim; { - return skip_chars (0, 0, string, lim); + return skip_chars (0, 0, string, lim, 1); } DEFUN ("skip-syntax-forward", Fskip_syntax_forward, Sskip_syntax_forward, 1, 2, 0, @@ -1342,7 +1353,7 @@ This function returns the distance traveled, either zero or positive. */) (syntax, lim) Lisp_Object syntax, lim; { - return skip_chars (1, 1, syntax, lim); + return skip_chars (1, 1, syntax, lim, 0); } DEFUN ("skip-syntax-backward", Fskip_syntax_backward, Sskip_syntax_backward, 1, 2, 0, @@ -1354,13 +1365,14 @@ This function returns the distance traveled, either zero or negative. */) (syntax, lim) Lisp_Object syntax, lim; { - return skip_chars (0, 1, syntax, lim); + return skip_chars (0, 1, syntax, lim, 0); } static Lisp_Object -skip_chars (forwardp, syntaxp, string, lim) +skip_chars (forwardp, syntaxp, string, lim, handle_iso_classes) int forwardp, syntaxp; Lisp_Object string, lim; + int handle_iso_classes; { register unsigned int c; unsigned char fastmap[0400]; @@ -1376,12 +1388,14 @@ skip_chars (forwardp, syntaxp, string, lim) int size_byte; const unsigned char *str; int len; + Lisp_Object iso_classes; CHECK_STRING (string); char_ranges = (int *) alloca (SCHARS (string) * (sizeof (int)) * 2); string_multibyte = STRING_MULTIBYTE (string); str = SDATA (string); size_byte = SBYTES (string); + iso_classes = Qnil; /* Adjust the multibyteness of the string to that of the buffer. */ if (multibyte != string_multibyte) @@ -1437,15 +1451,54 @@ skip_chars (forwardp, syntaxp, string, lim) fastmap[syntax_spec_code[c & 0377]] = 1; else { + if (handle_iso_classes && c == '[' + && i_byte < size_byte + && STRING_CHAR (str + i_byte, size_byte - i_byte) == ':') + { + const unsigned char *class_beg = str + i_byte + 1; + const unsigned char *class_end = class_beg; + const unsigned char *class_limit = str + size_byte - 2; + /* Leave room for the null. */ + unsigned char class_name[CHAR_CLASS_MAX_LENGTH + 1]; + re_wctype_t cc; + + if (class_limit - class_beg > CHAR_CLASS_MAX_LENGTH) + class_limit = class_beg + CHAR_CLASS_MAX_LENGTH; + + while (class_end < class_limit + && *class_end >= 'a' && *class_end <= 'z') + class_end++; + + if (class_end == class_beg + || *class_end != ':' || class_end[1] != ']') + goto not_a_class_name; + + bcopy (class_beg, class_name, class_end - class_beg); + class_name[class_end - class_beg] = 0; + + cc = re_wctype (class_name); + if (cc == 0) + error ("Invalid ISO C character class"); + + iso_classes = Fcons (make_number (cc), iso_classes); + + i_byte = class_end + 2 - str; + continue; + } + + not_a_class_name: if (c == '\\') { if (i_byte == size_byte) break; - c = STRING_CHAR_AND_LENGTH (str+i_byte, size_byte-i_byte, len); + c = STRING_CHAR_AND_LENGTH (str + i_byte, + size_byte - i_byte, len); i_byte += len; } - if (i_byte < size_byte + /* Treat `-' as range character only if another character + follows. */ + if (i_byte + 1 < size_byte && str[i_byte] == '-') { unsigned int c2; @@ -1453,11 +1506,9 @@ skip_chars (forwardp, syntaxp, string, lim) /* Skip over the dash. */ i_byte++; - if (i_byte == size_byte) - break; - /* Get the end of the range. */ - c2 =STRING_CHAR_AND_LENGTH (str+i_byte, size_byte-i_byte, len); + c2 = STRING_CHAR_AND_LENGTH (str + i_byte, + size_byte - i_byte, len); i_byte += len; if (SINGLE_BYTE_CHAR_P (c)) @@ -1516,12 +1567,12 @@ skip_chars (forwardp, syntaxp, string, lim) if (forwardp) { endp = (XINT (lim) == GPT) ? GPT_ADDR : CHAR_POS_ADDR (XINT (lim)); - stop = (pos < GPT && GPT < XINT (lim)) ? GPT_ADDR : endp; + stop = (pos < GPT && GPT < XINT (lim)) ? GPT_ADDR : endp; } else { endp = CHAR_POS_ADDR (XINT (lim)); - stop = (pos >= GPT && GPT > XINT (lim)) ? GAP_END_ADDR : endp; + stop = (pos >= GPT && GPT > XINT (lim)) ? GAP_END_ADDR : endp; } immediate_quit = 1; @@ -1629,6 +1680,15 @@ skip_chars (forwardp, syntaxp, string, lim) stop = endp; } c = STRING_CHAR_AND_LENGTH (p, MAX_MULTIBYTE_LENGTH, nbytes); + + if (! NILP (iso_classes) && in_classes (c, iso_classes)) + { + if (negate) + break; + else + goto fwd_ok; + } + if (SINGLE_BYTE_CHAR_P (c)) { if (!fastmap[c]) @@ -1651,6 +1711,7 @@ skip_chars (forwardp, syntaxp, string, lim) if (!(negate ^ (i < n_char_ranges))) break; } + fwd_ok: p += nbytes, pos++, pos_byte += nbytes; } else @@ -1663,8 +1724,18 @@ skip_chars (forwardp, syntaxp, string, lim) p = GAP_END_ADDR; stop = endp; } + + if (!NILP (iso_classes) && in_classes (*p, iso_classes)) + { + if (negate) + break; + else + goto fwd_unibyte_ok; + } + if (!fastmap[*p]) break; + fwd_unibyte_ok: p++, pos++; } } @@ -1690,6 +1761,15 @@ skip_chars (forwardp, syntaxp, string, lim) p = prev_p - 1, c = *p, nbytes = 1; else c = STRING_CHAR (p, MAX_MULTIBYTE_LENGTH); + + if (! NILP (iso_classes) && in_classes (c, iso_classes)) + { + if (negate) + break; + else + goto back_ok; + } + if (SINGLE_BYTE_CHAR_P (c)) { if (!fastmap[c]) @@ -1704,6 +1784,7 @@ skip_chars (forwardp, syntaxp, string, lim) if (!(negate ^ (i < n_char_ranges))) break; } + back_ok: pos--, pos_byte -= nbytes; } else @@ -1716,8 +1797,18 @@ skip_chars (forwardp, syntaxp, string, lim) p = GPT_ADDR; stop = endp; } + + if (! NILP (iso_classes) && in_classes (p[-1], iso_classes)) + { + if (negate) + break; + else + goto back_unibyte_ok; + } + if (!fastmap[p[-1]]) break; + back_unibyte_ok: p--, pos--; } } @@ -1740,6 +1831,30 @@ skip_chars (forwardp, syntaxp, string, lim) return make_number (PT - start_point); } } + +/* Return 1 if character C belongs to one of the ISO classes + in the list ISO_CLASSES. Each class is represented by an + integer which is its type according to re_wctype. */ + +static int +in_classes (c, iso_classes) + int c; + Lisp_Object iso_classes; +{ + int fits_class = 0; + + while (! NILP (iso_classes)) + { + Lisp_Object elt; + elt = XCAR (iso_classes); + iso_classes = XCDR (iso_classes); + + if (re_iswctype (c, XFASTINT (elt))) + fits_class = 1; + } + + return fits_class; +} /* Jump over a comment, assuming we are at the beginning of one. FROM is the current position. @@ -2123,7 +2238,7 @@ scan_lists (from, count, depth, sexpflag) INC_BOTH (from, from_byte); UPDATE_SYNTAX_TABLE_FORWARD (from); if (from < stop && comstart_first - && SYNTAX_COMSTART_SECOND (FETCH_CHAR (from_byte)) + && (c = FETCH_CHAR (from_byte), SYNTAX_COMSTART_SECOND (c)) && parse_sexp_ignore_comments) { /* we have encountered a comment start sequence and we @@ -2397,8 +2512,8 @@ scan_lists (from, count, depth, sexpflag) case Sstring_fence: while (1) { - DEC_BOTH (from, from_byte); if (from == stop) goto lose; + DEC_BOTH (from, from_byte); UPDATE_SYNTAX_TABLE_BACKWARD (from); if (!char_quoted (from, from_byte) && (c = FETCH_CHAR (from_byte), @@ -2413,19 +2528,13 @@ scan_lists (from, count, depth, sexpflag) while (1) { if (from == stop) goto lose; - temp_pos = from_byte; - if (! NILP (current_buffer->enable_multibyte_characters)) - DEC_POS (temp_pos); - else - temp_pos--; - UPDATE_SYNTAX_TABLE_BACKWARD (from - 1); - if (!char_quoted (from - 1, temp_pos) - && stringterm == (c = FETCH_CHAR (temp_pos)) + DEC_BOTH (from, from_byte); + UPDATE_SYNTAX_TABLE_BACKWARD (from); + if (!char_quoted (from, from_byte) + && stringterm == (c = FETCH_CHAR (from_byte)) && SYNTAX_WITH_MULTIBYTE_CHECK (c) == Sstring) break; - DEC_BOTH (from, from_byte); } - DEC_BOTH (from, from_byte); if (!depth && sexpflag) goto done2; break; default: @@ -2454,7 +2563,7 @@ scan_lists (from, count, depth, sexpflag) Fcons (build_string ("Unbalanced parentheses"), Fcons (make_number (last_good), Fcons (make_number (from), Qnil)))); - + abort (); /* NOTREACHED */ } @@ -2593,8 +2702,8 @@ scan_sexps_forward (stateptr, from, from_byte, end, targetdepth, #define INC_FROM \ do { prev_from = from; \ prev_from_byte = from_byte; \ - prev_from_syntax \ - = SYNTAX_WITH_FLAGS (FETCH_CHAR (prev_from_byte)); \ + temp = FETCH_CHAR (prev_from_byte); \ + prev_from_syntax = SYNTAX_WITH_FLAGS (temp); \ INC_BOTH (from, from_byte); \ if (from < end) \ UPDATE_SYNTAX_TABLE_FORWARD (from); \ @@ -2669,7 +2778,8 @@ do { prev_from = from; \ curlevel->last = -1; SETUP_SYNTAX_TABLE (prev_from, 1); - prev_from_syntax = SYNTAX_WITH_FLAGS (FETCH_CHAR (prev_from_byte)); + temp = FETCH_CHAR (prev_from_byte); + prev_from_syntax = SYNTAX_WITH_FLAGS (temp); UPDATE_SYNTAX_TABLE_FORWARD (from); /* Enter the loop at a place appropriate for initial state. */ @@ -2691,12 +2801,23 @@ do { prev_from = from; \ INC_FROM; code = prev_from_syntax & 0xff; - if (code == Scomment) + if (from < end + && SYNTAX_FLAGS_COMSTART_FIRST (prev_from_syntax) + && (c1 = FETCH_CHAR (from_byte), + SYNTAX_COMSTART_SECOND (c1))) + /* Duplicate code to avoid a complex if-expression + which causes trouble for the SGI compiler. */ { - state.comstyle = SYNTAX_FLAGS_COMMENT_STYLE (prev_from_syntax); - state.incomment = (SYNTAX_FLAGS_COMMENT_NESTED (prev_from_syntax) ? - 1 : -1); + /* 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 = SYNTAX_COMMENT_STYLE (c1); + comnested = SYNTAX_FLAGS_COMMENT_NESTED (prev_from_syntax); + comnested = comnested || SYNTAX_COMMENT_NESTED (c1); + state.incomment = comnested ? 1 : -1; state.comstr_start = prev_from; + INC_FROM; + code = Scomment; } else if (code == Scomment_fence) { @@ -2708,24 +2829,13 @@ do { prev_from = from; \ state.comstr_start = prev_from; code = Scomment; } - else if (from < end) - if (SYNTAX_FLAGS_COMSTART_FIRST (prev_from_syntax)) - 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 = SYNTAX_COMMENT_STYLE (c1); - comnested = SYNTAX_FLAGS_COMMENT_NESTED (prev_from_syntax); - comnested = comnested || SYNTAX_COMMENT_NESTED (c1); - state.incomment = comnested ? 1 : -1; - state.comstr_start = prev_from; - INC_FROM; - code = Scomment; - } + else if (code == Scomment) + { + state.comstyle = SYNTAX_FLAGS_COMMENT_STYLE (prev_from_syntax); + state.incomment = (SYNTAX_FLAGS_COMMENT_NESTED (prev_from_syntax) ? + 1 : -1); + state.comstr_start = prev_from; + } if (SYNTAX_FLAGS_PREFIX (prev_from_syntax)) continue; @@ -2748,7 +2858,8 @@ do { prev_from = from; \ while (from < end) { /* Some compilers can't handle this inside the switch. */ - temp = SYNTAX (FETCH_CHAR (from_byte)); + temp = FETCH_CHAR (from_byte); + temp = SYNTAX (temp); switch (temp) { case Scharquote: