js/src/jsregexp.h

   1 /* -*- Mode: C; tab-width: 8; indent-tabs-mode: nil; c-basic-offset: 4 -*-
   2  *
   3  * ***** BEGIN LICENSE BLOCK *****
   4  * Version: MPL 1.1/GPL 2.0/LGPL 2.1
   5  *
   6  * The contents of this file are subject to the Mozilla Public License Version
   7  * 1.1 (the "License"); you may not use this file except in compliance with
   8  * the License. You may obtain a copy of the License at
   9  * http://www.mozilla.org/MPL/
  10  *
  11  * Software distributed under the License is distributed on an "AS IS" basis,
  12  * WITHOUT WARRANTY OF ANY KIND, either express or implied. See the License
  13  * for the specific language governing rights and limitations under the
  14  * License.
  15  *
  16  * The Original Code is Mozilla Communicator client code, released
  17  * March 31, 1998.
  18  *
  19  * The Initial Developer of the Original Code is
  20  * Netscape Communications Corporation.
  21  * Portions created by the Initial Developer are Copyright (C) 1998
  22  * the Initial Developer. All Rights Reserved.
  23  *
  24  * Contributor(s):
  25  *
  26  * Alternatively, the contents of this file may be used under the terms of
  27  * either of the GNU General Public License Version 2 or later (the "GPL"),
  28  * or the GNU Lesser General Public License Version 2.1 or later (the "LGPL"),
  29  * in which case the provisions of the GPL or the LGPL are applicable instead
  30  * of those above. If you wish to allow use of your version of this file only
  31  * under the terms of either the GPL or the LGPL, and not to allow others to
  32  * use your version of this file under the terms of the MPL, indicate your
  33  * decision by deleting the provisions above and replace them with the notice
  34  * and other provisions required by the GPL or the LGPL. If you do not delete
  35  * the provisions above, a recipient may use your version of this file under
  36  * the terms of any one of the MPL, the GPL or the LGPL.
  37  *
  38  * ***** END LICENSE BLOCK ***** */
  39
  40 #ifndef jsregexp_h___
  41 #define jsregexp_h___
  42 /*
  43  * JS regular expression interface.
  44  */
  45 #include <stddef.h>
  46 #include "jspubtd.h"
  47 #include "jsstr.h"
  48
  49 #ifdef JS_THREADSAFE
  50 #include "jsdhash.h"
  51 #endif
  52
  53 JS_BEGIN_EXTERN_C
  54
  55 struct JSRegExpStatics {
  56     JSString    *input;         /* input string to match (perl $_, GC root) */
  57     JSBool      multiline;      /* whether input contains newlines (perl $*) */
  58     uint16      parenCount;     /* number of valid elements in parens[] */
  59     uint16      moreLength;     /* number of allocated elements in moreParens */
  60     JSSubString parens[9];      /* last set of parens matched (perl $1, $2) */
  61     JSSubString *moreParens;    /* null or realloc'd vector for $10, etc. */
  62     JSSubString lastMatch;      /* last string matched (perl $&) */
  63     JSSubString lastParen;      /* last paren matched (perl $+) */
  64     JSSubString leftContext;    /* input to left of last match (perl $`) */
  65     JSSubString rightContext;   /* input to right of last match (perl $') */
  66 };
  67
  68 /*
  69  * This struct holds a bitmap representation of a class from a regexp.
  70  * There's a list of these referenced by the classList field in the JSRegExp
  71  * struct below. The initial state has startIndex set to the offset in the
  72  * original regexp source of the beginning of the class contents. The first
  73  * use of the class converts the source representation into a bitmap.
  74  *
  75  */
  76 typedef struct RECharSet {
  77     JSPackedBool    converted;
  78     JSPackedBool    sense;
  79     uint16          length;
  80     union {
  81         uint8       *bits;
  82         struct {
  83             size_t  startIndex;
  84             size_t  length;
  85         } src;
  86     } u;
  87 } RECharSet;
  88
  89 /*
  90  * This macro is safe because moreParens is guaranteed to be allocated and big
  91  * enough to hold parenCount, or else be null when parenCount is 0.
  92  */
  93 #define REGEXP_PAREN_SUBSTRING(res, num)                                      \
  94     (((jsuint)(num) < (jsuint)(res)->parenCount)                              \
  95      ? ((jsuint)(num) < 9)                                                    \
  96        ? &(res)->parens[num]                                                  \
  97        : &(res)->moreParens[(num) - 9]                                        \
  98      : &js_EmptySubString)
  99
 100 typedef struct RENode RENode;
 101
 102 struct JSRegExp {
 103     jsrefcount   nrefs;         /* reference count */
 104     uint16       flags;         /* flags, see jsapi.h's JSREG_* defines */
 105     size_t       parenCount;    /* number of parenthesized submatches */
 106     size_t       classCount;    /* count [...] bitmaps */
 107     RECharSet    *classList;    /* list of [...] bitmaps */
 108     JSString     *source;       /* locked source string, sans // */
 109     jsbytecode   program[1];    /* regular expression bytecode */
 110 };
 111
 112 extern JSRegExp *
 113 js_NewRegExp(JSContext *cx, JSTokenStream *ts,
 114              JSString *str, uintN flags, JSBool flat);
 115
 116 extern JSRegExp *
 117 js_NewRegExpOpt(JSContext *cx, JSString *str, JSString *opt, JSBool flat);
 118
 119 #define HOLD_REGEXP(cx, re) JS_ATOMIC_INCREMENT(&(re)->nrefs)
 120 #define DROP_REGEXP(cx, re) js_DestroyRegExp(cx, re)
 121
 122 extern void
 123 js_DestroyRegExp(JSContext *cx, JSRegExp *re);
 124
 125 /*
 126  * Execute re on input str at *indexp, returning null in *rval on mismatch.
 127  * On match, return true if test is true, otherwise return an array object.
 128  * Update *indexp and cx->regExpStatics always on match.
 129  */
 130 extern JSBool
 131 js_ExecuteRegExp(JSContext *cx, JSRegExp *re, JSString *str, size_t *indexp,
 132                  JSBool test, jsval *rval);
 133
 134 /*
 135  * These two add and remove GC roots, respectively, so their calls must be
 136  * well-ordered.
 137  */
 138 extern JSBool
 139 js_InitRegExpStatics(JSContext *cx, JSRegExpStatics *res);
 140
 141 extern void
 142 js_FreeRegExpStatics(JSContext *cx, JSRegExpStatics *res);
 143
 144 #define VALUE_IS_REGEXP(cx, v)                                                \
 145     (JSVAL_IS_OBJECT(v) && JSVAL_TO_OBJECT(v) &&                              \
 146      OBJ_GET_CLASS(cx, JSVAL_TO_OBJECT(v)) == &js_RegExpClass)
 147
 148 extern JSClass js_RegExpClass;
 149
 150 enum regexp_tinyid {
 151     REGEXP_SOURCE       = -1,
 152     REGEXP_GLOBAL       = -2,
 153     REGEXP_IGNORE_CASE  = -3,
 154     REGEXP_LAST_INDEX   = -4,
 155     REGEXP_MULTILINE    = -5,
 156     REGEXP_STICKY       = -6
 157 };
 158
 159 extern JSObject *
 160 js_InitRegExpClass(JSContext *cx, JSObject *obj);
 161
 162 /*
 163  * Export js_regexp_toString to the decompiler.
 164  */
 165 extern JSBool
 166 js_regexp_toString(JSContext *cx, JSObject *obj, jsval *vp);
 167
 168 /*
 169  * Create, serialize/deserialize, or clone a RegExp object.
 170  */
 171 extern JSObject *
 172 js_NewRegExpObject(JSContext *cx, JSTokenStream *ts,
 173                    jschar *chars, size_t length, uintN flags);
 174
 175 extern JSBool
 176 js_XDRRegExp(JSXDRState *xdr, JSObject **objp);
 177
 178 extern JSObject *
 179 js_CloneRegExpObject(JSContext *cx, JSObject *obj, JSObject *parent);
 180
 181 /*
 182  * Get and set the per-object (clone or clone-parent) lastIndex slot.
 183  */
 184 extern JSBool
 185 js_GetLastIndex(JSContext *cx, JSObject *obj, jsdouble *lastIndex);
 186
 187 extern JSBool
 188 js_SetLastIndex(JSContext *cx, JSObject *obj, jsdouble lastIndex);
 189
 190 JS_END_EXTERN_C
 191
 192 #endif /* jsregexp_h___ */