1 #ifndef REGPARSE_H
2 #define REGPARSE_H
3 /**********************************************************************
4   regparse.h -  Oniguruma (regular expression library)
5 **********************************************************************/
6 /*-
7  * Copyright (c) 2002-2007  K.Kosako  <sndgk393 AT ybb DOT ne DOT jp>
8  * All rights reserved.
9  *
10  * (C) Copyright 2015 Hewlett Packard Enterprise Development LP<BR>
11  *
12  * Redistribution and use in source and binary forms, with or without
13  * modification, are permitted provided that the following conditions
14  * are met:
15  * 1. Redistributions of source code must retain the above copyright
16  *    notice, this list of conditions and the following disclaimer.
17  * 2. Redistributions in binary form must reproduce the above copyright
18  *    notice, this list of conditions and the following disclaimer in the
19  *    documentation and/or other materials provided with the distribution.
20  *
21  * THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND
22  * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
23  * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
24  * ARE DISCLAIMED.  IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE
25  * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
26  * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
27  * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
28  * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
29  * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
30  * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
31  * SUCH DAMAGE.
32  */
33 
34 #include "regint.h"
35 
36 /* node type */
37 #define NT_STR         0
38 #define NT_CCLASS      1
39 #define NT_CTYPE       2
40 #define NT_CANY        3
41 #define NT_BREF        4
42 #define NT_QTFR        5
43 #define NT_ENCLOSE     6
44 #define NT_ANCHOR      7
45 #define NT_LIST        8
46 #define NT_ALT         9
47 #define NT_CALL       10
48 
49 /* node type bit */
50 #define NTYPE2BIT(type)      (1<<(type))
51 
52 #define BIT_NT_STR        NTYPE2BIT(NT_STR)
53 #define BIT_NT_CCLASS     NTYPE2BIT(NT_CCLASS)
54 #define BIT_NT_CTYPE      NTYPE2BIT(NT_CTYPE)
55 #define BIT_NT_CANY       NTYPE2BIT(NT_CANY)
56 #define BIT_NT_BREF       NTYPE2BIT(NT_BREF)
57 #define BIT_NT_QTFR       NTYPE2BIT(NT_QTFR)
58 #define BIT_NT_ENCLOSE    NTYPE2BIT(NT_ENCLOSE)
59 #define BIT_NT_ANCHOR     NTYPE2BIT(NT_ANCHOR)
60 #define BIT_NT_LIST       NTYPE2BIT(NT_LIST)
61 #define BIT_NT_ALT        NTYPE2BIT(NT_ALT)
62 #define BIT_NT_CALL       NTYPE2BIT(NT_CALL)
63 
64 #define IS_NODE_TYPE_SIMPLE(type) \
65   ((NTYPE2BIT(type) & (BIT_NT_STR | BIT_NT_CCLASS | BIT_NT_CTYPE |\
66                        BIT_NT_CANY | BIT_NT_BREF)) != 0)
67 
68 #define NTYPE(node)             ((node)->u.base.type)
69 #define SET_NTYPE(node, ntype)   (node)->u.base.type = (ntype)
70 
71 #define NSTR(node)         (&((node)->u.str))
72 #define NCCLASS(node)      (&((node)->u.cclass))
73 #define NCTYPE(node)       (&((node)->u.ctype))
74 #define NBREF(node)        (&((node)->u.bref))
75 #define NQTFR(node)        (&((node)->u.qtfr))
76 #define NENCLOSE(node)     (&((node)->u.enclose))
77 #define NANCHOR(node)      (&((node)->u.anchor))
78 #define NCONS(node)        (&((node)->u.cons))
79 #define NCALL(node)        (&((node)->u.call))
80 
81 #define NCAR(node)         (NCONS(node)->car)
82 #define NCDR(node)         (NCONS(node)->cdr)
83 
84 
85 
86 #define ANCHOR_ANYCHAR_STAR_MASK (ANCHOR_ANYCHAR_STAR | ANCHOR_ANYCHAR_STAR_ML)
87 #define ANCHOR_END_BUF_MASK      (ANCHOR_END_BUF | ANCHOR_SEMI_END_BUF)
88 
89 #define ENCLOSE_MEMORY           (1<<0)
90 #define ENCLOSE_OPTION           (1<<1)
91 #define ENCLOSE_STOP_BACKTRACK   (1<<2)
92 
93 #define NODE_STR_MARGIN         16
94 #define NODE_STR_BUF_SIZE       24  /* sizeof(CClassNode) - sizeof(int)*4 */
95 #define NODE_BACKREFS_SIZE       6
96 
97 #define NSTR_RAW                (1<<0) /* by backslashed number */
98 #define NSTR_AMBIG              (1<<1)
99 #define NSTR_DONT_GET_OPT_INFO  (1<<2)
100 
101 #define NSTRING_LEN(node)             ((int)((node)->u.str.end - (node)->u.str.s))
102 #define NSTRING_SET_RAW(node)          (node)->u.str.flag |= NSTR_RAW
103 #define NSTRING_CLEAR_RAW(node)        (node)->u.str.flag &= ~NSTR_RAW
104 #define NSTRING_SET_AMBIG(node)        (node)->u.str.flag |= NSTR_AMBIG
105 #define NSTRING_SET_DONT_GET_OPT_INFO(node) \
106   (node)->u.str.flag |= NSTR_DONT_GET_OPT_INFO
107 #define NSTRING_IS_RAW(node)          (((node)->u.str.flag & NSTR_RAW)   != 0)
108 #define NSTRING_IS_AMBIG(node)        (((node)->u.str.flag & NSTR_AMBIG) != 0)
109 #define NSTRING_IS_DONT_GET_OPT_INFO(node) \
110   (((node)->u.str.flag & NSTR_DONT_GET_OPT_INFO) != 0)
111 
112 #define BACKREFS_P(br) \
113   (IS_NOT_NULL((br)->back_dynamic) ? (br)->back_dynamic : (br)->back_static);
114 
115 #define NQ_TARGET_ISNOT_EMPTY     0
116 #define NQ_TARGET_IS_EMPTY        1
117 #define NQ_TARGET_IS_EMPTY_MEM    2
118 #define NQ_TARGET_IS_EMPTY_REC    3
119 
120 /* status bits */
121 #define NST_MIN_FIXED             (1<<0)
122 #define NST_MAX_FIXED             (1<<1)
123 #define NST_CLEN_FIXED            (1<<2)
124 #define NST_MARK1                 (1<<3)
125 #define NST_MARK2                 (1<<4)
126 #define NST_MEM_BACKREFED         (1<<5)
127 #define NST_STOP_BT_SIMPLE_REPEAT (1<<6)
128 #define NST_RECURSION             (1<<7)
129 #define NST_CALLED                (1<<8)
130 #define NST_ADDR_FIXED            (1<<9)
131 #define NST_NAMED_GROUP           (1<<10)
132 #define NST_NAME_REF              (1<<11)
133 #define NST_IN_REPEAT             (1<<12) /* STK_REPEAT is nested in stack. */
134 #define NST_NEST_LEVEL            (1<<13)
135 #define NST_BY_NUMBER             (1<<14) /* {n,m} */
136 
137 #define SET_ENCLOSE_STATUS(node,f)      (node)->u.enclose.state |=  (f)
138 #define CLEAR_ENCLOSE_STATUS(node,f)    (node)->u.enclose.state &= ~(f)
139 
140 #define IS_ENCLOSE_CALLED(en)          (((en)->state & NST_CALLED)        != 0)
141 #define IS_ENCLOSE_ADDR_FIXED(en)      (((en)->state & NST_ADDR_FIXED)    != 0)
142 #define IS_ENCLOSE_RECURSION(en)       (((en)->state & NST_RECURSION)     != 0)
143 #define IS_ENCLOSE_MARK1(en)           (((en)->state & NST_MARK1)         != 0)
144 #define IS_ENCLOSE_MARK2(en)           (((en)->state & NST_MARK2)         != 0)
145 #define IS_ENCLOSE_MIN_FIXED(en)       (((en)->state & NST_MIN_FIXED)     != 0)
146 #define IS_ENCLOSE_MAX_FIXED(en)       (((en)->state & NST_MAX_FIXED)     != 0)
147 #define IS_ENCLOSE_CLEN_FIXED(en)      (((en)->state & NST_CLEN_FIXED)    != 0)
148 #define IS_ENCLOSE_STOP_BT_SIMPLE_REPEAT(en) \
149     (((en)->state & NST_STOP_BT_SIMPLE_REPEAT) != 0)
150 #define IS_ENCLOSE_NAMED_GROUP(en)     (((en)->state & NST_NAMED_GROUP)   != 0)
151 
152 #define SET_CALL_RECURSION(node)       (node)->u.call.state |= NST_RECURSION
153 #define IS_CALL_RECURSION(cn)          (((cn)->state & NST_RECURSION)  != 0)
154 #define IS_CALL_NAME_REF(cn)           (((cn)->state & NST_NAME_REF)   != 0)
155 #define IS_BACKREF_NAME_REF(bn)        (((bn)->state & NST_NAME_REF)   != 0)
156 #define IS_BACKREF_NEST_LEVEL(bn)      (((bn)->state & NST_NEST_LEVEL) != 0)
157 #define IS_QUANTIFIER_IN_REPEAT(qn)    (((qn)->state & NST_IN_REPEAT)  != 0)
158 #define IS_QUANTIFIER_BY_NUMBER(qn)    (((qn)->state & NST_BY_NUMBER)  != 0)
159 
160 #define CALLNODE_REFNUM_UNDEF  -1
161 
162 typedef struct {
163   NodeBase base;
164   UChar* s;
165   UChar* end;
166   unsigned int flag;
167   int    capa;    /* (allocated size - 1) or 0: use buf[] */
168   UChar  buf[NODE_STR_BUF_SIZE];
169 } StrNode;
170 
171 typedef struct {
172   NodeBase base;
173   int state;
174   struct _Node* target;
175   int lower;
176   int upper;
177   int greedy;
178   int target_empty_info;
179   struct _Node* head_exact;
180   struct _Node* next_head_exact;
181   int is_refered;     /* include called node. don't eliminate even if {0} */
182 #ifdef USE_COMBINATION_EXPLOSION_CHECK
183   int comb_exp_check_num;  /* 1,2,3...: check,  0: no check  */
184 #endif
185 } QtfrNode;
186 
187 typedef struct {
188   NodeBase base;
189   int state;
190   int type;
191   int regnum;
192   OnigOptionType option;
193   struct _Node*  target;
194   AbsAddrType    call_addr;
195   /* for multiple call reference */
196   OnigDistance min_len; /* min length (byte) */
197   OnigDistance max_len; /* max length (byte) */
198   int char_len;         /* character length  */
199   int opt_count;        /* referenced count in optimize_node_left() */
200 } EncloseNode;
201 
202 #ifdef USE_SUBEXP_CALL
203 
204 typedef struct {
205   int           offset;
206   struct _Node* target;
207 } UnsetAddr;
208 
209 typedef struct {
210   int        num;
211   int        alloc;
212   UnsetAddr* us;
213 } UnsetAddrList;
214 
215 typedef struct {
216   NodeBase base;
217   int     state;
218   int     group_num;
219   UChar*  name;
220   UChar*  name_end;
221   struct _Node*  target;  /* EncloseNode : ENCLOSE_MEMORY */
222   UnsetAddrList* unset_addr_list;
223 } CallNode;
224 
225 #endif
226 
227 typedef struct {
228   NodeBase base;
229   int  state;
230   int  back_num;
231   int  back_static[NODE_BACKREFS_SIZE];
232   int* back_dynamic;
233   int  nest_level;
234 } BRefNode;
235 
236 typedef struct {
237   NodeBase base;
238   int type;
239   struct _Node* target;
240   int char_len;
241 } AnchorNode;
242 
243 typedef struct {
244   NodeBase base;
245   struct _Node* car;
246   struct _Node* cdr;
247 } ConsAltNode;
248 
249 typedef struct {
250   NodeBase base;
251   int ctype;
252   int not;
253 } CtypeNode;
254 
255 typedef struct _Node {
256   union {
257     NodeBase     base;
258     StrNode      str;
259     CClassNode   cclass;
260     QtfrNode     qtfr;
261     EncloseNode  enclose;
262     BRefNode     bref;
263     AnchorNode   anchor;
264     ConsAltNode  cons;
265     CtypeNode    ctype;
266 #ifdef USE_SUBEXP_CALL
267     CallNode     call;
268 #endif
269   } u;
270 } Node;
271 
272 
273 #define NULL_NODE  ((Node* )0)
274 
275 #define SCANENV_MEMNODES_SIZE               8
276 #define SCANENV_MEM_NODES(senv)   \
277  (IS_NOT_NULL((senv)->mem_nodes_dynamic) ? \
278     (senv)->mem_nodes_dynamic : (senv)->mem_nodes_static)
279 
280 typedef struct {
281   OnigOptionType   option;
282   OnigCaseFoldType case_fold_flag;
283   OnigEncoding     enc;
284   OnigSyntaxType*  syntax;
285   BitStatusType    capture_history;
286   BitStatusType    bt_mem_start;
287   BitStatusType    bt_mem_end;
288   BitStatusType    backrefed_mem;
289   UChar*           pattern;
290   UChar*           pattern_end;
291   UChar*           error;
292   UChar*           error_end;
293   regex_t*         reg;       /* for reg->names only */
294   int              num_call;
295 #ifdef USE_SUBEXP_CALL
296   UnsetAddrList*   unset_addr_list;
297 #endif
298   int              num_mem;
299 #ifdef USE_NAMED_GROUP
300   int              num_named;
301 #endif
302   int              mem_alloc;
303   Node*            mem_nodes_static[SCANENV_MEMNODES_SIZE];
304   Node**           mem_nodes_dynamic;
305 #ifdef USE_COMBINATION_EXPLOSION_CHECK
306   int num_comb_exp_check;
307   int comb_exp_max_regnum;
308   int curr_max_regnum;
309   int has_recursion;
310 #endif
311 } ScanEnv;
312 
313 
314 #define IS_SYNTAX_OP(syn, opm)    (((syn)->op  & (opm)) != 0)
315 #define IS_SYNTAX_OP2(syn, opm)   (((syn)->op2 & (opm)) != 0)
316 #define IS_SYNTAX_BV(syn, bvm)    (((syn)->behavior & (bvm)) != 0)
317 
318 #ifdef USE_NAMED_GROUP
319 typedef struct {
320   int new_val;
321 } GroupNumRemap;
322 
323 extern int    onig_renumber_name_table P_((regex_t* reg, GroupNumRemap* map));
324 #endif
325 
326 extern int    onig_strncmp P_((const UChar* s1, const UChar* s2, int n));
327 extern void   onig_strcpy P_((UChar* dest, const UChar* src, const UChar* end));
328 extern void   onig_scan_env_set_error_string P_((ScanEnv* env, int ecode, UChar* arg, UChar* arg_end));
329 extern int    onig_scan_unsigned_number P_((UChar** src, const UChar* end, OnigEncoding enc));
330 extern void   onig_reduce_nested_quantifier P_((Node* pnode, Node* cnode));
331 extern void   onig_node_conv_to_str_node P_((Node* node, int raw));
332 extern int    onig_node_str_cat P_((Node* node, const UChar* s, const UChar* end));
333 extern int    onig_node_str_set P_((Node* node, const UChar* s, const UChar* end));
334 extern void   onig_node_free P_((Node* node));
335 extern Node*  onig_node_new_enclose P_((int type));
336 extern Node*  onig_node_new_anchor P_((int type));
337 extern Node*  onig_node_new_str P_((const UChar* s, const UChar* end));
338 extern Node*  onig_node_new_list P_((Node* left, Node* right));
339 extern Node*  onig_node_list_add P_((Node* list, Node* x));
340 extern Node*  onig_node_new_alt P_((Node* left, Node* right));
341 extern void   onig_node_str_clear P_((Node* node));
342 extern int    onig_free_node_list P_((void));
343 extern int    onig_names_free P_((regex_t* reg));
344 extern int    onig_parse_make_tree P_((Node** root, const UChar* pattern, const UChar* end, regex_t* reg, ScanEnv* env));
345 extern int    onig_free_shared_cclass_table P_((void));
346 
347 #ifdef ONIG_DEBUG
348 #ifdef USE_NAMED_GROUP
349 extern int onig_print_names(FILE*, regex_t*);
350 #endif
351 #endif
352 
353 #endif /* REGPARSE_H */
354