OpenVAS Libraries  4.0+rc3.SVN
nasl_regex.h
1 /* Definitions for data structures and routines for the regular
2  expression library, version 0.12.
3 
4  Copyright (C) 1985, 1989, 1990, 1991, 1992, 1993 Free Software Foundation, Inc.
5 
6  This program is free software; you can redistribute it and/or modify
7  it under the terms of the GNU General Public License as published by
8  the Free Software Foundation, version 2.
9 
10  This program is distributed in the hope that it will be useful,
11  but WITHOUT ANY WARRANTY; without even the implied warranty of
12  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
13  GNU General Public License for more details.
14 
15  You should have received a copy of the GNU General Public License
16  along with this program; if not, write to the Free Software
17  Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA. */
18 
19 #ifdef HAVE_REGEX_SUPPORT
20 #include <regex.h>
21 #else
22 #ifndef __REGEXP_LIBRARY_H__
23 #define __REGEXP_LIBRARY_H__
24 
25 /* POSIX says that <sys/types.h> must be included (by the caller) before
26  <regex.h>. */
27 
28 #include <sys/types.h> /* for size_t */
29 #ifdef VMS
30 /* VMS doesn't have `size_t' in <sys/types.h>, even though POSIX says it
31  should be there. */
32 #include <stddef.h>
33 #endif
34 
35 
36 /* The following bits are used to determine the regexp syntax we
37  recognize. The set/not-set meanings are chosen so that Emacs syntax
38  remains the value 0. The bits are given in alphabetical order, and
39  the definitions shifted by one from the previous bit; thus, when we
40  add or remove a bit, only one other definition need change. */
41 typedef unsigned reg_syntax_t;
42 
43 /* If this bit is not set, then \ inside a bracket expression is literal.
44  If set, then such a \ quotes the following character. */
45 #define RE_BACKSLASH_ESCAPE_IN_LISTS (1)
46 
47 /* If this bit is not set, then + and ? are operators, and \+ and \? are
48  literals.
49  If set, then \+ and \? are operators and + and ? are literals. */
50 #define RE_BK_PLUS_QM (RE_BACKSLASH_ESCAPE_IN_LISTS << 1)
51 
52 /* If this bit is set, then character classes are supported. They are:
53  [:alpha:], [:upper:], [:lower:], [:digit:], [:alnum:], [:xdigit:],
54  [:space:], [:print:], [:punct:], [:graph:], and [:cntrl:].
55  If not set, then character classes are not supported. */
56 #define RE_CHAR_CLASSES (RE_BK_PLUS_QM << 1)
57 
58 /* If this bit is set, then ^ and $ are always anchors (outside bracket
59  expressions, of course).
60  If this bit is not set, then it depends:
61  ^ is an anchor if it is at the beginning of a regular
62  expression or after an open-group or an alternation operator;
63  $ is an anchor if it is at the end of a regular expression, or
64  before a close-group or an alternation operator.
65 
66  This bit could be (re)combined with RE_CONTEXT_INDEP_OPS, because
67  POSIX draft 11.2 says that * etc. in leading positions is undefined.
68  We already implemented a previous draft which made those constructs
69  invalid, though, so we haven't changed the code back. */
70 #define RE_CONTEXT_INDEP_ANCHORS (RE_CHAR_CLASSES << 1)
71 
72 /* If this bit is set, then special characters are always special
73  regardless of where they are in the pattern.
74  If this bit is not set, then special characters are special only in
75  some contexts; otherwise they are ordinary. Specifically,
76  * + ? and intervals are only special when not after the beginning,
77  open-group, or alternation operator. */
78 #define RE_CONTEXT_INDEP_OPS (RE_CONTEXT_INDEP_ANCHORS << 1)
79 
80 /* If this bit is set, then *, +, ?, and { cannot be first in an re or
81  immediately after an alternation or begin-group operator. */
82 #define RE_CONTEXT_INVALID_OPS (RE_CONTEXT_INDEP_OPS << 1)
83 
84 /* If this bit is set, then . matches newline.
85  If not set, then it doesn't. */
86 #define RE_DOT_NEWLINE (RE_CONTEXT_INVALID_OPS << 1)
87 
88 /* If this bit is set, then . doesn't match NUL.
89  If not set, then it does. */
90 #define RE_DOT_NOT_NULL (RE_DOT_NEWLINE << 1)
91 
92 /* If this bit is set, nonmatching lists [^...] do not match newline.
93  If not set, they do. */
94 #define RE_HAT_LISTS_NOT_NEWLINE (RE_DOT_NOT_NULL << 1)
95 
96 /* If this bit is set, either \{...\} or {...} defines an
97  interval, depending on RE_NO_BK_BRACES.
98  If not set, \{, \}, {, and } are literals. */
99 #define RE_INTERVALS (RE_HAT_LISTS_NOT_NEWLINE << 1)
100 
101 /* If this bit is set, +, ? and | aren't recognized as operators.
102  If not set, they are. */
103 #define RE_LIMITED_OPS (RE_INTERVALS << 1)
104 
105 /* If this bit is set, newline is an alternation operator.
106  If not set, newline is literal. */
107 #define RE_NEWLINE_ALT (RE_LIMITED_OPS << 1)
108 
109 /* If this bit is set, then `{...}' defines an interval, and \{ and \}
110  are literals.
111  If not set, then `\{...\}' defines an interval. */
112 #define RE_NO_BK_BRACES (RE_NEWLINE_ALT << 1)
113 
114 /* If this bit is set, (...) defines a group, and \( and \) are literals.
115  If not set, \(...\) defines a group, and ( and ) are literals. */
116 #define RE_NO_BK_PARENS (RE_NO_BK_BRACES << 1)
117 
118 /* If this bit is set, then <digit> matches <digit>.
119  If not set, then <digit> is a back-reference. */
120 #define RE_NO_BK_REFS (RE_NO_BK_PARENS << 1)
121 
122 /* If this bit is set, then | is an alternation operator, and \| is literal.
123  If not set, then \| is an alternation operator, and | is literal. */
124 #define RE_NO_BK_VBAR (RE_NO_BK_REFS << 1)
125 
126 /* If this bit is set, then an ending range point collating higher
127  than the starting range point, as in [z-a], is invalid.
128  If not set, then when ending range point collates higher than the
129  starting range point, the range is ignored. */
130 #define RE_NO_EMPTY_RANGES (RE_NO_BK_VBAR << 1)
131 
132 /* If this bit is set, then an unmatched ) is ordinary.
133  If not set, then an unmatched ) is invalid. */
134 #define RE_UNMATCHED_RIGHT_PAREN_ORD (RE_NO_EMPTY_RANGES << 1)
135 
136 /* This global variable defines the particular regexp syntax to use (for
137  some interfaces). When a regexp is compiled, the syntax used is
138  stored in the pattern buffer, so changing this does not affect
139  already-compiled regexps. */
140 extern reg_syntax_t re_syntax_options;
141 
142 /* Define combinations of the above bits for the standard possibilities.
143  (The [[[ comments delimit what gets put into the Texinfo file, so
144  don't delete them!) */
145 /* [[[begin syntaxes]]] */
146 #define RE_SYNTAX_EMACS 0
147 
148 #define RE_SYNTAX_AWK \
149  (RE_BACKSLASH_ESCAPE_IN_LISTS | RE_DOT_NOT_NULL \
150  | RE_NO_BK_PARENS | RE_NO_BK_REFS \
151  | RE_NO_BK_VBAR | RE_NO_EMPTY_RANGES \
152  | RE_UNMATCHED_RIGHT_PAREN_ORD)
153 
154 #define RE_SYNTAX_POSIX_AWK \
155  (RE_SYNTAX_POSIX_EXTENDED | RE_BACKSLASH_ESCAPE_IN_LISTS)
156 
157 #define RE_SYNTAX_GREP \
158  (RE_BK_PLUS_QM | RE_CHAR_CLASSES \
159  | RE_HAT_LISTS_NOT_NEWLINE | RE_INTERVALS \
160  | RE_NEWLINE_ALT)
161 
162 #define RE_SYNTAX_EGREP \
163  (RE_CHAR_CLASSES | RE_CONTEXT_INDEP_ANCHORS \
164  | RE_CONTEXT_INDEP_OPS | RE_HAT_LISTS_NOT_NEWLINE \
165  | RE_NEWLINE_ALT | RE_NO_BK_PARENS \
166  | RE_NO_BK_VBAR)
167 
168 #define RE_SYNTAX_POSIX_EGREP \
169  (RE_SYNTAX_EGREP | RE_INTERVALS | RE_NO_BK_BRACES)
170 
171 /* P1003.2/D11.2, section 4.20.7.1, lines 5078ff. */
172 #define RE_SYNTAX_ED RE_SYNTAX_POSIX_BASIC
173 
174 #define RE_SYNTAX_SED RE_SYNTAX_POSIX_BASIC
175 
176 /* Syntax bits common to both basic and extended POSIX regex syntax. */
177 #define _RE_SYNTAX_POSIX_COMMON \
178  (RE_CHAR_CLASSES | RE_DOT_NEWLINE | RE_DOT_NOT_NULL \
179  | RE_INTERVALS | RE_NO_EMPTY_RANGES)
180 
181 #define RE_SYNTAX_POSIX_BASIC \
182  (_RE_SYNTAX_POSIX_COMMON | RE_BK_PLUS_QM)
183 
184 /* Differs from ..._POSIX_BASIC only in that RE_BK_PLUS_QM becomes
185  RE_LIMITED_OPS, i.e., \? \+ \| are not recognized. Actually, this
186  isn't minimal, since other operators, such as \`, aren't disabled. */
187 #define RE_SYNTAX_POSIX_MINIMAL_BASIC \
188  (_RE_SYNTAX_POSIX_COMMON | RE_LIMITED_OPS)
189 
190 #define RE_SYNTAX_POSIX_EXTENDED \
191  (_RE_SYNTAX_POSIX_COMMON | RE_CONTEXT_INDEP_ANCHORS \
192  | RE_CONTEXT_INDEP_OPS | RE_NO_BK_BRACES \
193  | RE_NO_BK_PARENS | RE_NO_BK_VBAR \
194  | RE_UNMATCHED_RIGHT_PAREN_ORD)
195 
196 /* Differs from ..._POSIX_EXTENDED in that RE_CONTEXT_INVALID_OPS
197  replaces RE_CONTEXT_INDEP_OPS and RE_NO_BK_REFS is added. */
198 #define RE_SYNTAX_POSIX_MINIMAL_EXTENDED \
199  (_RE_SYNTAX_POSIX_COMMON | RE_CONTEXT_INDEP_ANCHORS \
200  | RE_CONTEXT_INVALID_OPS | RE_NO_BK_BRACES \
201  | RE_NO_BK_PARENS | RE_NO_BK_REFS \
202  | RE_NO_BK_VBAR | RE_UNMATCHED_RIGHT_PAREN_ORD)
203 /* [[[end syntaxes]]] */
204 
205 /* Maximum number of duplicates an interval can allow. Some systems
206  (erroneously) define this in other header files, but we want our
207  value, so remove any previous define. */
208 #ifdef RE_DUP_MAX
209 #undef RE_DUP_MAX
210 #endif
211 #define RE_DUP_MAX ((1 << 15) - 1)
212 
213 
214 /* POSIX `cflags' bits (i.e., information for `regcomp'). */
215 
216 /* If this bit is set, then use extended regular expression syntax.
217  If not set, then use basic regular expression syntax. */
218 #define REG_EXTENDED 1
219 
220 /* If this bit is set, then ignore case when matching.
221  If not set, then case is significant. */
222 #define REG_ICASE (REG_EXTENDED << 1)
223 
224 /* If this bit is set, then anchors do not match at newline
225  characters in the string.
226  If not set, then anchors do match at newlines. */
227 #define REG_NEWLINE (REG_ICASE << 1)
228 
229 /* If this bit is set, then report only success or fail in regexec.
230  If not set, then returns differ between not matching and errors. */
231 #define REG_NOSUB (REG_NEWLINE << 1)
232 
233 
234 /* POSIX `eflags' bits (i.e., information for regexec). */
235 
236 /* If this bit is set, then the beginning-of-line operator doesn't match
237  the beginning of the string (presumably because it's not the
238  beginning of a line).
239  If not set, then the beginning-of-line operator does match the
240  beginning of the string. */
241 #define REG_NOTBOL 1
242 
243 /* Like REG_NOTBOL, except for the end-of-line. */
244 #define REG_NOTEOL (1 << 1)
245 
246 
247 /* If any error codes are removed, changed, or added, update the
248  `re_error_msg' table in regex.c. */
249 typedef enum
250 {
251  REG_NOERROR = 0, /* Success. */
252  REG_NOMATCH, /* Didn't find a match (for regexec). */
253 
254  /* POSIX regcomp return error codes. (In the order listed in the
255  standard.) */
256  REG_BADPAT, /* Invalid pattern. */
257  REG_ECOLLATE, /* Not implemented. */
258  REG_ECTYPE, /* Invalid character class name. */
259  REG_EESCAPE, /* Trailing backslash. */
260  REG_ESUBREG, /* Invalid back reference. */
261  REG_EBRACK, /* Unmatched left bracket. */
262  REG_EPAREN, /* Parenthesis imbalance. */
263  REG_EBRACE, /* Unmatched \{. */
264  REG_BADBR, /* Invalid contents of \{\}. */
265  REG_ERANGE, /* Invalid range end. */
266  REG_ESPACE, /* Ran out of memory. */
267  REG_BADRPT, /* No preceding re for repetition op. */
268 
269  /* Error codes we've added. */
270  REG_EEND, /* Premature end. */
271  REG_ESIZE, /* Compiled pattern bigger than 2^16 bytes. */
272  REG_ERPAREN /* Unmatched ) or \); not returned from regcomp. */
273 } reg_errcode_t;
274 
275 /* This data structure represents a compiled pattern. Before calling
276  the pattern compiler, the fields `buffer', `allocated', `fastmap',
277  `translate', and `no_sub' can be set. After the pattern has been
278  compiled, the `re_nsub' field is available. All other fields are
279  private to the regex routines. */
280 
282 {
283 /* [[[begin pattern_buffer]]] */
284  /* Space that holds the compiled pattern. It is declared as
285  `unsigned char *' because its elements are
286  sometimes used as array indexes. */
287  unsigned char *buffer;
288 
289  /* Number of bytes to which `buffer' points. */
290  unsigned long allocated;
291 
292  /* Number of bytes actually used in `buffer'. */
293  unsigned long used;
294 
295  /* Syntax setting with which the pattern was compiled. */
296  reg_syntax_t syntax;
297 
298  /* Pointer to a fastmap, if any, otherwise zero. re_search uses
299  the fastmap, if there is one, to skip over impossible
300  starting points for matches. */
301  char *fastmap;
302 
303  /* Either a translate table to apply to all characters before
304  comparing them, or zero for no translation. The translation
305  is applied to a pattern when it is compiled and to a string
306  when it is matched. */
307  char *translate;
308 
309  /* Number of subexpressions found by the compiler. */
310  size_t re_nsub;
311 
312  /* Zero if this pattern cannot match the empty string, one else.
313  Well, in truth it's used only in `re_search_2', to see
314  whether or not we should use the fastmap, so we don't set
315  this absolutely perfectly; see `re_compile_fastmap' (the
316  `duplicate' case). */
317  unsigned can_be_null : 1;
318 
319  /* If REGS_UNALLOCATED, allocate space in the `regs' structure
320  for `max (RE_NREGS, re_nsub + 1)' groups.
321  If REGS_REALLOCATE, reallocate space if necessary.
322  If REGS_FIXED, use what's there. */
323 #define REGS_UNALLOCATED 0
324 #define REGS_REALLOCATE 1
325 #define REGS_FIXED 2
326  unsigned regs_allocated : 2;
327 
328  /* Set to zero when `regex_compile' compiles a pattern; set to one
329  by `re_compile_fastmap' if it updates the fastmap. */
330  unsigned fastmap_accurate : 1;
331 
332  /* If set, `re_match_2' does not return information about
333  subexpressions. */
334  unsigned no_sub : 1;
335 
336  /* If set, a beginning-of-line anchor doesn't match at the
337  beginning of the string. */
338  unsigned not_bol : 1;
339 
340  /* Similarly for an end-of-line anchor. */
341  unsigned not_eol : 1;
342 
343  /* If true, an anchor at a newline matches. */
344  unsigned newline_anchor : 1;
345 
346  char * orig;
347 /* [[[end pattern_buffer]]] */
348 };
349 
350 typedef struct re_pattern_buffer regex_t;
351 
352 
353 /* search.c (search_buffer) in Emacs needs this one opcode value. It is
354  defined both in `regex.c' and here. */
355 #define RE_EXACTN_VALUE 1
356 
357 /* Type for byte offsets within the string. POSIX mandates this. */
358 typedef int regoff_t;
359 
360 
361 /* This is the structure we store register match data in. See
362  regex.texinfo for a full description of what registers match. */
364 {
365  unsigned num_regs;
366  regoff_t *start;
367  regoff_t *end;
368 };
369 
370 
371 /* If `regs_allocated' is REGS_UNALLOCATED in the pattern buffer,
372  `re_match_2' returns information about at least this many registers
373  the first time a `regs' structure is passed. */
374 #ifndef RE_NREGS
375 #define RE_NREGS 30
376 #endif
377 
378 
379 /* POSIX specification for registers. Aside from the different names than
380  `re_registers', POSIX uses an array of structures, instead of a
381  structure of arrays. */
382 typedef struct
383 {
384  regoff_t rm_so; /* Byte offset from string's start to substring's start. */
385  regoff_t rm_eo; /* Byte offset from string's start to substring's end. */
386 } regmatch_t;
387 
388 /* Declarations for routines. */
389 
390 /* To avoid duplicating every routine declaration -- once with a
391  prototype (if we are ANSI), and once without (if we aren't) -- we
392  use the following macro to declare argument types. This
393  unfortunately clutters up the declarations a bit, but I think it's
394  worth it. */
395 
396 #if __STDC__
397 
398 #define _RE_ARGS(args) args
399 
400 #else /* not __STDC__ */
401 
402 #define _RE_ARGS(args) ()
403 
404 #endif /* not __STDC__ */
405 
406 /* Sets the current default syntax to SYNTAX, and return the old syntax.
407  You can also simply assign to the `re_syntax_options' variable. */
408 extern reg_syntax_t re_set_syntax _RE_ARGS ((reg_syntax_t syntax));
409 
410 /* Compile the regular expression PATTERN, with length LENGTH
411  and syntax given by the global `re_syntax_options', into the buffer
412  BUFFER. Return NULL if successful, and an error string if not. */
413 extern const char *re_compile_pattern
414  _RE_ARGS ((const char *pattern, int length,
415  struct re_pattern_buffer *buffer));
416 
417 
418 /* Compile a fastmap for the compiled pattern in BUFFER; used to
419  accelerate searches. Return 0 if successful and -2 if was an
420  internal error. */
421 extern int re_compile_fastmap _RE_ARGS ((struct re_pattern_buffer *buffer));
422 
423 
424 /* Search in the string STRING (with length LENGTH) for the pattern
425  compiled into BUFFER. Start searching at position START, for RANGE
426  characters. Return the starting position of the match, -1 for no
427  match, or -2 for an internal error. Also return register
428  information in REGS (if REGS and BUFFER->no_sub are nonzero). */
429 extern int re_search
430  _RE_ARGS ((struct re_pattern_buffer *buffer, const char *string,
431  int length, int start, int range, struct re_registers *regs));
432 
433 
434 /* Like `re_search', but search in the concatenation of STRING1 and
435  STRING2. Also, stop searching at index START + STOP. */
436 extern int re_search_2
437  _RE_ARGS ((struct re_pattern_buffer *buffer, const char *string1,
438  int length1, const char *string2, int length2,
439  int start, int range, struct re_registers *regs, int stop));
440 
441 
442 /* Like `re_search', but return how many characters in STRING the regexp
443  in BUFFER matched, starting at position START. */
444 extern int re_match
445  _RE_ARGS ((struct re_pattern_buffer *buffer, const char *string,
446  int length, int start, struct re_registers *regs));
447 
448 
449 /* Relates to `re_match' as `re_search_2' relates to `re_search'. */
450 extern int re_match_2
451  _RE_ARGS ((struct re_pattern_buffer *buffer, const char *string1,
452  int length1, const char *string2, int length2,
453  int start, struct re_registers *regs, int stop));
454 
455 
456 /* Set REGS to hold NUM_REGS registers, storing them in STARTS and
457  ENDS. Subsequent matches using BUFFER and REGS will use this memory
458  for recording register information. STARTS and ENDS must be
459  allocated with malloc, and must each be at least `NUM_REGS * sizeof
460  (regoff_t)' bytes long.
461 
462  If NUM_REGS == 0, then subsequent matches should allocate their own
463  register data.
464 
465  Unless this function is called, the first search or match using
466  PATTERN_BUFFER will allocate its own register data, without
467  freeing the old data. */
468 extern void re_set_registers
469  _RE_ARGS ((struct re_pattern_buffer *buffer, struct re_registers *regs,
470  unsigned num_regs, regoff_t *starts, regoff_t *ends));
471 
472 /* 4.2 bsd compatibility. */
473 extern char *re_comp _RE_ARGS ((const char *));
474 extern int re_exec _RE_ARGS ((const char *));
475 
476 /* POSIX compatibility. */
477 extern int regcomp _RE_ARGS ((regex_t *preg, const char *pattern, int cflags));
478 extern int regexec
479  _RE_ARGS ((const regex_t *preg, const char *string, size_t nmatch,
480  regmatch_t pmatch[], int eflags));
481 extern size_t regerror
482  _RE_ARGS ((int errcode, const regex_t *preg, char *errbuf,
483  size_t errbuf_size));
484 extern void regfree _RE_ARGS ((regex_t *preg));
485 
486 #endif /* not __REGEXP_LIBRARY_H__ */
487 
488 /*
489 Local variables:
490 make-backup-files: t
491 version-control: t
492 trim-versions-without-asking: nil
493 End:
494 */
495 #endif
496 
497 
498 extern int nasl_regcomp (regex_t *preg, const char *pattern, int cflags);
499 extern int nasl_regexec (const regex_t *preg, const char *string, size_t nmatch,
500  regmatch_t pmatch[], int eflags);
501 extern size_t nasl_regerror (int errcode, const regex_t *preg, char *errbuf,
502  size_t errbuf_size);
503 extern reg_syntax_t nasl_re_set_syntax (reg_syntax_t syntax);
504 extern char * nasl_regorig(regex_t * preg);
505 extern void nasl_regfree (regex_t *preg);