2 This file is part of GNUnet
3 (C) 2012 Christian Grothoff (and other contributing authors)
5 GNUnet is free software; you can redistribute it and/or modify
6 it under the terms of the GNU General Public License as published
7 by the Free Software Foundation; either version 3, or (at your
8 option) any later version.
10 GNUnet is distributed in the hope that it will be useful, but
11 WITHOUT ANY WARRANTY; without even the implied warranty of
12 MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
13 General Public License for more details.
15 You should have received a copy of the GNU General Public License
16 along with GNUnet; see the file COPYING. If not, write to the
17 Free Software Foundation, Inc., 59 Temple Place - Suite 330,
18 Boston, MA 02111-1307, USA.
21 * @file include/gnunet_regex_lib.h
22 * @brief library to parse regular expressions into dfa
23 * @author Maximilian Szengel
27 #ifndef GNUNET_REGEX_LIB_H
28 #define GNUNET_REGEX_LIB_H
30 #include "gnunet_util_lib.h"
35 #if 0 /* keep Emacsens' auto-indent happy */
42 * Constant for how many bits the initial string regex should have.
44 #define GNUNET_REGEX_INITIAL_BYTES 24
48 * Maximum regex string length for use with GNUNET_REGEX_ipv4toregex
50 #define GNUNET_REGEX_IPV4_REGEXLEN 32 + 6
54 * Maximum regex string length for use with GNUNET_REGEX_ipv6toregex
56 #define GNUNET_REGEX_IPV6_REGEXLEN 128 + 6
60 * Automaton (NFA/DFA) representation.
62 struct GNUNET_REGEX_Automaton;
66 * Edge representation.
68 struct GNUNET_REGEX_Edge
71 * Label of the edge. FIXME: might want to not consume exactly multiples of 8 bits, need length?
76 * Destionation of the edge.
78 struct GNUNET_HashCode destination;
83 * Construct DFA for the given 'regex' of length 'len'.
85 * Path compression means, that for example a DFA o -> a -> b -> c -> o will be
86 * compressed to o -> abc -> o. Note that this parameter influences the
87 * non-determinism of states of the resulting NFA in the DHT (number of outgoing
88 * edges with the same label). For example for an application that stores IPv4
89 * addresses as bitstrings it could make sense to limit the path compression to
92 * @param regex regular expression string.
93 * @param len length of the regular expression.
94 * @param max_path_len limit the path compression length to the
95 * given value. If set to 1, no path compression is applied. Set to 0 for
96 * maximal possible path compression (generally not desireable).
97 * @return DFA, needs to be freed using GNUNET_REGEX_automaton_destroy.
99 struct GNUNET_REGEX_Automaton *
100 GNUNET_REGEX_construct_dfa (const char *regex, const size_t len,
105 * Free the memory allocated by constructing the GNUNET_REGEX_Automaton.
108 * @param a automaton to be destroyed.
111 GNUNET_REGEX_automaton_destroy (struct GNUNET_REGEX_Automaton *a);
115 * Options for graph creation function
116 * GNUNET_REGEX_automaton_save_graph.
118 enum GNUNET_REGEX_GraphSavingOptions
121 * Default. Do nothing special.
123 GNUNET_REGEX_GRAPH_DEFAULT = 0,
126 * The generated graph will include extra information such as the NFA states
127 * that were used to generate the DFA state.
129 GNUNET_REGEX_GRAPH_VERBOSE = 1,
132 * Enable graph coloring. Will color each SCC in a different color.
134 GNUNET_REGEX_GRAPH_COLORING = 2
139 * Save the given automaton as a GraphViz dot file.
141 * @param a the automaton to be saved.
142 * @param filename where to save the file.
143 * @param options options for graph generation that include coloring or verbose
147 GNUNET_REGEX_automaton_save_graph (struct GNUNET_REGEX_Automaton *a,
148 const char *filename,
149 enum GNUNET_REGEX_GraphSavingOptions options);
153 * Evaluates the given 'string' against the given compiled regex.
155 * @param a automaton.
156 * @param string string to check.
158 * @return 0 if string matches, non 0 otherwise.
161 GNUNET_REGEX_eval (struct GNUNET_REGEX_Automaton *a,
166 * Get the first key for the given 'input_string'. This hashes
167 * the first x bits of the 'input_string'.
169 * @param input_string string.
170 * @param string_len length of the 'input_string'.
171 * @param key pointer to where to write the hash code.
173 * @return number of bits of 'input_string' that have been consumed
174 * to construct the key
177 GNUNET_REGEX_get_first_key (const char *input_string, size_t string_len,
178 struct GNUNET_HashCode * key);
182 * Check if the given 'proof' matches the given 'key'.
184 * @param proof partial regex of a state.
185 * @param key hash of a state.
187 * @return GNUNET_OK if the proof is valid for the given key.
190 GNUNET_REGEX_check_proof (const char *proof,
191 const struct GNUNET_HashCode *key);
195 * Iterator callback function.
197 * @param cls closure.
198 * @param key hash for current state.
199 * @param proof proof for current state.
200 * @param accepting GNUNET_YES if this is an accepting state, GNUNET_NO if not.
201 * @param num_edges number of edges leaving current state.
202 * @param edges edges leaving current state.
204 typedef void (*GNUNET_REGEX_KeyIterator)(void *cls,
205 const struct GNUNET_HashCode *key,
208 unsigned int num_edges,
209 const struct GNUNET_REGEX_Edge *edges);
213 * Iterate over all edges starting from start state of automaton 'a'. Calling
214 * iterator for each edge.
216 * @param a automaton.
217 * @param iterator iterator called for each edge.
218 * @param iterator_cls closure.
221 GNUNET_REGEX_iterate_all_edges (struct GNUNET_REGEX_Automaton *a,
222 GNUNET_REGEX_KeyIterator iterator,
227 * Create a regex in 'rxstr' from the given 'ip' and 'netmask'.
229 * @param ip IPv4 representation.
230 * @param netmask netmask for the ip.
231 * @param rxstr generated regex, must be at least GNUNET_REGEX_IPV4_REGEXLEN
235 GNUNET_REGEX_ipv4toregex (const struct in_addr *ip, const char *netmask,
240 * Create a regex in 'rxstr' from the given 'ipv6' and 'prefixlen'.
242 * @param ipv6 IPv6 representation.
243 * @param prefixlen length of the ipv6 prefix.
244 * @param rxstr generated regex, must be at least GNUNET_REGEX_IPV6_REGEXLEN
248 GNUNET_REGEX_ipv6toregex (const struct in6_addr *ipv6,
249 unsigned int prefixlen, char *rxstr);
252 #if 0 /* keep Emacsens' auto-indent happy */
259 /* end of gnunet_regex_lib.h */