2 * InspIRCd -- Internet Relay Chat Daemon
4 * Copyright (C) 2011 Adam <Adam@anope.org>
5 * Copyright (C) 2009 Daniel De Graaf <danieldg@inspircd.org>
6 * Copyright (C) 2007, 2009 Dennis Friis <peavey@inspircd.org>
7 * Copyright (C) 2005-2009 Craig Edwards <craigedwards@brainbox.cc>
8 * Copyright (C) 2007-2008 Robin Burchell <robin+git@viroteck.net>
9 * Copyright (C) 2008 Pippijn van Steenhoven <pip88nl@gmail.com>
11 * This file is part of InspIRCd. InspIRCd is free software: you can
12 * redistribute it and/or modify it under the terms of the GNU General Public
13 * License as published by the Free Software Foundation, version 2.
15 * This program is distributed in the hope that it will be useful, but WITHOUT
16 * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS
17 * FOR A PARTICULAR PURPOSE. See the GNU General Public License for more
20 * You should have received a copy of the GNU General Public License
21 * along with this program. If not, see <http://www.gnu.org/licenses/>.
36 /*******************************************************
37 * This file contains classes and templates that deal
38 * with the comparison and hashing of 'irc strings'.
39 * An 'irc string' is a string which compares in a
40 * case insensitive manner, and as per RFC 1459 will
41 * treat [ identical to {, ] identical to }, and \
44 * Our hashing functions are designed to accept
45 * std::string and compare/hash them as type irc::string
46 * by converting them internally. This makes them
47 * backwards compatible with other code which is not
48 * aware of irc::string.
49 *******************************************************/
51 /** Seperate from the other casemap tables so that code *can* still exclusively rely on RFC casemapping
54 * This is provided as a pointer so that modules can change it to their custom mapping tables,
55 * e.g. for national character support.
57 CoreExport extern unsigned const char *national_case_insensitive_map;
59 /** A mapping of uppercase to lowercase, including scandinavian
60 * 'oddities' as specified by RFC1459, e.g. { -> [, and | -> \
62 CoreExport extern unsigned const char rfc_case_insensitive_map[256];
64 /** Case insensitive map, ASCII rules.
68 CoreExport extern unsigned const char ascii_case_insensitive_map[256];
70 /** Case sensitive (identity) map.
72 CoreExport extern unsigned const char rfc_case_sensitive_map[256];
74 template<typename T> const T& SearchAndReplace(T& text, const T& pattern, const T& replace)
77 if ((!pattern.empty()) && (!text.empty()))
79 for (std::string::size_type n = 0; n != text.length(); ++n)
81 if (text.length() >= pattern.length() && text.substr(n, pattern.length()) == pattern)
83 /* Found the pattern in the text, replace it, and advance */
84 replacement.append(replace);
85 n = n + pattern.length() - 1;
89 replacement += text[n];
97 /** The irc namespace contains a number of helper classes.
102 /** This class returns true if two strings match.
103 * Case sensitivity is ignored, and the RFC 'character set'
106 struct CoreExport StrHashComp
108 /** The operator () does the actual comparison in hash_map
110 bool operator()(const std::string& s1, const std::string& s2) const;
113 /** The irc_char_traits class is used for RFC-style comparison of strings.
114 * This class is used to implement irc::string, a case-insensitive, RFC-
115 * comparing string class.
117 struct irc_char_traits : std::char_traits<char> {
119 /** Check if two chars match.
120 * @param c1st First character
121 * @param c2nd Second character
122 * @return true if the characters are equal
124 static bool eq(char c1st, char c2nd);
126 /** Check if two chars do NOT match.
127 * @param c1st First character
128 * @param c2nd Second character
129 * @return true if the characters are unequal
131 static bool ne(char c1st, char c2nd);
133 /** Check if one char is less than another.
134 * @param c1st First character
135 * @param c2nd Second character
136 * @return true if c1st is less than c2nd
138 static bool lt(char c1st, char c2nd);
140 /** Compare two strings of size n.
141 * @param str1 First string
142 * @param str2 Second string
143 * @param n Length to compare to
144 * @return similar to strcmp, zero for equal, less than zero for str1
145 * being less and greater than zero for str1 being greater than str2.
147 static CoreExport int compare(const char* str1, const char* str2, size_t n);
149 /** Find a char within a string up to position n.
150 * @param s1 String to find in
151 * @param n Position to search up to
152 * @param c Character to search for
153 * @return Pointer to the first occurance of c in s1
155 static CoreExport const char* find(const char* s1, int n, char c);
158 /** Compose a hex string from raw data.
159 * @param raw The raw data to compose hex from
160 * @param rawsz The size of the raw data buffer
161 * @return The hex string.
163 CoreExport std::string hex(const unsigned char *raw, size_t rawsz);
165 /** This typedef declares irc::string based upon irc_char_traits.
167 typedef std::basic_string<char, irc_char_traits, std::allocator<char> > string;
169 /** irc::stringjoiner joins string lists into a string, using
170 * the given seperator string.
171 * This class can join a vector of std::string, a deque of
172 * std::string, or a const char* const* array, using overloaded
175 class CoreExport stringjoiner
185 /** Join elements of a vector, between (and including) begin and end
186 * @param seperator The string to seperate values with
187 * @param sequence One or more items to seperate
188 * @param begin The starting element in the sequence to be joined
189 * @param end The ending element in the sequence to be joined
191 stringjoiner(const std::string &seperator, const std::vector<std::string> &sequence, int begin, int end);
193 /** Join elements of a deque, between (and including) begin and end
194 * @param seperator The string to seperate values with
195 * @param sequence One or more items to seperate
196 * @param begin The starting element in the sequence to be joined
197 * @param end The ending element in the sequence to be joined
199 stringjoiner(const std::string &seperator, const std::deque<std::string> &sequence, int begin, int end);
201 /** Join elements of an array of char arrays, between (and including) begin and end
202 * @param seperator The string to seperate values with
203 * @param sequence One or more items to seperate
204 * @param begin The starting element in the sequence to be joined
205 * @param end The ending element in the sequence to be joined
207 stringjoiner(const std::string &seperator, const char* const* sequence, int begin, int end);
209 /** Get the joined sequence
210 * @return A reference to the joined string
212 std::string& GetJoined();
215 /** irc::modestacker stacks mode sequences into a list.
216 * It can then reproduce this list, clamped to a maximum of MAXMODES
219 class CoreExport modestacker
222 /** The mode sequence and its parameters
224 std::deque<std::string> sequence;
226 /** True if the mode sequence is initially adding
227 * characters, false if it is initially removing
233 /** Construct a new modestacker.
234 * @param add True if the stack is adding modes,
235 * false if it is removing them
237 modestacker(bool add);
239 /** Push a modeletter and its parameter onto the stack.
240 * No checking is performed as to if this mode actually
241 * requires a parameter. If you stack invalid mode
242 * sequences, they will be tidied if and when they are
243 * passed to a mode parser.
244 * @param modeletter The mode letter to insert
245 * @param parameter The parameter for the mode
247 void Push(char modeletter, const std::string ¶meter);
249 /** Push a modeletter without parameter onto the stack.
250 * No checking is performed as to if this mode actually
251 * requires a parameter. If you stack invalid mode
252 * sequences, they will be tidied if and when they are
253 * passed to a mode parser.
254 * @param modeletter The mode letter to insert
256 void Push(char modeletter);
258 /** Push a '+' symbol onto the stack.
262 /** Push a '-' symbol onto the stack.
266 /** Return zero or more elements which form the
267 * mode line. This will be clamped to a max of
268 * MAXMODES items (MAXMODES-1 mode parameters and
269 * one mode sequence string), and max_line_size
270 * characters. As specified below, this function
271 * should be called in a loop until it returns zero,
272 * indicating there are no more modes to return.
273 * @param result The vector to populate. This will not
274 * be cleared before it is used.
275 * @param max_line_size The maximum size of the line
276 * to build, in characters, seperate to MAXMODES.
277 * @return The number of elements in the deque.
278 * The function should be called repeatedly until it
279 * returns 0, in case there are multiple lines of
280 * mode changes to be obtained.
282 int GetStackedLine(std::vector<std::string> &result, int max_line_size = 360);
286 /** irc::tokenstream reads a string formatted as per RFC1459 and RFC2812.
287 * It will split the string into 'tokens' each containing one parameter
289 * For instance, if it is instantiated with the string:
290 * "PRIVMSG #test :foo bar baz qux"
291 * then each successive call to tokenstream::GetToken() will return
292 * "PRIVMSG", "#test", "foo bar baz qux", "".
293 * Note that if the whole string starts with a colon this is not taken
294 * to mean the string is all one parameter, and the first item in the
295 * list will be ":item". This is to allow for parsing 'source' fields
298 class CoreExport tokenstream
306 /** Last position of a seperator token
308 std::string::iterator last_starting_position;
310 /** Current string position
312 std::string::iterator n;
314 /** True if the last value was an ending value
319 /** Create a tokenstream and fill it with the provided data
321 tokenstream(const std::string &source);
327 /** Fetch the next token from the stream as a std::string
328 * @param token The next token available, or an empty string if none remain
329 * @return True if tokens are left to be read, false if the last token was just retrieved.
331 bool GetToken(std::string &token);
333 /** Fetch the next token from the stream as an irc::string
334 * @param token The next token available, or an empty string if none remain
335 * @return True if tokens are left to be read, false if the last token was just retrieved.
337 bool GetToken(irc::string &token);
339 /** Fetch the next token from the stream as an integer
340 * @param token The next token available, or undefined if none remain
341 * @return True if tokens are left to be read, false if the last token was just retrieved.
343 bool GetToken(int &token);
345 /** Fetch the next token from the stream as a long integer
346 * @param token The next token available, or undefined if none remain
347 * @return True if tokens are left to be read, false if the last token was just retrieved.
349 bool GetToken(long &token);
352 /** irc::sepstream allows for splitting token seperated lists.
353 * Each successive call to sepstream::GetToken() returns
354 * the next token, until none remain, at which point the method returns
357 class CoreExport sepstream
363 /** Last position of a seperator token
365 std::string::iterator last_starting_position;
366 /** Current string position
368 std::string::iterator n;
373 /** Create a sepstream and fill it with the provided data
375 sepstream(const std::string &source, char seperator);
379 virtual ~sepstream();
381 /** Fetch the next token from the stream
382 * @param token The next token from the stream is placed here
383 * @return True if tokens still remain, false if there are none left
385 virtual bool GetToken(std::string &token);
387 /** Fetch the entire remaining stream, without tokenizing
388 * @return The remaining part of the stream
390 virtual const std::string GetRemaining();
392 /** Returns true if the end of the stream has been reached
393 * @return True if the end of the stream has been reached, otherwise false
395 virtual bool StreamEnd();
398 /** A derived form of sepstream, which seperates on commas
400 class CoreExport commasepstream : public sepstream
403 /** Initialize with comma seperator
405 commasepstream(const std::string &source) : sepstream(source, ',')
410 /** A derived form of sepstream, which seperates on spaces
412 class CoreExport spacesepstream : public sepstream
415 /** Initialize with space seperator
417 spacesepstream(const std::string &source) : sepstream(source, ' ')
422 /** The portparser class seperates out a port range into integers.
423 * A port range may be specified in the input string in the form
424 * "6660,6661,6662-6669,7020". The end of the stream is indicated by
425 * a return value of 0 from portparser::GetToken(). If you attempt
426 * to specify an illegal range (e.g. one where start >= end, or
427 * start or end < 0) then GetToken() will return the first element
428 * of the pair of numbers.
430 class CoreExport portparser
434 /** Used to split on commas
438 /** Current position in a range of ports
442 /** Starting port in a range of ports
446 /** Ending port in a range of ports
450 /** Allow overlapped port ranges
454 /** Used to determine overlapping of ports
455 * without O(n) algorithm being used
457 std::set<long> overlap_set;
459 /** Returns true if val overlaps an existing range
461 bool Overlaps(long val);
464 /** Create a portparser and fill it with the provided data
465 * @param source The source text to parse from
466 * @param allow_overlapped Allow overlapped ranges
468 portparser(const std::string &source, bool allow_overlapped = true);
470 /** Fetch the next token from the stream
471 * @return The next port number is returned, or 0 if none remain
476 /** Turn _ characters in a string into spaces
477 * @param n String to translate
478 * @return The new value with _ translated to space.
480 CoreExport const char* Spacify(const char* n);
484 /** Hash an irc::string using RFC1459 case sensitivity rules
485 * @param s A string to hash
486 * @return The hash value
488 size_t CoreExport operator()(const irc::string &s) const;
492 /* Define operators for using >> and << with irc::string to an ostream on an istream. */
493 /* This was endless fun. No. Really. */
494 /* It was also the first core change Ommeh made, if anyone cares */
496 /** Operator << for irc::string
498 inline std::ostream& operator<<(std::ostream &os, const irc::string &str) { return os << str.c_str(); }
500 /** Operator >> for irc::string
502 inline std::istream& operator>>(std::istream &is, irc::string &str)
510 /* Define operators for + and == with irc::string to std::string for easy assignment
515 inline std::string operator+ (std::string& leftval, irc::string& rightval)
517 return leftval + std::string(rightval.c_str());
520 /* Define operators for + and == with irc::string to std::string for easy assignment
525 inline irc::string operator+ (irc::string& leftval, std::string& rightval)
527 return leftval + irc::string(rightval.c_str());
530 /* Define operators for + and == with irc::string to std::string for easy assignment
535 inline bool operator== (const std::string& leftval, const irc::string& rightval)
537 return (leftval.c_str() == rightval);
540 /* Define operators for + and == with irc::string to std::string for easy assignment
545 inline bool operator== (const irc::string& leftval, const std::string& rightval)
547 return (leftval == rightval.c_str());
550 /* Define operators != for irc::string to std::string for easy comparison
552 inline bool operator!= (const irc::string& leftval, const std::string& rightval)
554 return !(leftval == rightval.c_str());
557 /* Define operators != for std::string to irc::string for easy comparison
559 inline bool operator!= (const std::string& leftval, const irc::string& rightval)
561 return !(leftval.c_str() == rightval);
564 /** Assign an irc::string to a std::string.
566 inline std::string assign(const irc::string &other) { return other.c_str(); }
568 /** Assign a std::string to an irc::string.
570 inline irc::string assign(const std::string &other) { return other.c_str(); }
572 /** Trim the leading and trailing spaces from a std::string.
574 inline std::string& trim(std::string &str)
576 std::string::size_type start = str.find_first_not_of(" ");
577 std::string::size_type end = str.find_last_not_of(" ");
578 if (start == std::string::npos || end == std::string::npos)
581 str = str.substr(start, end-start+1);
586 /** Hashing stuff is totally different on vc++'s hash_map implementation, so to save a buttload of
587 * \#ifdefs we'll just do it all at once. Except, of course, with TR1, when it's the same as GCC.
589 BEGIN_HASHMAP_NAMESPACE
591 /** Hashing function to hash irc::string
593 #if defined(_WIN32) && !defined(HAS_TR1_UNORDERED)
594 template<> class CoreExport hash_compare<irc::string, std::less<irc::string> >
597 enum { bucket_size = 4, min_buckets = 8 }; /* Got these numbers from the CRT source, if anyone wants to change them feel free. */
599 /** Compare two irc::string values for hashing in hash_map
601 bool operator()(const irc::string & s1, const irc::string & s2) const
603 if(s1.length() != s2.length()) return true;
604 return (irc::irc_char_traits::compare(s1.c_str(), s2.c_str(), (size_t)s1.length()) < 0);
607 /** Hash an irc::string value for hash_map
609 size_t operator()(const irc::string & s) const;
612 template<> class CoreExport hash_compare<std::string, std::less<std::string> >
615 enum { bucket_size = 4, min_buckets = 8 }; /* Again, from the CRT source */
617 /** Compare two std::string values for hashing in hash_map
619 bool operator()(const std::string & s1, const std::string & s2) const
621 if(s1.length() != s2.length()) return true;
622 return (irc::irc_char_traits::compare(s1.c_str(), s2.c_str(), (size_t)s1.length()) < 0);
625 /** Hash a std::string using RFC1459 case sensitivity rules
626 * @param s A string to hash
627 * @return The hash value
629 size_t operator()(const std::string & s) const;
633 /* XXX FIXME: Implement a hash function overriding std::string's that works with TR1! */
635 #ifdef HASHMAP_DEPRECATED
638 CoreExport template<> struct hash<std::string>
641 size_t CoreExport operator()(const std::string &s) const;
646 /** Convert a string to lower case respecting RFC1459
647 * @param n A string to lowercase
649 void strlower(char *n);
651 END_HASHMAP_NAMESPACE