X-Git-Url: https://git.netwichtig.de/gitweb/?a=blobdiff_plain;f=src%2Fhashcomp.cpp;h=4ed98712a809e1d74e632e104ad35fef2996605f;hb=78db7544d26cdeffeb2bd8045529fe90bd5d852d;hp=8d19d059a6d8e1de107e367938c8f052443f02a1;hpb=ba5a5b345cdfbc73c0faf9a6cb5fd8d96474ae41;p=user%2Fhenk%2Fcode%2Finspircd.git diff --git a/src/hashcomp.cpp b/src/hashcomp.cpp index 8d19d059a..4ed98712a 100644 --- a/src/hashcomp.cpp +++ b/src/hashcomp.cpp @@ -1,25 +1,32 @@ -/* +------------------------------------+ - * | Inspire Internet Relay Chat Daemon | - * +------------------------------------+ +/* + * InspIRCd -- Internet Relay Chat Daemon * - * InspIRCd is copyright (C) 2002-2006 ChatSpike-Dev. - * E-mail: - * - * + * Copyright (C) 2019 linuxdaemon + * Copyright (C) 2013, 2018-2019 Sadie Powell + * Copyright (C) 2013 Adam + * Copyright (C) 2012-2013, 2015-2016 Attila Molnar + * Copyright (C) 2012 Robby + * Copyright (C) 2009 Uli Schlachter + * Copyright (C) 2009 Daniel De Graaf + * Copyright (C) 2008 Robin Burchell + * Copyright (C) 2007 Dennis Friis + * Copyright (C) 2006-2007, 2010 Craig Edwards * - * Written by Craig Edwards, Craig McLure, and others. - * This program is free but copyrighted software; see - * the file COPYING for details. + * This file is part of InspIRCd. InspIRCd is free software: you can + * redistribute it and/or modify it under the terms of the GNU General Public + * License as published by the Free Software Foundation, version 2. * - * --------------------------------------------------- + * This program is distributed in the hope that it will be useful, but WITHOUT + * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS + * FOR A PARTICULAR PURPOSE. See the GNU General Public License for more + * details. + * + * You should have received a copy of the GNU General Public License + * along with this program. If not, see . */ -using namespace std; #include "inspircd.h" -#include "hashcomp.h" -#include -#define nspace __gnu_cxx /****************************************************** * @@ -30,7 +37,7 @@ using namespace std; * scene spend a lot of time debating (arguing) about * the best way to write hash functions to hash irc * nicknames, channels etc. - * We are lucky as C++ developers as hash_map does + * We are lucky as C++ developers as unordered_map does * a lot of this for us. It does intellegent memory * requests, bucketing, search functions, insertion * and deletion etc. All we have to do is write some @@ -44,268 +51,327 @@ using namespace std; * Scandanavian Comparisons: The characters [, ], \ will * be considered the lowercase of {, } and |. * - * This file also contains hashing methods for hashing - * insp_inaddr structs, we use this if we want to cache IP - * addresses. - * ******************************************************/ -using namespace std; -using namespace irc::sockets; -/* convert a string to lowercase. Note following special circumstances - * taken from RFC 1459. Many "official" server branches still hold to this - * rule so i will too; - * - * Because of IRC's scandanavian origin, the characters {}| are - * considered to be the lower case equivalents of the characters []\, - * respectively. This is a critical issue when determining the - * equivalence of two nicknames. +/** + * A case insensitive mapping of characters from upper case to lower case for + * the ASCII character set. */ -void nspace::strlower(char *n) -{ - if (n) - { - for (char* t = n; *t; t++) - *t = lowermap[(unsigned char)*t]; - } -} - -size_t nspace::hash::operator()(const insp_inaddr &a) const +unsigned const char ascii_case_insensitive_map[256] = { + 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, // 0-9 + 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, // 10-19 + 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, // 20-29 + 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, // 30-39 + 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, // 40-49 + 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, // 50-59 + 60, 61, 62, 63, 64, 97, 98, 99, 100, 101, // 60-69 + 102, 103, 104, 105, 106, 107, 108, 109, 110, 111, // 70-79 + 112, 113, 114, 115, 116, 117, 118, 119, 120, 121, // 80-89 + 122, 91, 92, 93, 94, 95, 96, 97, 98, 99, // 90-99 + 100, 101, 102, 103, 104, 105, 106, 107, 108, 109, // 100-109 + 110, 111, 112, 113, 114, 115, 116, 117, 118, 119, // 110-119 + 120, 121, 122, 123, 124, 125, 126, 127, 128, 129, // 120-129 + 130, 131, 132, 133, 134, 135, 136, 137, 138, 139, // 130-139 + 140, 141, 142, 143, 144, 145, 146, 147, 148, 149, // 140-149 + 150, 151, 152, 153, 154, 155, 156, 157, 158, 159, // 150-159 + 160, 161, 162, 163, 164, 165, 166, 167, 168, 169, // 160-169 + 170, 171, 172, 173, 174, 175, 176, 177, 178, 179, // 170-179 + 180, 181, 182, 183, 184, 185, 186, 187, 188, 189, // 180-189 + 190, 191, 192, 193, 194, 195, 196, 197, 198, 199, // 190-199 + 200, 201, 202, 203, 204, 205, 206, 207, 208, 209, // 200-209 + 210, 211, 212, 213, 214, 215, 216, 217, 218, 219, // 210-219 + 220, 221, 222, 223, 224, 225, 226, 227, 228, 229, // 220-229 + 230, 231, 232, 233, 234, 235, 236, 237, 238, 239, // 230-249 + 240, 241, 242, 243, 244, 245, 246, 247, 248, 249, // 240-249 + 250, 251, 252, 253, 254, 255, // 250-255 +}; + + + +/** + * A case insensitive mapping of characters from upper case to lower case for + * the character set of RFC 1459. This is identical to ASCII with the small + * exception of {}| being considered to be the lower case equivalents of the + * characters []\ respectively. + */ +unsigned const char rfc_case_insensitive_map[256] = { + 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, // 0-9 + 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, // 10-19 + 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, // 20-29 + 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, // 30-39 + 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, // 40-49 + 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, // 50-59 + 60, 61, 62, 63, 64, 97, 98, 99, 100, 101, // 60-69 + 102, 103, 104, 105, 106, 107, 108, 109, 110, 111, // 70-79 + 112, 113, 114, 115, 116, 117, 118, 119, 120, 121, // 80-89 + 122, 123, 124, 125, 94, 95, 96, 97, 98, 99, // 90-99 + 100, 101, 102, 103, 104, 105, 106, 107, 108, 109, // 100-109 + 110, 111, 112, 113, 114, 115, 116, 117, 118, 119, // 110-119 + 120, 121, 122, 123, 124, 125, 126, 127, 128, 129, // 120-129 + 130, 131, 132, 133, 134, 135, 136, 137, 138, 139, // 130-139 + 140, 141, 142, 143, 144, 145, 146, 147, 148, 149, // 140-149 + 150, 151, 152, 153, 154, 155, 156, 157, 158, 159, // 150-159 + 160, 161, 162, 163, 164, 165, 166, 167, 168, 169, // 160-169 + 170, 171, 172, 173, 174, 175, 176, 177, 178, 179, // 170-179 + 180, 181, 182, 183, 184, 185, 186, 187, 188, 189, // 180-189 + 190, 191, 192, 193, 194, 195, 196, 197, 198, 199, // 190-199 + 200, 201, 202, 203, 204, 205, 206, 207, 208, 209, // 200-209 + 210, 211, 212, 213, 214, 215, 216, 217, 218, 219, // 210-219 + 220, 221, 222, 223, 224, 225, 226, 227, 228, 229, // 220-229 + 230, 231, 232, 233, 234, 235, 236, 237, 238, 239, // 230-239 + 240, 241, 242, 243, 244, 245, 246, 247, 248, 249, // 240-249 + 250, 251, 252, 253, 254, 255, // 250-255 +}; + +bool irc::equals(const std::string& s1, const std::string& s2) { - size_t q; - memcpy(&q,&a,sizeof(size_t)); - return q; + const unsigned char* n1 = (const unsigned char*)s1.c_str(); + const unsigned char* n2 = (const unsigned char*)s2.c_str(); + for (; *n1 && *n2; n1++, n2++) + if (national_case_insensitive_map[*n1] != national_case_insensitive_map[*n2]) + return false; + return (national_case_insensitive_map[*n1] == national_case_insensitive_map[*n2]); } -size_t nspace::hash::operator()(const string &s) const +size_t irc::find(const std::string& haystack, const std::string& needle) { - char a[s.length()]; - size_t t = 0; - static struct hash strhash; + // The haystack can't contain the needle if it is smaller than it. + if (needle.length() > haystack.length()) + return std::string::npos; - for (const char* x = s.c_str(); *x; x++) /* Faster to do it this way than */ - a[t++] = lowermap[(unsigned char)*x]; /* Seperate strlcpy and strlower */ + // The inner loop checks the characters between haystack_last and the end of the haystack. + size_t haystack_last = haystack.length() - needle.length(); + for (size_t hpos = 0; hpos <= haystack_last; ++hpos) + { + // Check for the needle at the current haystack position. + bool found = true; + for (size_t npos = 0; npos < needle.length(); ++npos) + { + if (national_case_insensitive_map[(unsigned char)needle[npos]] != national_case_insensitive_map[(unsigned char)haystack[hpos + npos]]) + { + // Uh-oh, characters at the current haystack position don't match. + found = false; + break; + } + } - a[t] = 0; + // The entire needle was found in the haystack! + if (found) + return hpos; + } - return strhash(a); + // We didn't find anything. + return std::string::npos; } -bool irc::StrHashComp::operator()(const std::string& s1, const std::string& s2) const -{ - unsigned char* n1 = (unsigned char*)s1.c_str(); - unsigned char* n2 = (unsigned char*)s2.c_str(); - for (; *n1 && *n2; n1++, n2++) - if (lowermap[*n1] != lowermap[*n2]) - return false; - return (lowermap[*n1] == lowermap[*n2]); -} -bool irc::InAddr_HashComp::operator()(const insp_inaddr &s1, const insp_inaddr &s2) const +bool irc::insensitive_swo::operator()(const std::string& a, const std::string& b) const { -#ifdef IPV6 - for (int n = 0; n < 16; n++) - if (s2.s6_addr[n] != s1.s6_addr[n]) + const unsigned char* charmap = national_case_insensitive_map; + std::string::size_type asize = a.size(); + std::string::size_type bsize = b.size(); + std::string::size_type maxsize = std::min(asize, bsize); + + for (std::string::size_type i = 0; i < maxsize; i++) + { + unsigned char A = charmap[(unsigned char)a[i]]; + unsigned char B = charmap[(unsigned char)b[i]]; + if (A > B) return false; - return true; -#else - return (s1.s_addr == s1.s_addr); -#endif + else if (A < B) + return true; + } + return (asize < bsize); } -/****************************************************** - * - * This is the implementation of our special irc::string - * class which is a case-insensitive equivalent to - * std::string which is not only case-insensitive but - * can also do scandanavian comparisons, e.g. { = [, etc. - * - * This class depends on the const array 'lowermap'. - * - ******************************************************/ - -bool irc::irc_char_traits::eq(char c1st, char c2nd) +size_t irc::insensitive::operator()(const std::string &s) const { - return lowermap[(unsigned char)c1st] == lowermap[(unsigned char)c2nd]; + /* XXX: NO DATA COPIES! :) + * The hash function here is practically + * a copy of the one in STL's hash_fun.h, + * only with *x replaced with national_case_insensitive_map[*x]. + * This avoids a copy to use hash + */ + size_t t = 0; + for (std::string::const_iterator x = s.begin(); x != s.end(); ++x) /* ++x not x++, as its faster */ + t = 5 * t + national_case_insensitive_map[(unsigned char)*x]; + return t; } -bool irc::irc_char_traits::ne(char c1st, char c2nd) +irc::tokenstream::tokenstream(const std::string& msg, size_t start, size_t end) + : message(msg, start, end) + , position(0) { - return lowermap[(unsigned char)c1st] != lowermap[(unsigned char)c2nd]; } -bool irc::irc_char_traits::lt(char c1st, char c2nd) +bool irc::tokenstream::GetMiddle(std::string& token) { - return lowermap[(unsigned char)c1st] < lowermap[(unsigned char)c2nd]; -} + // If we are past the end of the string we can't do anything. + if (position >= message.length()) + { + token.clear(); + return false; + } -int irc::irc_char_traits::compare(const char* str1, const char* str2, size_t n) -{ - for(unsigned int i = 0; i < n; i++) + // If we can't find another separator this is the last token in the message. + size_t separator = message.find(' ', position); + if (separator == std::string::npos) { - if(lowermap[(unsigned char)*str1] > lowermap[(unsigned char)*str2]) - return 1; + token.assign(message, position, std::string::npos); + position = message.length(); + return true; + } - if(lowermap[(unsigned char)*str1] < lowermap[(unsigned char)*str2]) - return -1; + token.assign(message, position, separator - position); + position = message.find_first_not_of(' ', separator); + return true; +} - if(*str1 == 0 || *str2 == 0) - return 0; +bool irc::tokenstream::GetTrailing(std::string& token) +{ + // If we are past the end of the string we can't do anything. + if (position >= message.length()) + { + token.clear(); + return false; + } - str1++; - str2++; + // If this is true then we have a token! + if (message[position] == ':') + { + token.assign(message, position + 1, std::string::npos); + position = message.length(); + return true; } - return 0; -} -std::string operator+ (std::string& leftval, irc::string& rightval) -{ - return leftval + std::string(rightval.c_str()); + // There is no token so it must be a token. + return GetMiddle(token); } -irc::string operator+ (irc::string& leftval, std::string& rightval) +irc::sepstream::sepstream(const std::string& source, char separator, bool allowempty) + : tokens(source), sep(separator), pos(0), allow_empty(allowempty) { - return leftval + irc::string(rightval.c_str()); } -bool operator== (std::string& leftval, irc::string& rightval) +bool irc::sepstream::GetToken(std::string &token) { - return (leftval == std::string(rightval.c_str())); -} + if (this->StreamEnd()) + { + token.clear(); + return false; + } -bool operator== (irc::string& leftval, std::string& rightval) -{ - return (rightval == std::string(leftval.c_str())); + if (!this->allow_empty) + { + this->pos = this->tokens.find_first_not_of(this->sep, this->pos); + if (this->pos == std::string::npos) + { + this->pos = this->tokens.length() + 1; + token.clear(); + return false; + } + } + + size_t p = this->tokens.find(this->sep, this->pos); + if (p == std::string::npos) + p = this->tokens.length(); + + token.assign(tokens, this->pos, p - this->pos); + this->pos = p + 1; + + return true; } -const char* irc::irc_char_traits::find(const char* s1, int n, char c) +const std::string irc::sepstream::GetRemaining() { - while(n-- > 0 && lowermap[(unsigned char)*s1] != lowermap[(unsigned char)c]) - s1++; - return s1; + return !this->StreamEnd() ? this->tokens.substr(this->pos) : ""; } -/* See hashcomp.h if you care about these... */ -std::ostream& operator<<(std::ostream &os, const irc::string &str) +bool irc::sepstream::StreamEnd() { - return os << str.c_str(); + return this->pos > this->tokens.length(); } -std::istream& operator>>(std::istream &is, irc::string &str) +bool irc::sepstream::Contains(const std::string& value) { - std::string tmp; - is >> tmp; - str = tmp.c_str(); - return is; + std::string token; + while (GetToken(token)) + if (value == token) + return true; + + return false; } -irc::tokenstream::tokenstream(const std::string &source) : tokens(source), last_pushed(false) +irc::portparser::portparser(const std::string &source, bool allow_overlapped) + : sep(source), in_range(0), range_begin(0), range_end(0), overlapped(allow_overlapped) { - /* Record starting position and current position */ - last_starting_position = tokens.begin(); - n = tokens.begin(); } -irc::tokenstream::~tokenstream() +bool irc::portparser::Overlaps(long val) { + if (overlapped) + return false; + + return (!overlap_set.insert(val).second); } -const std::string irc::tokenstream::GetToken() +long irc::portparser::GetToken() { - std::string::iterator lsp = last_starting_position; - - while (n != tokens.end()) + if (in_range > 0) { - if ((last_pushed) && (*n == ':')) + in_range++; + if (in_range <= range_end) { - /* If we find a token thats not the first and starts with :, - * this is the last token on the line - */ - std::string::iterator curr = ++n; - n = tokens.end(); - return std::string(curr, tokens.end()); + if (!Overlaps(in_range)) + { + return in_range; + } + else + { + while (((Overlaps(in_range)) && (in_range <= range_end))) + in_range++; + + if (in_range <= range_end) + return in_range; + } } - - last_pushed = false; - - if ((*n == ' ') || (n+1 == tokens.end())) - { - /* If we find a space, or end of string, this is the end of a token. - */ - last_starting_position = n+1; - last_pushed = true; - - std::string strip(lsp, n+1 == tokens.end() ? n+1 : n++); - while ((strip.length()) && (strip.find_last_of(' ') == strip.length() - 1)) - strip.erase(strip.end() - 1); - - return strip; - } - - n++; + else + in_range = 0; } - return ""; -} -irc::commasepstream::commasepstream(const std::string &source) : tokens(source) -{ - last_starting_position = tokens.begin(); - n = tokens.begin(); -} + std::string x; + sep.GetToken(x); -const std::string irc::commasepstream::GetToken() -{ - std::string::iterator lsp = last_starting_position; + if (x.empty()) + return 0; - while (n != tokens.end()) + while (Overlaps(ConvToNum(x))) { - if ((*n == ',') || (n+1 == tokens.end())) - { - last_starting_position = n+1; - std::string strip = std::string(lsp, n+1 == tokens.end() ? n+1 : n++); + if (!sep.GetToken(x)) + return 0; + } - while ((strip.length()) && (strip.find_last_of(',') == strip.length() - 1)) - strip.erase(strip.end() - 1); + std::string::size_type dash = x.rfind('-'); + if (dash != std::string::npos) + { + std::string sbegin(x, 0, dash); + range_begin = ConvToNum(sbegin); + range_end = ConvToNum(x.c_str() + dash + 1); - return strip; + if ((range_begin > 0) && (range_end > 0) && (range_begin < 65536) && (range_end < 65536) && (range_begin < range_end)) + { + in_range = range_begin; + return in_range; + } + else + { + /* Assume its just the one port */ + return ConvToNum(sbegin); } - - n++; } - - return ""; -} - -irc::commasepstream::~commasepstream() -{ -} - -std::string irc::hex(const unsigned char *raw, size_t rawsz) -{ - if (!rawsz) - return ""; - - char buf[rawsz*2+1]; - size_t i; - - for (i = 0; i < rawsz; i++) + else { - sprintf (&(buf[i*2]), "%02x", raw[i]); + return ConvToNum(x); } - buf[i*2] = 0; - - return buf; } - -const char* irc::Spacify(char* n) -{ - static char x[MAXBUF]; - strlcpy(x,n,MAXBUF); - for (char* y = x; *y; y++) - if (*y == '_') - *y = ' '; - return x; -} - -