X-Git-Url: https://git.netwichtig.de/gitweb/?a=blobdiff_plain;f=src%2Fhashcomp.cpp;h=5ca032229a71c6e06138954e21f07a850cbc7e3a;hb=d494fa6e094e85cd29235e995fb2b447d6e1f168;hp=0d036d528b9447c811d396a376df7452d4159553;hpb=a3fb932831ca09b2a931616f1701ea39429356c2;p=user%2Fhenk%2Fcode%2Finspircd.git diff --git a/src/hashcomp.cpp b/src/hashcomp.cpp index 0d036d528..5ca032229 100644 --- a/src/hashcomp.cpp +++ b/src/hashcomp.cpp @@ -1,21 +1,32 @@ -/* +------------------------------------+ - * | Inspire Internet Relay Chat Daemon | - * +------------------------------------+ +/* + * InspIRCd -- Internet Relay Chat Daemon * - * InspIRCd: (C) 2002-2009 InspIRCd Development Team - * See: http://www.inspircd.org/wiki/index.php/Credits + * 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) 2005-2007, 2010 Craig Edwards * - * 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 . */ -/* $Core */ #include "inspircd.h" -#include "hashcomp.h" -#include "hash_map.h" /****************************************************** * @@ -26,7 +37,7 @@ * 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 @@ -42,60 +53,78 @@ * ******************************************************/ -/* 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. - */ -void nspace::strlower(char *n) -{ - if (n) - { - for (char* t = n; *t; t++) - *t = national_case_insensitive_map[(unsigned char)*t]; - } -} -#ifndef WIN32 - #ifdef HASHMAP_DEPRECATED - size_t nspace::insensitive::operator()(const std::string &s) const - #else - size_t nspace::hash::operator()(const std::string &s) const - #endif -#else - size_t nspace::hash_compare >::operator()(const std::string &s) const -#endif -{ - /* 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 - */ - register 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; -} - - -#ifndef WIN32 -size_t nspace::hash::operator()(const irc::string &s) const -#else -size_t nspace::hash_compare >::operator()(const irc::string &s) const -#endif -{ - register size_t t = 0; - for (irc::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::StrHashComp::operator()(const std::string& s1, const std::string& s2) const +/** + * A case insensitive mapping of characters from upper case to lower case for + * the ASCII character set. + */ +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) { const unsigned char* n1 = (const unsigned char*)s1.c_str(); const unsigned char* n2 = (const unsigned char*)s2.c_str(); @@ -105,339 +134,186 @@ bool irc::StrHashComp::operator()(const std::string& s1, const std::string& s2) return (national_case_insensitive_map[*n1] == national_case_insensitive_map[*n2]); } -/****************************************************** - * - * 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 'national_case_insensitive_map'. - * - ******************************************************/ - -bool irc::irc_char_traits::eq(char c1st, char c2nd) -{ - return national_case_insensitive_map[(unsigned char)c1st] == national_case_insensitive_map[(unsigned char)c2nd]; -} - -bool irc::irc_char_traits::ne(char c1st, char c2nd) +size_t irc::find(const std::string& haystack, const std::string& needle) { - return national_case_insensitive_map[(unsigned char)c1st] != national_case_insensitive_map[(unsigned char)c2nd]; -} + // The haystack can't contain the needle if it is smaller than it. + if (needle.length() > haystack.length()) + return std::string::npos; -bool irc::irc_char_traits::lt(char c1st, char c2nd) -{ - return national_case_insensitive_map[(unsigned char)c1st] < national_case_insensitive_map[(unsigned char)c2nd]; -} - -int irc::irc_char_traits::compare(const char* str1, const char* str2, size_t n) -{ - for(unsigned int i = 0; i < n; i++) + // 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) { - if(national_case_insensitive_map[(unsigned char)*str1] > national_case_insensitive_map[(unsigned char)*str2]) - return 1; - - if(national_case_insensitive_map[(unsigned char)*str1] < national_case_insensitive_map[(unsigned char)*str2]) - return -1; - - if(*str1 == 0 || *str2 == 0) - return 0; + // 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; + } + } - str1++; - str2++; + // The entire needle was found in the haystack! + if (found) + return hpos; } - return 0; -} - -const char* irc::irc_char_traits::find(const char* s1, int n, char c) -{ - while(n-- > 0 && national_case_insensitive_map[(unsigned char)*s1] != national_case_insensitive_map[(unsigned char)c]) - s1++; - return s1; -} -irc::tokenstream::tokenstream(const std::string &source) : tokens(source), last_pushed(false) -{ - /* Record starting position and current position */ - last_starting_position = tokens.begin(); - n = tokens.begin(); + // We didn't find anything. + return std::string::npos; } -irc::tokenstream::~tokenstream() -{ -} -bool irc::tokenstream::GetToken(std::string &token) +bool irc::insensitive_swo::operator()(const std::string& a, const std::string& b) const { - std::string::iterator lsp = last_starting_position; + 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); - while (n != tokens.end()) + for (std::string::size_type i = 0; i < maxsize; i++) { - /** Skip multi space, converting " " into " " - */ - while ((n+1 != tokens.end()) && (*n == ' ') && (*(n+1) == ' ')) - n++; - - if ((last_pushed) && (*n == ':')) - { - /* 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(); - token = std::string(curr, tokens.end()); + unsigned char A = charmap[(unsigned char)a[i]]; + unsigned char B = charmap[(unsigned char)b[i]]; + if (A > B) + return false; + else if (A < B) return true; - } - - 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); - - token = strip; - return !token.empty(); - } - - n++; } - token.clear(); - return false; -} - -bool irc::tokenstream::GetToken(irc::string &token) -{ - std::string stdstring; - bool returnval = GetToken(stdstring); - token = assign(stdstring); - return returnval; -} - -bool irc::tokenstream::GetToken(int &token) -{ - std::string tok; - bool returnval = GetToken(tok); - token = ConvToInt(tok); - return returnval; + return (asize < bsize); } -bool irc::tokenstream::GetToken(long &token) +size_t irc::insensitive::operator()(const std::string &s) const { - std::string tok; - bool returnval = GetToken(tok); - token = ConvToInt(tok); - return returnval; + /* 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; } -irc::sepstream::sepstream(const std::string &source, char seperator) : tokens(source), sep(seperator) +irc::tokenstream::tokenstream(const std::string& msg, size_t start, size_t end) + : message(msg, start, end) + , position(0) { - last_starting_position = tokens.begin(); - n = tokens.begin(); } -bool irc::sepstream::GetToken(std::string &token) +bool irc::tokenstream::GetMiddle(std::string& token) { - std::string::iterator lsp = last_starting_position; - - while (n != tokens.end()) + // If we are past the end of the string we can't do anything. + if (position >= message.length()) { - if ((*n == sep) || (n+1 == tokens.end())) - { - last_starting_position = n+1; - token = std::string(lsp, n+1 == tokens.end() ? n+1 : n++); - - while ((token.length()) && (token.find_last_of(sep) == token.length() - 1)) - token.erase(token.end() - 1); - - if (token.empty()) - n++; - - return n == tokens.end() ? false : true; - } - - n++; + token.clear(); + return false; } - token = ""; - return false; -} - -const std::string irc::sepstream::GetRemaining() -{ - return std::string(n, tokens.end()); -} - -bool irc::sepstream::StreamEnd() -{ - return ((n + 1) == tokens.end()); -} - -irc::sepstream::~sepstream() -{ -} - -std::string irc::hex(const unsigned char *raw, size_t rawsz) -{ - if (!rawsz) - return ""; - - /* EWW! This used to be using sprintf, which is WAY inefficient. -Special */ - - const char *hex = "0123456789abcdef"; - static char hexbuf[MAXBUF]; - - size_t i, j; - for (i = 0, j = 0; j < rawsz; ++j) + // 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) { - hexbuf[i++] = hex[raw[j] / 16]; - hexbuf[i++] = hex[raw[j] % 16]; + token.assign(message, position, std::string::npos); + position = message.length(); + return true; } - hexbuf[i] = 0; - - return hexbuf; -} - -CoreExport const char* irc::Spacify(const char* n) -{ - static char x[MAXBUF]; - strlcpy(x,n,MAXBUF); - for (char* y = x; *y; y++) - if (*y == '_') - *y = ' '; - return x; -} - -irc::modestacker::modestacker(InspIRCd* Instance, bool add) : ServerInstance(Instance), adding(add) -{ - sequence.clear(); - sequence.push_back(""); + token.assign(message, position, separator - position); + position = message.find_first_not_of(' ', separator); + return true; } -void irc::modestacker::Push(char modeletter, const std::string ¶meter) +bool irc::tokenstream::GetTrailing(std::string& token) { - *(sequence.begin()) += modeletter; - sequence.push_back(parameter); -} + // If we are past the end of the string we can't do anything. + if (position >= message.length()) + { + token.clear(); + return false; + } -void irc::modestacker::Push(char modeletter) -{ - this->Push(modeletter,""); -} + // 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; + } -void irc::modestacker::PushPlus() -{ - this->Push('+',""); + // There is no token so it must be a token. + return GetMiddle(token); } -void irc::modestacker::PushMinus() +irc::sepstream::sepstream(const std::string& source, char separator, bool allowempty) + : tokens(source), sep(separator), pos(0), allow_empty(allowempty) { - this->Push('-',""); } -int irc::modestacker::GetStackedLine(std::deque &result, int max_line_size) +bool irc::sepstream::GetToken(std::string &token) { - if (sequence.empty()) + if (this->StreamEnd()) { - result.clear(); - return 0; + token.clear(); + return false; } - int n = 0; - int size = 1; /* Account for initial +/- char */ - int nextsize = 0; - result.clear(); - result.push_back(adding ? "+" : "-"); - - if (sequence.size() > 1) - nextsize = sequence[1].length() + 2; - - while (!sequence[0].empty() && (sequence.size() > 1) && (result.size() < ServerInstance->Config->Limits.MaxModes) && ((size + nextsize) < max_line_size)) + if (!this->allow_empty) { - result[0] += *(sequence[0].begin()); - if (!sequence[1].empty()) + this->pos = this->tokens.find_first_not_of(this->sep, this->pos); + if (this->pos == std::string::npos) { - result.push_back(sequence[1]); - size += nextsize; /* Account for mode character and whitespace */ + this->pos = this->tokens.length() + 1; + token.clear(); + return false; } - sequence[0].erase(sequence[0].begin()); - sequence.erase(sequence.begin() + 1); - - if (sequence.size() > 1) - nextsize = sequence[1].length() + 2; - - n++; } - return n; -} + size_t p = this->tokens.find(this->sep, this->pos); + if (p == std::string::npos) + p = this->tokens.length(); -irc::stringjoiner::stringjoiner(const std::string &seperator, const std::vector &sequence, int begin, int end) -{ - if (end < begin) - throw "stringjoiner logic error, this causes problems."; + token.assign(tokens, this->pos, p - this->pos); + this->pos = p + 1; - for (int v = begin; v < end; v++) - joined.append(sequence[v]).append(seperator); - joined.append(sequence[end]); + return true; } -irc::stringjoiner::stringjoiner(const std::string &seperator, const std::deque &sequence, int begin, int end) +const std::string irc::sepstream::GetRemaining() { - if (end < begin) - throw "stringjoiner logic error, this causes problems."; - - for (int v = begin; v < end; v++) - joined.append(sequence[v]).append(seperator); - joined.append(sequence[end]); + return !this->StreamEnd() ? this->tokens.substr(this->pos) : ""; } -irc::stringjoiner::stringjoiner(const std::string &seperator, const char* const* sequence, int begin, int end) +bool irc::sepstream::StreamEnd() { - if (end < begin) - throw "stringjoiner logic error, this causes problems."; - - for (int v = begin; v < end; v++) - joined.append(sequence[v]).append(seperator); - joined.append(sequence[end]); + return this->pos > this->tokens.length(); } -std::string& irc::stringjoiner::GetJoined() +bool irc::sepstream::Contains(const std::string& value) { - return joined; -} + std::string token; + while (GetToken(token)) + if (value == token) + return true; -irc::portparser::portparser(const std::string &source, bool allow_overlapped) : in_range(0), range_begin(0), range_end(0), overlapped(allow_overlapped) -{ - sep = new irc::commasepstream(source); - overlap_set.clear(); + return false; } -irc::portparser::~portparser() +irc::portparser::portparser(const std::string &source, bool allow_overlapped) + : sep(source), in_range(0), range_begin(0), range_end(0), overlapped(allow_overlapped) { - delete sep; } bool irc::portparser::Overlaps(long val) { - if (!overlapped) + if (overlapped) return false; - if (overlap_set.find(val) == overlap_set.end()) - { - overlap_set[val] = true; - return false; - } - else - return true; + return (!overlap_set.insert(val).second); } long irc::portparser::GetToken() @@ -455,7 +331,7 @@ long irc::portparser::GetToken() { while (((Overlaps(in_range)) && (in_range <= range_end))) in_range++; - + if (in_range <= range_end) return in_range; } @@ -465,24 +341,23 @@ long irc::portparser::GetToken() } std::string x; - sep->GetToken(x); + sep.GetToken(x); if (x.empty()) return 0; - while (Overlaps(atoi(x.c_str()))) + while (Overlaps(ConvToNum(x))) { - if (!sep->GetToken(x)) + if (!sep.GetToken(x)) return 0; } std::string::size_type dash = x.rfind('-'); if (dash != std::string::npos) { - std::string sbegin = x.substr(0, dash); - std::string send = x.substr(dash+1, x.length()); - range_begin = atoi(sbegin.c_str()); - range_end = atoi(send.c_str()); + std::string sbegin(x, 0, dash); + range_begin = ConvToNum(sbegin); + range_end = ConvToNum(x.c_str() + dash + 1); if ((range_begin > 0) && (range_end > 0) && (range_begin < 65536) && (range_end < 65536) && (range_begin < range_end)) { @@ -492,35 +367,11 @@ long irc::portparser::GetToken() else { /* Assume its just the one port */ - return atoi(sbegin.c_str()); + return ConvToNum(sbegin); } } else { - return atoi(x.c_str()); - } -} - -const std::string& SearchAndReplace(std::string& text, const std::string& pattern, const std::string& replace) -{ - std::string replacement; - if ((!pattern.empty()) && (!text.empty())) - { - for (std::string::size_type n = 0; n != text.length(); ++n) - { - if (text.length() >= pattern.length() && text.substr(n, pattern.length()) == pattern) - { - /* Found the pattern in the text, replace it, and advance */ - replacement.append(replace); - n = n + pattern.length() - 1; - } - else - { - replacement += text[n]; - } - } + return ConvToNum(x); } - text = replacement; - return text; } -