X-Git-Url: https://git.netwichtig.de/gitweb/?a=blobdiff_plain;f=src%2Fhashcomp.cpp;h=04c24418c5424f7d0ac620ace63bf4fa4d4f2079;hb=219993bc9018d9f0d9568330d7a972b68b785d27;hp=3271e599cda9dfa3e2cd3daebf1fbd5794af9dba;hpb=5454b47bd5d5ce5fb08aed83bac0148eabab620e;p=user%2Fhenk%2Fcode%2Finspircd.git diff --git a/src/hashcomp.cpp b/src/hashcomp.cpp index 3271e599c..04c24418c 100644 --- a/src/hashcomp.cpp +++ b/src/hashcomp.cpp @@ -2,31 +2,20 @@ * | Inspire Internet Relay Chat Daemon | * +------------------------------------+ * - * InspIRCd is copyright (C) 2002-2006 ChatSpike-Dev. - * E-mail: - * - * + * InspIRCd: (C) 2002-2009 InspIRCd Development Team + * See: http://wiki.inspircd.org/Credits * - * Written by Craig Edwards, Craig McLure, and others. * This program is free but copyrighted software; see - * the file COPYING for details. + * the file COPYING for details. * * --------------------------------------------------- */ -using namespace std; +/* $Core */ -#include "inspircd_config.h" #include "inspircd.h" -#include #include "hashcomp.h" -#include "helperfuncs.h" -#include - -#define nspace __gnu_cxx - -// from helperfuncs.cpp -extern const char lowermap[255]; +#include "hash_map.h" /****************************************************** * @@ -51,43 +40,69 @@ extern const char lowermap[255]; * Scandanavian Comparisons: The characters [, ], \ will * be considered the lowercase of {, } and |. * - * This file also contains hashing methods for hashing - * in_addr structs, we use this if we want to cache IP - * addresses. - * ******************************************************/ -using namespace std; +/* 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]; + } +} -size_t nspace::hash::operator()(const struct in_addr &a) const +#if defined(WINDOWS) && !defined(HASHMAP_DEPRECATED) + size_t nspace::hash_compare >::operator()(const std::string &s) const +#else + #ifdef HASHMAP_DEPRECATED + size_t CoreExport nspace::insensitive::operator()(const std::string &s) const + #else + size_t nspace::hash::operator()(const std::string &s) const + #endif +#endif { - size_t q; - memcpy(&q,&a,sizeof(size_t)); - return q; + /* 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; } -size_t nspace::hash::operator()(const string &s) const + +#if defined(WINDOWS) && !defined(HASHMAP_DEPRECATED) + size_t nspace::hash_compare >::operator()(const irc::string &s) const +#else + size_t CoreExport nspace::hash::operator()(const irc::string &s) const +#endif { - char a[MAXBUF]; - static struct hash strhash; - strlcpy(a,s.c_str(),MAXBUF); - strlower(a); - return strhash(a); + 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 { - unsigned char* n1 = (unsigned char*)s1.c_str(); - unsigned char* n2 = (unsigned char*)s2.c_str(); + 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 (lowermap[*n1] != lowermap[*n2]) + if (national_case_insensitive_map[*n1] != national_case_insensitive_map[*n2]) return false; - return (lowermap[*n1] == lowermap[*n2]); -} - -bool irc::InAddr_HashComp::operator()(const in_addr &s1, const in_addr &s2) const -{ - return (s1.s_addr == s1.s_addr); + return (national_case_insensitive_map[*n1] == national_case_insensitive_map[*n2]); } /****************************************************** @@ -97,83 +112,414 @@ bool irc::InAddr_HashComp::operator()(const in_addr &s1, const in_addr &s2) cons * std::string which is not only case-insensitive but * can also do scandanavian comparisons, e.g. { = [, etc. * - * This class depends on the global 'lowermap' which is - * initialized at startup by inspircd.cpp, and contains - * the 'scandanavian' casemappings for fast irc compare. + * This class depends on the const array 'national_case_insensitive_map'. * ******************************************************/ bool irc::irc_char_traits::eq(char c1st, char c2nd) { - return lowermap[(unsigned char)c1st] == lowermap[(unsigned 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) { - return lowermap[(unsigned char)c1st] != lowermap[(unsigned char)c2nd]; + return national_case_insensitive_map[(unsigned char)c1st] != national_case_insensitive_map[(unsigned char)c2nd]; } bool irc::irc_char_traits::lt(char c1st, char c2nd) { - return lowermap[(unsigned char)c1st] < lowermap[(unsigned 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++) { - if(lowermap[(unsigned char)*str1] > lowermap[(unsigned char)*str2]) - return 1; + if(national_case_insensitive_map[(unsigned char)*str1] > national_case_insensitive_map[(unsigned char)*str2]) + return 1; - if(lowermap[(unsigned char)*str1] < lowermap[(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; + return 0; - str1++; + str1++; str2++; } return 0; } -std::string operator+ (std::string& leftval, irc::string& rightval) +const char* irc::irc_char_traits::find(const char* s1, int n, char c) { - return leftval + std::string(rightval.c_str()); + while(n-- > 0 && national_case_insensitive_map[(unsigned char)*s1] != national_case_insensitive_map[(unsigned char)c]) + s1++; + return (n >= 0) ? s1 : NULL; } -irc::string operator+ (irc::string& leftval, std::string& rightval) +irc::tokenstream::tokenstream(const std::string &source) : tokens(source), last_pushed(false) { - return leftval + irc::string(rightval.c_str()); + /* Record starting position and current position */ + last_starting_position = tokens.begin(); + n = tokens.begin(); } -bool operator== (std::string& leftval, irc::string& rightval) +irc::tokenstream::~tokenstream() { - return (leftval == std::string(rightval.c_str())); } -bool operator== (irc::string& leftval, std::string& rightval) +bool irc::tokenstream::GetToken(std::string &token) { - return (rightval == std::string(leftval.c_str())); + std::string::iterator lsp = last_starting_position; + + while (n != tokens.end()) + { + /** 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()); + 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; } -const char* irc::irc_char_traits::find(const char* s1, int n, char c) +bool irc::tokenstream::GetToken(irc::string &token) { - while(n-- > 0 && lowermap[(unsigned char)*s1] != lowermap[(unsigned char)c]) - s1++; - return s1; + 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; +} + +bool irc::tokenstream::GetToken(long &token) +{ + std::string tok; + bool returnval = GetToken(tok); + token = ConvToInt(tok); + return returnval; +} + +irc::sepstream::sepstream(const std::string &source, char seperator) : tokens(source), sep(seperator) +{ + last_starting_position = tokens.begin(); + n = tokens.begin(); +} + +bool irc::sepstream::GetToken(std::string &token) +{ + std::string::iterator lsp = last_starting_position; + + while (n != tokens.end()) + { + 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 = ""; + 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) + { + hexbuf[i++] = hex[raw[j] / 16]; + hexbuf[i++] = hex[raw[j] % 16]; + } + 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(""); +} + +void irc::modestacker::Push(char modeletter, const std::string ¶meter) +{ + *(sequence.begin()) += modeletter; + sequence.push_back(parameter); } -/* See hashcomp.h if you care about these... */ -std::ostream& operator<<(std::ostream &os, const irc::string &str) +void irc::modestacker::Push(char modeletter) { - return os << str.c_str(); + this->Push(modeletter,""); } -std::istream& operator>>(std::istream &is, irc::string &str) +void irc::modestacker::PushPlus() { - std::string tmp; - is >> tmp; - str = tmp.c_str(); - return is; + this->Push('+',""); } + +void irc::modestacker::PushMinus() +{ + this->Push('-',""); +} + +int irc::modestacker::GetStackedLine(std::vector &result, int max_line_size) +{ + if (sequence.empty()) + { + return 0; + } + + unsigned int n = 0; + int size = 1; /* Account for initial +/- char */ + int nextsize = 0; + int start = result.size(); + std::string modeline = adding ? "+" : "-"; + result.push_back(modeline); + + if (sequence.size() > 1) + nextsize = sequence[1].length() + 2; + + while (!sequence[0].empty() && (sequence.size() > 1) && (n < ServerInstance->Config->Limits.MaxModes) && ((size + nextsize) < max_line_size)) + { + modeline += *(sequence[0].begin()); + if (!sequence[1].empty()) + { + result.push_back(sequence[1]); + size += nextsize; /* Account for mode character and whitespace */ + } + sequence[0].erase(sequence[0].begin()); + sequence.erase(sequence.begin() + 1); + + if (sequence.size() > 1) + nextsize = sequence[1].length() + 2; + + n++; + } + result[start] = modeline; + + return n; +} + +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."; + + for (int v = begin; v < end; v++) + joined.append(sequence[v]).append(seperator); + joined.append(sequence[end]); +} + +irc::stringjoiner::stringjoiner(const std::string &seperator, const std::deque &sequence, int begin, int end) +{ + 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]); +} + +irc::stringjoiner::stringjoiner(const std::string &seperator, const char* const* sequence, int begin, int end) +{ + 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]); +} + +std::string& irc::stringjoiner::GetJoined() +{ + return joined; +} + +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(); +} + +irc::portparser::~portparser() +{ + delete sep; +} + +bool irc::portparser::Overlaps(long val) +{ + if (!overlapped) + return false; + + if (overlap_set.find(val) == overlap_set.end()) + { + overlap_set[val] = true; + return false; + } + else + return true; +} + +long irc::portparser::GetToken() +{ + if (in_range > 0) + { + in_range++; + if (in_range <= range_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; + } + } + else + in_range = 0; + } + + std::string x; + sep->GetToken(x); + + if (x.empty()) + return 0; + + while (Overlaps(atoi(x.c_str()))) + { + 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()); + + 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 atoi(sbegin.c_str()); + } + } + else + { + return atoi(x.c_str()); + } +} + +/*const std::basic_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) + { + replacement.append(replace); + n = n + pattern.length() - 1; + } + else + { + replacement += text[n]; + } + } + } + text = replacement; + return text; +}*/