Thread portable class Gate portable class Timer class Pool class Terimber 2.0 About C++

Downloads Products & Services Support Clients Open Source About Home / Open source / Terimber 2.0 fuzzyphonetic.cpp Go to the documentation of this file. /* * The Software License * ================================================================================= * Copyright (c) 2003-.The Terimber Corporation. All rights reserved. * ================================================================================= * Redistributions of source code must retain the above copyright notice, * this list of conditions and the following disclaimer. * Redistributions in binary form must reproduce the above copyright notice, * this list of conditions and the following disclaimer in the documentation * and/or other materials provided with the distribution. * The end-user documentation included with the redistribution, if any, * must include the following acknowledgment: * "This product includes software developed by the Terimber Corporation." * ================================================================================= * THIS SOFTWARE IS PROVIDED "AS IS" AND ANY EXPRESSED OR IMPLIED WARRANTIES, * INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY * AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. * IN NO EVENT SHALL THE TERIMBER CORPORATION OR ITS CONTRIBUTORS BE LIABLE FOR * ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; * LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON * ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. * ================================================================================ */ #include "fuzzy/fuzzyphonetic.hpp" #include "base/list.hpp" #include "base/map.hpp" #include "base/vector.hpp" #include "base/memory.hpp" BEGIN_TERIMBER_NAMESPACE #pragma pack(4) /* *********** BEGIN METAPHONE RULES *********** Lawrence Philips' RULES follow: The 16 consonant sounds: |--- ZERO represents "th" | B X S K J T F H L M N P R 0 W Y Exceptions: Beginning of word: "ae-", "gn", "kn-", "pn-", "wr-" ----> drop first letter "Aebersold", "Gnagy", "Knuth", "Pniewski", "Wright" Beginning of word: "x" ----> change to "s" as in "Deng Xiaopeng" Beginning of word: "wh-" ----> change to "w" as in "Whalen" Transformations: B ----> B unless at the end of word after "m", as in "dumb", "McComb" C ----> X (sh) if "-cia-" or "-ch-" S if "-ci-", "-ce-", or "-cy-" SILENT if "-sci-", "-sce-", or "-scy-" K otherwise, including "-sch-" D ----> J if in "-dge-", "-dgy-", or "-dgi-" T otherwise F ----> F G ----> SILENT if in "-gh-" and not at end or before a vowel in "-gn" or "-gned" in "-dge-" etc., as in above rule J if before "i", or "e", or "y"; if not, double "gg" K otherwise H ----> SILENT if after vowel and no vowel follows or after "-ch-", "-sh-", "-ph-", "-th-", "-gh-" H otherwise J ----> J K ----> SILENT if after "c" K otherwise L ----> L M ----> M N ----> N P ----> F if before "h" P otherwise Q ----> K R ----> R S ----> X (sh) if before "h" or in "-sio-" or "-sia-" S otherwise T ----> X (sh) if "-tia-" or "-tio-" 0 (th) if before "h" silent if in "-tch-" T otherwise V ----> F W ----> SILENT if not followed by a vowel W if followed by a vowel X ----> KS Y ----> SILENT if not followed by a vowel Y if followed by a vowel Z ----> S */ namespace fuzzyphonetic { const ub1_t empty_code = 0xff; const ub1_t blank_code = 0xfe; //#define DOUBLEMETAPHONE // adds sounds to list class metaphone_keeper { public: metaphone_keeper(_list< ub1_t >& backet, #ifdef DOUBLEMETAPHONE _list< ub1_t >& backet2, #endif byte_allocator& all) : _backet(backet), #ifdef DOUBLEMETAPHONE _backet2(backet), _alternative(false), #endif _all(all) { clear(); } inline void clear() { _code[0] = _code[1] = _alt[0] = _alt[1] = empty_code; } inline void metaph_add() { if (_code[0] != empty_code) { _backet.push_back(_all, _code[0]); #ifdef DOUBLEMETAPHONE _backet2.push_back(_all, _code[0]); #endif } if (_code[1] != empty_code) { _backet.push_back(_all, _code[1]); #ifdef DOUBLEMETAPHONE _backet2.push_back(_all, _code[1]); #endif } } inline void metaph_add_alt() { if (_code[0] != empty_code) _backet.push_back(_all, _code[0]); if (_code[1] != empty_code) _backet.push_back(_all, _code[1]); #ifdef DOUBLEMETAPHONE if (_alt[0] != empty_code) { _alternative = true; if (_alt[0] != blank_code) { _backet2.push_back(_all, _alt[0]); if (_alt[1] != empty_code) _backet2.push_back(_all, _alt[1]); } } else if (_code[0] != empty_code && _code[0] != blank_code) { _backet2.push_back(_all, _code[0]); if (_code[1] != empty_code) _backet2.push_back(_all, _code[1]); } #endif } public: _list< ub1_t >& _backet; #ifdef DOUBLEMETAPHONE bool _alternative; _list< ub1_t >& _backet2; #endif byte_allocator& _all; ub1_t _code[2]; ub1_t _alt[2]; }; // converts single word to the phonetic key // memory is allocated on internal memory factory is, external allocator is not provided metaphone_key convert_to_metaphone(const char* word, size_t len, byte_allocator& all) { _list< ub1_t > backet #ifdef DOUBLEMETAPHONE , backet2 #endif ; metaphone_keeper meta(backet #ifdef DOUBLEMETAPHONE , backet2 #endif , all); metaphone_key ret; const char* begin = word; register size_t index = 0; bool isSG = is_slavo_germanic(word, len); const size_t slot_size = 8; register char slot[slot_size] = {0}; // slot[4] == current register const char* pslot[slot_size] = {0}; const size_t prev4 = 0, prev3 = 1, prev2 = 2, prev = 3, curr = 4, next = 5, next2 = 6, next3 = 7; register size_t step = 0; slot[curr] = to_upper(*word); slot[next] = 1 == len ? 0 : to_upper(*(word + 1)); pslot[curr] = word; pslot[next] = 1 == len ? 0 : word + 1; register const char* pend = word + len; // analyzes the initial char(s) switch (slot[curr]) { case 'P': case 'K': case 'G': // PN, KN, GN -> N if (len > 1 && slot[next] == 'N') { ++word, ++index; memmove(slot, slot + 1, slot_size - 1); memmove(pslot, pslot + 1, (slot_size - 1) * sizeof(char*)); } break; case 'A': // AE -> E if (len > 1 && slot[next] == 'E') { ++word, ++index; memmove(slot, slot + 1, slot_size - 1); memmove(pslot, pslot + 1, (slot_size - 1) * sizeof(char*)); } break; case 'W': // WR -> R, WH -> H if (len > 1 && slot[next] == 'R') { ++word, ++index; memmove(slot, slot + 1, slot_size - 1); memmove(pslot, pslot + 1, (slot_size - 1) * sizeof(char*)); } break; case 'X': meta._code[0] = S_NOTE; meta._code[1] = empty_code; meta.metaph_add(); ++word, ++index; memmove(slot, slot + 1, slot_size - 1); memmove(pslot, pslot + 1, (slot_size - 1) * sizeof(char*)); break; } // switch while (index < len) { // converts to lower upper case slot[curr] = to_upper(*word); pslot[curr] = word; if (index + 3 < len) { slot[next3] = to_upper(*(word + 3)); slot[next2] = to_upper(*(word + 2)); slot[next] = to_upper(*(word + 1)); pslot[next3] = word + 3; pslot[next2] = word + 2; pslot[next] = word + 1; } else if (index + 2 < len) { slot[next3] = 0; slot[next2] = to_upper(*(word + 2)); slot[next] = to_upper(*(word + 1)); pslot[next3] = 0; pslot[next2] = word + 2; pslot[next] = word + 1; } else if (index + 1 < len) { slot[next3] = slot[next2] = 0; slot[next] = to_upper(*(word + 1)); pslot[next3] = pslot[next2] = 0; pslot[next] = word + 1; } meta.clear(); switch (slot[curr]) { case 0: // illegal char step = 1; break; case 'A': case 'E': case 'I': case 'O': case 'U': case 'Y': if (slot[prev] == 0) { meta._code[0] = A_NOTE; meta.metaph_add(); } step = 1; break; case 'B': // B -> P meta._code[0] = P_NOTE; meta.metaph_add(); if (slot[next] == 'B') step = 2; else step = 1; break; case 'C': //various germanic if(slot[prev2] != 0 && !is_vowel(slot[prev2]) && slot[prev] == 'A' && slot[next] == 'H' && (slot[next2] != 'I' && slot[next2] != 'E' || check_sub_string(pslot[prev2], pend, "BACHER", 6) || check_sub_string(pslot[prev2], pend, "MACHER", 6) ) ) { meta._code[0] = K_NOTE; meta.metaph_add(); step = 2; break; } // special case 'caesar' if (slot[prev] == 0 && check_sub_string(pslot[next], pend, "AESAR", 5)) { meta._code[0] = S_NOTE; meta.metaph_add(); step = 2; break; } if (slot[next] == 'H') { // italian 'chianti' if (check_sub_string(pslot[next2], pend, "IA", 2)) { meta._code[0] = K_NOTE; meta.metaph_add(); step = 2; break; } //finds 'michael' if(slot[prev] != 0 && check_sub_string(pslot[next2], pend, "AE", 2)) { meta._code[0] = K_NOTE; meta._alt[0] = X_NOTE; meta.metaph_add_alt(); step = 2; break; } //greek roots e.g. 'chemistry', 'chorus' if(slot[prev] == 0 && (check_sub_string(pslot[next2], pend, "ARAC", 5) || check_sub_string(pslot[next2], pend, "ARIS", 5) || check_sub_string(pslot[next2], pend, "OR", 3) || check_sub_string(pslot[next2], pend, "YM", 3) || check_sub_string(pslot[next2], pend, "IA", 3) || check_sub_string(pslot[next2], pend, "EM", 3) ) && !check_sub_string(begin, pend, "CHORE", 5)) { meta._code[0] = K_NOTE; meta.metaph_add(); step = 2; break; } // germanic, greek, or otherwise 'ch' for 'kh' sound if((check_sub_string(begin, pend, "VAN ", 4) || check_sub_string(begin, pend, "VON ", 4) || check_sub_string(begin, pend, "SCH", 3)) // 'architect but not 'arch', 'orchestra', 'orchid' || slot[prev2] && check_sub_string(pslot[prev2], pend, "ORCHES", 6) || slot[prev2] && check_sub_string(pslot[prev2], pend, "ARCHIT", 6) || slot[prev2] && check_sub_string(pslot[prev2], pend, "ORCHID", 6) || slot[next2] == 'T' || slot[next2] == 'S' || ((slot[prev] == 'A' || slot[prev] == 'O' || slot[prev] == 'U'|| slot[prev] == 'E' || slot[prev] == 0) && //e.g., 'wachtler', 'wechsler', but not 'tichner' (slot[next2] == 'L' || slot[next2] == 'R' || slot[next2] == 'N' || slot[next2] == 'M' || slot[next2] == 'B' || slot[next2] == 'H' || slot[next2] == 'F' || slot[next2] == 'V' || slot[next2] == 'W' || slot[next2] == ' ') ) ) { meta._code[0] = K_NOTE; meta.metaph_add(); } else { if (slot[prev] != 0) { if (check_sub_string(begin, pend, "MC", 2)) { //e.g., "McHugh" meta._code[0] = K_NOTE; meta.metaph_add(); } else { meta._code[0] = X_NOTE; meta._alt[0] = K_NOTE; meta.metaph_add_alt(); } } else { meta._code[0] = X_NOTE; meta.metaph_add(); } } step = 2; break; } // CH //e.g, 'czerny' if (slot[next] == 'Z' && !(slot[prev2] == 'W' && slot[prev] == 'I')) { meta._code[0] = S_NOTE; meta._alt[0] = X_NOTE; meta.metaph_add_alt(); step = 2; break; } //e.g., 'focaccia' if (slot[next] == 'I' && slot[next2] == 'A') { meta._code[0] = X_NOTE; meta.metaph_add(); step = 3; break; } //doubles 'C', but not if e.g. 'McClellan' if (slot[next] == 'C' && !(slot[prev2] == 0 && slot[prev] != 0 && to_upper(*begin) == 'M')) { //'bellocchio' but not 'bacchus' if ((slot[next2] == 'I' || slot[next2] == 'E' || slot[next2] == 'H') && !check_sub_string(pslot[next2], pend, "HU", 2) ) { //'accident', 'accede' 'succeed' if ((slot[prev2] == 0 && slot[prev] == 'A') || (slot[prev] == 'U' && (check_sub_string(pslot[next2], pend, "EE", 2) || check_sub_string(pslot[next2], pend, "ES", 2)) ) ) { meta._code[0] = K_NOTE; meta._code[1] = S_NOTE; meta._alt[0] = K_NOTE; meta._alt[1] = S_NOTE; meta.metaph_add_alt(); } else { meta._code[0] = X_NOTE; meta.metaph_add(); } step = 3; break; } else { //Pierce's rule meta._code[0] = K_NOTE; meta.metaph_add(); step = 2; break; } } if (slot[next] == 'K' || slot[next] == 'G' || slot[next] == 'Q' ) { meta._code[0] = K_NOTE; meta.metaph_add(); step = 2; break; } if (slot[next] == 'I' || slot[next] == 'E' || slot[next] == 'Y' ) { //italian vs. english if (slot[next2] == 'O' || slot[next2] == 'E' || slot[next2] == 'A' ) { meta._code[0] = S_NOTE; meta._alt[0] = X_NOTE; meta.metaph_add_alt(); } else { meta._code[0] = S_NOTE; meta.metaph_add(); } step = 2; break; } meta._code[0] = K_NOTE; meta.metaph_add(); // name sent in 'mac caffrey', 'mac gregor' if (slot[next] == ' ' && (slot[next2] == 'C' || slot[next2] == 'Q' || slot[next2] == 'G')) step = 3; else if ((slot[next] == 'C' || slot[next] == 'Q' || slot[next] == 'G') && !(slot[next] == 'E' || slot[next] == 'I')) step = 2; else step = 1; break; case 'D': if (slot[next] == 'G') { if (slot[next2] == 'I' || slot[next2] == 'E' || slot[next2] == 'Y' ) { //e.g. 'edge' meta._code[0] = J_NOTE; meta.metaph_add(); step = 3; break; } else { //e.g. 'edgar' meta._code[0] = T_NOTE; meta._code[1] = K_NOTE; meta.metaph_add(); step = 2; break; } } if (slot[next] == 'T' || slot[next] == 'D' ) { meta._code[0] = T_NOTE; meta.metaph_add(); step = 2; break; } meta._code[0] = T_NOTE; meta.metaph_add(); step = 1; break; case 'F': if (slot[next] == 'F') step = 2; else step = 1; meta._code[0] = F_NOTE; meta.metaph_add(); break; case 'G': if (slot[next] == 'H') { if (slot[prev] != 0 && !is_vowel(slot[prev])) { meta._code[0] = K_NOTE; meta.metaph_add(); step = 2; break; } if (slot[prev3] == 0 && slot[prev] == 0) { //'ghislane', 'ghiradelli' if(slot[next2] == 'I') { meta._code[0] = J_NOTE; meta.metaph_add(); } else { meta._code[0] = K_NOTE; meta.metaph_add(); } step = 2; break; } //Parker's rule (with some further refinements) - e.g., 'hugh' if (slot[prev2] == 'B' || slot[prev2] == 'H' || slot[prev2] == 'D' //e.g., 'bough' || slot[prev3] == 'B' || slot[prev3] == 'H' || slot[prev3] == 'D' //e.g., 'broughton' || slot[prev4] == 'B' || slot[prev4] == 'H' ) { step = 2; break; } else { //e.g., 'laugh', 'McLaughlin', 'cough', 'gough', 'rough', 'tough' if (slot[prev2] != 0 && slot[prev] == 'U' && (slot[prev3] == 'C' || slot[prev3] == 'G' || slot[prev3] == 'L' || slot[prev3] == 'R' || slot[prev3] == 'T') ) { meta._code[0] = F_NOTE; meta.metaph_add(); } else if (slot[prev] != 0 && slot[prev] != 'I') { meta._code[0] = K_NOTE; meta.metaph_add(); } step = 2; break; } } if (slot[next] == 'N') { if (slot[prev] != 0 && slot[prev2] == 0 && is_vowel(*begin) && !isSG) { meta._code[0] = K_NOTE; meta._code[1] = N_NOTE; meta._alt[0] = N_NOTE; meta.metaph_add_alt(); } else { //not e.g. 'cagney' if (!check_sub_string(pslot[next2], pend, "EY", 2) && slot[prev] != 'Y' && !isSG) { meta._code[0] = N_NOTE; meta._alt[0] = K_NOTE; meta._alt[1] = N_NOTE; meta.metaph_add_alt(); } else { meta._code[0] = K_NOTE; meta._code[1] = N_NOTE; meta.metaph_add(); } } step = 2; break; } //'tagliaro' if (slot[next] == 'L' && slot[next2] == 'I' && !isSG) { meta._code[0] = K_NOTE; meta._code[1] = L_NOTE; meta._alt[0] = L_NOTE; meta.metaph_add_alt(); step = 2; break; } //-ges-,-gep-,-gel-, -gie- at beginning if (slot[prev] == 0 && (slot[next] == 'Y' || (slot[next] == 'E' && (slot[next2] == 'S' || slot[next2] == 'P' || slot[next2] == 'B' || slot[next2] == 'L' || slot[next2] == 'Y' || slot[next2] == 'I' || slot[next2] == 'R' ) ) || (slot[next] == 'I' && (slot[next2] == 'B' || slot[next2] == 'L' || slot[next2] == 'N' || slot[next2] == 'E' ) ) ) ) { meta._code[0] = K_NOTE; meta._alt[0] = J_NOTE; meta.metaph_add_alt(); step = 2; break; } // -ger-, -gy- if ((slot[next] == 'E' && slot[next2] == 'R' || slot[next] == 'Y' ) && !(check_sub_string(begin, pend, "DANGER", 6) || check_sub_string(begin, pend, "RANGER", 6) || check_sub_string(begin, pend, "MANGER", 6)) && !(slot[prev] == 'E' || slot[prev] == 'I') && !((slot[prev] == 'R' || slot[prev] == 'O') && slot[prev2] == 'G' && slot[prev3] == 'Y') ) { meta._code[0] = K_NOTE; meta._alt[0] = J_NOTE; meta.metaph_add_alt(); step = 2; break; } // italian e.g, 'biaggi' if ((slot[next] == 'E' || slot[next] == 'I' || slot[next] == 'Y') || ((slot[prev] == 'A' || slot[prev] == 'O') && slot[next] == 'G' && slot[next2] == 'I') ) { //obvious germanic if ((check_sub_string(begin, pend, "VAN ", 4) || check_sub_string(begin, pend, "VON ", 4) || check_sub_string(begin, pend, "SCH", 3) ) || (slot[next] == 'E' && slot[next2] == 'T') ) { meta._code[0] = K_NOTE; meta.metaph_add(); } else { //always soft if french ending if (slot[next] == 'I' && slot[next2] == 'E' && slot[next3] == 'R') { meta._code[0] = J_NOTE; meta.metaph_add(); } else { meta._code[0] = J_NOTE; meta._alt[0] = K_NOTE; meta.metaph_add_alt(); } } step = 2; break; } if (slot[next] == 'G') step = 2; else step = 1; meta._code[0] = K_NOTE; meta.metaph_add(); break; case 'H': //only keep if first & before vowel or btw. 2 vowels if ((slot[prev] == 0 || is_vowel(prev)) && is_vowel(next) ) { meta._code[0] = H_NOTE; meta.metaph_add(); step = 2; } else //also takes care of 'HH' step = 1; break; case 'J': //obvious spanish, 'jose', 'san jacinto' if ((slot[next] == 'O' && slot[next2] == 'S' && slot[next3] == 'E') || check_sub_string(begin, pend, "SAN J", 5)) { if (index == 0) { meta._code[0] = H_NOTE; meta.metaph_add(); } else { meta._code[0] = J_NOTE; meta._alt[0] = H_NOTE; meta.metaph_add_alt(); } step = 1; break; } else { meta._code[0] = J_NOTE; meta._alt[0] = A_NOTE; meta.metaph_add_alt(); //Yankelovich/Jankelowicz if(slot[next] == 'J')//it could happen! { step = 2; } else { step = 1; } break; } if (slot[prev] == 0 && !(slot[next] == 'O' && slot[next2] == 'S' && slot[next3] == 'E')) { //Yankelovich/Jankelowicz meta._code[0] = J_NOTE; meta._alt[0] = A_NOTE; meta.metaph_add_alt(); } else { //spanish pron. of e.g. 'bajador' if (is_vowel(prev) && !isSG && (slot[next] == 'A' || slot[next] == 'O') ) { meta._code[0] = J_NOTE; meta._alt[0] = H_NOTE; meta.metaph_add_alt(); } else if (index + 1 == len) { meta._code[0] = J_NOTE; meta._alt[0] = blank_code; meta.metaph_add_alt(); } else if (!(slot[next] == 'L' || slot[next] == 'T' || slot[next] == 'K' || slot[next] == 'S' || slot[next] == 'N' || slot[next] == 'M' || slot[next] == 'B' || slot[next] == 'Z' ) && !(slot[prev] == 'S' || slot[prev] == 'K' || slot[prev] == 'L' ) ) { meta._code[0] = J_NOTE; meta.metaph_add(); } } if (slot[next] == 'J')//it could happen! { step = 2; } else { step = 1; } break; case 'K': if (slot[next] == 'K') { step = 2; } else { step = 1; } meta._code[0] = K_NOTE; meta.metaph_add(); break; case 'L': if (slot[next] == 'L') { //spanish e.g. 'cabrillo', 'gallegos' if ((index == len - 3 && ((slot[prev] == 'I' && slot[next] == 'L' && (slot[next2] == 'O' || slot[next2] == 'A')) || (slot[prev] == 'A' && slot[next] == 'L' && slot[next2] == 'E')) ) || ((check_sub_string(begin + len - 2, pend, "AS", 2) || check_sub_string(begin + len - 2, pend, "OS", 2) || to_upper(*(begin + len - 1)) == 'A' || to_upper(*(begin + len - 1)) == 'O') && slot[prev] && (slot[prev] == 'A' && slot[next] == 'L' && slot[next2] == 'E')) ) { meta._code[0] = L_NOTE; meta._alt[0] = blank_code; meta.metaph_add_alt(); step = 2; break; } step = 2; } else step = 1; meta._code[0] = L_NOTE; meta.metaph_add(); break; case 'M': if ((slot[prev] == 'U' && slot[next] == 'B') && ((index + 1 == len) || check_sub_string(pslot[next2], pend, "ER", 2)) //'dumb','thumb' || slot[next] == 'M' ) { step = 2; } else { step = 1; } meta._code[0] = M_NOTE; meta.metaph_add(); break; case 'N': if (slot[next] == 'N') { step = 2; } else { step = 1; } meta._code[0] = N_NOTE; meta.metaph_add(); break; case 'P': if (slot[next] == 'H') { meta._code[0] = F_NOTE; meta.metaph_add(); step = 2; break; } //also account for "campbell", "raspberry" if (slot[next] == 'P' || slot[next] == 'B') { step = 2; } else { step = 1; } meta._code[0] = P_NOTE; meta.metaph_add(); break; case 'Q': if (slot[next] == 'Q') { step = 2; } else { step = 1; } meta._code[0] = K_NOTE; meta.metaph_add(); break; case 'R': //french e.g. 'rogier', but exclude 'hochmeier' if(index + 1 == len && !isSG && (slot[prev2] == 'I' && slot[prev] == 'E') && !(slot[prev4] == 'M' && slot[prev3] == 'E') || (slot[prev4] == 'M' && slot[prev3] == 'A') ) { meta._alt[0] = R_NOTE; meta.metaph_add_alt(); } else { meta._code[0] = R_NOTE; meta.metaph_add(); } if (slot[next] == 'R') { step = 2; } else { step = 1; } break; case 'S': //special cases 'island', 'isle', 'carlisle', 'carlysle' if ((slot[prev] == 'I' || slot[prev] == 'Y') && slot[next] == 'L') { step = 1; break; } //special case 'sugar-' if (slot[prev] == 0 && check_sub_string(pslot[next], pend, "UGAR", 4)) { meta._code[0] = X_NOTE; meta._alt[0] = S_NOTE; meta.metaph_add_alt(); step = 1; break; } if (slot[next] == 'H') { //germanic if (check_sub_string(pslot[next2], pend, "EIM", 3) || check_sub_string(pslot[next2], pend, "OEK", 3) || check_sub_string(pslot[next2], pend, "OLM", 3) || check_sub_string(pslot[next2], pend, "OLZ", 3) ) { meta._code[0] = S_NOTE; meta.metaph_add(); } else { meta._code[0] = X_NOTE; meta.metaph_add(); } step = 2; break; } //italian & armenian if (slot[next] == 'I' && (slot[next2] == 'O' || slot[next2] == 'A' && (slot[next3] == 0 || slot[next3] == 'N'))) { if (!isSG) { meta._code[0] = S_NOTE; meta._alt[0] = X_NOTE; meta.metaph_add_alt(); } else { meta._code[0] = S_NOTE; meta.metaph_add(); } step = 3; break; } //german & anglicisations, e.g. 'smith' match 'schmidt', 'snider' match 'schneider' //also, -sz- in slavic language altho in hungarian it is pronounced 's' if ((slot[prev] == 0 && (slot[next] == 'M' || slot[next] == 'N' || slot[next] == 'L' || slot[next] == 'W' ) ) || (slot[next] == 'Z') ) { meta._code[0] = S_NOTE; meta._alt[0] = X_NOTE; meta.metaph_add_alt(); if (slot[next] == 'Z') { step = 2; } else { step = 1; } break; } if (slot[next] == 'C') { //Schlesinger's rule if (slot[next2] == 'H') { //dutch origin, e.g. 'school', 'schooner' if (check_sub_string(pslot[next3], pend, "OO", 2) || check_sub_string(pslot[next3], pend, "ER", 2) || check_sub_string(pslot[next3], pend, "EN", 2) || check_sub_string(pslot[next3], pend, "UY", 2) || check_sub_string(pslot[next3], pend, "ED", 2) || check_sub_string(pslot[next3], pend, "EM", 2) ) { //'schermerhorn', 'schenker' if (check_sub_string(pslot[next3], pend, "ER", 2) || check_sub_string(pslot[next3], pend, "EN", 2) ) { meta._code[0] = X_NOTE; meta._alt[0] = S_NOTE; meta._alt[1] = K_NOTE; meta.metaph_add_alt(); } else { meta._code[0] = S_NOTE; meta._code[1] = K_NOTE; meta.metaph_add(); } step = 3; break; } else { if (slot[prev] == 0 && !is_vowel(*(begin + 3)) && len > 3 && to_upper(*(begin + 3)) != 'W') { meta._code[0] = X_NOTE; meta._alt[0] = S_NOTE; meta.metaph_add_alt(); } else { meta._code[0] = X_NOTE; meta.metaph_add(); } step = 3; break; } } if (slot[next2] == 'I' || slot[next2] == 'E' || slot[next2] == 'Y' ) { meta._code[0] = S_NOTE; meta.metaph_add(); step = 3; break; } meta._code[0] = S_NOTE; meta._code[1] = K_NOTE; meta.metaph_add(); step = 3; break; } //french e.g. 'resnais', 'artois' if (index + 1 == len && (((slot[prev2] == 'A' || slot[prev2] == 'O') && slot[prev] == 'I'))) { meta._alt[0] = S_NOTE; meta.metaph_add_alt(); } else { meta._code[0] = S_NOTE; meta.metaph_add(); } if (slot[next] == 'S' || slot[next] == 'Z') { step = 2; } else { step = 1; } break; case 'T': if (slot[next] == 'I' && slot[next2] == 'O' && slot[next3] == 'N') { meta._code[0] = X_NOTE; meta.metaph_add(); step = 3; break; } if (slot[next] == 'I' && slot[next2] == 'A' || slot[next] == 'C' && slot[next2] == 'H') { meta._code[0] = X_NOTE; meta.metaph_add(); step = 3; break; } if (slot[next] == 'H' || slot[next] == 'T' && slot[next2] == 'H') { //special case 'thomas', 'thames' or germanic if (check_sub_string(pslot[next2], pend, "OM", 2) || check_sub_string(pslot[next2], pend, "AM", 2) || check_sub_string(begin, pend, "VAN ", 4) || check_sub_string(begin, pend, "VON ", 4) || check_sub_string(begin, pend, "SCH", 3) ) { meta._code[0] = T_NOTE; meta.metaph_add(); } else { meta._code[0] = TH_NOTE; meta._alt[0] = T_NOTE; meta.metaph_add_alt(); } step = 2; break; } if (slot[next] == 'T' || slot[next] == 'D') { step = 2; } else { step = 1; } meta._code[0] = T_NOTE; meta.metaph_add(); break; case 'V': if (slot[next] == 'V') { step = 2; } else { step = 1; } meta._code[0] = F_NOTE; meta.metaph_add(); break; case 'W': //can also be in the middle of the word if (slot[next] == 'R') { meta._code[0] = R_NOTE; meta.metaph_add(); step = 2; break; } if (slot[prev] == 0 && (is_vowel(next) || slot[next] == 'H')) { //Wasserman should match Vasserman if (is_vowel(next)) { meta._code[0] = A_NOTE; meta._alt[0] = F_NOTE; meta.metaph_add_alt(); } else { //need Uomo to match Womo meta._code[0] = A_NOTE; meta.metaph_add(); } } //Arnow should match Arnoff if ((index + 1 == len && is_vowel(prev)) || (slot[prev] == 'E' && slot[next] == 'S' && slot[next2] == 'K' && slot[next3] == 'I') || (slot[prev] == 'O' && slot[next] == 'S' && slot[next2] == 'K' && slot[next3] == 'I') || (slot[prev] == 'O' && slot[next] == 'S' && slot[next2] == 'K' && slot[next3] == 'Y') || check_sub_string(begin, pend, "SCH", 3) ) { meta._code[0] = empty_code; meta._code[1] = empty_code; meta._alt[0] = F_NOTE; meta._alt[1] = empty_code; meta.metaph_add_alt(); step = 1; break; } //polish e.g. 'filipowicz' if (slot[next] == 'I' && (slot[next2] == 'C' || slot[next2] == 'T') && slot[next3] == 'Z') { meta._code[0] = T_NOTE; meta._code[1] = S_NOTE; meta._alt[0] = F_NOTE; meta._alt[1] = X_NOTE; meta.metaph_add_alt(); step = 4; break; } //else skips it step = 1; break; case 'X': //french e.g. breaux if (!(index + 1 == len && (slot[prev3] == 'I' && slot[prev2] == 'A' && slot[prev] == 'U') || (slot[prev3] == 'E' && slot[prev2] == 'A' && slot[prev] == 'U') || (slot[prev2] == 'A' && slot[prev] == 'U') || (slot[prev2] == 'O' && slot[prev] == 'U') ) ) { meta._code[0] = K_NOTE; meta._code[1] = S_NOTE; meta.metaph_add(); } if (slot[next] == 'C' || slot[next] == 'X') { step = 2; } else { step = 1; } break; case 'Z': //chinese pinyin e.g. 'zhao' if (slot[next] == 'H') { meta._code[0] = J_NOTE; meta.metaph_add(); step = 2; break; } else { if ((slot[next] == 'Z' && (slot[next2] == 'O' || slot[next2] == 'I' || slot[next2] == 'A')) || (isSG && slot[prev] != 0 && slot[prev] != 'T') ) { meta._code[0] = S_NOTE; meta._alt[0] = T_NOTE; meta._alt[1] = S_NOTE; meta.metaph_add_alt(); } else { meta._code[0] = S_NOTE; meta.metaph_add(); } } if (slot[next] == 'Z') { step = 2; } else { step = 1; } break; default: step = 1; break; } // switch // how many we skipped assert(step > 0 && step < 7); memmove(slot, slot + step, slot_size - step); memmove(pslot, pslot + step, (slot_size - step) * sizeof(char*)); word += step, index += step; } // allocates memory ret._length = (ub4_t)backet.size(); if (ret._length) { ret._array = (ub1_t*)all.allocate(ret._length); // copies memory index = 0; for (_list< ub1_t >::const_iterator it = backet.begin(); it != backet.end(); ++it, ++index) { // converts phonet to code //assert(*it < metaphone_reflection::REFSIZE); ret._array[index] = *it; } } return ret; } void convert_to_reflection(const char* word, size_t len, byte_allocator& all, reflection_key& reflection) // output reflection { char* buf = (char*)all.allocate(len + 1); size_t offset = 0, shift = 0; register char ch; while (shift < len) { switch (ch = to_upper(word[shift])) { case 0: break; default: buf[offset++] = ch; } ++shift; } buf[offset] = 0; metaphone_key phonet = convert_to_metaphone(buf, offset, all); for (size_t i = 0; i < phonet._length; ++i) { assert(phonet._array[i] < reflection_key::REFSIZE); if (reflection._array[phonet._array[i]] < reflection_key::REFMAX) ++reflection._array[phonet._array[i]]; } } } #pragma pack() END_TERIMBER_NAMESPACE

© Copyright Terimber 2003-.