diff options
author | Tobias Weimer <wishmaster51@googlemail.com> | 2016-05-13 19:32:21 +0000 |
---|---|---|
committer | Tobias Weimer <wishmaster51@googlemail.com> | 2016-05-13 19:32:21 +0000 |
commit | 37c98eaad76b7f1bf86c75fe2c32cf6aa11f7c6f (patch) | |
tree | 32aede144aa0cd0f2dd058b003cdbd534a2f969c | |
parent | e73bb3845517a31fa795e8d2174fcc8572835b33 (diff) |
SpellChecker: Updated hunspell to 1.4.1
git-svn-id: http://svn.miranda-ng.org/main/trunk@16828 1316c22d-e87f-b044-9b9b-93d7a3e3ba9c
-rw-r--r-- | libs/hunspell/src/affentry.c++ | 201 | ||||
-rw-r--r-- | libs/hunspell/src/affentry.hxx | 8 | ||||
-rw-r--r-- | libs/hunspell/src/affixmgr.c++ | 381 | ||||
-rw-r--r-- | libs/hunspell/src/affixmgr.hxx | 24 | ||||
-rw-r--r-- | libs/hunspell/src/atypes.hxx | 13 | ||||
-rw-r--r-- | libs/hunspell/src/baseaffix.hxx | 2 | ||||
-rw-r--r-- | libs/hunspell/src/config.h | 7 | ||||
-rw-r--r-- | libs/hunspell/src/csutil.c++ | 520 | ||||
-rw-r--r-- | libs/hunspell/src/csutil.hxx | 121 | ||||
-rw-r--r-- | libs/hunspell/src/filemgr.c++ | 10 | ||||
-rw-r--r-- | libs/hunspell/src/hashmgr.c++ | 193 | ||||
-rw-r--r-- | libs/hunspell/src/hashmgr.hxx | 15 | ||||
-rw-r--r-- | libs/hunspell/src/hunspell.c++ | 1554 | ||||
-rw-r--r-- | libs/hunspell/src/hunspell.hxx | 52 | ||||
-rw-r--r-- | libs/hunspell/src/phonet.c++ | 59 | ||||
-rw-r--r-- | libs/hunspell/src/phonet.hxx | 6 | ||||
-rw-r--r-- | libs/hunspell/src/replist.c++ | 20 | ||||
-rw-r--r-- | libs/hunspell/src/replist.hxx | 4 | ||||
-rw-r--r-- | libs/hunspell/src/suggestmgr.c++ | 812 | ||||
-rw-r--r-- | libs/hunspell/src/suggestmgr.hxx | 7 | ||||
-rw-r--r-- | libs/hunspell/src/w_char.hxx | 20 | ||||
-rw-r--r-- | plugins/SpellChecker/src/Version.h | 4 | ||||
-rw-r--r-- | plugins/SpellChecker/src/dictionary.cpp | 5 |
23 files changed, 1481 insertions, 2557 deletions
diff --git a/libs/hunspell/src/affentry.c++ b/libs/hunspell/src/affentry.c++ index 983fe2c1ec..bd28274368 100644 --- a/libs/hunspell/src/affentry.c++ +++ b/libs/hunspell/src/affentry.c++ @@ -79,8 +79,6 @@ #include "affentry.hxx" #include "csutil.hxx" -#define MAXTEMPWORDLEN (MAXWORDUTF8LEN + 4) - PfxEntry::PfxEntry(AffixMgr* pmgr, affentry* dp) // register affix manager : pmyMgr(pmgr), @@ -117,11 +115,10 @@ PfxEntry::~PfxEntry() { } // add prefix to this word assuming conditions hold -char* PfxEntry::add(const char* word, int len) { +char* PfxEntry::add(const char* word, size_t len) { if ((len > strip.size() || (len == 0 && pmyMgr->get_fullstrip())) && (len >= numconds) && test_condition(word) && - (!strip.size() || (strncmp(word, strip.c_str(), strip.size()) == 0)) && - ((MAXTEMPWORDLEN) > (len + appnd.size() - strip.size()))) { + (!strip.size() || (strncmp(word, strip.c_str(), strip.size()) == 0))) { /* we have a match so add prefix */ std::string tword(appnd); tword.append(word + strip.size()); @@ -233,26 +230,21 @@ struct hentry* PfxEntry::checkword(const char* word, int len, char in_compound, const FLAG needflag) { - int tmpl; // length of tmpword struct hentry* he; // hash entry of root word or NULL - char tmpword[MAXTEMPWORDLEN]; // on entry prefix is 0 length or already matches the beginning of the word. // So if the remaining root word has positive length // and if there are enough chars in root word and added back strip chars // to meet the number of characters conditions, then test it - tmpl = len - appnd.size(); + int tmpl = len - appnd.size(); // length of tmpword if (tmpl > 0 || (tmpl == 0 && pmyMgr->get_fullstrip())) { // generate new root word by removing prefix and adding // back any characters that would have been stripped - if (strip.size()) { - strncpy(tmpword, strip.c_str(), MAXTEMPWORDLEN - 1); - tmpword[MAXTEMPWORDLEN - 1] = '\0'; - } - strcpy((tmpword + strip.size()), (word + appnd.size())); + std::string tmpword(strip); + tmpword.append(word + appnd.size()); // now make sure all of the conditions on characters // are met. Please see the appendix at the end of @@ -262,9 +254,9 @@ struct hentry* PfxEntry::checkword(const char* word, // if all conditions are met then check if resulting // root word in the dictionary - if (test_condition(tmpword)) { + if (test_condition(tmpword.c_str())) { tmpl += strip.size(); - if ((he = pmyMgr->lookup(tmpword)) != NULL) { + if ((he = pmyMgr->lookup(tmpword.c_str())) != NULL) { do { if (TESTAFF(he->astr, aflag, he->alen) && // forbid single prefixes with needaffix flag @@ -283,8 +275,9 @@ struct hentry* PfxEntry::checkword(const char* word, // if ((opts & aeXPRODUCT) && in_compound) { if ((opts & aeXPRODUCT)) { - he = pmyMgr->suffix_check(tmpword, tmpl, aeXPRODUCT, this, NULL, 0, - NULL, FLAG_NULL, needflag, in_compound); + he = pmyMgr->suffix_check(tmpword.c_str(), tmpl, aeXPRODUCT, this, + NULL, 0, NULL, FLAG_NULL, needflag, + in_compound); if (he) return he; } @@ -298,27 +291,22 @@ struct hentry* PfxEntry::check_twosfx(const char* word, int len, char in_compound, const FLAG needflag) { - int tmpl; // length of tmpword struct hentry* he; // hash entry of root word or NULL - char tmpword[MAXTEMPWORDLEN]; // on entry prefix is 0 length or already matches the beginning of the word. // So if the remaining root word has positive length // and if there are enough chars in root word and added back strip chars // to meet the number of characters conditions, then test it - tmpl = len - appnd.size(); + int tmpl = len - appnd.size(); // length of tmpword if ((tmpl > 0 || (tmpl == 0 && pmyMgr->get_fullstrip())) && (tmpl + strip.size() >= numconds)) { // generate new root word by removing prefix and adding // back any characters that would have been stripped - if (strip.size()) { - strncpy(tmpword, strip.c_str(), MAXTEMPWORDLEN - 1); - tmpword[MAXTEMPWORDLEN - 1] = '\0'; - } - strcpy((tmpword + strip.size()), (word + appnd.size())); + std::string tmpword(strip); + tmpword.append(word + appnd.size()); // now make sure all of the conditions on characters // are met. Please see the appendix at the end of @@ -328,7 +316,7 @@ struct hentry* PfxEntry::check_twosfx(const char* word, // if all conditions are met then check if resulting // root word in the dictionary - if (test_condition(tmpword)) { + if (test_condition(tmpword.c_str())) { tmpl += strip.size(); // prefix matched but no root word was found @@ -336,7 +324,7 @@ struct hentry* PfxEntry::check_twosfx(const char* word, // cross checked combined with a suffix if ((opts & aeXPRODUCT) && (in_compound != IN_CPD_BEGIN)) { - he = pmyMgr->suffix_check_twosfx(tmpword, tmpl, aeXPRODUCT, this, + he = pmyMgr->suffix_check_twosfx(tmpword.c_str(), tmpl, aeXPRODUCT, this, needflag); if (he) return he; @@ -351,26 +339,20 @@ char* PfxEntry::check_twosfx_morph(const char* word, int len, char in_compound, const FLAG needflag) { - int tmpl; // length of tmpword - char tmpword[MAXTEMPWORDLEN]; - // on entry prefix is 0 length or already matches the beginning of the word. // So if the remaining root word has positive length // and if there are enough chars in root word and added back strip chars // to meet the number of characters conditions, then test it - tmpl = len - appnd.size(); + int tmpl = len - appnd.size(); // length of tmpword if ((tmpl > 0 || (tmpl == 0 && pmyMgr->get_fullstrip())) && (tmpl + strip.size() >= numconds)) { // generate new root word by removing prefix and adding // back any characters that would have been stripped - if (strip.size()) { - strncpy(tmpword, strip.c_str(), MAXTEMPWORDLEN - 1); - tmpword[MAXTEMPWORDLEN - 1] = '\0'; - } - strcpy((tmpword + strip.size()), (word + appnd.size())); + std::string tmpword(strip); + tmpword.append(word + appnd.size()); // now make sure all of the conditions on characters // are met. Please see the appendix at the end of @@ -380,7 +362,7 @@ char* PfxEntry::check_twosfx_morph(const char* word, // if all conditions are met then check if resulting // root word in the dictionary - if (test_condition(tmpword)) { + if (test_condition(tmpword.c_str())) { tmpl += strip.size(); // prefix matched but no root word was found @@ -388,7 +370,8 @@ char* PfxEntry::check_twosfx_morph(const char* word, // ross checked combined with a suffix if ((opts & aeXPRODUCT) && (in_compound != IN_CPD_BEGIN)) { - return pmyMgr->suffix_check_twosfx_morph(tmpword, tmpl, aeXPRODUCT, + return pmyMgr->suffix_check_twosfx_morph(tmpword.c_str(), tmpl, + aeXPRODUCT, this, needflag); } } @@ -401,31 +384,23 @@ char* PfxEntry::check_morph(const char* word, int len, char in_compound, const FLAG needflag) { - int tmpl; // length of tmpword struct hentry* he; // hash entry of root word or NULL - char tmpword[MAXTEMPWORDLEN]; - char result[MAXLNLEN]; char* st; - *result = '\0'; - // on entry prefix is 0 length or already matches the beginning of the word. // So if the remaining root word has positive length // and if there are enough chars in root word and added back strip chars // to meet the number of characters conditions, then test it - tmpl = len - appnd.size(); + int tmpl = len - appnd.size(); // length of tmpword if ((tmpl > 0 || (tmpl == 0 && pmyMgr->get_fullstrip())) && (tmpl + strip.size() >= numconds)) { // generate new root word by removing prefix and adding // back any characters that would have been stripped - if (strip.size()) { - strncpy(tmpword, strip.c_str(), MAXTEMPWORDLEN - 1); - tmpword[MAXTEMPWORDLEN - 1] = '\0'; - } - strcpy(tmpword + strip.size(), word + appnd.size()); + std::string tmpword(strip); + tmpword.append(word + appnd.size()); // now make sure all of the conditions on characters // are met. Please see the appendix at the end of @@ -435,9 +410,11 @@ char* PfxEntry::check_morph(const char* word, // if all conditions are met then check if resulting // root word in the dictionary - if (test_condition(tmpword)) { + if (test_condition(tmpword.c_str())) { + std::string result; + tmpl += strip.size(); - if ((he = pmyMgr->lookup(tmpword)) != NULL) { + if ((he = pmyMgr->lookup(tmpword.c_str())) != NULL) { do { if (TESTAFF(he->astr, aflag, he->alen) && // forbid single prefixes with needaffix flag @@ -446,28 +423,28 @@ char* PfxEntry::check_morph(const char* word, ((!needflag) || TESTAFF(he->astr, needflag, he->alen) || (contclass && TESTAFF(contclass, needflag, contclasslen)))) { if (morphcode) { - mystrcat(result, " ", MAXLNLEN); - mystrcat(result, morphcode, MAXLNLEN); + result.append(" "); + result.append(morphcode); } else - mystrcat(result, getKey(), MAXLNLEN); + result.append(getKey()); if (!HENTRY_FIND(he, MORPH_STEM)) { - mystrcat(result, " ", MAXLNLEN); - mystrcat(result, MORPH_STEM, MAXLNLEN); - mystrcat(result, HENTRY_WORD(he), MAXLNLEN); + result.append(" "); + result.append(MORPH_STEM); + result.append(HENTRY_WORD(he)); } // store the pointer of the hash entry if (HENTRY_DATA(he)) { - mystrcat(result, " ", MAXLNLEN); - mystrcat(result, HENTRY_DATA2(he), MAXLNLEN); + result.append(" "); + result.append(HENTRY_DATA2(he)); } else { // return with debug information char* flag = pmyMgr->encode_flag(getFlag()); - mystrcat(result, " ", MAXLNLEN); - mystrcat(result, MORPH_FLAG, MAXLNLEN); - mystrcat(result, flag, MAXLNLEN); + result.append(" "); + result.append(MORPH_FLAG); + result.append(flag); free(flag); } - mystrcat(result, "\n", MAXLNLEN); + result.append("\n"); } he = he->next_homonym; } while (he); @@ -478,18 +455,19 @@ char* PfxEntry::check_morph(const char* word, // ross checked combined with a suffix if ((opts & aeXPRODUCT) && (in_compound != IN_CPD_BEGIN)) { - st = pmyMgr->suffix_check_morph(tmpword, tmpl, aeXPRODUCT, this, + st = pmyMgr->suffix_check_morph(tmpword.c_str(), tmpl, aeXPRODUCT, this, FLAG_NULL, needflag); if (st) { - mystrcat(result, st, MAXLNLEN); + result.append(st); free(st); } } + + if (!result.empty()) + return mystrdup(result.c_str()); } } - if (*result) - return mystrdup(result); return NULL; } @@ -516,7 +494,8 @@ SfxEntry::SfxEntry(AffixMgr* pmgr, affentry* dp) c.l.conds2 = dp->c.l.conds2; } else memcpy(c.conds, dp->c.conds, MAXCONDLEN); - rappnd = myrevstrdup(appnd.c_str()); + rappnd = appnd; + reverseword(rappnd); morphcode = dp->morphcode; contclass = dp->contclass; contclasslen = dp->contclasslen; @@ -524,8 +503,6 @@ SfxEntry::SfxEntry(AffixMgr* pmgr, affentry* dp) SfxEntry::~SfxEntry() { aflag = 0; - if (rappnd) - free(rappnd); pmyMgr = NULL; if (opts & aeLONGCOND) free(c.l.conds2); @@ -536,13 +513,12 @@ SfxEntry::~SfxEntry() { } // add suffix to this word assuming conditions hold -char* SfxEntry::add(const char* word, int len) { +char* SfxEntry::add(const char* word, size_t len) { /* make sure all conditions match */ if ((len > strip.size() || (len == 0 && pmyMgr->get_fullstrip())) && (len >= numconds) && test_condition(word + len, word) && (!strip.size() || - (strcmp(word + len - strip.size(), strip.c_str()) == 0)) && - ((MAXTEMPWORDLEN) > (len + appnd.size() - strip.size()))) { + (strcmp(word + len - strip.size(), strip.c_str()) == 0))) { std::string tword(word); /* we have a match so add suffix */ tword.replace(len - strip.size(), std::string::npos, appnd); @@ -699,10 +675,7 @@ struct hentry* SfxEntry::checkword(const char* word, const FLAG cclass, const FLAG needflag, const FLAG badflag) { - int tmpl; // length of tmpword struct hentry* he; // hash entry pointer - unsigned char* cp; - char tmpword[MAXTEMPWORDLEN]; PfxEntry* ep = ppfx; // if this suffix is being cross checked with a prefix @@ -716,7 +689,7 @@ struct hentry* SfxEntry::checkword(const char* word, // and if there are enough chars in root word and added back strip chars // to meet the number of characters conditions, then test it - tmpl = len - appnd.size(); + int tmpl = len - appnd.size(); // length of tmpword // the second condition is not enough for UTF-8 strings // it checked in test_condition() @@ -726,15 +699,13 @@ struct hentry* SfxEntry::checkword(const char* word, // back any characters that would have been stripped or // or null terminating the shorter string - strncpy(tmpword, word, MAXTEMPWORDLEN - 1); - tmpword[MAXTEMPWORDLEN - 1] = '\0'; - cp = (unsigned char*)(tmpword + tmpl); + std::string tmpstring(word, tmpl); if (strip.size()) { - strcpy((char*)cp, strip.c_str()); - tmpl += strip.size(); - cp = (unsigned char*)(tmpword + tmpl); - } else - *cp = '\0'; + tmpstring.append(strip); + } + + const char* tmpword = tmpstring.c_str(); + const char* endword = tmpword + tmpstring.size(); // now make sure all of the conditions on characters // are met. Please see the appendix at the end of @@ -744,7 +715,7 @@ struct hentry* SfxEntry::checkword(const char* word, // if all conditions are met then check if resulting // root word in the dictionary - if (test_condition((char*)cp, (char*)tmpword)) { + if (test_condition(endword, tmpword)) { #ifdef SZOSZABLYA_POSSIBLE_ROOTS fprintf(stdout, "%s %s %c\n", word, tmpword, aflag); #endif @@ -804,10 +775,7 @@ struct hentry* SfxEntry::check_twosfx(const char* word, int optflags, PfxEntry* ppfx, const FLAG needflag) { - int tmpl; // length of tmpword struct hentry* he; // hash entry pointer - unsigned char* cp; - char tmpword[MAXTEMPWORDLEN]; PfxEntry* ep = ppfx; // if this suffix is being cross checked with a prefix @@ -821,7 +789,7 @@ struct hentry* SfxEntry::check_twosfx(const char* word, // and if there are enough chars in root word and added back strip chars // to meet the number of characters conditions, then test it - tmpl = len - appnd.size(); + int tmpl = len - appnd.size(); // length of tmpword if ((tmpl > 0 || (tmpl == 0 && pmyMgr->get_fullstrip())) && (tmpl + strip.size() >= numconds)) { @@ -829,15 +797,13 @@ struct hentry* SfxEntry::check_twosfx(const char* word, // back any characters that would have been stripped or // or null terminating the shorter string - strncpy(tmpword, word, MAXTEMPWORDLEN - 1); - tmpword[MAXTEMPWORDLEN - 1] = '\0'; - cp = (unsigned char*)(tmpword + tmpl); - if (strip.size()) { - strcpy((char*)cp, strip.c_str()); - tmpl += strip.size(); - cp = (unsigned char*)(tmpword + tmpl); - } else - *cp = '\0'; + std::string tmpword(word); + tmpword.resize(tmpl); + tmpword.append(strip); + tmpl += strip.size(); + + const char* beg = tmpword.c_str(); + const char* end = beg + tmpl; // now make sure all of the conditions on characters // are met. Please see the appendix at the end of @@ -846,17 +812,17 @@ struct hentry* SfxEntry::check_twosfx(const char* word, // if all conditions are met then recall suffix_check - if (test_condition((char*)cp, (char*)tmpword)) { + if (test_condition(end, beg)) { if (ppfx) { // handle conditional suffix if ((contclass) && TESTAFF(contclass, ep->getFlag(), contclasslen)) - he = pmyMgr->suffix_check(tmpword, tmpl, 0, NULL, NULL, 0, NULL, + he = pmyMgr->suffix_check(tmpword.c_str(), tmpl, 0, NULL, NULL, 0, NULL, (FLAG)aflag, needflag); else - he = pmyMgr->suffix_check(tmpword, tmpl, optflags, ppfx, NULL, 0, + he = pmyMgr->suffix_check(tmpword.c_str(), tmpl, optflags, ppfx, NULL, 0, NULL, (FLAG)aflag, needflag); } else { - he = pmyMgr->suffix_check(tmpword, tmpl, 0, NULL, NULL, 0, NULL, + he = pmyMgr->suffix_check(tmpword.c_str(), tmpl, 0, NULL, NULL, 0, NULL, (FLAG)aflag, needflag); } if (he) @@ -872,9 +838,6 @@ char* SfxEntry::check_twosfx_morph(const char* word, int optflags, PfxEntry* ppfx, const FLAG needflag) { - int tmpl; // length of tmpword - unsigned char* cp; - char tmpword[MAXTEMPWORDLEN]; PfxEntry* ep = ppfx; char* st; @@ -893,7 +856,7 @@ char* SfxEntry::check_twosfx_morph(const char* word, // and if there are enough chars in root word and added back strip chars // to meet the number of characters conditions, then test it - tmpl = len - appnd.size(); + int tmpl = len - appnd.size(); // length of tmpword if ((tmpl > 0 || (tmpl == 0 && pmyMgr->get_fullstrip())) && (tmpl + strip.size() >= numconds)) { @@ -901,15 +864,13 @@ char* SfxEntry::check_twosfx_morph(const char* word, // back any characters that would have been stripped or // or null terminating the shorter string - strncpy(tmpword, word, MAXTEMPWORDLEN - 1); - tmpword[MAXTEMPWORDLEN - 1] = '\0'; - cp = (unsigned char*)(tmpword + tmpl); - if (strip.size()) { - strcpy((char*)cp, strip.c_str()); - tmpl += strip.size(); - cp = (unsigned char*)(tmpword + tmpl); - } else - *cp = '\0'; + std::string tmpword(word); + tmpword.resize(tmpl); + tmpword.append(strip); + tmpl += strip.size(); + + const char* beg = tmpword.c_str(); + const char* end = beg + tmpl; // now make sure all of the conditions on characters // are met. Please see the appendix at the end of @@ -918,11 +879,11 @@ char* SfxEntry::check_twosfx_morph(const char* word, // if all conditions are met then recall suffix_check - if (test_condition((char*)cp, (char*)tmpword)) { + if (test_condition(end, beg)) { if (ppfx) { // handle conditional suffix if ((contclass) && TESTAFF(contclass, ep->getFlag(), contclasslen)) { - st = pmyMgr->suffix_check_morph(tmpword, tmpl, 0, NULL, aflag, + st = pmyMgr->suffix_check_morph(tmpword.c_str(), tmpl, 0, NULL, aflag, needflag); if (st) { if (ppfx->getMorph()) { @@ -934,7 +895,7 @@ char* SfxEntry::check_twosfx_morph(const char* word, mychomp(result); } } else { - st = pmyMgr->suffix_check_morph(tmpword, tmpl, optflags, ppfx, aflag, + st = pmyMgr->suffix_check_morph(tmpword.c_str(), tmpl, optflags, ppfx, aflag, needflag); if (st) { mystrcat(result, st, MAXLNLEN); @@ -944,7 +905,7 @@ char* SfxEntry::check_twosfx_morph(const char* word, } } else { st = - pmyMgr->suffix_check_morph(tmpword, tmpl, 0, NULL, aflag, needflag); + pmyMgr->suffix_check_morph(tmpword.c_str(), tmpl, 0, NULL, aflag, needflag); if (st) { mystrcat(result, st, MAXLNLEN); free(st); diff --git a/libs/hunspell/src/affentry.hxx b/libs/hunspell/src/affentry.hxx index f3db20013c..6311d83fff 100644 --- a/libs/hunspell/src/affentry.hxx +++ b/libs/hunspell/src/affentry.hxx @@ -122,7 +122,7 @@ class LIBHUNSPELL_DLL_EXPORTED PfxEntry : protected AffEntry { inline FLAG getFlag() { return aflag; } inline const char* getKey() { return appnd.c_str(); } - char* add(const char* word, int len); + char* add(const char* word, size_t len); inline short getKeyLen() { return appnd.size(); } @@ -154,7 +154,7 @@ class LIBHUNSPELL_DLL_EXPORTED SfxEntry : protected AffEntry { private: AffixMgr* pmyMgr; - char* rappnd; + std::string rappnd; SfxEntry* next; SfxEntry* nexteq; @@ -200,8 +200,8 @@ class LIBHUNSPELL_DLL_EXPORTED SfxEntry : protected AffEntry { const FLAG needflag); inline FLAG getFlag() { return aflag; } - inline const char* getKey() { return rappnd; } - char* add(const char* word, int len); + inline const char* getKey() { return rappnd.c_str(); } + char* add(const char* word, size_t len); inline const char* getMorph() { return morphcode; } diff --git a/libs/hunspell/src/affixmgr.c++ b/libs/hunspell/src/affixmgr.c++ index d21ff49573..d6bb677982 100644 --- a/libs/hunspell/src/affixmgr.c++ +++ b/libs/hunspell/src/affixmgr.c++ @@ -72,13 +72,13 @@ */ #include <stdlib.h> -#include <string> #include <string.h> #include <stdio.h> #include <ctype.h> +#include <algorithm> #include <limits> - +#include <string> #include <vector> #include "affixmgr.hxx" @@ -152,11 +152,7 @@ AffixMgr::AffixMgr(const char* affpath, cpdsyllablenum = NULL; // syllable count incrementing flag checknum = 0; // checking numbers, and word with numbers wordchars = NULL; // letters + spec. word characters - wordchars_utf16 = NULL; // letters + spec. word characters - wordchars_utf16_len = 0; // letters + spec. word characters ignorechars = NULL; // letters + spec. word characters - ignorechars_utf16 = NULL; // letters + spec. word characters - ignorechars_utf16_len = 0; // letters + spec. word characters version = NULL; // affix and dictionary file version string havecontclass = 0; // flags of possible continuing classes (double affix) // LEMMA_PRESENT: not put root into the morphological output. Lemma presents @@ -336,12 +332,8 @@ AffixMgr::~AffixMgr() { free(lang); if (wordchars) free(wordchars); - if (wordchars_utf16) - free(wordchars_utf16); if (ignorechars) free(ignorechars); - if (ignorechars_utf16) - free(ignorechars_utf16); if (version) free(version); checknum = 0; @@ -632,8 +624,8 @@ int AffixMgr::parse_file(const char* affpath, const char* key) { /* parse in the extra word characters */ if (strncmp(line, "WORDCHARS", 9) == 0) { - if (parse_array(line, &wordchars, &wordchars_utf16, &wordchars_utf16_len, - utf8, afflst->getlinenum())) { + if (!parse_array(line, &wordchars, wordchars_utf16, + utf8, afflst->getlinenum())) { finishFileMgr(afflst); return 1; } @@ -642,8 +634,8 @@ int AffixMgr::parse_file(const char* affpath, const char* key) { /* parse in the ignored characters (for example, Arabic optional diacretics * charachters */ if (strncmp(line, "IGNORE", 6) == 0) { - if (parse_array(line, &ignorechars, &ignorechars_utf16, - &ignorechars_utf16_len, utf8, afflst->getlinenum())) { + if (!parse_array(line, &ignorechars, ignorechars_utf16, + utf8, afflst->getlinenum())) { finishFileMgr(afflst); return 1; } @@ -1174,7 +1166,7 @@ std::string& AffixMgr::debugflag(std::string& result, unsigned short flag) { } // calculate the character length of the condition -int AffixMgr::condlen(char* st) { +int AffixMgr::condlen(const char* st) { int l = 0; bool group = false; for (; *st; st++) { @@ -1189,7 +1181,7 @@ int AffixMgr::condlen(char* st) { return l; } -int AffixMgr::encodeit(affentry& entry, char* cs) { +int AffixMgr::encodeit(affentry& entry, const char* cs) { if (strcmp(cs, ".") != 0) { entry.numconds = (char)condlen(cs); // coverity[buffer_size_warning] - deliberate use of lack of end of conds @@ -1328,7 +1320,6 @@ char* AffixMgr::prefix_check_morph(const char* word, int len, char in_compound, const FLAG needflag) { - char* st; char result[MAXLNLEN]; result[0] = '\0'; @@ -1340,7 +1331,7 @@ char* AffixMgr::prefix_check_morph(const char* word, // first handle the special case of 0 length prefixes PfxEntry* pe = pStart[0]; while (pe) { - st = pe->check_morph(word, len, in_compound, needflag); + char* st = pe->check_morph(word, len, in_compound, needflag); if (st) { mystrcat(result, st, MAXLNLEN); free(st); @@ -1355,7 +1346,7 @@ char* AffixMgr::prefix_check_morph(const char* word, while (pptr) { if (isSubset(pptr->getKey(), word)) { - st = pptr->check_morph(word, len, in_compound, needflag); + char* st = pptr->check_morph(word, len, in_compound, needflag); if (st) { // fogemorpheme if ((in_compound != IN_CPD_NOT) || @@ -1382,8 +1373,6 @@ char* AffixMgr::prefix_check_twosfx_morph(const char* word, int len, char in_compound, const FLAG needflag) { - char* st; - char result[MAXLNLEN]; result[0] = '\0'; @@ -1394,7 +1383,7 @@ char* AffixMgr::prefix_check_twosfx_morph(const char* word, // first handle the special case of 0 length prefixes PfxEntry* pe = pStart[0]; while (pe) { - st = pe->check_twosfx_morph(word, len, in_compound, needflag); + char* st = pe->check_twosfx_morph(word, len, in_compound, needflag); if (st) { mystrcat(result, st, MAXLNLEN); free(st); @@ -1408,7 +1397,7 @@ char* AffixMgr::prefix_check_twosfx_morph(const char* word, while (pptr) { if (isSubset(pptr->getKey(), word)) { - st = pptr->check_twosfx_morph(word, len, in_compound, needflag); + char* st = pptr->check_twosfx_morph(word, len, in_compound, needflag); if (st) { mystrcat(result, st, MAXLNLEN); free(st); @@ -1427,13 +1416,12 @@ char* AffixMgr::prefix_check_twosfx_morph(const char* word, // Is word a non compound with a REP substitution (see checkcompoundrep)? int AffixMgr::cpdrep_check(const char* word, int wl) { - const char* r; if ((wl < 2) || !numrep) return 0; for (int i = 0; i < numrep; i++) { - r = word; + const char* r = word; int lenp = strlen(reptable[i].pattern); // search every occurence of the pattern in the word while ((r = strstr(r, reptable[i].pattern)) != NULL) { @@ -1478,14 +1466,14 @@ int AffixMgr::cpdpat_check(const char* word, // bounds int AffixMgr::cpdcase_check(const char* word, int pos) { if (utf8) { - w_char u, w; const char* p; - u8_u16(&u, 1, word + pos); for (p = word + pos - 1; (*p & 0xc0) == 0x80; p--) ; - u8_u16(&w, 1, p); - unsigned short a = (u.h << 8) + u.l; - unsigned short b = (w.h << 8) + w.l; + std::string pair(p); + std::vector<w_char> pair_u; + u8_u16(pair_u, pair); + unsigned short a = pair_u.size() > 1 ? ((pair_u[1].h << 8) + pair_u[1].l) : 0; + unsigned short b = !pair_u.empty() ? ((pair_u[0].h << 8) + pair_u[0].l) : 0; if (((unicodetoupper(a, langnum) == a) || (unicodetoupper(b, langnum) == b)) && (a != '-') && (b != '-')) @@ -1499,20 +1487,18 @@ int AffixMgr::cpdcase_check(const char* word, int pos) { return 0; } +struct metachar_data { + signed short btpp; // metacharacter (*, ?) position for backtracking + signed short btwp; // word position for metacharacters + int btnum; // number of matched characters in metacharacter +}; + // check compound patterns int AffixMgr::defcpd_check(hentry*** words, short wnum, hentry* rv, hentry** def, char all) { - signed short - btpp[MAXWORDLEN]; // metacharacter (*, ?) positions for backtracking - signed short btwp[MAXWORDLEN]; // word positions for metacharacters - int btnum[MAXWORDLEN]; // number of matched characters in metacharacter - // positions - short bt = 0; - int i, j; - int ok; int w = 0; if (!*words) { @@ -1524,6 +1510,11 @@ int AffixMgr::defcpd_check(hentry*** words, return 0; } + std::vector<metachar_data> btinfo(1); + + short bt = 0; + int i, j; + (*words)[wnum] = rv; // has the last word COMPOUNDRULE flag? @@ -1533,7 +1524,7 @@ int AffixMgr::defcpd_check(hentry*** words, *words = NULL; return 0; } - ok = 0; + int ok = 0; for (i = 0; i < numdefcpd; i++) { for (j = 0; j < defcpdtable[i].len; j++) { if (defcpdtable[i].def[j] != '*' && defcpdtable[i].def[j] != '?' && @@ -1564,8 +1555,8 @@ int AffixMgr::defcpd_check(hentry*** words, int wend = (defcpdtable[i].def[pp + 1] == '?') ? wp : wnum; ok2 = 1; pp += 2; - btpp[bt] = pp; - btwp[bt] = wp; + btinfo[bt].btpp = pp; + btinfo[bt].btwp = wp; while (wp <= wend) { if (!(*words)[wp]->alen || !TESTAFF((*words)[wp]->astr, defcpdtable[i].def[pp - 2], @@ -1577,9 +1568,11 @@ int AffixMgr::defcpd_check(hentry*** words, } if (wp <= wnum) ok2 = 0; - btnum[bt] = wp - btwp[bt]; - if (btnum[bt] > 0) - bt++; + btinfo[bt].btnum = wp - btinfo[bt].btwp; + if (btinfo[bt].btnum > 0) { + ++bt; + btinfo.resize(bt+1); + } if (ok2) break; } else { @@ -1609,10 +1602,10 @@ int AffixMgr::defcpd_check(hentry*** words, if (bt) do { ok = 1; - btnum[bt - 1]--; - pp = btpp[bt - 1]; - wp = btwp[bt - 1] + (signed short)btnum[bt - 1]; - } while ((btnum[bt - 1] < 0) && --bt); + btinfo[bt - 1].btnum--; + pp = btinfo[bt - 1].btpp; + wp = btinfo[bt - 1].btwp + (signed short)btinfo[bt - 1].btnum; + } while ((btinfo[bt - 1].btnum < 0) && --bt); } while (bt); if (ok && ok2 && (!all || (defcpdtable[i].len <= pp))) @@ -1650,24 +1643,26 @@ inline int AffixMgr::candidate_check(const char* word, int len) { } // calculate number of syllable for compound-checking -short AffixMgr::get_syllable(const char* word, int wlen) { +short AffixMgr::get_syllable(const std::string& word) { if (cpdmaxsyllable == 0) return 0; short num = 0; if (!utf8) { - for (int i = 0; i < wlen; i++) { + for (size_t i = 0; i < word.size(); ++i) { if (strchr(cpdvowels, word[i])) num++; } } else if (cpdvowels_utf16) { - w_char w[MAXWORDUTF8LEN]; - int i = u8_u16(w, MAXWORDUTF8LEN, word); + std::vector<w_char> w; + int i = u8_u16(w, word); for (; i > 0; i--) { - if (flag_bsearch((unsigned short*)cpdvowels_utf16, - ((unsigned short*)w)[i - 1], cpdvowels_utf16_len)) - num++; + if (std::binary_search(cpdvowels_utf16, + cpdvowels_utf16 + cpdvowels_utf16_len, + w[i - 1])) { + ++num; + } } } return num; @@ -1676,12 +1671,12 @@ short AffixMgr::get_syllable(const char* word, int wlen) { void AffixMgr::setcminmax(int* cmin, int* cmax, const char* word, int len) { if (utf8) { int i; - for (*cmin = 0, i = 0; (i < cpdmin) && word[*cmin]; i++) { - for ((*cmin)++; (word[*cmin] & 0xc0) == 0x80; (*cmin)++) + for (*cmin = 0, i = 0; (i < cpdmin) && *cmin < len; i++) { + for ((*cmin)++; *cmin < len && (word[*cmin] & 0xc0) == 0x80; (*cmin)++) ; } - for (*cmax = len, i = 0; (i < (cpdmin - 1)) && *cmax; i++) { - for ((*cmax)--; (word[*cmax] & 0xc0) == 0x80; (*cmax)--) + for (*cmax = len, i = 0; (i < (cpdmin - 1)) && *cmax >= 0; i++) { + for ((*cmax)--; *cmax >= 0 && (word[*cmax] & 0xc0) == 0x80; (*cmax)--) ; } } else { @@ -1699,6 +1694,7 @@ struct hentry* AffixMgr::compound_check(const char* word, short maxwordnum, short wnum, hentry** words = NULL, + hentry** rwords = NULL, char hu_mov_rule = 0, char is_sug = 0, int* info = NULL) { @@ -1706,8 +1702,7 @@ struct hentry* AffixMgr::compound_check(const char* word, short oldnumsyllable, oldnumsyllable2, oldwordnum, oldwordnum2; struct hentry* rv = NULL; struct hentry* rv_first; - struct hentry* rwords[MAXWORDLEN]; // buffer for COMPOUND pattern checking - char st[MAXWORDUTF8LEN + 4]; + std::string st; char ch = '\0'; int cmin; int cmax; @@ -1726,7 +1721,7 @@ struct hentry* AffixMgr::compound_check(const char* word, setcminmax(&cmin, &cmax, word, len); - strcpy(st, word); + st.assign(word); for (i = cmin; i < cmax; i++) { // go to end of the UTF-8 character @@ -1758,11 +1753,11 @@ struct hentry* AffixMgr::compound_check(const char* word, if (scpd > numcheckcpd) break; // break simplified checkcompoundpattern loop - strcpy(st + i, checkcpdtable[scpd - 1].pattern); + st.replace(i, std::string::npos, checkcpdtable[scpd - 1].pattern); soldi = i; i += strlen(checkcpdtable[scpd - 1].pattern); - strcpy(st + i, checkcpdtable[scpd - 1].pattern2); - strcpy(st + i + strlen(checkcpdtable[scpd - 1].pattern2), + st.replace(i, std::string::npos, checkcpdtable[scpd - 1].pattern2); + st.replace(i + strlen(checkcpdtable[scpd - 1].pattern2), std::string::npos, word + soldi + strlen(checkcpdtable[scpd - 1].pattern3)); oldlen = len; @@ -1771,7 +1766,7 @@ struct hentry* AffixMgr::compound_check(const char* word, strlen(checkcpdtable[scpd - 1].pattern3); oldcmin = cmin; oldcmax = cmax; - setcminmax(&cmin, &cmax, st, len); + setcminmax(&cmin, &cmax, st.c_str(), len); cmax = len - cpdmin + 1; } @@ -1785,7 +1780,7 @@ struct hentry* AffixMgr::compound_check(const char* word, // FIRST WORD affixed = 1; - rv = lookup(st); // perhaps without prefix + rv = lookup(st.c_str()); // perhaps without prefix // search homonym with compound flag while ((rv) && !hu_mov_rule && @@ -1798,9 +1793,9 @@ struct hentry* AffixMgr::compound_check(const char* word, TESTAFF(rv->astr, compoundmiddle, rv->alen)) || (numdefcpd && onlycpdrule && ((!words && !wordnum && - defcpd_check(&words, wnum, rv, (hentry**)&rwords, 0)) || + defcpd_check(&words, wnum, rv, rwords, 0)) || (words && - defcpd_check(&words, wnum, rv, (hentry**)&rwords, 0))))) || + defcpd_check(&words, wnum, rv, rwords, 0))))) || (scpd != 0 && checkcpdtable[scpd - 1].cond != FLAG_NULL && !TESTAFF(rv->astr, checkcpdtable[scpd - 1].cond, rv->alen)))) { rv = rv->next_homonym; @@ -1813,14 +1808,14 @@ struct hentry* AffixMgr::compound_check(const char* word, if (onlycpdrule) break; if (compoundflag && - !(rv = prefix_check(st, i, + !(rv = prefix_check(st.c_str(), i, hu_mov_rule ? IN_CPD_OTHER : IN_CPD_BEGIN, compoundflag))) { if (((rv = suffix_check( - st, i, 0, NULL, NULL, 0, NULL, FLAG_NULL, compoundflag, + st.c_str(), i, 0, NULL, NULL, 0, NULL, FLAG_NULL, compoundflag, hu_mov_rule ? IN_CPD_OTHER : IN_CPD_BEGIN)) || (compoundmoresuffixes && - (rv = suffix_check_twosfx(st, i, 0, NULL, compoundflag)))) && + (rv = suffix_check_twosfx(st.c_str(), i, 0, NULL, compoundflag)))) && !hu_mov_rule && sfx->getCont() && ((compoundforbidflag && TESTAFF(sfx->getCont(), compoundforbidflag, @@ -1834,24 +1829,24 @@ struct hentry* AffixMgr::compound_check(const char* word, if (rv || (((wordnum == 0) && compoundbegin && ((rv = suffix_check( - st, i, 0, NULL, NULL, 0, NULL, FLAG_NULL, compoundbegin, + st.c_str(), i, 0, NULL, NULL, 0, NULL, FLAG_NULL, compoundbegin, hu_mov_rule ? IN_CPD_OTHER : IN_CPD_BEGIN)) || (compoundmoresuffixes && (rv = suffix_check_twosfx( - st, i, 0, NULL, + st.c_str(), i, 0, NULL, compoundbegin))) || // twofold suffixes + compound - (rv = prefix_check(st, i, + (rv = prefix_check(st.c_str(), i, hu_mov_rule ? IN_CPD_OTHER : IN_CPD_BEGIN, compoundbegin)))) || ((wordnum > 0) && compoundmiddle && ((rv = suffix_check( - st, i, 0, NULL, NULL, 0, NULL, FLAG_NULL, compoundmiddle, + st.c_str(), i, 0, NULL, NULL, 0, NULL, FLAG_NULL, compoundmiddle, hu_mov_rule ? IN_CPD_OTHER : IN_CPD_BEGIN)) || (compoundmoresuffixes && (rv = suffix_check_twosfx( - st, i, 0, NULL, + st.c_str(), i, 0, NULL, compoundmiddle))) || // twofold suffixes + compound - (rv = prefix_check(st, i, + (rv = prefix_check(st.c_str(), i, hu_mov_rule ? IN_CPD_OTHER : IN_CPD_BEGIN, compoundmiddle)))))) checked_prefix = 1; @@ -1942,7 +1937,7 @@ struct hentry* AffixMgr::compound_check(const char* word, cpdcase_check(word, i)))) // LANG_hu section: spec. Hungarian rule || ((!rv) && (langnum == LANG_hu) && hu_mov_rule && - (rv = affix_check(st, i)) && + (rv = affix_check(st.c_str(), i)) && (sfx && sfx->getCont() && ( // XXX hardwired Hungarian dic. codes TESTAFF(sfx->getCont(), (unsigned short)'x', @@ -1954,10 +1949,10 @@ struct hentry* AffixMgr::compound_check(const char* word, // LANG_hu section: spec. Hungarian rule if (langnum == LANG_hu) { // calculate syllable number of the word - numsyllable += get_syllable(st, i); + numsyllable += get_syllable(st.substr(i)); // + 1 word, if syllable number of the prefix > 1 (hungarian // convention) - if (pfx && (get_syllable(pfx->getKey(), strlen(pfx->getKey())) > 1)) + if (pfx && (get_syllable(pfx->getKey()) > 1)) wordnum++; } // END of LANG_hu section @@ -1977,7 +1972,7 @@ struct hentry* AffixMgr::compound_check(const char* word, striple = 1; } - rv = lookup((st + i)); // perhaps without prefix + rv = lookup(st.c_str() + i); // perhaps without prefix // search homonym with compound flag while ((rv) && @@ -2039,7 +2034,7 @@ struct hentry* AffixMgr::compound_check(const char* word, (compoundend && TESTAFF(rv->astr, compoundend, rv->alen))) && (((cpdwordmax == -1) || (wordnum + 1 < cpdwordmax)) || ((cpdmaxsyllable != 0) && - (numsyllable + get_syllable(HENTRY_WORD(rv), rv->clen) <= + (numsyllable + get_syllable(std::string(HENTRY_WORD(rv), rv->clen)) <= cpdmaxsyllable))) && ( // test CHECKCOMPOUNDPATTERN @@ -2123,20 +2118,19 @@ struct hentry* AffixMgr::compound_check(const char* word, if (langnum == LANG_hu) { // calculate syllable number of the word - numsyllable += get_syllable(word + i, strlen(word + i)); + numsyllable += get_syllable(word + i); // - affix syllable num. // XXX only second suffix (inflections, not derivations) if (sfxappnd) { - char* tmp = myrevstrdup(sfxappnd); - numsyllable -= get_syllable(tmp, strlen(tmp)) + sfxextra; - free(tmp); + std::string tmp(sfxappnd); + reverseword(tmp); + numsyllable -= get_syllable(tmp) + sfxextra; } // + 1 word, if syllable number of the prefix > 1 (hungarian // convention) - if (pfx && - (get_syllable(pfx->getKey(), strlen(pfx->getKey())) > 1)) + if (pfx && (get_syllable(pfx->getKey()) > 1)) wordnum++; // increment syllable num, if last word has a SYLLABLENUM flag @@ -2187,8 +2181,8 @@ struct hentry* AffixMgr::compound_check(const char* word, // perhaps second word is a compound word (recursive call) if (wordnum < maxwordnum) { - rv = compound_check((st + i), strlen(st + i), wordnum + 1, - numsyllable, maxwordnum, wnum + 1, words, 0, + rv = compound_check(st.c_str() + i, strlen(st.c_str() + i), wordnum + 1, + numsyllable, maxwordnum, wnum + 1, words, rwords, 0, is_sug, info); if (rv && numcheckcpd && @@ -2211,11 +2205,11 @@ struct hentry* AffixMgr::compound_check(const char* word, // check first part if (strncmp(rv->word, word + i, rv->blen) == 0) { - char r = *(st + i + rv->blen); - *(st + i + rv->blen) = '\0'; + char r = st[i + rv->blen]; + st[i + rv->blen] = '\0'; - if (checkcompoundrep && cpdrep_check(st, i + rv->blen)) { - *(st + i + rv->blen) = r; + if (checkcompoundrep && cpdrep_check(st.c_str(), i + rv->blen)) { + st[ + i + rv->blen] = r; continue; } @@ -2225,11 +2219,11 @@ struct hentry* AffixMgr::compound_check(const char* word, rv2 = affix_check(word, len); if (rv2 && rv2->astr && TESTAFF(rv2->astr, forbiddenword, rv2->alen) && - (strncmp(rv2->word, st, i + rv->blen) == 0)) { + (strncmp(rv2->word, st.c_str(), i + rv->blen) == 0)) { return NULL; } } - *(st + i + rv->blen) = r; + st[i + rv->blen] = r; } } return rv_first; @@ -2262,7 +2256,7 @@ struct hentry* AffixMgr::compound_check(const char* word, if (soldi != 0) { i = soldi; - strcpy(st, word); // XXX add more optim. + st.assign(word); // XXX add more optim. soldi = 0; } else st[i] = ch; @@ -2283,6 +2277,7 @@ int AffixMgr::compound_check_morph(const char* word, short maxwordnum, short wnum, hentry** words, + hentry** rwords, char hu_mov_rule = 0, char** result = NULL, char* partresult = NULL) { @@ -2292,8 +2287,7 @@ int AffixMgr::compound_check_morph(const char* word, struct hentry* rv = NULL; struct hentry* rv_first; - struct hentry* rwords[MAXWORDLEN]; // buffer for COMPOUND pattern checking - char st[MAXWORDUTF8LEN + 4]; + std::string st; char ch; int checked_prefix; @@ -2308,7 +2302,7 @@ int AffixMgr::compound_check_morph(const char* word, setcminmax(&cmin, &cmax, word, len); - strcpy(st, word); + st.assign(word); for (i = cmin; i < cmax; i++) { // go to end of the UTF-8 character @@ -2340,7 +2334,7 @@ int AffixMgr::compound_check_morph(const char* word, if (partresult) mystrcat(presult, partresult, MAXLNLEN); - rv = lookup(st); // perhaps without prefix + rv = lookup(st.c_str()); // perhaps without prefix // search homonym with compound flag while ((rv) && !hu_mov_rule && @@ -2353,9 +2347,9 @@ int AffixMgr::compound_check_morph(const char* word, TESTAFF(rv->astr, compoundmiddle, rv->alen)) || (numdefcpd && onlycpdrule && ((!words && !wordnum && - defcpd_check(&words, wnum, rv, (hentry**)&rwords, 0)) || + defcpd_check(&words, wnum, rv, rwords, 0)) || (words && - defcpd_check(&words, wnum, rv, (hentry**)&rwords, 0))))))) { + defcpd_check(&words, wnum, rv, rwords, 0))))))) { rv = rv->next_homonym; } @@ -2363,10 +2357,10 @@ int AffixMgr::compound_check_morph(const char* word, affixed = 0; if (rv) { - sprintf(presult + strlen(presult), "%c%s%s", MSEP_FLD, MORPH_PART, st); + sprintf(presult + strlen(presult), "%c%s%s", MSEP_FLD, MORPH_PART, st.c_str()); if (!HENTRY_FIND(rv, MORPH_STEM)) { sprintf(presult + strlen(presult), "%c%s%s", MSEP_FLD, MORPH_STEM, - st); + st.c_str()); } // store the pointer of the hash entry // sprintf(presult + strlen(presult), "%c%s%p", MSEP_FLD, @@ -2382,13 +2376,13 @@ int AffixMgr::compound_check_morph(const char* word, break; if (compoundflag && !(rv = - prefix_check(st, i, hu_mov_rule ? IN_CPD_OTHER : IN_CPD_BEGIN, + prefix_check(st.c_str(), i, hu_mov_rule ? IN_CPD_OTHER : IN_CPD_BEGIN, compoundflag))) { - if (((rv = suffix_check(st, i, 0, NULL, NULL, 0, NULL, FLAG_NULL, + if (((rv = suffix_check(st.c_str(), i, 0, NULL, NULL, 0, NULL, FLAG_NULL, compoundflag, hu_mov_rule ? IN_CPD_OTHER : IN_CPD_BEGIN)) || (compoundmoresuffixes && - (rv = suffix_check_twosfx(st, i, 0, NULL, compoundflag)))) && + (rv = suffix_check_twosfx(st.c_str(), i, 0, NULL, compoundflag)))) && !hu_mov_rule && sfx->getCont() && ((compoundforbidflag && TESTAFF(sfx->getCont(), compoundforbidflag, @@ -2401,44 +2395,44 @@ int AffixMgr::compound_check_morph(const char* word, if (rv || (((wordnum == 0) && compoundbegin && - ((rv = suffix_check(st, i, 0, NULL, NULL, 0, NULL, FLAG_NULL, + ((rv = suffix_check(st.c_str(), i, 0, NULL, NULL, 0, NULL, FLAG_NULL, compoundbegin, hu_mov_rule ? IN_CPD_OTHER : IN_CPD_BEGIN)) || (compoundmoresuffixes && (rv = suffix_check_twosfx( - st, i, 0, NULL, + st.c_str(), i, 0, NULL, compoundbegin))) || // twofold suffix+compound - (rv = prefix_check(st, i, + (rv = prefix_check(st.c_str(), i, hu_mov_rule ? IN_CPD_OTHER : IN_CPD_BEGIN, compoundbegin)))) || ((wordnum > 0) && compoundmiddle && - ((rv = suffix_check(st, i, 0, NULL, NULL, 0, NULL, FLAG_NULL, + ((rv = suffix_check(st.c_str(), i, 0, NULL, NULL, 0, NULL, FLAG_NULL, compoundmiddle, hu_mov_rule ? IN_CPD_OTHER : IN_CPD_BEGIN)) || (compoundmoresuffixes && (rv = suffix_check_twosfx( - st, i, 0, NULL, + st.c_str(), i, 0, NULL, compoundmiddle))) || // twofold suffix+compound - (rv = prefix_check(st, i, + (rv = prefix_check(st.c_str(), i, hu_mov_rule ? IN_CPD_OTHER : IN_CPD_BEGIN, compoundmiddle)))))) { // char * p = prefix_check_morph(st, i, 0, compound); char* p = NULL; if (compoundflag) - p = affix_check_morph(st, i, compoundflag); + p = affix_check_morph(st.c_str(), i, compoundflag); if (!p || (*p == '\0')) { if (p) free(p); p = NULL; if ((wordnum == 0) && compoundbegin) { - p = affix_check_morph(st, i, compoundbegin); + p = affix_check_morph(st.c_str(), i, compoundbegin); } else if ((wordnum > 0) && compoundmiddle) { - p = affix_check_morph(st, i, compoundmiddle); + p = affix_check_morph(st.c_str(), i, compoundmiddle); } } if (p && (*p != '\0')) { sprintf(presult + strlen(presult), "%c%s%s%s", MSEP_FLD, MORPH_PART, - st, line_uniq_app(&p, MSEP_REC)); + st.c_str(), line_uniq_app(&p, MSEP_REC)); } if (p) free(p); @@ -2519,7 +2513,7 @@ int AffixMgr::compound_check_morph(const char* word, // LANG_hu section: spec. Hungarian rule || ((!rv) && (langnum == LANG_hu) && hu_mov_rule && - (rv = affix_check(st, i)) && + (rv = affix_check(st.c_str(), i)) && (sfx && sfx->getCont() && (TESTAFF(sfx->getCont(), (unsigned short)'x', sfx->getContLen()) || TESTAFF(sfx->getCont(), (unsigned short)'%', sfx->getContLen())))) @@ -2528,11 +2522,11 @@ int AffixMgr::compound_check_morph(const char* word, // LANG_hu section: spec. Hungarian rule if (langnum == LANG_hu) { // calculate syllable number of the word - numsyllable += get_syllable(st, i); + numsyllable += get_syllable(st.substr(i)); // + 1 word, if syllable number of the prefix > 1 (hungarian // convention) - if (pfx && (get_syllable(pfx->getKey(), strlen(pfx->getKey())) > 1)) + if (pfx && (get_syllable(pfx->getKey()) > 1)) wordnum++; } // END of LANG_hu section @@ -2608,7 +2602,7 @@ int AffixMgr::compound_check_morph(const char* word, (compoundend && TESTAFF(rv->astr, compoundend, rv->alen))) && (((cpdwordmax == -1) || (wordnum + 1 < cpdwordmax)) || ((cpdmaxsyllable != 0) && - (numsyllable + get_syllable(HENTRY_WORD(rv), rv->blen) <= + (numsyllable + get_syllable(std::string(HENTRY_WORD(rv), rv->blen)) <= cpdmaxsyllable))) && ((!checkcompounddup || (rv != rv_first)))) { // bad compound word @@ -2701,19 +2695,19 @@ int AffixMgr::compound_check_morph(const char* word, if (langnum == LANG_hu) { // calculate syllable number of the word - numsyllable += get_syllable(word + i, strlen(word + i)); + numsyllable += get_syllable(word + i); // - affix syllable num. // XXX only second suffix (inflections, not derivations) if (sfxappnd) { - char* tmp = myrevstrdup(sfxappnd); - numsyllable -= get_syllable(tmp, strlen(tmp)) + sfxextra; - free(tmp); + std::string tmp(sfxappnd); + reverseword(tmp); + numsyllable -= get_syllable(tmp) + sfxextra; } // + 1 word, if syllable number of the prefix > 1 (hungarian // convention) - if (pfx && (get_syllable(pfx->getKey(), strlen(pfx->getKey())) > 1)) + if (pfx && (get_syllable(pfx->getKey()) > 1)) wordnum++; // increment syllable num, if last word has a SYLLABLENUM flag @@ -2779,7 +2773,7 @@ int AffixMgr::compound_check_morph(const char* word, // perhaps second word is a compound word (recursive call) if ((wordnum < maxwordnum) && (ok == 0)) { compound_check_morph((word + i), strlen(word + i), wordnum + 1, - numsyllable, maxwordnum, wnum + 1, words, 0, + numsyllable, maxwordnum, wnum + 1, words, rwords, 0, result, presult); } else { rv = NULL; @@ -2795,6 +2789,7 @@ int AffixMgr::compound_check_morph(const char* word, return 0; } + // return 1 if s1 (reversed) is a leading subset of end of s2 /* inline int AffixMgr::isRevSubset(const char * s1, const char * end_of_s2, int len) @@ -3402,7 +3397,7 @@ int AffixMgr::expand_rootword(struct guessword* wlst, unsigned short al, const char* bad, int badl, - char* phon) { + const char* phon) { int nh = 0; // first add root word to list if ((nh < maxn) && @@ -3653,8 +3648,7 @@ char* AffixMgr::get_ignore() const { } // return the preferred ignore string for suggestions -unsigned short* AffixMgr::get_ignore_utf16(int* len) const { - *len = ignorechars_utf16_len; +const std::vector<w_char>& AffixMgr::get_ignore_utf16() const { return ignorechars_utf16; } @@ -3677,8 +3671,7 @@ const char* AffixMgr::get_wordchars() const { return wordchars; } -unsigned short* AffixMgr::get_wordchars_utf16(int* len) const { - *len = wordchars_utf16_len; +const std::vector<w_char>& AffixMgr::get_wordchars_utf16() const { return wordchars_utf16; } @@ -3840,7 +3833,6 @@ int AffixMgr::parse_cpdsyllable(char* line, FileMgr* af) { char* piece; int i = 0; int np = 0; - w_char w[MAXWORDLEN]; piece = mystrsep(&tp, 0); while (piece) { if (*piece != '\0') { @@ -3858,15 +3850,16 @@ int AffixMgr::parse_cpdsyllable(char* line, FileMgr* af) { if (!utf8) { cpdvowels = mystrdup(piece); } else { - int n = u8_u16(w, MAXWORDLEN, piece); - if (n > 0) { - flag_qsort((unsigned short*)w, 0, n); - cpdvowels_utf16 = (w_char*)malloc(n * sizeof(w_char)); + std::vector<w_char> w; + u8_u16(w, piece); + if (!w.empty()) { + std::sort(w.begin(), w.end()); + cpdvowels_utf16 = (w_char*)malloc(w.size() * sizeof(w_char)); if (!cpdvowels_utf16) return 1; - memcpy(cpdvowels_utf16, w, n * sizeof(w_char)); + memcpy(cpdvowels_utf16, &w[0], w.size()); } - cpdvowels_utf16_len = n; + cpdvowels_utf16_len = w.size(); } np++; break; @@ -4636,13 +4629,16 @@ int AffixMgr::parse_breaktable(char* line, FileMgr* af) { return 0; } -void AffixMgr::reverse_condition(char* piece) { +void AffixMgr::reverse_condition(std::string& piece) { + if (piece.empty()) + return; + int neg = 0; - for (char* k = piece + strlen(piece) - 1; k >= piece; k--) { + for (std::string::reverse_iterator k = piece.rbegin(); k != piece.rend(); ++k) { switch (*k) { case '[': { if (neg) - *(k + 1) = '['; + *(k - 1) = '['; else *k = ']'; break; @@ -4650,20 +4646,20 @@ void AffixMgr::reverse_condition(char* piece) { case ']': { *k = '['; if (neg) - *(k + 1) = '^'; + *(k - 1) = '^'; neg = 0; break; } case '^': { - if (*(k + 1) == ']') + if (*(k - 1) == ']') neg = 1; else - *(k + 1) = *k; + *(k - 1) = *k; break; } default: { if (neg) - *(k + 1) = *k; + *(k - 1) = *k; } } } @@ -4731,8 +4727,8 @@ int AffixMgr::parse_affix(char* line, case 3: { np++; numents = atoi(piece); - if ((numents <= 0) || ((::std::numeric_limits<size_t>::max() / - sizeof(struct affentry)) < numents)) { + if ((numents <= 0) || ((std::numeric_limits<size_t>::max() / + sizeof(struct affentry)) < static_cast<size_t>(numents))) { char* err = pHMgr->encode_flag(aflag); if (err) { HUNSPELL_WARNING(stderr, "error: line %d: bad entry number\n", @@ -4817,18 +4813,15 @@ int AffixMgr::parse_affix(char* line, // piece 3 - is string to strip or 0 for null case 2: { np++; + entry->strip = piece; if (complexprefixes) { if (utf8) - reverseword_utf(piece); + reverseword_utf(entry->strip); else - reverseword(piece); + reverseword(entry->strip); } - entry->strip = mystrdup(piece); - entry->stripl = (unsigned char)strlen(entry->strip); - if (strcmp(entry->strip, "0") == 0) { - free(entry->strip); - entry->strip = mystrdup(""); - entry->stripl = 0; + if (entry->strip.compare("0") == 0) { + entry->strip.clear(); } break; } @@ -4844,22 +4837,22 @@ int AffixMgr::parse_affix(char* line, if (dash) { *dash = '\0'; + entry->appnd = piece; + if (ignorechars) { if (utf8) { - remove_ignored_chars_utf(piece, ignorechars_utf16, - ignorechars_utf16_len); + remove_ignored_chars_utf(entry->appnd, ignorechars_utf16); } else { - remove_ignored_chars(piece, ignorechars); + remove_ignored_chars(entry->appnd, ignorechars); } } if (complexprefixes) { if (utf8) - reverseword_utf(piece); + reverseword_utf(entry->appnd); else - reverseword(piece); + reverseword(entry->appnd); } - entry->appnd = mystrdup(piece); if (pHMgr->is_aliasf()) { int index = atoi(dash + 1); @@ -4872,7 +4865,7 @@ int AffixMgr::parse_affix(char* line, } else { entry->contclasslen = (unsigned short)pHMgr->decode_flags( &(entry->contclass), dash + 1, af); - flag_qsort(entry->contclass, 0, entry->contclasslen); + std::sort(entry->contclass, entry->contclass + entry->contclasslen); } *dash = '/'; @@ -4881,74 +4874,74 @@ int AffixMgr::parse_affix(char* line, contclasses[(entry->contclass)[_i]] = 1; } } else { + entry->appnd = piece; + if (ignorechars) { if (utf8) { - remove_ignored_chars_utf(piece, ignorechars_utf16, - ignorechars_utf16_len); + remove_ignored_chars_utf(entry->appnd, ignorechars_utf16); } else { - remove_ignored_chars(piece, ignorechars); + remove_ignored_chars(entry->appnd, ignorechars); } } if (complexprefixes) { if (utf8) - reverseword_utf(piece); + reverseword_utf(entry->appnd); else - reverseword(piece); + reverseword(entry->appnd); } - entry->appnd = mystrdup(piece); } - entry->appndl = (unsigned char)strlen(entry->appnd); - if (strcmp(entry->appnd, "0") == 0) { - free(entry->appnd); - entry->appnd = mystrdup(""); - entry->appndl = 0; + if (entry->appnd.compare("0") == 0) { + entry->appnd.clear(); } break; } // piece 5 - is the conditions descriptions case 4: { + std::string chunk(piece); np++; if (complexprefixes) { if (utf8) - reverseword_utf(piece); + reverseword_utf(chunk); else - reverseword(piece); - reverse_condition(piece); + reverseword(chunk); + reverse_condition(chunk); } - if (entry->stripl && (strcmp(piece, ".") != 0) && - redundant_condition(at, entry->strip, entry->stripl, piece, + if (!entry->strip.empty() && chunk != "." && + redundant_condition(at, entry->strip.c_str(), entry->strip.size(), chunk.c_str(), af->getlinenum())) - strcpy(piece, "."); + chunk = "."; if (at == 'S') { - reverseword(piece); - reverse_condition(piece); + reverseword(chunk); + reverse_condition(chunk); } - if (encodeit(*entry, piece)) + if (encodeit(*entry, chunk.c_str())) return 1; break; } case 5: { + std::string chunk(piece); np++; if (pHMgr->is_aliasm()) { - int index = atoi(piece); + int index = atoi(chunk.c_str()); entry->morphcode = pHMgr->get_aliasm(index); } else { if (complexprefixes) { // XXX - fix me for morph. gen. if (utf8) - reverseword_utf(piece); + reverseword_utf(chunk); else - reverseword(piece); + reverseword(chunk); } // add the remaining of the line if (*tp) { *(tp - 1) = ' '; - tp = tp + strlen(tp); + chunk.push_back(' '); + chunk.append(tp); } - entry->morphcode = mystrdup(piece); + entry->morphcode = mystrdup(chunk.c_str()); if (!entry->morphcode) return 1; } @@ -5002,7 +4995,7 @@ int AffixMgr::parse_affix(char* line, } int AffixMgr::redundant_condition(char ft, - char* strip, + const char* strip, int stripl, const char* cond, int linenum) { @@ -5112,11 +5105,7 @@ int AffixMgr::get_suffix_words(short unsigned* suff, hentry* ht = ptr->checkword(nw.c_str(), nw.size(), 0, NULL, NULL, 0, NULL, 0, 0, 0); if (ht) { - slst[suff_words_cnt] = (char*)malloc(MAXWORDUTF8LEN * sizeof(char)); - if (slst[suff_words_cnt]) { - strcpy(slst[suff_words_cnt], nw.c_str()); - suff_words_cnt++; - } + slst[suff_words_cnt++] = mystrdup(nw.c_str()); } } suff++; diff --git a/libs/hunspell/src/affixmgr.hxx b/libs/hunspell/src/affixmgr.hxx index ca376953cd..d70e853388 100644 --- a/libs/hunspell/src/affixmgr.hxx +++ b/libs/hunspell/src/affixmgr.hxx @@ -160,11 +160,9 @@ class LIBHUNSPELL_DLL_EXPORTED AffixMgr { PfxEntry* pfx; // BUG: not stateless int checknum; char* wordchars; - unsigned short* wordchars_utf16; - int wordchars_utf16_len; + std::vector<w_char> wordchars_utf16; char* ignorechars; - unsigned short* ignorechars_utf16; - int ignorechars_utf16_len; + std::vector<w_char> ignorechars_utf16; char* version; char* lang; int langnum; @@ -258,9 +256,9 @@ class LIBHUNSPELL_DLL_EXPORTED AffixMgr { unsigned short al, const char* bad, int, - char*); + const char*); - short get_syllable(const char* word, int wlen); + short get_syllable(const std::string& word); int cpdrep_check(const char* word, int len); int cpdpat_check(const char* word, int len, @@ -282,6 +280,7 @@ class LIBHUNSPELL_DLL_EXPORTED AffixMgr { short maxwordnum, short wnum, hentry** words, + hentry** rwords, char hu_mov_rule, char is_sug, int* info); @@ -293,6 +292,7 @@ class LIBHUNSPELL_DLL_EXPORTED AffixMgr { short maxwordnum, short wnum, hentry** words, + hentry** rwords, char hu_mov_rule, char** result, char* partresult); @@ -317,9 +317,9 @@ class LIBHUNSPELL_DLL_EXPORTED AffixMgr { char* get_key_string(); char* get_try_string() const; const char* get_wordchars() const; - unsigned short* get_wordchars_utf16(int* len) const; + const std::vector<w_char>& get_wordchars_utf16() const; char* get_ignore() const; - unsigned short* get_ignore_utf16(int* len) const; + const std::vector<w_char>& get_ignore_utf16() const; int get_compound() const; FLAG get_compoundflag() const; FLAG get_compoundbegin() const; @@ -370,11 +370,11 @@ class LIBHUNSPELL_DLL_EXPORTED AffixMgr { int parse_defcpdtable(char* line, FileMgr* af); int parse_affix(char* line, const char at, FileMgr* af, char* dupflags); - void reverse_condition(char*); + void reverse_condition(std::string&); void debugflag(char* result, unsigned short flag); std::string& debugflag(std::string& result, unsigned short flag); - int condlen(char*); - int encodeit(affentry& entry, char* cs); + int condlen(const char*); + int encodeit(affentry& entry, const char* cs); int build_pfxtree(PfxEntry* pfxptr); int build_sfxtree(SfxEntry* sfxptr); int process_pfx_order(); @@ -383,7 +383,7 @@ class LIBHUNSPELL_DLL_EXPORTED AffixMgr { SfxEntry* process_sfx_in_order(SfxEntry* ptr, SfxEntry* nptr); int process_pfx_tree_to_list(); int process_sfx_tree_to_list(); - int redundant_condition(char, char* strip, int stripl, const char* cond, int); + int redundant_condition(char, const char* strip, int stripl, const char* cond, int); void finishFileMgr(FileMgr* afflst); }; diff --git a/libs/hunspell/src/atypes.hxx b/libs/hunspell/src/atypes.hxx index d71f62a32d..60826af20e 100644 --- a/libs/hunspell/src/atypes.hxx +++ b/libs/hunspell/src/atypes.hxx @@ -57,11 +57,11 @@ static inline void HUNSPELL_WARNING(FILE*, const char*, ...) {} #include "hashmgr.hxx" #include "w_char.hxx" +#include <algorithm> +#include <string> #define SETSIZE 256 #define CONTSIZE 65536 -#define MAXWORDLEN 100 -#define MAXWORDUTF8LEN 256 // affentry options #define aeXPRODUCT (1 << 0) @@ -98,14 +98,11 @@ static inline void HUNSPELL_WARNING(FILE*, const char*, ...) {} #define FLAG_NULL 0x00 #define FREE_FLAG(a) a = 0 -#define TESTAFF(a, b, c) \ - (flag_bsearch((unsigned short*)a, (unsigned short)b, c)) +#define TESTAFF(a, b, c) (std::binary_search(a, a + c, b)) struct affentry { - char* strip; - char* appnd; - unsigned char stripl; - unsigned char appndl; + std::string strip; + std::string appnd; char numconds; char opts; unsigned short aflag; diff --git a/libs/hunspell/src/baseaffix.hxx b/libs/hunspell/src/baseaffix.hxx index cfc6b71ca4..59256e92f3 100644 --- a/libs/hunspell/src/baseaffix.hxx +++ b/libs/hunspell/src/baseaffix.hxx @@ -59,7 +59,7 @@ class LIBHUNSPELL_DLL_EXPORTED AffEntry { contclasslen(0) {} std::string appnd; std::string strip; - char numconds; + unsigned char numconds; char opts; unsigned short aflag; union { diff --git a/libs/hunspell/src/config.h b/libs/hunspell/src/config.h index f1963148b6..1230ed0be7 100644 --- a/libs/hunspell/src/config.h +++ b/libs/hunspell/src/config.h @@ -179,9 +179,6 @@ /* Define to 1 if you have the `__argz_stringify' function. */ #define HAVE___ARGZ_STRINGIFY 1 -/* "Define if you use exterimental functions" */ -//#define HUNSPELL_EXPERIMENTAL 1 - /* "Define if you need warning messages" */ #define HUNSPELL_WARNING_ON @@ -204,5 +201,5 @@ #define PACKAGE_TARNAME /* Define to the version of this package. */ -#define PACKAGE_VERSION "1.3.4" -#define VERSION "1.3.4" +#define PACKAGE_VERSION "1.4.0" +#define VERSION "1.4.0" diff --git a/libs/hunspell/src/csutil.c++ b/libs/hunspell/src/csutil.c++ index d7411bb216..1948e4a3b3 100644 --- a/libs/hunspell/src/csutil.c++ +++ b/libs/hunspell/src/csutil.c++ @@ -144,53 +144,6 @@ FILE* myfopen(const char* path, const char* mode) { return fopen(path, mode); } -/* only UTF-16 (BMP) implementation */ -char* u16_u8(char* dest, int size, const w_char* src, int srclen) { - signed char* u8 = (signed char*)dest; - signed char* u8_max = (signed char*)(u8 + size); - const w_char* u2 = src; - const w_char* u2_max = src + srclen; - while ((u2 < u2_max) && (u8 < u8_max)) { - if (u2->h) { // > 0xFF - // XXX 4-byte haven't implemented yet. - if (u2->h >= 0x08) { // >= 0x800 (3-byte UTF-8 character) - *u8 = 0xe0 + (u2->h >> 4); - u8++; - if (u8 < u8_max) { - *u8 = 0x80 + ((u2->h & 0xf) << 2) + (u2->l >> 6); - u8++; - if (u8 < u8_max) { - *u8 = 0x80 + (u2->l & 0x3f); - u8++; - } - } - } else { // < 0x800 (2-byte UTF-8 character) - *u8 = 0xc0 + (u2->h << 2) + (u2->l >> 6); - u8++; - if (u8 < u8_max) { - *u8 = 0x80 + (u2->l & 0x3f); - u8++; - } - } - } else { // <= 0xFF - if (u2->l & 0x80) { // >0x80 (2-byte UTF-8 character) - *u8 = 0xc0 + (u2->l >> 6); - u8++; - if (u8 < u8_max) { - *u8 = 0x80 + (u2->l & 0x3f); - u8++; - } - } else { // < 0x80 (1-byte UTF-8 character) - *u8 = u2->l; - u8++; - } - } - u2++; - } - *u8 = '\0'; - return dest; -} - std::string& u16_u8(std::string& dest, const std::vector<w_char>& src) { dest.clear(); std::vector<w_char>::const_iterator u2 = src.begin(); @@ -228,93 +181,6 @@ std::string& u16_u8(std::string& dest, const std::vector<w_char>& src) { return dest; } -/* only UTF-16 (BMP) implementation */ -int u8_u16(w_char* dest, int size, const char* src) { - const signed char* u8 = (const signed char*)src; - w_char* u2 = dest; - w_char* u2_max = u2 + size; - - while ((u2 < u2_max) && *u8) { - switch ((*u8) & 0xf0) { - case 0x00: - case 0x10: - case 0x20: - case 0x30: - case 0x40: - case 0x50: - case 0x60: - case 0x70: { - u2->h = 0; - u2->l = *u8; - break; - } - case 0x80: - case 0x90: - case 0xa0: - case 0xb0: { - HUNSPELL_WARNING(stderr, - "UTF-8 encoding error. Unexpected continuation bytes " - "in %ld. character position\n%s\n", - static_cast<long>(u8 - (signed char*)src), src); - u2->h = 0xff; - u2->l = 0xfd; - break; - } - case 0xc0: - case 0xd0: { // 2-byte UTF-8 codes - if ((*(u8 + 1) & 0xc0) == 0x80) { - u2->h = (*u8 & 0x1f) >> 2; - u2->l = (*u8 << 6) + (*(u8 + 1) & 0x3f); - u8++; - } else { - HUNSPELL_WARNING(stderr, - "UTF-8 encoding error. Missing continuation byte in " - "%ld. character position:\n%s\n", - static_cast<long>(u8 - (signed char*)src), src); - u2->h = 0xff; - u2->l = 0xfd; - } - break; - } - case 0xe0: { // 3-byte UTF-8 codes - if ((*(u8 + 1) & 0xc0) == 0x80) { - u2->h = ((*u8 & 0x0f) << 4) + ((*(u8 + 1) & 0x3f) >> 2); - u8++; - if ((*(u8 + 1) & 0xc0) == 0x80) { - u2->l = (*u8 << 6) + (*(u8 + 1) & 0x3f); - u8++; - } else { - HUNSPELL_WARNING(stderr, - "UTF-8 encoding error. Missing continuation byte " - "in %ld. character position:\n%s\n", - static_cast<long>(u8 - (signed char*)src), src); - u2->h = 0xff; - u2->l = 0xfd; - } - } else { - HUNSPELL_WARNING(stderr, - "UTF-8 encoding error. Missing continuation byte in " - "%ld. character position:\n%s\n", - static_cast<long>(u8 - (signed char*)src), src); - u2->h = 0xff; - u2->l = 0xfd; - } - break; - } - case 0xf0: { // 4 or more byte UTF-8 codes - HUNSPELL_WARNING( - stderr, "This UTF-8 encoding can't convert to UTF-16:\n%s\n", src); - u2->h = 0xff; - u2->l = 0xfd; - return -1; - } - } - u8++; - u2++; - } - return (int)(u2 - dest); -} - int u8_u16(std::vector<w_char>& dest, const std::string& src) { dest.clear(); std::string::const_iterator u8 = src.begin(); @@ -370,7 +236,7 @@ int u8_u16(std::vector<w_char>& dest, const std::string& src) { u2.h = ((*u8 & 0x0f) << 4) + ((*(u8 + 1) & 0x3f) >> 2); ++u8; if ((*(u8 + 1) & 0xc0) == 0x80) { - u2.l = (*u8 << 6) + (*(u8 + 1) & 0x3f); + u2.l = (static_cast<unsigned char>(*u8) << 6) + (*(u8 + 1) & 0x3f); ++u8; } else { HUNSPELL_WARNING(stderr, @@ -409,48 +275,6 @@ int u8_u16(std::vector<w_char>& dest, const std::string& src) { return dest.size(); } -void flag_qsort(unsigned short flags[], int begin, int end) { - unsigned short reg; - if (end > begin) { - unsigned short pivot = flags[begin]; - int l = begin + 1; - int r = end; - while (l < r) { - if (flags[l] <= pivot) { - l++; - } else { - r--; - reg = flags[l]; - flags[l] = flags[r]; - flags[r] = reg; - } - } - l--; - reg = flags[begin]; - flags[begin] = flags[l]; - flags[l] = reg; - - flag_qsort(flags, begin, l); - flag_qsort(flags, r, end); - } -} - -int flag_bsearch(unsigned short flags[], unsigned short flag, int length) { - int mid; - int left = 0; - int right = length - 1; - while (left <= right) { - mid = (left + right) / 2; - if (flags[mid] == flag) - return 1; - if (flag < flags[mid]) - right = mid - 1; - else - left = mid + 1; - } - return 0; -} - // strip strings into token based on single char delimiter // acts like strsep() but only uses a delim char and not // a delim string @@ -519,25 +343,6 @@ void mychomp(char* s) { *(s + k - 2) = '\0'; } -// does an ansi strdup of the reverse of a string -char* myrevstrdup(const char* s) { - char* d = NULL; - if (s) { - size_t sl = strlen(s); - d = (char*)malloc(sl + 1); - if (d) { - const char* p = s + sl - 1; - char* q = d; - while (p >= s) - *q++ = *p--; - *q = '\0'; - } else { - HUNSPELL_WARNING(stderr, "Can't allocate memory.\n"); - } - } - return d; -} - // break text to lines // return number of lines int line_tok(const char* text, char*** lines, char breakchar) { @@ -654,26 +459,6 @@ char* line_uniq_app(char** text, char breakchar) { } // append s to ends of every lines in text -void strlinecat(char* dest, const char* s) { - char* dup = mystrdup(dest); - char* source = dup; - int len = strlen(s); - if (dup) { - while (*source) { - if (*source == '\n') { - strncpy(dest, s, len); - dest += len; - } - *dest = *source; - source++; - dest++; - } - strcpy(dest, s); - free(dup); - } -} - -// append s to ends of every lines in text std::string& strlinecat(std::string& str, const std::string& apd) { size_t pos = 0; while ((pos = str.find('\n', pos)) != std::string::npos) { @@ -684,15 +469,6 @@ std::string& strlinecat(std::string& str, const std::string& apd) { return str; } -// change \n to char c -char* tr(char* text, char oldc, char newc) { - char* p; - for (p = text; *p; p++) - if (*p == oldc) - *p = newc; - return text; -} - // morphcmp(): compare MORPH_DERI_SFX, MORPH_INFL_SFX and MORPH_TERM_SFX fields // in the first line of the inputs // return 0, if inputs equal @@ -807,23 +583,6 @@ int fieldlen(const char* r) { return n; } -char* copy_field(char* dest, const char* morph, const char* var) { - if (!morph) - return NULL; - const char* beg = strstr(morph, var); - if (beg) { - char* d = dest; - for (beg += MORPH_TAG_LEN; - *beg != ' ' && *beg != '\t' && *beg != '\n' && *beg != '\0'; - d++, beg++) { - *d = *beg; - } - *d = '\0'; - return dest; - } - return NULL; -} - bool copy_field(std::string& dest, const std::string& morph, const std::string& var) { @@ -884,47 +643,18 @@ char* mystrrep(char* word, const char* pat, const char* rep) { } // reverse word -int reverseword(char* word) { - char r; - for (char *dest = word + strlen(word) - 1; word < dest; word++, dest--) { - r = *word; - *word = *dest; - *dest = r; - } - return 0; -} - -// reverse word -std::string& reverseword(std::string& word) { +size_t reverseword(std::string& word) { std::reverse(word.begin(), word.end()); - return word; -} - -// reverse word (error: 1) -int reverseword_utf(char* word) { - w_char w[MAXWORDLEN]; - w_char* p; - w_char r; - int l = u8_u16(w, MAXWORDLEN, word); - if (l == -1) - return 1; - p = w; - for (w_char *dest = w + l - 1; p < dest; p++, dest--) { - r = *p; - *p = *dest; - *dest = r; - } - u16_u8(word, MAXWORDUTF8LEN, w, l); - return 0; + return word.size(); } // reverse word -std::string& reverseword_utf(std::string& word) { +size_t reverseword_utf(std::string& word) { std::vector<w_char> w; u8_u16(w, word); std::reverse(w.begin(), w.end()); u16_u8(word, w); - return word; + return w.size(); } int uniqlist(char** list, int n) { @@ -978,12 +708,22 @@ unsigned char ccase(const struct cs_info* csconv, int nIndex) { } } -// convert null terminated string to all caps -void mkallcap(char* p, const struct cs_info* csconv) { - while (*p != '\0') { - *p = cupper(csconv, static_cast<unsigned char>(*p)); - p++; +w_char upper_utf(w_char u, int langnum) { + unsigned short idx = (u.h << 8) + u.l; + if (idx != unicodetoupper(idx, langnum)) { + u.h = (unsigned char)(unicodetoupper(idx, langnum) >> 8); + u.l = (unsigned char)(unicodetoupper(idx, langnum) & 0x00FF); + } + return u; +} + +w_char lower_utf(w_char u, int langnum) { + unsigned short idx = (u.h << 8) + u.l; + if (idx != unicodetolower(idx, langnum)) { + u.h = (unsigned char)(unicodetolower(idx, langnum) >> 8); + u.l = (unsigned char)(unicodetolower(idx, langnum) & 0x00FF); } + return u; } // convert std::string to all caps @@ -994,14 +734,6 @@ std::string& mkallcap(std::string& s, const struct cs_info* csconv) { return s; } -// convert null terminated string to all little -void mkallsmall(char* p, const struct cs_info* csconv) { - while (*p != '\0') { - *p = clower(csconv, static_cast<unsigned char>(*p)); - p++; - } -} - // convert std::string to all little std::string& mkallsmall(std::string& s, const struct cs_info* csconv) { for (std::string::iterator aI = s.begin(), aEnd = s.end(); aI != aEnd; ++aI) { @@ -1010,20 +742,9 @@ std::string& mkallsmall(std::string& s, const struct cs_info* csconv) { return s; } -void mkallsmall_utf(w_char* u, int nc, int langnum) { - for (int i = 0; i < nc; i++) { - unsigned short idx = (u[i].h << 8) + u[i].l; - if (idx != unicodetolower(idx, langnum)) { - u[i].h = (unsigned char)(unicodetolower(idx, langnum) >> 8); - u[i].l = (unsigned char)(unicodetolower(idx, langnum) & 0x00FF); - } - } -} - std::vector<w_char>& mkallsmall_utf(std::vector<w_char>& u, - int nc, int langnum) { - for (int i = 0; i < nc; i++) { + for (size_t i = 0; i < u.size(); ++i) { unsigned short idx = (u[i].h << 8) + u[i].l; if (idx != unicodetolower(idx, langnum)) { u[i].h = (unsigned char)(unicodetolower(idx, langnum) >> 8); @@ -1033,31 +754,51 @@ std::vector<w_char>& mkallsmall_utf(std::vector<w_char>& u, return u; } -void mkallcap_utf(w_char* u, int nc, int langnum) { - for (int i = 0; i < nc; i++) { +std::vector<w_char>& mkallcap_utf(std::vector<w_char>& u, int langnum) { + for (size_t i = 0; i < u.size(); i++) { unsigned short idx = (u[i].h << 8) + u[i].l; if (idx != unicodetoupper(idx, langnum)) { u[i].h = (unsigned char)(unicodetoupper(idx, langnum) >> 8); u[i].l = (unsigned char)(unicodetoupper(idx, langnum) & 0x00FF); } } + return u; } -std::vector<w_char>& mkallcap_utf(std::vector<w_char>& u, int nc, int langnum) { - for (int i = 0; i < nc; i++) { - unsigned short idx = (u[i].h << 8) + u[i].l; +std::string& mkinitcap(std::string& s, const struct cs_info* csconv) { + if (!s.empty()) { + s[0] = cupper(csconv, static_cast<unsigned char>(s[0])); + } + return s; +} + +std::vector<w_char>& mkinitcap_utf(std::vector<w_char>& u, int langnum) { + if (!u.empty()) { + unsigned short idx = (u[0].h << 8) + u[0].l; if (idx != unicodetoupper(idx, langnum)) { - u[i].h = (unsigned char)(unicodetoupper(idx, langnum) >> 8); - u[i].l = (unsigned char)(unicodetoupper(idx, langnum) & 0x00FF); + u[0].h = (unsigned char)(unicodetoupper(idx, langnum) >> 8); + u[0].l = (unsigned char)(unicodetoupper(idx, langnum) & 0x00FF); } } return u; } -// convert null terminated string to have initial capital -void mkinitcap(char* p, const struct cs_info* csconv) { - if (*p != '\0') - *p = cupper(csconv, static_cast<unsigned char>(*p)); +std::string& mkinitsmall(std::string& s, const struct cs_info* csconv) { + if (!s.empty()) { + s[0] = clower(csconv, static_cast<unsigned char>(s[0])); + } + return s; +} + +std::vector<w_char>& mkinitsmall_utf(std::vector<w_char>& u, int langnum) { + if (!u.empty()) { + unsigned short idx = (u[0].h << 8) + u[0].l; + if (idx != unicodetolower(idx, langnum)) { + u[0].h = (unsigned char)(unicodetolower(idx, langnum) >> 8); + u[0].l = (unsigned char)(unicodetolower(idx, langnum) & 0x00FF); + } + } + return u; } // conversion function for protected memory @@ -1073,35 +814,6 @@ char* get_stored_pointer(const char* s) { } #ifndef MOZILLA_CLIENT -// convert null terminated string to all caps using encoding -void enmkallcap(char* d, const char* p, const char* encoding) - -{ - struct cs_info* csconv = get_current_cs(encoding); - while (*p != '\0') { - *d++ = cupper(csconv, static_cast<unsigned char>(*p)); - p++; - } - *d = '\0'; -} - -// convert null terminated string to all little using encoding -void enmkallsmall(char* d, const char* p, const char* encoding) { - struct cs_info* csconv = get_current_cs(encoding); - while (*p != '\0') { - *d++ = clower(csconv, static_cast<unsigned char>(*p)); - p++; - } - *d = '\0'; -} - -// convert null terminated string to have initial capital using encoding -void enmkinitcap(char* d, const char* p, const char* encoding) { - struct cs_info* csconv = get_current_cs(encoding); - memcpy(d, p, (strlen(p) + 1)); - if (*p != '\0') - *d = cupper(csconv, static_cast<unsigned char>(*p)); -} // these are simple character mappings for the // encodings supported @@ -2982,14 +2694,14 @@ int unicodeisalpha(unsigned short c) { } /* get type of capitalization */ -int get_captype(char* word, int nl, cs_info* csconv) { +int get_captype(const std::string& word, cs_info* csconv) { // now determine the capitalization type of the first nl letters - int ncap = 0; - int nneutral = 0; - int firstcap = 0; + size_t ncap = 0; + size_t nneutral = 0; + size_t firstcap = 0; if (csconv == NULL) return NOCAP; - for (char* q = word; *q != '\0'; q++) { + for (std::string::const_iterator q = word.begin(); q != word.end(); ++q) { unsigned char nIndex = static_cast<unsigned char>(*q); if (ccase(csconv, nIndex)) ncap++; @@ -3006,7 +2718,7 @@ int get_captype(char* word, int nl, cs_info* csconv) { return NOCAP; } else if ((ncap == 1) && firstcap) { return INITCAP; - } else if ((ncap == nl) || ((ncap + nneutral) == nl)) { + } else if ((ncap == word.size()) || ((ncap + nneutral) == word.size())) { return ALLCAP; } else if ((ncap > 1) && firstcap) { return HUHINITCAP; @@ -3014,27 +2726,20 @@ int get_captype(char* word, int nl, cs_info* csconv) { return HUHCAP; } -int get_captype_utf8(w_char* word, int nl, int langnum) { +int get_captype_utf8(const std::vector<w_char>& word, int langnum) { // now determine the capitalization type of the first nl letters - int ncap = 0; - int nneutral = 0; - int firstcap = 0; - unsigned short idx; - // don't check too long words - if (nl >= MAXWORDLEN) - return 0; - // big Unicode character (non BMP area) - if (nl == -1) - return NOCAP; - for (int i = 0; i < nl; i++) { - idx = (word[i].h << 8) + word[i].l; + size_t ncap = 0; + size_t nneutral = 0; + size_t firstcap = 0; + for (size_t i = 0; i < word.size(); ++i) { + unsigned short idx = (word[i].h << 8) + word[i].l; if (idx != unicodetolower(idx, langnum)) ncap++; if (unicodetoupper(idx, langnum) == unicodetolower(idx, langnum)) nneutral++; } if (ncap) { - idx = (word[0].h << 8) + word[0].l; + unsigned short idx = (word[0].h << 8) + word[0].l; firstcap = (idx != unicodetolower(idx, langnum)); } @@ -3043,7 +2748,7 @@ int get_captype_utf8(w_char* word, int nl, int langnum) { return NOCAP; } else if ((ncap == 1) && firstcap) { return INITCAP; - } else if ((ncap == nl) || ((ncap + nneutral) == nl)) { + } else if ((ncap == word.size()) || ((ncap + nneutral) == word.size())) { return ALLCAP; } else if ((ncap > 1) && firstcap) { return HUHINITCAP; @@ -3052,63 +2757,22 @@ int get_captype_utf8(w_char* word, int nl, int langnum) { } // strip all ignored characters in the string -void remove_ignored_chars_utf(char* word, - unsigned short ignored_chars[], - int ignored_len) { - w_char w[MAXWORDLEN]; - w_char w2[MAXWORDLEN]; - int i; - int j; - int len = u8_u16(w, MAXWORDLEN, word); - for (i = 0, j = 0; i < len; i++) { - if (!flag_bsearch(ignored_chars, ((unsigned short*)w)[i], ignored_len)) { - w2[j] = w[i]; - j++; - } - } - if (j < i) - u16_u8(word, MAXWORDUTF8LEN, w2, j); -} - -namespace { -union w_s { - w_char w; - unsigned short s; -}; - -unsigned short asushort(w_char in) { - w_s c; - c.w = in; - return c.s; -} -} - -// strip all ignored characters in the string -std::string& remove_ignored_chars_utf(std::string& word, - unsigned short ignored_chars[], - int ignored_len) { +size_t remove_ignored_chars_utf(std::string& word, + const std::vector<w_char>& ignored_chars) { std::vector<w_char> w; std::vector<w_char> w2; u8_u16(w, word); for (size_t i = 0; i < w.size(); ++i) { - if (!flag_bsearch(ignored_chars, asushort(w[i]), ignored_len)) + if (!std::binary_search(ignored_chars.begin(), + ignored_chars.end(), + w[i])) { w2.push_back(w[i]); + } } u16_u8(word, w2); - return word; -} - -// strip all ignored characters in the string -void remove_ignored_chars(char* word, char* ignored_chars) { - for (char* p = word; *p != '\0'; p++) { - if (!strchr(ignored_chars, *p)) { - *word = *p; - word++; - } - } - *word = '\0'; + return w2.size(); } namespace { @@ -3119,16 +2783,17 @@ class is_any_of { bool operator()(char c) { return chars.find(c) != std::string::npos; } private: - const std::string& chars; + std::string chars; }; } // strip all ignored characters in the string -std::string& remove_ignored_chars(std::string& word, - const std::string& ignored_chars) { +size_t remove_ignored_chars(std::string& word, + const std::string& ignored_chars) { word.erase( - std::remove_if(word.begin(), word.end(), is_any_of(ignored_chars))); - return word; + std::remove_if(word.begin(), word.end(), is_any_of(ignored_chars)), + word.end()); + return word.size(); } int parse_string(char* line, char** out, int ln) { @@ -3170,25 +2835,16 @@ int parse_string(char* line, char** out, int ln) { return 0; } -int parse_array(char* line, - char** out, - unsigned short** out_utf16, - int* out_utf16_len, - int utf8, - int ln) { +bool parse_array(char* line, + char** out, + std::vector<w_char>& out_utf16, + int utf8, + int ln) { if (parse_string(line, out, ln)) - return 1; + return false; if (utf8) { - w_char w[MAXWORDLEN]; - int n = u8_u16(w, MAXWORDLEN, *out); - if (n > 0) { - flag_qsort((unsigned short*)w, 0, n); - *out_utf16 = (unsigned short*)malloc(n * sizeof(unsigned short)); - if (!*out_utf16) - return 1; - memcpy(*out_utf16, w, n * sizeof(unsigned short)); - } - *out_utf16_len = n; + u8_u16(out_utf16, *out); + std::sort(out_utf16.begin(), out_utf16.end()); } - return 0; + return true; } diff --git a/libs/hunspell/src/csutil.hxx b/libs/hunspell/src/csutil.hxx index cd582933b1..ce7091df55 100644 --- a/libs/hunspell/src/csutil.hxx +++ b/libs/hunspell/src/csutil.hxx @@ -131,30 +131,13 @@ LIBHUNSPELL_DLL_EXPORTED FILE* myfopen(const char* path, const char* mode); // convert UTF-16 characters to UTF-8 -LIBHUNSPELL_DLL_EXPORTED char* u16_u8(char* dest, - int size, - const w_char* src, - int srclen); -// convert UTF-16 characters to UTF-8 LIBHUNSPELL_DLL_EXPORTED std::string& u16_u8(std::string& dest, const std::vector<w_char>& src); // convert UTF-8 characters to UTF-16 -LIBHUNSPELL_DLL_EXPORTED int u8_u16(w_char* dest, int size, const char* src); -// convert UTF-8 characters to UTF-16 LIBHUNSPELL_DLL_EXPORTED int u8_u16(std::vector<w_char>& dest, const std::string& src); -// sort 2-byte vector -LIBHUNSPELL_DLL_EXPORTED void flag_qsort(unsigned short flags[], - int begin, - int end); - -// binary search in 2-byte vector -LIBHUNSPELL_DLL_EXPORTED int flag_bsearch(unsigned short flags[], - unsigned short flag, - int right); - // remove end of line char(s) LIBHUNSPELL_DLL_EXPORTED void mychomp(char* s); @@ -164,13 +147,8 @@ LIBHUNSPELL_DLL_EXPORTED char* mystrdup(const char* s); // strcat for limited length destination string LIBHUNSPELL_DLL_EXPORTED char* mystrcat(char* dest, const char* st, int max); -// duplicate reverse of string -LIBHUNSPELL_DLL_EXPORTED char* myrevstrdup(const char* s); - // parse into tokens with char delimiter LIBHUNSPELL_DLL_EXPORTED char* mystrsep(char** sptr, const char delim); -// parse into tokens with char delimiter -LIBHUNSPELL_DLL_EXPORTED char* mystrsep2(char** sptr, const char delim); // replace pat by rep in word and return word LIBHUNSPELL_DLL_EXPORTED char* mystrrep(char* word, @@ -181,9 +159,6 @@ LIBHUNSPELL_DLL_EXPORTED std::string& mystrrep(std::string& str, const std::string& replace); // append s to ends of every lines in text -LIBHUNSPELL_DLL_EXPORTED void strlinecat(char* lines, const char* s); - -// append s to ends of every lines in text LIBHUNSPELL_DLL_EXPORTED std::string& strlinecat(std::string& str, const std::string& apd); @@ -196,18 +171,11 @@ LIBHUNSPELL_DLL_EXPORTED int line_tok(const char* text, LIBHUNSPELL_DLL_EXPORTED char* line_uniq(char* text, char breakchar); LIBHUNSPELL_DLL_EXPORTED char* line_uniq_app(char** text, char breakchar); -// change oldchar to newchar in place -LIBHUNSPELL_DLL_EXPORTED char* tr(char* text, char oldc, char newc); - // reverse word -LIBHUNSPELL_DLL_EXPORTED int reverseword(char*); -// reverse word -LIBHUNSPELL_DLL_EXPORTED std::string& reverseword(std::string& word); +LIBHUNSPELL_DLL_EXPORTED size_t reverseword(std::string& word); // reverse word -LIBHUNSPELL_DLL_EXPORTED int reverseword_utf(char*); -// reverse word -LIBHUNSPELL_DLL_EXPORTED std::string& reverseword_utf(std::string&); +LIBHUNSPELL_DLL_EXPORTED size_t reverseword_utf(std::string&); // remove duplicates LIBHUNSPELL_DLL_EXPORTED int uniqlist(char** list, int n); @@ -226,6 +194,8 @@ LIBHUNSPELL_DLL_EXPORTED int initialize_utf_tbl(); LIBHUNSPELL_DLL_EXPORTED void free_utf_tbl(); LIBHUNSPELL_DLL_EXPORTED unsigned short unicodetoupper(unsigned short c, int langnum); +LIBHUNSPELL_DLL_EXPORTED w_char upper_utf(w_char u, int langnum); +LIBHUNSPELL_DLL_EXPORTED w_char lower_utf(w_char u, int langnum); LIBHUNSPELL_DLL_EXPORTED unsigned short unicodetolower(unsigned short c, int langnum); LIBHUNSPELL_DLL_EXPORTED int unicodeisalpha(unsigned short c); @@ -238,87 +208,64 @@ LIBHUNSPELL_DLL_EXPORTED int get_lang_num(const char* lang); // get characters of the given 8bit encoding with lower- and uppercase forms LIBHUNSPELL_DLL_EXPORTED char* get_casechars(const char* enc); -// convert null terminated string to all caps using encoding -LIBHUNSPELL_DLL_EXPORTED void enmkallcap(char* d, - const char* p, - const char* encoding); - -// convert null terminated string to all little using encoding -LIBHUNSPELL_DLL_EXPORTED void enmkallsmall(char* d, - const char* p, - const char* encoding); - -// convert null terminated string to have initial capital using encoding -LIBHUNSPELL_DLL_EXPORTED void enmkinitcap(char* d, - const char* p, - const char* encoding); - -// convert null terminated string to all caps -LIBHUNSPELL_DLL_EXPORTED void mkallcap(char* p, const struct cs_info* csconv); // convert std::string to all caps LIBHUNSPELL_DLL_EXPORTED std::string& mkallcap(std::string& s, const struct cs_info* csconv); // convert null terminated string to all little -LIBHUNSPELL_DLL_EXPORTED void mkallsmall(char* p, const struct cs_info* csconv); -// convert null terminated string to all little LIBHUNSPELL_DLL_EXPORTED std::string& mkallsmall(std::string& s, const struct cs_info* csconv); -// convert null terminated string to have initial capital -LIBHUNSPELL_DLL_EXPORTED void mkinitcap(char* p, const struct cs_info* csconv); +// convert first letter of string to little +LIBHUNSPELL_DLL_EXPORTED std::string& mkinitsmall(std::string& s, + const struct cs_info* csconv); + +// convert first letter of string to capital +LIBHUNSPELL_DLL_EXPORTED std::string& mkinitcap(std::string& s, + const struct cs_info* csconv); + +// convert first letter of UTF-8 string to capital +LIBHUNSPELL_DLL_EXPORTED std::vector<w_char>& +mkinitcap_utf(std::vector<w_char>& u, int langnum); + +// convert UTF-8 string to little +LIBHUNSPELL_DLL_EXPORTED std::vector<w_char>& +mkallsmall_utf(std::vector<w_char>& u, int langnum); -// convert first nc characters of UTF-8 string to little -LIBHUNSPELL_DLL_EXPORTED void mkallsmall_utf(w_char* u, int nc, int langnum); -// convert first nc characters of UTF-8 string to little +// convert first letter of UTF-8 string to little LIBHUNSPELL_DLL_EXPORTED std::vector<w_char>& -mkallsmall_utf(std::vector<w_char>& u, int nc, int langnum); +mkinitsmall_utf(std::vector<w_char>& u, int langnum); -// convert first nc characters of UTF-8 string to capital -LIBHUNSPELL_DLL_EXPORTED void mkallcap_utf(w_char* u, int nc, int langnum); -// convert first nc characters of UTF-8 string to capital +// convert UTF-8 string to capital LIBHUNSPELL_DLL_EXPORTED std::vector<w_char>& -mkallcap_utf(std::vector<w_char>& u, int nc, int langnum); +mkallcap_utf(std::vector<w_char>& u, int langnum); // get type of capitalization -LIBHUNSPELL_DLL_EXPORTED int get_captype(char* q, int nl, cs_info*); +LIBHUNSPELL_DLL_EXPORTED int get_captype(const std::string& q, cs_info*); // get type of capitalization (UTF-8) -LIBHUNSPELL_DLL_EXPORTED int get_captype_utf8(w_char* q, int nl, int langnum); +LIBHUNSPELL_DLL_EXPORTED int get_captype_utf8(const std::vector<w_char>& q, int langnum); // strip all ignored characters in the string -LIBHUNSPELL_DLL_EXPORTED void remove_ignored_chars_utf( - char* word, - unsigned short ignored_chars[], - int ignored_len); -// strip all ignored characters in the string -LIBHUNSPELL_DLL_EXPORTED std::string& remove_ignored_chars_utf( +LIBHUNSPELL_DLL_EXPORTED size_t remove_ignored_chars_utf( std::string& word, - unsigned short ignored_chars[], - int ignored_len); + const std::vector<w_char>& ignored_chars); // strip all ignored characters in the string -LIBHUNSPELL_DLL_EXPORTED void remove_ignored_chars(char* word, - char* ignored_chars); -// strip all ignored characters in the string -LIBHUNSPELL_DLL_EXPORTED std::string& remove_ignored_chars( +LIBHUNSPELL_DLL_EXPORTED size_t remove_ignored_chars( std::string& word, const std::string& ignored_chars); LIBHUNSPELL_DLL_EXPORTED int parse_string(char* line, char** out, int ln); -LIBHUNSPELL_DLL_EXPORTED int parse_array(char* line, - char** out, - unsigned short** out_utf16, - int* out_utf16_len, - int utf8, - int ln); +LIBHUNSPELL_DLL_EXPORTED bool parse_array(char* line, + char** out, + std::vector<w_char>& out_utf16, + int utf8, + int ln); LIBHUNSPELL_DLL_EXPORTED int fieldlen(const char* r); -LIBHUNSPELL_DLL_EXPORTED char* copy_field(char* dest, - const char* morph, - const char* var); LIBHUNSPELL_DLL_EXPORTED bool copy_field(std::string& dest, const std::string& morph, const std::string& var); @@ -375,6 +322,4 @@ LIBHUNSPELL_DLL_EXPORTED inline char* HENTRY_FIND(struct hentry* h, return (HENTRY_DATA(h) ? strstr(HENTRY_DATA(h), p) : NULL); } -#define w_char_eq(a, b) (((a).l == (b).l) && ((a).h == (b).h)) - #endif diff --git a/libs/hunspell/src/filemgr.c++ b/libs/hunspell/src/filemgr.c++ index cbe41c577b..2218bc79e1 100644 --- a/libs/hunspell/src/filemgr.c++ +++ b/libs/hunspell/src/filemgr.c++ @@ -89,13 +89,9 @@ FileMgr::FileMgr(const char* file, const char* key) : hin(NULL), linenum(0) { fin = myfopen(file, "r"); if (!fin) { // check hzipped file - char* st = (char*)malloc(strlen(file) + strlen(HZIP_EXTENSION) + 1); - if (st) { - strcpy(st, file); - strcat(st, HZIP_EXTENSION); - hin = new Hunzip(st, key); - free(st); - } + std::string st(file); + st.append(HZIP_EXTENSION); + hin = new Hunzip(st.c_str(), key); } if (!fin && !hin) fail(MSG_OPEN, file); diff --git a/libs/hunspell/src/hashmgr.c++ b/libs/hunspell/src/hashmgr.c++ index dbcf56a51c..c3cd95420f 100644 --- a/libs/hunspell/src/hashmgr.c++ +++ b/libs/hunspell/src/hashmgr.c++ @@ -76,6 +76,7 @@ #include <stdio.h> #include <ctype.h> #include <limits> +#include <sstream> #include "hashmgr.hxx" #include "csutil.hxx" @@ -101,8 +102,6 @@ HashMgr::HashMgr(const char* tpath, const char* apath, const char* key) enc = NULL; csconv = 0; ignorechars = NULL; - ignorechars_utf16 = NULL; - ignorechars_utf16_len = 0; load_config(apath, key); int ec = load_tables(tpath, key); if (ec) { @@ -167,8 +166,6 @@ HashMgr::~HashMgr() { if (ignorechars) free(ignorechars); - if (ignorechars_utf16) - free(ignorechars_utf16); #ifdef MOZILLA_CLIENT delete[] csconv; @@ -199,28 +196,56 @@ int HashMgr::add_word(const char* word, int al, const char* desc, bool onlyupcase) { + + std::string *word_copy = NULL; + std::string *desc_copy = NULL; + if (ignorechars || complexprefixes) { + word_copy = new std::string(word, wbl); + + if (ignorechars != NULL) { + if (utf8) { + wcl = remove_ignored_chars_utf(*word_copy, ignorechars_utf16); + } else { + remove_ignored_chars(*word_copy, ignorechars); + } + } + + if (complexprefixes) { + if (utf8) + wcl = reverseword_utf(*word_copy); + else + reverseword(*word_copy); + + if (desc && !aliasm) { + desc_copy = new std::string(desc); + + if (complexprefixes) { + if (utf8) + reverseword_utf(*desc_copy); + else + reverseword(*desc_copy); + } + desc = desc_copy->c_str(); + } + } + + wbl = word_copy->size(); + word = word_copy->c_str(); + } + bool upcasehomonym = false; int descl = desc ? (aliasm ? sizeof(char*) : strlen(desc) + 1) : 0; // variable-length hash record with word and optional fields struct hentry* hp = (struct hentry*)malloc(sizeof(struct hentry) + wbl + descl); - if (!hp) + if (!hp) { + delete desc_copy; + delete word_copy; return 1; + } + char* hpw = hp->word; strcpy(hpw, word); - if (ignorechars != NULL) { - if (utf8) { - remove_ignored_chars_utf(hpw, ignorechars_utf16, ignorechars_utf16_len); - } else { - remove_ignored_chars(hpw, ignorechars); - } - } - if (complexprefixes) { - if (utf8) - reverseword_utf(hpw); - else - reverseword(hpw); - } int i = hash(hpw); @@ -239,12 +264,6 @@ int HashMgr::add_word(const char* word, store_pointer(hpw + wbl + 1, get_aliasm(atoi(desc))); } else { strcpy(hpw + wbl + 1, desc); - if (complexprefixes) { - if (utf8) - reverseword_utf(HENTRY_DATA(hp)); - else - reverseword(HENTRY_DATA(hp)); - } } if (strstr(HENTRY_DATA(hp), MORPH_PHON)) hp->var += H_OPT_PHON; @@ -254,6 +273,8 @@ int HashMgr::add_word(const char* word, struct hentry* dp = tableptr[i]; if (!dp) { tableptr[i] = hp; + delete desc_copy; + delete word_copy; return 0; } while (dp->next != NULL) { @@ -265,6 +286,8 @@ int HashMgr::add_word(const char* word, dp->astr = hp->astr; dp->alen = hp->alen; free(hp); + delete desc_copy; + delete word_copy; return 0; } else { dp->next_homonym = hp; @@ -283,6 +306,8 @@ int HashMgr::add_word(const char* word, dp->astr = hp->astr; dp->alen = hp->alen; free(hp); + delete desc_copy; + delete word_copy; return 0; } else { dp->next_homonym = hp; @@ -299,11 +324,13 @@ int HashMgr::add_word(const char* word, free(hp->astr); free(hp); } + + delete desc_copy; + delete word_copy; return 0; } -int HashMgr::add_hidden_capitalized_word(char* word, - int wbl, +int HashMgr::add_hidden_capitalized_word(const std::string& word, int wcl, unsigned short* flags, int flagslen, @@ -326,32 +353,34 @@ int HashMgr::add_hidden_capitalized_word(char* word, memcpy(flags2, flags, flagslen * sizeof(unsigned short)); flags2[flagslen] = ONLYUPCASEFLAG; if (utf8) { - char st[BUFSIZE]; - w_char w[BUFSIZE]; - int wlen = u8_u16(w, BUFSIZE, word); - mkallsmall_utf(w, wlen, langnum); - mkallcap_utf(w, 1, langnum); - u16_u8(st, BUFSIZE, w, wlen); - return add_word(st, wbl, wcl, flags2, flagslen + 1, dp, true); + std::string st; + std::vector<w_char> w; + u8_u16(w, word); + mkallsmall_utf(w, langnum); + mkinitcap_utf(w, langnum); + u16_u8(st, w); + return add_word(st.c_str(), st.size(), wcl, flags2, flagslen + 1, dp, true); } else { - mkallsmall(word, csconv); - mkinitcap(word, csconv); - return add_word(word, wbl, wcl, flags2, flagslen + 1, dp, true); + std::string new_word(word); + mkallsmall(new_word, csconv); + mkinitcap(new_word, csconv); + int ret = add_word(new_word.c_str(), new_word.size(), wcl, flags2, flagslen + 1, dp, true); + return ret; } } return 0; } // detect captype and modify word length for UTF-8 encoding -int HashMgr::get_clen_and_captype(const char* word, int wbl, int* captype) { +int HashMgr::get_clen_and_captype(const std::string& word, int* captype) { int len; if (utf8) { - w_char dest_utf[BUFSIZE]; - len = u8_u16(dest_utf, BUFSIZE, word); - *captype = get_captype_utf8(dest_utf, len, langnum); + std::vector<w_char> dest_utf; + len = u8_u16(dest_utf, word); + *captype = get_captype_utf8(dest_utf, langnum); } else { - len = wbl; - *captype = get_captype((char*)word, len, csconv); + len = word.size(); + *captype = get_captype(word, csconv); } return len; } @@ -370,7 +399,7 @@ int HashMgr::remove(const char* word) { flags[dp->alen] = forbiddenword; dp->astr = flags; dp->alen++; - flag_qsort(flags, 0, dp->alen); + std::sort(flags, flags + dp->alen); } dp = dp->next_homonym; } @@ -378,8 +407,8 @@ int HashMgr::remove(const char* word) { } /* remove forbidden flag to add a personal word to the hash */ -int HashMgr::remove_forbidden_flag(const char* word) { - struct hentry* dp = lookup(word); +int HashMgr::remove_forbidden_flag(const std::string& word) { + struct hentry* dp = lookup(word.c_str()); if (!dp) return 1; while (dp) { @@ -406,15 +435,15 @@ int HashMgr::remove_forbidden_flag(const char* word) { } // add a custom dic. word to the hash table (public) -int HashMgr::add(const char* word) { +int HashMgr::add(const std::string& word) { unsigned short* flags = NULL; int al = 0; if (remove_forbidden_flag(word)) { int captype; - int wbl = strlen(word); - int wcl = get_clen_and_captype(word, wbl, &captype); - add_word(word, wbl, wcl, flags, al, NULL, false); - return add_hidden_capitalized_word((char*)word, wbl, wcl, flags, al, NULL, + int wbl = word.size(); + int wcl = get_clen_and_captype(word, &captype); + add_word(word.c_str(), wbl, wcl, flags, al, NULL, false); + return add_hidden_capitalized_word(word, wcl, flags, al, NULL, captype); } return 0; @@ -427,7 +456,7 @@ int HashMgr::add_with_affix(const char* word, const char* example) { if (dp && dp->astr) { int captype; int wbl = strlen(word); - int wcl = get_clen_and_captype(word, wbl, &captype); + int wcl = get_clen_and_captype(word, &captype); if (aliasf) { add_word(word, wbl, wcl, dp->astr, dp->alen, NULL, false); } else { @@ -440,7 +469,7 @@ int HashMgr::add_with_affix(const char* word, const char* example) { } else return 1; } - return add_hidden_capitalized_word((char*)word, wbl, wcl, dp->astr, + return add_hidden_capitalized_word(word, wcl, dp->astr, dp->alen, NULL, captype); } return 1; @@ -574,7 +603,7 @@ int HashMgr::load_tables(const char* tpath, const char* key) { delete dict; return 6; } - flag_qsort(flags, 0, al); + std::sort(flags, flags + al); } } else { al = 0; @@ -584,10 +613,10 @@ int HashMgr::load_tables(const char* tpath, const char* key) { int captype; int wbl = strlen(ts); - int wcl = get_clen_and_captype(ts, wbl, &captype); + int wcl = get_clen_and_captype(ts, &captype); // add the word and its index plus its capitalized form optionally if (add_word(ts, wbl, wcl, flags, al, dp, false) || - add_hidden_capitalized_word(ts, wbl, wcl, flags, al, dp, captype)) { + add_hidden_capitalized_word(ts, wcl, flags, al, dp, captype)) { delete dict; return 5; } @@ -674,12 +703,13 @@ int HashMgr::decode_flags(unsigned short** result, char* flags, FileMgr* af) { break; } case FLAG_UNI: { // UTF-8 characters - w_char w[BUFSIZE / 2]; - len = u8_u16(w, BUFSIZE / 2, flags); + std::vector<w_char> w; + u8_u16(w, flags); + len = w.size(); *result = (unsigned short*)malloc(len * sizeof(unsigned short)); if (!*result) return -1; - memcpy(*result, w, len * sizeof(short)); + memcpy(*result, &w[0], len * sizeof(short)); break; } default: { // Ispell's one-character flags (erfg -> e r f g) @@ -712,9 +742,13 @@ unsigned short HashMgr::decode_flag(const char* f) { i, DEFAULTFLAGS - 1); s = (unsigned short)i; break; - case FLAG_UNI: - u8_u16((w_char*)&s, 1, f); + case FLAG_UNI: { + std::vector<w_char> w; + u8_u16(w, f); + if (!w.empty()) + memcpy(&s, &w[0], 1 * sizeof(short)); break; + } default: s = (unsigned short)*((unsigned char*)f); } @@ -724,22 +758,24 @@ unsigned short HashMgr::decode_flag(const char* f) { } char* HashMgr::encode_flag(unsigned short f) { - unsigned char ch[10]; if (f == 0) return mystrdup("(NULL)"); + std::string ch; if (flag_mode == FLAG_LONG) { - ch[0] = (unsigned char)(f >> 8); - ch[1] = (unsigned char)(f - ((f >> 8) << 8)); - ch[2] = '\0'; + ch.push_back((unsigned char)(f >> 8)); + ch.push_back((unsigned char)(f - ((f >> 8) << 8))); } else if (flag_mode == FLAG_NUM) { - sprintf((char*)ch, "%d", f); + std::ostringstream stream; + stream << f; + ch = stream.str(); } else if (flag_mode == FLAG_UNI) { - u16_u8((char*)&ch, 10, (w_char*)&f, 1); + const w_char* w_c = (const w_char*)&f; + std::vector<w_char> w(w_c, w_c + 1); + u16_u8(ch, w); } else { - ch[0] = (unsigned char)(f); - ch[1] = '\0'; + ch.push_back((unsigned char)(f)); } - return mystrdup((char*)ch); + return mystrdup(ch.c_str()); } // read in aff file and set flag mode @@ -824,8 +860,8 @@ int HashMgr::load_config(const char* affpath, const char* key) { /* parse in the ignored characters (for example, Arabic optional diacritics * characters */ if (strncmp(line, "IGNORE", 6) == 0) { - if (parse_array(line, &ignorechars, &ignorechars_utf16, - &ignorechars_utf16_len, utf8, afflst->getlinenum())) { + if (!parse_array(line, &ignorechars, ignorechars_utf16, + utf8, afflst->getlinenum())) { delete afflst; return 1; } @@ -951,7 +987,7 @@ int HashMgr::parse_aliasf(char* line, FileMgr* af) { case 1: { aliasflen[j] = (unsigned short)decode_flags(&(aliasf[j]), piece, af); - flag_qsort(aliasf[j], 0, aliasflen[j]); + std::sort(aliasf[j], aliasf[j] + aliasflen[j]); break; } default: @@ -1070,19 +1106,14 @@ int HashMgr::parse_aliasm(char* line, FileMgr* af) { *(tp - 1) = ' '; tp = tp + strlen(tp); } + std::string chunk(piece); if (complexprefixes) { if (utf8) - reverseword_utf(piece); + reverseword_utf(chunk); else - reverseword(piece); - } - aliasm[j] = mystrdup(piece); - if (!aliasm[j]) { - numaliasm = 0; - free(aliasm); - aliasm = NULL; - return 1; + reverseword(chunk); } + aliasm[j] = mystrdup(chunk.c_str()); break; } default: diff --git a/libs/hunspell/src/hashmgr.hxx b/libs/hunspell/src/hashmgr.hxx index c6d72f3c40..95b06b13f9 100644 --- a/libs/hunspell/src/hashmgr.hxx +++ b/libs/hunspell/src/hashmgr.hxx @@ -77,9 +77,12 @@ #include "hunvisapi.h" #include <stdio.h> +#include <string> +#include <vector> #include "htypes.hxx" #include "filemgr.hxx" +#include "w_char.hxx" enum flag { FLAG_CHAR, FLAG_LONG, FLAG_NUM, FLAG_UNI }; @@ -95,8 +98,7 @@ class LIBHUNSPELL_DLL_EXPORTED HashMgr { char* lang; struct cs_info* csconv; char* ignorechars; - unsigned short* ignorechars_utf16; - int ignorechars_utf16_len; + std::vector<w_char> ignorechars_utf16; int numaliasf; // flag vector `compression' with aliases unsigned short** aliasf; unsigned short* aliasflen; @@ -111,7 +113,7 @@ class LIBHUNSPELL_DLL_EXPORTED HashMgr { int hash(const char*) const; struct hentry* walk_hashtable(int& col, struct hentry* hp) const; - int add(const char* word); + int add(const std::string& word); int add_with_affix(const char* word, const char* pattern); int remove(const char* word); int decode_flags(unsigned short** result, char* flags, FileMgr* af); @@ -123,7 +125,7 @@ class LIBHUNSPELL_DLL_EXPORTED HashMgr { char* get_aliasm(int index); private: - int get_clen_and_captype(const char* word, int wbl, int* captype); + int get_clen_and_captype(const std::string& word, int* captype); int load_tables(const char* tpath, const char* key); int add_word(const char* word, int wbl, @@ -134,15 +136,14 @@ class LIBHUNSPELL_DLL_EXPORTED HashMgr { bool onlyupcase); int load_config(const char* affpath, const char* key); int parse_aliasf(char* line, FileMgr* af); - int add_hidden_capitalized_word(char* word, - int wbl, + int add_hidden_capitalized_word(const std::string& word, int wcl, unsigned short* flags, int al, char* dp, int captype); int parse_aliasm(char* line, FileMgr* af); - int remove_forbidden_flag(const char* word); + int remove_forbidden_flag(const std::string& word); }; #endif diff --git a/libs/hunspell/src/hunspell.c++ b/libs/hunspell/src/hunspell.c++ index 726c72931a..f7c1581087 100644 --- a/libs/hunspell/src/hunspell.c++ +++ b/libs/hunspell/src/hunspell.c++ @@ -85,6 +85,9 @@ #include <limits> #include <string> +#define MAXWORDLEN 176 +#define MAXWORDUTF8LEN (MAXWORDLEN * 3) + Hunspell::Hunspell(const char* affpath, const char* dpath, const char* key) { encoding = NULL; csconv = NULL; @@ -158,14 +161,16 @@ int Hunspell::add_dic(const char* dpath, const char* key) { // set the capitalization type // return the length of the "cleaned" (and UTF-8 encoded) word -int Hunspell::cleanword2(char* dest, +size_t Hunspell::cleanword2(std::string& dest, + std::vector<w_char>& dest_utf, const char* src, - w_char* dest_utf, int* nc, int* pcaptype, - int* pabbrev) { - unsigned char* p = (unsigned char*)dest; - const unsigned char* q = (const unsigned char*)src; + size_t* pabbrev) { + dest.clear(); + dest_utf.clear(); + + const char* q = src; // first skip over any leading blanks while ((*q != '\0') && (*q == ' ')) @@ -173,7 +178,7 @@ int Hunspell::cleanword2(char* dest, // now strip off any trailing periods (recording their presence) *pabbrev = 0; - int nl = strlen((const char*)q); + int nl = strlen(q); while ((nl > 0) && (*(q + nl - 1) == '.')) { nl--; (*pabbrev)++; @@ -182,35 +187,26 @@ int Hunspell::cleanword2(char* dest, // if no characters are left it can't be capitalized if (nl <= 0) { *pcaptype = NOCAP; - *p = '\0'; return 0; } - strncpy(dest, (char*)q, nl); - *(dest + nl) = '\0'; - nl = strlen(dest); + dest.append(q, nl); + nl = dest.size(); if (utf8) { - *nc = u8_u16(dest_utf, MAXWORDLEN, dest); - // don't check too long words - if (*nc >= MAXWORDLEN) - return 0; - if (*nc == -1) { // big Unicode character (non BMP area) - *pcaptype = NOCAP; - return nl; - } - *pcaptype = get_captype_utf8(dest_utf, *nc, langnum); + *nc = u8_u16(dest_utf, dest); + *pcaptype = get_captype_utf8(dest_utf, langnum); } else { - *pcaptype = get_captype(dest, nl, csconv); + *pcaptype = get_captype(dest, csconv); *nc = nl; } return nl; } -int Hunspell::cleanword(char* dest, +void Hunspell::cleanword(std::string& dest, const char* src, int* pcaptype, int* pabbrev) { - unsigned char* p = (unsigned char*)dest; + dest.clear(); const unsigned char* q = (const unsigned char*)src; int firstcap = 0; @@ -229,8 +225,7 @@ int Hunspell::cleanword(char* dest, // if no characters are left it can't be capitalized if (nl <= 0) { *pcaptype = NOCAP; - *p = '\0'; - return 0; + return; } // now determine the capitalization type of the first nl letters @@ -245,27 +240,25 @@ int Hunspell::cleanword(char* dest, ncap++; if (csconv[(*q)].cupper == csconv[(*q)].clower) nneutral++; - *p++ = *q++; + dest.push_back(*q++); nl--; } // remember to terminate the destination string - *p = '\0'; - firstcap = csconv[(unsigned char)(*dest)].ccase; + firstcap = csconv[static_cast<unsigned char>(dest[0])].ccase; } else { - unsigned short idx; - w_char t[MAXWORDLEN]; - nc = u8_u16(t, MAXWORDLEN, src); - for (int i = 0; i < nc; i++) { - idx = (t[i].h << 8) + t[i].l; + std::vector<w_char> t; + u8_u16(t, src); + for (size_t i = 0; i < t.size(); ++i) { + unsigned short idx = (t[i].h << 8) + t[i].l; unsigned short low = unicodetolower(idx, langnum); if (idx != low) ncap++; if (unicodetoupper(idx, langnum) == low) nneutral++; } - u16_u8(dest, MAXWORDUTF8LEN, t, nc); + u16_u8(dest, t); if (ncap) { - idx = (t[0].h << 8) + t[0].l; + unsigned short idx = (t[0].h << 8) + t[0].l; firstcap = (idx != unicodetolower(idx, langnum)); } } @@ -282,117 +275,60 @@ int Hunspell::cleanword(char* dest, } else { *pcaptype = HUHCAP; } - return strlen(dest); } -void Hunspell::mkallcap(char* p) { +void Hunspell::mkallcap(std::string& u8) { if (utf8) { - w_char u[MAXWORDLEN]; - int nc = u8_u16(u, MAXWORDLEN, p); - unsigned short idx; - for (int i = 0; i < nc; i++) { - idx = (u[i].h << 8) + u[i].l; - if (idx != unicodetoupper(idx, langnum)) { - u[i].h = (unsigned char)(unicodetoupper(idx, langnum) >> 8); - u[i].l = (unsigned char)(unicodetoupper(idx, langnum) & 0x00FF); - } - } - u16_u8(p, MAXWORDUTF8LEN, u, nc); + std::vector<w_char> u16; + u8_u16(u16, u8); + ::mkallcap_utf(u16, langnum); + u16_u8(u8, u16); } else { - while (*p != '\0') { - *p = csconv[((unsigned char)*p)].cupper; - p++; - } - } -} - -int Hunspell::mkallcap2(char* p, w_char* u, int nc) { - if (utf8) { - unsigned short idx; - for (int i = 0; i < nc; i++) { - idx = (u[i].h << 8) + u[i].l; - unsigned short up = unicodetoupper(idx, langnum); - if (idx != up) { - u[i].h = (unsigned char)(up >> 8); - u[i].l = (unsigned char)(up & 0x00FF); - } - } - u16_u8(p, MAXWORDUTF8LEN, u, nc); - return strlen(p); - } else { - while (*p != '\0') { - *p = csconv[((unsigned char)*p)].cupper; - p++; - } - } - return nc; -} - -void Hunspell::mkallsmall(char* p) { - while (*p != '\0') { - *p = csconv[((unsigned char)*p)].clower; - p++; + ::mkallcap(u8, csconv); } } -int Hunspell::mkallsmall2(char* p, w_char* u, int nc) { +int Hunspell::mkallsmall2(std::string& u8, std::vector<w_char>& u16) { if (utf8) { - unsigned short idx; - for (int i = 0; i < nc; i++) { - idx = (u[i].h << 8) + u[i].l; - unsigned short low = unicodetolower(idx, langnum); - if (idx != low) { - u[i].h = (unsigned char)(low >> 8); - u[i].l = (unsigned char)(low & 0x00FF); - } - } - u16_u8(p, MAXWORDUTF8LEN, u, nc); - return strlen(p); + ::mkallsmall_utf(u16, langnum); + u16_u8(u8, u16); } else { - while (*p != '\0') { - *p = csconv[((unsigned char)*p)].clower; - p++; - } + ::mkallsmall(u8, csconv); } - return nc; + return u8.size(); } // convert UTF-8 sharp S codes to latin 1 -char* Hunspell::sharps_u8_l1(char* dest, char* source) { - char* p = dest; - *p = *source; - for (p++, source++; *(source - 1); p++, source++) { - *p = *source; - if (*source == '\x9F') - *--p = '\xDF'; - } +std::string Hunspell::sharps_u8_l1(const std::string& source) { + std::string dest(source); + mystrrep(dest, "\xC3\x9F", "\xDF"); return dest; } // recursive search for right ss - sharp s permutations -hentry* Hunspell::spellsharps(char* base, - char* pos, +hentry* Hunspell::spellsharps(std::string& base, + size_t n_pos, int n, int repnum, - char* tmp, int* info, char** root) { - pos = strstr(pos, "ss"); - if (pos && (n < MAXSHARPS)) { - *pos = '\xC3'; - *(pos + 1) = '\x9F'; - hentry* h = spellsharps(base, pos + 2, n + 1, repnum + 1, tmp, info, root); + size_t pos = base.find("ss", n_pos); + if (pos != std::string::npos && (n < MAXSHARPS)) { + base[pos] = '\xC3'; + base[pos + 1] = '\x9F'; + hentry* h = spellsharps(base, pos + 2, n + 1, repnum + 1, info, root); if (h) return h; - *pos = 's'; - *(pos + 1) = 's'; - h = spellsharps(base, pos + 2, n + 1, repnum, tmp, info, root); + base[pos] = 's'; + base[pos + 1] = 's'; + h = spellsharps(base, pos + 2, n + 1, repnum, info, root); if (h) return h; } else if (repnum > 0) { if (utf8) - return checkword(base, info, root); - return checkword(sharps_u8_l1(tmp, base), info, root); + return checkword(base.c_str(), info, root); + std::string tmp(sharps_u8_l1(base)); + return checkword(tmp.c_str(), info, root); } return NULL; } @@ -403,7 +339,7 @@ int Hunspell::is_keepcase(const hentry* rv) { } /* insert a word to the beginning of the suggestion array and return ns */ -int Hunspell::insert_sug(char*** slst, char* word, int ns) { +int Hunspell::insert_sug(char*** slst, const char* word, int ns) { if (!*slst) return ns; char* dup = mystrdup(word); @@ -421,11 +357,6 @@ int Hunspell::insert_sug(char*** slst, char* word, int ns) { int Hunspell::spell(const char* word, int* info, char** root) { struct hentry* rv = NULL; - // need larger vector. For example, Turkish capital letter I converted a - // 2-byte UTF-8 character (dotless i) by mkallsmall. - char cw[MAXWORDUTF8LEN]; - char wspace[MAXWORDUTF8LEN]; - w_char unicw[MAXWORDLEN]; int info2 = 0; if (!info) @@ -437,7 +368,6 @@ int Hunspell::spell(const char* word, int* info, char** root) { if (strcmp(word, SPELL_XML) == 0) return 1; int nc = strlen(word); - int wl2 = 0; if (utf8) { if (nc >= MAXWORDUTF8LEN) return 0; @@ -445,19 +375,26 @@ int Hunspell::spell(const char* word, int* info, char** root) { if (nc >= MAXWORDLEN) return 0; } - int captype = 0; - int abbv = 0; - int wl = 0; + int captype = NOCAP; + size_t abbv = 0; + size_t wl = 0; + + std::string scw; + std::vector<w_char> sunicw; // input conversion RepList* rl = (pAMgr) ? pAMgr->get_iconvtable() : NULL; - int convstatus = rl ? rl->conv(word, wspace, MAXWORDUTF8LEN) : 0; - if (convstatus < 0) - return 0; - else if (convstatus > 0) - wl = cleanword2(cw, wspace, unicw, &nc, &captype, &abbv); - else - wl = cleanword2(cw, word, unicw, &nc, &captype, &abbv); + { + std::string wspace; + + int convstatus = rl ? rl->conv(word, wspace) : 0; + if (convstatus < 0) + return 0; + else if (convstatus > 0) + wl = cleanword2(scw, sunicw, wspace.c_str(), &nc, &captype, &abbv); + else + wl = cleanword2(scw, sunicw, word, &nc, &captype, &abbv); + } #ifdef MOZILLA_CLIENT // accept the abbreviated words without dots @@ -474,12 +411,12 @@ int Hunspell::spell(const char* word, int* info, char** root) { // "..", "--" etc.) enum { NBEGIN, NNUM, NSEP }; int nstate = NBEGIN; - int i; + size_t i; for (i = 0; (i < wl); i++) { - if ((cw[i] <= '9') && (cw[i] >= '0')) { + if ((scw[i] <= '9') && (scw[i] >= '0')) { nstate = NNUM; - } else if ((cw[i] == ',') || (cw[i] == '.') || (cw[i] == '-')) { + } else if ((scw[i] == ',') || (scw[i] == '.') || (scw[i] == '-')) { if ((nstate == NSEP) || (i == 0)) break; nstate = NSEP; @@ -496,75 +433,75 @@ int Hunspell::spell(const char* word, int* info, char** root) { *info += SPELL_ORIGCAP; /* FALLTHROUGH */ case NOCAP: - rv = checkword(cw, info, root); + rv = checkword(scw.c_str(), info, root); if ((abbv) && !(rv)) { - memcpy(wspace, cw, wl); - *(wspace + wl) = '.'; - *(wspace + wl + 1) = '\0'; - rv = checkword(wspace, info, root); + std::string u8buffer(scw); + u8buffer.push_back('.'); + rv = checkword(u8buffer.c_str(), info, root); } break; case ALLCAP: { *info += SPELL_ORIGCAP; - rv = checkword(cw, info, root); + rv = checkword(scw.c_str(), info, root); if (rv) break; if (abbv) { - memcpy(wspace, cw, wl); - *(wspace + wl) = '.'; - *(wspace + wl + 1) = '\0'; - rv = checkword(wspace, info, root); + std::string u8buffer(scw); + u8buffer.push_back('.'); + rv = checkword(u8buffer.c_str(), info, root); if (rv) break; } // Spec. prefix handling for Catalan, French, Italian: // prefixes separated by apostrophe (SANT'ELIA -> Sant'+Elia). - if (pAMgr && strchr(cw, '\'')) { - mkallsmall2(cw, unicw, nc); - // There are no really sane circumstances where this could fail, - // but anyway... - if (char* apostrophe = strchr(cw, '\'')) { + size_t apos = pAMgr ? scw.find('\'') : std::string::npos; + if (apos != std::string::npos) { + mkallsmall2(scw, sunicw); + //conversion may result in string with different len to pre-mkallsmall2 + //so re-scan + if (apos != std::string::npos && apos < scw.size() - 1) { + std::string part1 = scw.substr(0, apos+1); + std::string part2 = scw.substr(apos+1); if (utf8) { - w_char tmpword[MAXWORDLEN]; - *apostrophe = '\0'; - wl2 = u8_u16(tmpword, MAXWORDLEN, cw); - *apostrophe = '\''; - if (wl2 >= 0 && wl2 < nc) { - mkinitcap2(apostrophe + 1, unicw + wl2 + 1, nc - wl2 - 1); - rv = checkword(cw, info, root); - if (rv) - break; - } + std::vector<w_char> part1u, part2u; + u8_u16(part1u, part1); + u8_u16(part2u, part2); + mkinitcap2(part2, part2u); + scw = part1 + part2; + sunicw = part1u; + sunicw.insert(sunicw.end(), part2u.begin(), part2u.end()); + rv = checkword(scw.c_str(), info, root); + if (rv) + break; } else { - mkinitcap2(apostrophe + 1, unicw, nc); - rv = checkword(cw, info, root); + mkinitcap2(part2, sunicw); + scw = part1 + part2; + rv = checkword(scw.c_str(), info, root); if (rv) break; } + mkinitcap2(scw, sunicw); + rv = checkword(scw.c_str(), info, root); + if (rv) + break; } - mkinitcap2(cw, unicw, nc); - rv = checkword(cw, info, root); - if (rv) - break; } - if (pAMgr && pAMgr->get_checksharps() && strstr(cw, "SS")) { - char tmpword[MAXWORDUTF8LEN]; - wl = mkallsmall2(cw, unicw, nc); - memcpy(wspace, cw, (wl + 1)); - rv = spellsharps(wspace, wspace, 0, 0, tmpword, info, root); + if (pAMgr && pAMgr->get_checksharps() && scw.find("SS") != std::string::npos) { + + mkallsmall2(scw, sunicw); + std::string u8buffer(scw); + rv = spellsharps(u8buffer, 0, 0, 0, info, root); if (!rv) { - wl2 = mkinitcap2(cw, unicw, nc); - rv = spellsharps(cw, cw, 0, 0, tmpword, info, root); + mkinitcap2(scw, sunicw); + rv = spellsharps(scw, 0, 0, 0, info, root); } if ((abbv) && !(rv)) { - *(wspace + wl) = '.'; - *(wspace + wl + 1) = '\0'; - rv = spellsharps(wspace, wspace, 0, 0, tmpword, info, root); + u8buffer.push_back('.'); + rv = spellsharps(u8buffer, 0, 0, 0, info, root); if (!rv) { - memcpy(wspace, cw, wl2); - *(wspace + wl2) = '.'; - *(wspace + wl2 + 1) = '\0'; - rv = spellsharps(wspace, wspace, 0, 0, tmpword, info, root); + u8buffer = std::string(scw); + u8buffer.push_back('.'); + rv = spellsharps(u8buffer, 0, 0, 0, info, root); } } if (rv) @@ -572,13 +509,14 @@ int Hunspell::spell(const char* word, int* info, char** root) { } } case INITCAP: { + *info += SPELL_ORIGCAP; - wl = mkallsmall2(cw, unicw, nc); - memcpy(wspace, cw, (wl + 1)); - wl2 = mkinitcap2(cw, unicw, nc); + mkallsmall2(scw, sunicw); + std::string u8buffer(scw); + mkinitcap2(scw, sunicw); if (captype == INITCAP) *info += SPELL_INITCAP; - rv = checkword(cw, info, root); + rv = checkword(scw.c_str(), info, root); if (captype == INITCAP) *info -= SPELL_INITCAP; // forbid bad capitalization @@ -593,18 +531,16 @@ int Hunspell::spell(const char* word, int* info, char** root) { if (rv) break; - rv = checkword(wspace, info, root); + rv = checkword(u8buffer.c_str(), info, root); if (abbv && !rv) { - *(wspace + wl) = '.'; - *(wspace + wl + 1) = '\0'; - rv = checkword(wspace, info, root); + u8buffer.push_back('.'); + rv = checkword(u8buffer.c_str(), info, root); if (!rv) { - memcpy(wspace, cw, wl2); - *(wspace + wl2) = '.'; - *(wspace + wl2 + 1) = '\0'; + u8buffer = scw; + u8buffer.push_back('.'); if (captype == INITCAP) *info += SPELL_INITCAP; - rv = checkword(wspace, info, root); + rv = checkword(u8buffer.c_str(), info, root); if (captype == INITCAP) *info -= SPELL_INITCAP; if (rv && is_keepcase(rv) && (captype == ALLCAP)) @@ -617,8 +553,8 @@ int Hunspell::spell(const char* word, int* info, char** root) { // if CHECKSHARPS: KEEPCASE words with \xDF are allowed // in INITCAP form, too. !(pAMgr->get_checksharps() && - ((utf8 && strstr(wspace, "\xC3\x9F")) || - (!utf8 && strchr(wspace, '\xDF')))))) + ((utf8 && u8buffer.find("\xC3\x9F") != std::string::npos) || + (!utf8 && u8buffer.find('\xDF') != std::string::npos))))) rv = NULL; break; } @@ -637,67 +573,66 @@ int Hunspell::spell(const char* word, int* info, char** root) { // recursive breaking at break points if (wordbreak) { - char* s; - char r; + int nbr = 0; - wl = strlen(cw); + wl = scw.size(); int numbreak = pAMgr ? pAMgr->get_numbreak() : 0; // calculate break points for recursion limit for (int j = 0; j < numbreak; j++) { - s = cw; - do { - s = (char*)strstr(s, wordbreak[j]); - if (s) { - nbr++; - s++; - } - } while (s); + size_t len = strlen(wordbreak[j]); + size_t pos = 0; + while ((pos = scw.find(wordbreak[j], pos, len)) != std::string::npos) { + ++nbr; + pos += len; + } } if (nbr >= 10) return 0; // check boundary patterns (^begin and end$) for (int j = 0; j < numbreak; j++) { - int plen = strlen(wordbreak[j]); + size_t plen = strlen(wordbreak[j]); if (plen == 1 || plen > wl) continue; + if (wordbreak[j][0] == '^' && - strncmp(cw, wordbreak[j] + 1, plen - 1) == 0 && spell(cw + plen - 1)) + scw.compare(0, plen - 1, wordbreak[j] + 1, plen -1) == 0 && spell(scw.c_str() + plen - 1)) return 1; + if (wordbreak[j][plen - 1] == '$' && - strncmp(cw + wl - plen + 1, wordbreak[j], plen - 1) == 0) { - r = cw[wl - plen + 1]; - cw[wl - plen + 1] = '\0'; - if (spell(cw)) + scw.compare(wl - plen + 1, plen - 1, wordbreak[j], plen - 1) == 0) { + char r = scw[wl - plen + 1]; + scw[wl - plen + 1] = '\0'; + if (spell(scw.c_str())) return 1; - cw[wl - plen + 1] = r; + scw[wl - plen + 1] = r; } } // other patterns for (int j = 0; j < numbreak; j++) { - int plen = strlen(wordbreak[j]); - s = (char*)strstr(cw, wordbreak[j]); - if (s && (s > cw) && (s < cw + wl - plen)) { - if (!spell(s + plen)) + size_t plen = strlen(wordbreak[j]); + size_t found = scw.find(wordbreak[j]); + if ((found > 0) && (found < wl - plen)) { + if (!spell(scw.c_str() + found + plen)) continue; - r = *s; - *s = '\0'; + char r = scw[found]; + scw[found] = '\0'; // examine 2 sides of the break point - if (spell(cw)) + if (spell(scw.c_str())) return 1; - *s = r; + scw[found] = r; // LANG_hu: spec. dash rule if (langnum == LANG_hu && strcmp(wordbreak[j], "-") == 0) { - r = s[1]; - s[1] = '\0'; - if (spell(cw)) + r = scw[found + 1]; + scw[found + 1] = '\0'; + if (spell(scw.c_str())) return 1; // check the first part with dash - s[1] = r; + scw[found + 1] = r; } - // end of LANG speficic region + // end of LANG specific region } } } @@ -716,10 +651,9 @@ struct hentry* Hunspell::checkword(const char* w, int* info, char** root) { if (ignoredchars != NULL) { w2.assign(w); if (utf8) { - int ignoredchars_utf16_len; - unsigned short* ignoredchars_utf16 = - pAMgr->get_ignore_utf16(&ignoredchars_utf16_len); - remove_ignored_chars_utf(w2, ignoredchars_utf16, ignoredchars_utf16_len); + const std::vector<w_char>& ignoredchars_utf16 = + pAMgr->get_ignore_utf16(); + remove_ignored_chars_utf(w2, ignoredchars_utf16); } else { remove_ignored_chars(w2, ignoredchars); } @@ -802,37 +736,40 @@ struct hentry* Hunspell::checkword(const char* w, int* info, char** root) { return NULL; } if (root) { - *root = mystrdup(he->word); - if (*root && complexprefixes) { + std::string word_root(he->word); + if (complexprefixes) { if (utf8) - reverseword_utf(*root); + reverseword_utf(word_root); else - reverseword(*root); + reverseword(word_root); } + *root = mystrdup(word_root.c_str()); } // try check compound word } else if (pAMgr->get_compound()) { - he = pAMgr->compound_check(word, len, 0, 0, 100, 0, NULL, 0, 0, info); + struct hentry* rwords[100]; // buffer for COMPOUND pattern checking + he = pAMgr->compound_check(word, len, 0, 0, 100, 0, NULL, (hentry**)&rwords, 0, 0, info); // LANG_hu section: `moving rule' with last dash if ((!he) && (langnum == LANG_hu) && (word[len - 1] == '-')) { char* dup = mystrdup(word); if (!dup) return NULL; dup[len - 1] = '\0'; - he = pAMgr->compound_check(dup, len - 1, -5, 0, 100, 0, NULL, 1, 0, + he = pAMgr->compound_check(dup, len - 1, -5, 0, 100, 0, NULL, (hentry**)&rwords, 1, 0, info); free(dup); } - // end of LANG speficic region + // end of LANG specific region if (he) { if (root) { - *root = mystrdup(he->word); - if (*root && complexprefixes) { + std::string word_root(he->word); + if (complexprefixes) { if (utf8) - reverseword_utf(*root); + reverseword_utf(word_root); else - reverseword(*root); + reverseword(word_root); } + *root = mystrdup(word_root.c_str()); } if (info) *info += SPELL_COMPOUND; @@ -845,11 +782,8 @@ struct hentry* Hunspell::checkword(const char* w, int* info, char** root) { int Hunspell::suggest(char*** slst, const char* word) { int onlycmpdsug = 0; - char cw[MAXWORDUTF8LEN]; - char wspace[MAXWORDUTF8LEN]; if (!pSMgr || maxdic == 0) return 0; - w_char unicw[MAXWORDLEN]; *slst = NULL; // process XML input of the simplified API (see manual) if (strncmp(word, SPELL_XML, sizeof(SPELL_XML) - 3) == 0) { @@ -863,130 +797,132 @@ int Hunspell::suggest(char*** slst, const char* word) { if (nc >= MAXWORDLEN) return 0; } - int captype = 0; - int abbv = 0; - int wl = 0; + int captype = NOCAP; + size_t abbv = 0; + size_t wl = 0; + + std::string scw; + std::vector<w_char> sunicw; // input conversion RepList* rl = (pAMgr) ? pAMgr->get_iconvtable() : NULL; - int convstatus = rl ? rl->conv(word, wspace, MAXWORDUTF8LEN) : 0; - if (convstatus < 0) - return 0; - else if (convstatus > 0) - wl = cleanword2(cw, wspace, unicw, &nc, &captype, &abbv); - else - wl = cleanword2(cw, word, unicw, &nc, &captype, &abbv); + { + std::string wspace; + + int convstatus = rl ? rl->conv(word, wspace) : 0; + if (convstatus < 0) + return 0; + else if (convstatus > 0) + wl = cleanword2(scw, sunicw, wspace.c_str(), &nc, &captype, &abbv); + else + wl = cleanword2(scw, sunicw, word, &nc, &captype, &abbv); + + if (wl == 0) + return 0; + } - if (wl == 0) - return 0; int ns = 0; int capwords = 0; // check capitalized form for FORCEUCASE if (pAMgr && captype == NOCAP && pAMgr->get_forceucase()) { int info = SPELL_ORIGCAP; - char** wlst; - if (checkword(cw, &info, NULL)) { - if (*slst) { - wlst = *slst; - } else { - wlst = (char**)malloc(MAXSUGGESTION * sizeof(char*)); - if (wlst == NULL) - return -1; - *slst = wlst; - for (int i = 0; i < MAXSUGGESTION; i++) { - wlst[i] = NULL; - } + if (checkword(scw.c_str(), &info, NULL)) { + std::string form(scw); + mkinitcap(form); + + char** wlst = (char**)malloc(MAXSUGGESTION * sizeof(char*)); + if (wlst == NULL) + return -1; + *slst = wlst; + wlst[0] = mystrdup(form.c_str()); + for (int i = 1; i < MAXSUGGESTION; ++i) { + wlst[i] = NULL; } - wlst[0] = mystrdup(cw); - mkinitcap(wlst[0]); + return 1; } } switch (captype) { case NOCAP: { - ns = pSMgr->suggest(slst, cw, ns, &onlycmpdsug); + ns = pSMgr->suggest(slst, scw.c_str(), ns, &onlycmpdsug); break; } case INITCAP: { capwords = 1; - ns = pSMgr->suggest(slst, cw, ns, &onlycmpdsug); + ns = pSMgr->suggest(slst, scw.c_str(), ns, &onlycmpdsug); if (ns == -1) break; - memcpy(wspace, cw, (wl + 1)); - mkallsmall2(wspace, unicw, nc); - ns = pSMgr->suggest(slst, wspace, ns, &onlycmpdsug); + std::string wspace(scw); + mkallsmall2(wspace, sunicw); + ns = pSMgr->suggest(slst, wspace.c_str(), ns, &onlycmpdsug); break; } case HUHINITCAP: capwords = 1; case HUHCAP: { - ns = pSMgr->suggest(slst, cw, ns, &onlycmpdsug); + ns = pSMgr->suggest(slst, scw.c_str(), ns, &onlycmpdsug); if (ns != -1) { - int prevns; // something.The -> something. The - char* dot = strchr(cw, '.'); - if (dot && (dot > cw)) { + size_t dot_pos = scw.find('.'); + if (dot_pos != std::string::npos) { + std::string postdot = scw.substr(dot_pos + 1); int captype_; if (utf8) { - w_char w_[MAXWORDLEN]; - int wl_ = u8_u16(w_, MAXWORDLEN, dot + 1); - captype_ = get_captype_utf8(w_, wl_, langnum); - } else - captype_ = get_captype(dot + 1, strlen(dot + 1), csconv); + std::vector<w_char> postdotu; + u8_u16(postdotu, postdot); + captype_ = get_captype_utf8(postdotu, langnum); + } else { + captype_ = get_captype(postdot, csconv); + } if (captype_ == INITCAP) { - char* st = mystrdup(cw); - if (st) { - char* newst = (char*)realloc(st, wl + 2); - if (newst == NULL) - free(st); - st = newst; - } - if (st) { - st[(dot - cw) + 1] = ' '; - strcpy(st + (dot - cw) + 2, dot + 1); - ns = insert_sug(slst, st, ns); - free(st); - } + std::string str(scw); + str.insert(dot_pos + 1, 1, ' '); + ns = insert_sug(slst, str.c_str(), ns); } } + + std::string wspace; + if (captype == HUHINITCAP) { // TheOpenOffice.org -> The OpenOffice.org - memcpy(wspace, cw, (wl + 1)); - mkinitsmall2(wspace, unicw, nc); - ns = pSMgr->suggest(slst, wspace, ns, &onlycmpdsug); + wspace = scw; + mkinitsmall2(wspace, sunicw); + ns = pSMgr->suggest(slst, wspace.c_str(), ns, &onlycmpdsug); } - memcpy(wspace, cw, (wl + 1)); - mkallsmall2(wspace, unicw, nc); - if (spell(wspace)) - ns = insert_sug(slst, wspace, ns); - prevns = ns; - ns = pSMgr->suggest(slst, wspace, ns, &onlycmpdsug); + wspace = scw; + mkallsmall2(wspace, sunicw); + if (spell(wspace.c_str())) + ns = insert_sug(slst, wspace.c_str(), ns); + int prevns = ns; + ns = pSMgr->suggest(slst, wspace.c_str(), ns, &onlycmpdsug); if (captype == HUHINITCAP) { - mkinitcap2(wspace, unicw, nc); - if (spell(wspace)) - ns = insert_sug(slst, wspace, ns); - ns = pSMgr->suggest(slst, wspace, ns, &onlycmpdsug); + mkinitcap2(wspace, sunicw); + if (spell(wspace.c_str())) + ns = insert_sug(slst, wspace.c_str(), ns); + ns = pSMgr->suggest(slst, wspace.c_str(), ns, &onlycmpdsug); } // aNew -> "a New" (instead of "a new") for (int j = prevns; j < ns; j++) { char* space = strchr((*slst)[j], ' '); if (space) { - int slen = strlen(space + 1); + size_t slen = strlen(space + 1); // different case after space (need capitalisation) - if ((slen < wl) && strcmp(cw + wl - slen, space + 1)) { - w_char w[MAXWORDLEN]; - int wc = 0; - char* r = (*slst)[j]; + if ((slen < wl) && strcmp(scw.c_str() + wl - slen, space + 1)) { + std::string first((*slst)[j], space + 1); + std::string second(space + 1); + std::vector<w_char> w; if (utf8) - wc = u8_u16(w, MAXWORDLEN, space + 1); - mkinitcap2(space + 1, w, wc); + u8_u16(w, second); + mkinitcap2(second, w); // set as first suggestion + char* r = (*slst)[j]; for (int k = j; k > 0; k--) (*slst)[k] = (*slst)[k - 1]; - (*slst)[0] = r; + free(r); + (*slst)[0] = mystrdup((first + second).c_str()); } } } @@ -995,35 +931,30 @@ int Hunspell::suggest(char*** slst, const char* word) { } case ALLCAP: { - memcpy(wspace, cw, (wl + 1)); - mkallsmall2(wspace, unicw, nc); - ns = pSMgr->suggest(slst, wspace, ns, &onlycmpdsug); + std::string wspace(scw); + mkallsmall2(wspace, sunicw); + ns = pSMgr->suggest(slst, wspace.c_str(), ns, &onlycmpdsug); if (ns == -1) break; - if (pAMgr && pAMgr->get_keepcase() && spell(wspace)) - ns = insert_sug(slst, wspace, ns); - mkinitcap2(wspace, unicw, nc); - ns = pSMgr->suggest(slst, wspace, ns, &onlycmpdsug); + if (pAMgr && pAMgr->get_keepcase() && spell(wspace.c_str())) + ns = insert_sug(slst, wspace.c_str(), ns); + mkinitcap2(wspace, sunicw); + ns = pSMgr->suggest(slst, wspace.c_str(), ns, &onlycmpdsug); for (int j = 0; j < ns; j++) { - mkallcap((*slst)[j]); + std::string form((*slst)[j]); + mkallcap(form); + if (pAMgr && pAMgr->get_checksharps()) { - char* pos; if (utf8) { - pos = strstr((*slst)[j], "\xC3\x9F"); - while (pos) { - *pos = 'S'; - *(pos + 1) = 'S'; - pos = strstr(pos + 2, "\xC3\x9F"); - } + mystrrep(form, "\xC3\x9F", "SS"); } else { - pos = strchr((*slst)[j], '\xDF'); - while (pos) { - (*slst)[j] = (char*)realloc((*slst)[j], strlen((*slst)[j]) + 2); - mystrrep((*slst)[j], "\xDF", "SS"); - pos = strchr((*slst)[j], '\xDF'); - } + mystrrep(form, "\xDF", "SS"); } } + + free((*slst)[j]); + (*slst)[j] = mystrdup(form.c_str()); + } break; } @@ -1035,11 +966,10 @@ int Hunspell::suggest(char*** slst, const char* word) { char* pos = strchr((*slst)[j], '-'); if (pos) { int info; - char w[MAXWORDUTF8LEN]; *pos = '\0'; - strcpy(w, (*slst)[j]); - strcat(w, pos + 1); - (void)spell(w, &info, NULL); + std::string w((*slst)[j]); + w.append(pos + 1); + (void)spell(w.c_str(), &info, NULL); if ((info & SPELL_COMPOUND) && (info & SPELL_FORBIDDEN)) { *pos = ' '; } else @@ -1054,64 +984,67 @@ int Hunspell::suggest(char*** slst, const char* word) { (*slst)) { switch (captype) { case NOCAP: { - ns = pSMgr->ngsuggest(*slst, cw, ns, pHMgr, maxdic); + ns = pSMgr->ngsuggest(*slst, scw.c_str(), ns, pHMgr, maxdic); break; } case HUHINITCAP: capwords = 1; case HUHCAP: { - memcpy(wspace, cw, (wl + 1)); - mkallsmall2(wspace, unicw, nc); - ns = pSMgr->ngsuggest(*slst, wspace, ns, pHMgr, maxdic); + std::string wspace(scw); + mkallsmall2(wspace, sunicw); + ns = pSMgr->ngsuggest(*slst, wspace.c_str(), ns, pHMgr, maxdic); break; } case INITCAP: { capwords = 1; - memcpy(wspace, cw, (wl + 1)); - mkallsmall2(wspace, unicw, nc); - ns = pSMgr->ngsuggest(*slst, wspace, ns, pHMgr, maxdic); + std::string wspace(scw); + mkallsmall2(wspace, sunicw); + ns = pSMgr->ngsuggest(*slst, wspace.c_str(), ns, pHMgr, maxdic); break; } case ALLCAP: { - memcpy(wspace, cw, (wl + 1)); - mkallsmall2(wspace, unicw, nc); + std::string wspace(scw); + mkallsmall2(wspace, sunicw); int oldns = ns; - ns = pSMgr->ngsuggest(*slst, wspace, ns, pHMgr, maxdic); - for (int j = oldns; j < ns; j++) - mkallcap((*slst)[j]); + ns = pSMgr->ngsuggest(*slst, wspace.c_str(), ns, pHMgr, maxdic); + for (int j = oldns; j < ns; j++) { + std::string form((*slst)[j]); + mkallcap(form); + free((*slst)[j]); + (*slst)[j] = mystrdup(form.c_str()); + } break; } } } // try dash suggestion (Afo-American -> Afro-American) - if (char* pos = strchr(cw, '-')) { - char* ppos = cw; + size_t dash_pos = scw.find('-'); + if (dash_pos != std::string::npos) { int nodashsug = 1; - char** nlst = NULL; - int nn = 0; - int last = 0; - if (*slst) { - for (int j = 0; j < ns && nodashsug == 1; j++) { - if (strchr((*slst)[j], '-')) - nodashsug = 0; - } + for (int j = 0; j < ns && nodashsug == 1; j++) { + if (strchr((*slst)[j], '-')) + nodashsug = 0; } + + size_t prev_pos = 0; + bool last = false; + while (nodashsug && !last) { - if (*pos == '\0') + if (dash_pos == scw.size()) last = 1; - else - *pos = '\0'; - if (!spell(ppos)) { - nn = suggest(&nlst, ppos); + std::string chunk = scw.substr(prev_pos, dash_pos - prev_pos); + if (!spell(chunk.c_str())) { + char** nlst = NULL; + int nn = suggest(&nlst, chunk.c_str()); for (int j = nn - 1; j >= 0; j--) { - strncpy(wspace, cw, ppos - cw); - strcpy(wspace + (ppos - cw), nlst[j]); + std::string wspace = scw.substr(0, prev_pos); + wspace.append(nlst[j]); if (!last) { - strcat(wspace, "-"); - strcat(wspace, pos + 1); + wspace.append("-"); + wspace.append(scw.substr(dash_pos + 1)); } - ns = insert_sug(slst, wspace, ns); + ns = insert_sug(slst, wspace.c_str(), ns); free(nlst[j]); } if (nlst != NULL) @@ -1119,29 +1052,34 @@ int Hunspell::suggest(char*** slst, const char* word) { nodashsug = 0; } if (!last) { - *pos = '-'; - ppos = pos + 1; - pos = strchr(ppos, '-'); + prev_pos = dash_pos + 1; + dash_pos = scw.find('-', prev_pos); } - if (!pos) - pos = cw + strlen(cw); + if (dash_pos == std::string::npos) + dash_pos = scw.size(); } } // word reversing wrapper for complex prefixes if (complexprefixes) { for (int j = 0; j < ns; j++) { + std::string root((*slst)[j]); + free((*slst)[j]); if (utf8) - reverseword_utf((*slst)[j]); + reverseword_utf(root); else - reverseword((*slst)[j]); + reverseword(root); + (*slst)[j] = mystrdup(root.c_str()); } } // capitalize if (capwords) for (int j = 0; j < ns; j++) { - mkinitcap((*slst)[j]); + std::string form((*slst)[j]); + free((*slst)[j]); + mkinitcap(form); + (*slst)[j] = mystrdup(form.c_str()); } // expand suggestions with dot(s) @@ -1160,25 +1098,23 @@ int Hunspell::suggest(char*** slst, const char* word) { int l = 0; for (int j = 0; j < ns; j++) { if (!strchr((*slst)[j], ' ') && !spell((*slst)[j])) { - char s[MAXSWUTF8L]; - w_char w[MAXSWL]; - int len; + std::string s; + std::vector<w_char> w; if (utf8) { - len = u8_u16(w, MAXSWL, (*slst)[j]); + u8_u16(w, (*slst)[j]); } else { - strcpy(s, (*slst)[j]); - len = strlen(s); + s = (*slst)[j]; } - mkallsmall2(s, w, len); + mkallsmall2(s, w); free((*slst)[j]); - if (spell(s)) { - (*slst)[l] = mystrdup(s); + if (spell(s.c_str())) { + (*slst)[l] = mystrdup(s.c_str()); if ((*slst)[l]) l++; } else { - mkinitcap2(s, w, len); - if (spell(s)) { - (*slst)[l] = mystrdup(s); + mkinitcap2(s, w); + if (spell(s.c_str())) { + (*slst)[l] = mystrdup(s.c_str()); if ((*slst)[l]) l++; } @@ -1211,9 +1147,10 @@ int Hunspell::suggest(char*** slst, const char* word) { // output conversion rl = (pAMgr) ? pAMgr->get_oconvtable() : NULL; for (int j = 0; rl && j < ns; j++) { - if (rl->conv((*slst)[j], wspace, MAXWORDUTF8LEN) > 0) { + std::string wspace; + if (rl->conv((*slst)[j], wspace) > 0) { free((*slst)[j]); - (*slst)[j] = mystrdup(wspace); + (*slst)[j] = mystrdup(wspace.c_str()); } } @@ -1233,151 +1170,25 @@ char* Hunspell::get_dic_encoding() { return encoding; } -#ifdef HUNSPELL_EXPERIMENTAL -// XXX UTF-8 support is OK? -int Hunspell::suggest_auto(char*** slst, const char* word) { - char cw[MAXWORDUTF8LEN]; - char wspace[MAXWORDUTF8LEN]; - if (!pSMgr || maxdic == 0) - return 0; - w_char unicw[MAXWORDLEN]; - int nc = strlen(word); - if (utf8) { - if (nc >= MAXWORDUTF8LEN) - return 0; - } else { - if (nc >= MAXWORDLEN) - return 0; - } - int captype = 0; - int abbv = 0; - int wl = 0; - - // input conversion - RepList* rl = (pAMgr) ? pAMgr->get_iconvtable() : NULL; - int convstatus = rl ? rl->conv(word, wspace) : 0; - if (convstatus < 0) - return 0; - else if (convstatus > 0) - wl = cleanword2(cw, wspace, unicw, &nc, &captype, &abbv); - else - wl = cleanword2(cw, word, unicw, &nc, &captype, &abbv); - - if (wl == 0) - return 0; - int ns = 0; - *slst = NULL; // HU, nsug in pSMgr->suggest - - switch (captype) { - case NOCAP: { - ns = pSMgr->suggest_auto(slst, cw, ns); - if (ns > 0) - break; - break; - } - - case INITCAP: { - memcpy(wspace, cw, (wl + 1)); - mkallsmall2(wspace, unicw, nc); - ns = pSMgr->suggest_auto(slst, wspace, ns); - for (int j = 0; j < ns; j++) - mkinitcap((*slst)[j]); - ns = pSMgr->suggest_auto(slst, cw, ns); - break; - } - - case HUHINITCAP: - case HUHCAP: { - ns = pSMgr->suggest_auto(slst, cw, ns); - if (ns == 0) { - memcpy(wspace, cw, (wl + 1)); - mkallsmall2(wspace, unicw, nc); - ns = pSMgr->suggest_auto(slst, wspace, ns); - } - break; - } - - case ALLCAP: { - memcpy(wspace, cw, (wl + 1)); - mkallsmall2(wspace, unicw, nc); - ns = pSMgr->suggest_auto(slst, wspace, ns); - - mkinitcap(wspace); - ns = pSMgr->suggest_auto(slst, wspace, ns); - - for (int j = 0; j < ns; j++) - mkallcap((*slst)[j]); - break; - } - } - - // word reversing wrapper for complex prefixes - if (complexprefixes) { - for (int j = 0; j < ns; j++) { - if (utf8) - reverseword_utf((*slst)[j]); - else - reverseword((*slst)[j]); - } - } - - // expand suggestions with dot(s) - if (abbv && pAMgr && pAMgr->get_sugswithdots()) { - for (int j = 0; j < ns; j++) { - (*slst)[j] = (char*)realloc((*slst)[j], strlen((*slst)[j]) + 1 + abbv); - strcat((*slst)[j], word + strlen(word) - abbv); - } - } - - // LANG_hu section: replace '-' with ' ' in Hungarian - if (langnum == LANG_hu) { - for (int j = 0; j < ns; j++) { - char* pos = strchr((*slst)[j], '-'); - if (pos) { - int info; - char w[MAXWORDUTF8LEN]; - *pos = '\0'; - strcpy(w, (*slst)[j]); - strcat(w, pos + 1); - spell(w, &info, NULL); - if ((info & SPELL_COMPOUND) && (info & SPELL_FORBIDDEN)) { - *pos = ' '; - } else - *pos = '-'; - } - } - } - - // output conversion - rl = (pAMgr) ? pAMgr->get_oconvtable() : NULL; - for (int j = 0; rl && j < ns; j++) { - if (rl->conv((*slst)[j], wspace) > 0) { - free((*slst)[j]); - (*slst)[j] = mystrdup(wspace); - } - } - - // END OF LANG_hu section - return ns; -} -#endif - int Hunspell::stem(char*** slst, char** desc, int n) { - char result[MAXLNLEN]; - char result2[MAXLNLEN]; + + std::string result2; *slst = NULL; if (n == 0) return 0; - *result2 = '\0'; for (int i = 0; i < n; i++) { - *result = '\0'; + + std::string result; + // add compound word parts (except the last one) char* s = (char*)desc[i]; char* part = strstr(s, MORPH_PART); if (part) { char* nextpart = strstr(part + 1, MORPH_PART); while (nextpart) { - copy_field(result + strlen(result), part, MORPH_PART); + std::string field; + copy_field(field, part, MORPH_PART); + result.append(field); part = nextpart; nextpart = strstr(part + 1, MORPH_PART); } @@ -1404,22 +1215,28 @@ int Hunspell::stem(char*** slst, char** desc, int n) { int genl = line_tok(sg, &gen, MSEP_REC); free(sg); for (int j = 0; j < genl; j++) { - sprintf(result2 + strlen(result2), "%c%s%s", MSEP_REC, result, - gen[j]); + result2.push_back(MSEP_REC); + result2.append(result); + result2.append(gen[j]); } freelist(&gen, genl); } } else { - sprintf(result2 + strlen(result2), "%c%s", MSEP_REC, result); + result2.push_back(MSEP_REC); + result2.append(result); if (strstr(pl[k], MORPH_SURF_PFX)) { - copy_field(result2 + strlen(result2), pl[k], MORPH_SURF_PFX); + std::string field; + copy_field(field, pl[k], MORPH_SURF_PFX); + result2.append(field); } - copy_field(result2 + strlen(result2), pl[k], MORPH_STEM); + std::string field; + copy_field(field, pl[k], MORPH_STEM); + result2.append(field); } } freelist(&pl, pln); } - int sln = line_tok(result2, slst, MSEP_REC); + int sln = line_tok(result2.c_str(), slst, MSEP_REC); return uniqlist(*slst, sln); } @@ -1431,148 +1248,43 @@ int Hunspell::stem(char*** slst, const char* word) { return pln2; } -#ifdef HUNSPELL_EXPERIMENTAL -int Hunspell::suggest_pos_stems(char*** slst, const char* word) { - char cw[MAXWORDUTF8LEN]; - char wspace[MAXWORDUTF8LEN]; - if (!pSMgr || maxdic == 0) - return 0; - w_char unicw[MAXWORDLEN]; - int nc = strlen(word); - if (utf8) { - if (nc >= MAXWORDUTF8LEN) - return 0; - } else { - if (nc >= MAXWORDLEN) - return 0; - } - int captype = 0; - int abbv = 0; - int wl = 0; - - // input conversion - RepList* rl = (pAMgr) ? pAMgr->get_iconvtable() : NULL; - int convstatus = rl ? rl->conv(word, wspace) : 0; - if (convstatus < 0) - return 0; - else if (convstatus > 0) - wl = cleanword2(cw, wspace, unicw, &nc, &captype, &abbv); - else - wl = cleanword2(cw, word, unicw, &nc, &captype, &abbv); - - if (wl == 0) - return 0; - - int ns = 0; // ns=0 = normalized input - - *slst = NULL; // HU, nsug in pSMgr->suggest - - switch (captype) { - case HUHCAP: - case NOCAP: { - ns = pSMgr->suggest_pos_stems(slst, cw, ns); - - if ((abbv) && (ns == 0)) { - memcpy(wspace, cw, wl); - *(wspace + wl) = '.'; - *(wspace + wl + 1) = '\0'; - ns = pSMgr->suggest_pos_stems(slst, wspace, ns); - } - - break; - } - - case INITCAP: { - ns = pSMgr->suggest_pos_stems(slst, cw, ns); - - if (ns == 0 || ((*slst)[0][0] == '#')) { - memcpy(wspace, cw, (wl + 1)); - mkallsmall2(wspace, unicw, nc); - ns = pSMgr->suggest_pos_stems(slst, wspace, ns); - } - - break; - } - - case ALLCAP: { - ns = pSMgr->suggest_pos_stems(slst, cw, ns); - if (ns != 0) - break; - - memcpy(wspace, cw, (wl + 1)); - mkallsmall2(wspace, unicw, nc); - ns = pSMgr->suggest_pos_stems(slst, wspace, ns); - - if (ns == 0) { - mkinitcap(wspace); - ns = pSMgr->suggest_pos_stems(slst, wspace, ns); - } - break; - } - } - - // output conversion - rl = (pAMgr) ? pAMgr->get_oconvtable() : NULL; - for (int j = 0; rl && j < ns; j++) { - if (rl->conv((*slst)[j], wspace) > 0) { - free((*slst)[j]); - (*slst)[j] = mystrdup(wspace); - } - } - - return ns; -} -#endif // END OF HUNSPELL_EXPERIMENTAL CODE - const char* Hunspell::get_wordchars() { return pAMgr->get_wordchars(); } -unsigned short* Hunspell::get_wordchars_utf16(int* len) { - return pAMgr->get_wordchars_utf16(len); +const std::vector<w_char>& Hunspell::get_wordchars_utf16() { + return pAMgr->get_wordchars_utf16(); } -void Hunspell::mkinitcap(char* p) { - if (!utf8) { - if (*p != '\0') - *p = csconv[((unsigned char)*p)].cupper; +void Hunspell::mkinitcap(std::string& u8) { + if (utf8) { + std::vector<w_char> u16; + u8_u16(u16, u8); + ::mkinitcap_utf(u16, langnum); + u16_u8(u8, u16); } else { - int len; - w_char u[MAXWORDLEN]; - len = u8_u16(u, MAXWORDLEN, p); - unsigned short i = unicodetoupper((u[0].h << 8) + u[0].l, langnum); - u[0].h = (unsigned char)(i >> 8); - u[0].l = (unsigned char)(i & 0x00FF); - u16_u8(p, MAXWORDUTF8LEN, u, len); + ::mkinitcap(u8, csconv); } } -int Hunspell::mkinitcap2(char* p, w_char* u, int nc) { - if (!utf8) { - if (*p != '\0') - *p = csconv[((unsigned char)*p)].cupper; - } else if (nc > 0) { - unsigned short i = unicodetoupper((u[0].h << 8) + u[0].l, langnum); - u[0].h = (unsigned char)(i >> 8); - u[0].l = (unsigned char)(i & 0x00FF); - u16_u8(p, MAXWORDUTF8LEN, u, nc); - return strlen(p); +int Hunspell::mkinitcap2(std::string& u8, std::vector<w_char>& u16) { + if (utf8) { + ::mkinitcap_utf(u16, langnum); + u16_u8(u8, u16); + } else { + ::mkinitcap(u8, csconv); } - return nc; + return u8.size(); } -int Hunspell::mkinitsmall2(char* p, w_char* u, int nc) { - if (!utf8) { - if (*p != '\0') - *p = csconv[((unsigned char)*p)].clower; - } else if (nc > 0) { - unsigned short i = unicodetolower((u[0].h << 8) + u[0].l, langnum); - u[0].h = (unsigned char)(i >> 8); - u[0].l = (unsigned char)(i & 0x00FF); - u16_u8(p, MAXWORDUTF8LEN, u, nc); - return strlen(p); +int Hunspell::mkinitsmall2(std::string& u8, std::vector<w_char>& u16) { + if (utf8) { + ::mkinitsmall_utf(u16, langnum); + u16_u8(u8, u16); + } else { + ::mkinitsmall(u8, csconv); } - return nc; + return u8.size(); } int Hunspell::add(const char* word) { @@ -1601,20 +1313,16 @@ struct cs_info* Hunspell::get_csconv() { return csconv; } -void Hunspell::cat_result(char* result, char* st) { +void Hunspell::cat_result(std::string& result, char* st) { if (st) { - if (*result) - mystrcat(result, "\n", MAXLNLEN); - mystrcat(result, st, MAXLNLEN); + if (!result.empty()) + result.append("\n"); + result.append(st); free(st); } } int Hunspell::analyze(char*** slst, const char* word) { - char cw[MAXWORDUTF8LEN]; - char wspace[MAXWORDUTF8LEN]; - w_char unicw[MAXWORDLEN]; - int wl2 = 0; *slst = NULL; if (!pSMgr || maxdic == 0) return 0; @@ -1626,48 +1334,52 @@ int Hunspell::analyze(char*** slst, const char* word) { if (nc >= MAXWORDLEN) return 0; } - int captype = 0; - int abbv = 0; - int wl = 0; + int captype = NOCAP; + size_t abbv = 0; + size_t wl = 0; + + std::string scw; + std::vector<w_char> sunicw; // input conversion RepList* rl = (pAMgr) ? pAMgr->get_iconvtable() : NULL; - int convstatus = rl ? rl->conv(word, wspace, MAXWORDUTF8LEN) : 0; - if (convstatus < 0) - return 0; - else if (convstatus > 0) - wl = cleanword2(cw, wspace, unicw, &nc, &captype, &abbv); - else - wl = cleanword2(cw, word, unicw, &nc, &captype, &abbv); + { + std::string wspace; + + int convstatus = rl ? rl->conv(word, wspace) : 0; + if (convstatus < 0) + return 0; + else if (convstatus > 0) + wl = cleanword2(scw, sunicw, wspace.c_str(), &nc, &captype, &abbv); + else + wl = cleanword2(scw, sunicw, word, &nc, &captype, &abbv); + } if (wl == 0) { if (abbv) { + scw.clear(); for (wl = 0; wl < abbv; wl++) - cw[wl] = '.'; - cw[wl] = '\0'; + scw.push_back('.'); abbv = 0; } else return 0; } - char result[MAXLNLEN]; - char* st = NULL; - - *result = '\0'; + std::string result; - int n = 0; - int n2 = 0; - int n3 = 0; + size_t n = 0; + size_t n2 = 0; + size_t n3 = 0; // test numbers // LANG_hu section: set dash information for suggestions if (langnum == LANG_hu) { - while ((n < wl) && (((cw[n] <= '9') && (cw[n] >= '0')) || - (((cw[n] == '.') || (cw[n] == ',')) && (n > 0)))) { + while ((n < wl) && (((scw[n] <= '9') && (scw[n] >= '0')) || + (((scw[n] == '.') || (scw[n] == ',')) && (n > 0)))) { n++; - if ((cw[n] == '.') || (cw[n] == ',')) { + if ((scw[n] == '.') || (scw[n] == ',')) { if (((n2 == 0) && (n > 3)) || - ((n2 > 0) && ((cw[n - 1] == '.') || (cw[n - 1] == ',')))) + ((n2 > 0) && ((scw[n - 1] == '.') || (scw[n - 1] == ',')))) break; n2++; n3 = n; @@ -1676,21 +1388,21 @@ int Hunspell::analyze(char*** slst, const char* word) { if ((n == wl) && (n3 > 0) && (n - n3 > 3)) return 0; - if ((n == wl) || ((n > 0) && ((cw[n] == '%') || (cw[n] == '\xB0')) && - checkword(cw + n, NULL, NULL))) { - mystrcat(result, cw, MAXLNLEN); - result[n - 1] = '\0'; + if ((n == wl) || ((n > 0) && ((scw[n] == '%') || (scw[n] == '\xB0')) && + checkword(scw.c_str() + n, NULL, NULL))) { + result.append(scw); + result.resize(n - 1); if (n == wl) - cat_result(result, pSMgr->suggest_morph(cw + n - 1)); + cat_result(result, pSMgr->suggest_morph(scw.c_str() + n - 1)); else { - char sign = cw[n]; - cw[n] = '\0'; - cat_result(result, pSMgr->suggest_morph(cw + n - 1)); - mystrcat(result, "+", MAXLNLEN); // XXX SPEC. MORPHCODE - cw[n] = sign; - cat_result(result, pSMgr->suggest_morph(cw + n)); + char sign = scw[n]; + scw[n] = '\0'; + cat_result(result, pSMgr->suggest_morph(scw.c_str() + n - 1)); + result.push_back('+'); // XXX SPEC. MORPHCODE + scw[n] = sign; + cat_result(result, pSMgr->suggest_morph(scw.c_str() + n)); } - return line_tok(result, slst, MSEP_REC); + return line_tok(result.c_str(), slst, MSEP_REC); } } // END OF LANG_hu section @@ -1699,64 +1411,58 @@ int Hunspell::analyze(char*** slst, const char* word) { case HUHCAP: case HUHINITCAP: case NOCAP: { - cat_result(result, pSMgr->suggest_morph(cw)); + cat_result(result, pSMgr->suggest_morph(scw.c_str())); if (abbv) { - memcpy(wspace, cw, wl); - *(wspace + wl) = '.'; - *(wspace + wl + 1) = '\0'; - cat_result(result, pSMgr->suggest_morph(wspace)); + std::string u8buffer(scw); + u8buffer.push_back('.'); + cat_result(result, pSMgr->suggest_morph(u8buffer.c_str())); } break; } case INITCAP: { - wl = mkallsmall2(cw, unicw, nc); - memcpy(wspace, cw, (wl + 1)); - wl2 = mkinitcap2(cw, unicw, nc); - cat_result(result, pSMgr->suggest_morph(wspace)); - cat_result(result, pSMgr->suggest_morph(cw)); + wl = mkallsmall2(scw, sunicw); + std::string u8buffer(scw); + mkinitcap2(scw, sunicw); + cat_result(result, pSMgr->suggest_morph(u8buffer.c_str())); + cat_result(result, pSMgr->suggest_morph(scw.c_str())); if (abbv) { - *(wspace + wl) = '.'; - *(wspace + wl + 1) = '\0'; - cat_result(result, pSMgr->suggest_morph(wspace)); + u8buffer.push_back('.'); + cat_result(result, pSMgr->suggest_morph(u8buffer.c_str())); - memcpy(wspace, cw, wl2); - *(wspace + wl2) = '.'; - *(wspace + wl2 + 1) = '\0'; + u8buffer = scw; + u8buffer.push_back('.'); - cat_result(result, pSMgr->suggest_morph(wspace)); + cat_result(result, pSMgr->suggest_morph(u8buffer.c_str())); } break; } case ALLCAP: { - cat_result(result, pSMgr->suggest_morph(cw)); + cat_result(result, pSMgr->suggest_morph(scw.c_str())); if (abbv) { - memcpy(wspace, cw, wl); - *(wspace + wl) = '.'; - *(wspace + wl + 1) = '\0'; - cat_result(result, pSMgr->suggest_morph(cw)); + std::string u8buffer(scw); + u8buffer.push_back('.'); + cat_result(result, pSMgr->suggest_morph(u8buffer.c_str())); } - wl = mkallsmall2(cw, unicw, nc); - memcpy(wspace, cw, (wl + 1)); - wl2 = mkinitcap2(cw, unicw, nc); + mkallsmall2(scw, sunicw); + std::string u8buffer(scw); + mkinitcap2(scw, sunicw); - cat_result(result, pSMgr->suggest_morph(wspace)); - cat_result(result, pSMgr->suggest_morph(cw)); + cat_result(result, pSMgr->suggest_morph(u8buffer.c_str())); + cat_result(result, pSMgr->suggest_morph(scw.c_str())); if (abbv) { - *(wspace + wl) = '.'; - *(wspace + wl + 1) = '\0'; - cat_result(result, pSMgr->suggest_morph(wspace)); + u8buffer.push_back('.'); + cat_result(result, pSMgr->suggest_morph(u8buffer.c_str())); - memcpy(wspace, cw, wl2); - *(wspace + wl2) = '.'; - *(wspace + wl2 + 1) = '\0'; + u8buffer = scw; + u8buffer.push_back('.'); - cat_result(result, pSMgr->suggest_morph(wspace)); + cat_result(result, pSMgr->suggest_morph(u8buffer.c_str())); } break; } } - if (*result) { + if (!result.empty()) { // word reversing wrapper for complex prefixes if (complexprefixes) { if (utf8) @@ -1764,95 +1470,94 @@ int Hunspell::analyze(char*** slst, const char* word) { else reverseword(result); } - return line_tok(result, slst, MSEP_REC); + return line_tok(result.c_str(), slst, MSEP_REC); } // compound word with dash (HU) I18n - char* dash = NULL; - int nresult = 0; // LANG_hu section: set dash information for suggestions - if (langnum == LANG_hu) - dash = (char*)strchr(cw, '-'); - if ((langnum == LANG_hu) && dash) { - *dash = '\0'; + + size_t dash_pos = langnum == LANG_hu ? scw.find('-') : std::string::npos; + int nresult = 0; + if (dash_pos != std::string::npos) { + std::string part1 = scw.substr(0, dash_pos); + std::string part2 = scw.substr(dash_pos+1); + // examine 2 sides of the dash - if (dash[1] == '\0') { // base word ending with dash - if (spell(cw)) { - char* p = pSMgr->suggest_morph(cw); + if (part2.empty()) { // base word ending with dash + if (spell(part1.c_str())) { + char* p = pSMgr->suggest_morph(part1.c_str()); if (p) { int ret = line_tok(p, slst, MSEP_REC); free(p); return ret; } } - } else if ((dash[1] == 'e') && (dash[2] == '\0')) { // XXX (HU) -e hat. - if (spell(cw) && (spell("-e"))) { - st = pSMgr->suggest_morph(cw); + } else if (part2.size() == 1 && part2[0] == 'e') { // XXX (HU) -e hat. + if (spell(part1.c_str()) && (spell("-e"))) { + char* st = pSMgr->suggest_morph(part1.c_str()); if (st) { - mystrcat(result, st, MAXLNLEN); + result.append(st); free(st); } - mystrcat(result, "+", MAXLNLEN); // XXX spec. separator in MORPHCODE + result.push_back('+'); // XXX spec. separator in MORPHCODE st = pSMgr->suggest_morph("-e"); if (st) { - mystrcat(result, st, MAXLNLEN); + result.append(st); free(st); } - return line_tok(result, slst, MSEP_REC); + return line_tok(result.c_str(), slst, MSEP_REC); } } else { // first word ending with dash: word- XXX ??? - char r2 = *(dash + 1); - dash[0] = '-'; - dash[1] = '\0'; - nresult = spell(cw); - dash[1] = r2; - dash[0] = '\0'; - if (nresult && spell(dash + 1) && - ((strlen(dash + 1) > 1) || ((dash[1] > '0') && (dash[1] < '9')))) { - st = pSMgr->suggest_morph(cw); + part1.push_back(' '); + nresult = spell(part1.c_str()); + part1.erase(part1.size() - 1); + if (nresult && spell(part2.c_str()) && + ((part2.size() > 1) || ((part2[0] > '0') && (part2[0] < '9')))) { + char* st = pSMgr->suggest_morph(part1.c_str()); if (st) { - mystrcat(result, st, MAXLNLEN); + result.append(st); free(st); - mystrcat(result, "+", MAXLNLEN); // XXX spec. separator in MORPHCODE + result.push_back('+'); // XXX spec. separator in MORPHCODE } - st = pSMgr->suggest_morph(dash + 1); + st = pSMgr->suggest_morph(part2.c_str()); if (st) { - mystrcat(result, st, MAXLNLEN); + result.append(st); free(st); } - return line_tok(result, slst, MSEP_REC); + return line_tok(result.c_str(), slst, MSEP_REC); } } // affixed number in correct word - if (nresult && (dash > cw) && - (((*(dash - 1) <= '9') && (*(dash - 1) >= '0')) || - (*(dash - 1) == '.'))) { - *dash = '-'; + if (nresult && (dash_pos > 0) && + (((scw[dash_pos - 1] <= '9') && (scw[dash_pos - 1] >= '0')) || + (scw[dash_pos - 1] == '.'))) { n = 1; - if (*(dash - n) == '.') + if (scw[dash_pos - n] == '.') n++; // search first not a number character to left from dash - while (((dash - n) >= cw) && ((*(dash - n) == '0') || (n < 3)) && + while ((dash_pos >= n) && ((scw[dash_pos - n] == '0') || (n < 3)) && (n < 6)) { n++; } - if ((dash - n) < cw) + if (dash_pos < n) n--; // numbers: valami1000000-hoz // examine 100000-hoz, 10000-hoz 1000-hoz, 10-hoz, // 56-hoz, 6-hoz for (; n >= 1; n--) { - if ((*(dash - n) >= '0') && (*(dash - n) <= '9') && - checkword(dash - n, NULL, NULL)) { - mystrcat(result, cw, MAXLNLEN); - result[dash - cw - n] = '\0'; - st = pSMgr->suggest_morph(dash - n); + if (scw[dash_pos - n] < '0' || scw[dash_pos - n] > '9') { + continue; + } + std::string chunk = scw.substr(dash_pos - n); + if (checkword(chunk.c_str(), NULL, NULL)) { + result.append(chunk); + char* st = pSMgr->suggest_morph(chunk.c_str()); if (st) { - mystrcat(result, st, MAXLNLEN); + result.append(st); free(st); } - return line_tok(result, slst, MSEP_REC); + return line_tok(result.c_str(), slst, MSEP_REC); } } } @@ -1866,30 +1571,33 @@ int Hunspell::generate(char*** slst, const char* word, char** pl, int pln) { return 0; char** pl2; int pl2n = analyze(&pl2, word); - int captype = 0; + int captype = NOCAP; int abbv = 0; - char cw[MAXWORDUTF8LEN]; + std::string cw; cleanword(cw, word, &captype, &abbv); - char result[MAXLNLEN]; - *result = '\0'; + std::string result; for (int i = 0; i < pln; i++) { cat_result(result, pSMgr->suggest_gen(pl2, pl2n, pl[i])); } freelist(&pl2, pl2n); - if (*result) { + if (!result.empty()) { // allcap if (captype == ALLCAP) mkallcap(result); // line split - int linenum = line_tok(result, slst, MSEP_REC); + int linenum = line_tok(result.c_str(), slst, MSEP_REC); // capitalize if (captype == INITCAP || captype == HUHINITCAP) { - for (int j = 0; j < linenum; j++) - mkinitcap((*slst)[j]); + for (int j = 0; j < linenum; j++) { + std::string form((*slst)[j]); + free((*slst)[j]); + mkinitcap(form); + (*slst)[j] = mystrdup(form.c_str()); + } } // temporary filtering of prefix related errors (eg. @@ -1923,22 +1631,21 @@ int Hunspell::generate(char*** slst, const char* word, const char* pattern) { } // minimal XML parser functions -int Hunspell::get_xml_par(char* dest, const char* par, int max) { - char* d = dest; +std::string Hunspell::get_xml_par(const char* par) { + std::string dest; if (!par) - return 0; + return dest; char end = *par; - char* dmax = dest + max; if (end == '>') end = '<'; else if (end != '\'' && end != '"') return 0; // bad XML - for (par++; d < dmax && *par != '\0' && *par != end; par++, d++) - *d = *par; - *d = '\0'; + for (par++; *par != '\0' && *par != end; ++par) { + dest.push_back(*par); + } mystrrep(dest, "<", "<"); mystrrep(dest, "&", "&"); - return (int)(d - dest); + return dest; } int Hunspell::get_langnum() const { @@ -1967,18 +1674,17 @@ const char* Hunspell::get_xml_pos(const char* s, const char* attr) { int Hunspell::check_xml_par(const char* q, const char* attr, const char* value) { - char cw[MAXWORDUTF8LEN]; - if (get_xml_par(cw, get_xml_pos(q, attr), MAXWORDUTF8LEN - 1) && - strcmp(cw, value) == 0) + std::string cw = get_xml_par(get_xml_pos(q, attr)); + if (cw == value) return 1; return 0; } -int Hunspell::get_xml_list(char*** slst, char* list, const char* tag) { - int n = 0; - char* p; +int Hunspell::get_xml_list(char*** slst, const char* list, const char* tag) { if (!list) return 0; + int n = 0; + const char* p; for (p = list; ((p = strstr(p, tag)) != NULL); p++) n++; if (n == 0) @@ -1987,25 +1693,20 @@ int Hunspell::get_xml_list(char*** slst, char* list, const char* tag) { if (!*slst) return 0; for (p = list, n = 0; ((p = strstr(p, tag)) != NULL); p++, n++) { - int l = strlen(p); - (*slst)[n] = (char*)malloc(l + 1); - if (!(*slst)[n]) - return n; - if (!get_xml_par((*slst)[n], p + strlen(tag) - 1, l)) { - free((*slst)[n]); + std::string cw = get_xml_par(p + strlen(tag) - 1); + if (cw.empty()) { break; } + (*slst)[n] = mystrdup(cw.c_str()); } return n; } int Hunspell::spellml(char*** slst, const char* word) { - char *q, *q2; - char cw[MAXWORDUTF8LEN], cw2[MAXWORDUTF8LEN]; - q = (char*)strstr(word, "<query"); + const char* q = strstr(word, "<query"); if (!q) return 0; // bad XML input - q2 = strchr(q, '>'); + const char* q2 = strchr(q, '>'); if (!q2) return 0; // bad XML input q2 = strstr(q2, "<word"); @@ -2013,8 +1714,9 @@ int Hunspell::spellml(char*** slst, const char* word) { return 0; // bad XML input if (check_xml_par(q, "type=", "analyze")) { int n = 0; - if (get_xml_par(cw, strchr(q2, '>'), MAXWORDUTF8LEN - 10)) - n = analyze(slst, cw); + std::string cw = get_xml_par(strchr(q2, '>')); + if (!cw.empty()) + n = analyze(slst, cw.c_str()); if (n == 0) return 0; // convert the result to <code><a>ana1</a><a>ana2</a></code> format @@ -2036,22 +1738,25 @@ int Hunspell::spellml(char*** slst, const char* word) { (*slst)[0] = mystrdup(r.c_str()); return 1; } else if (check_xml_par(q, "type=", "stem")) { - if (get_xml_par(cw, strchr(q2, '>'), MAXWORDUTF8LEN - 1)) - return stem(slst, cw); + std::string cw = get_xml_par(strchr(q2, '>')); + if (!cw.empty()) + return stem(slst, cw.c_str()); } else if (check_xml_par(q, "type=", "generate")) { - int n = get_xml_par(cw, strchr(q2, '>'), MAXWORDUTF8LEN - 1); - if (n == 0) + std::string cw = get_xml_par(strchr(q2, '>')); + if (cw.empty()) return 0; - char* q3 = strstr(q2 + 1, "<word"); + const char* q3 = strstr(q2 + 1, "<word"); if (q3) { - if (get_xml_par(cw2, strchr(q3, '>'), MAXWORDUTF8LEN - 1)) { - return generate(slst, cw, cw2); + std::string cw2 = get_xml_par(strchr(q3, '>')); + if (!cw2.empty()) { + return generate(slst, cw.c_str(), cw2.c_str()); } } else { if ((q2 = strstr(q2 + 1, "<code")) != NULL) { char** slst2; - if ((n = get_xml_list(&slst2, strchr(q2, '>'), "<a>")) != 0) { - int n2 = generate(slst, cw, slst2, n); + int n = get_xml_list(&slst2, strchr(q2, '>'), "<a>"); + if (n != 0) { + int n2 = generate(slst, cw.c_str(), slst2, n); freelist(&slst2, n); return uniqlist(*slst, n2); } @@ -2062,182 +1767,6 @@ int Hunspell::spellml(char*** slst, const char* word) { return 0; } -#ifdef HUNSPELL_EXPERIMENTAL -// XXX is UTF-8 support OK? -char* Hunspell::morph_with_correction(const char* word) { - char cw[MAXWORDUTF8LEN]; - char wspace[MAXWORDUTF8LEN]; - if (!pSMgr || maxdic == 0) - return NULL; - w_char unicw[MAXWORDLEN]; - int nc = strlen(word); - if (utf8) { - if (nc >= MAXWORDUTF8LEN) - return NULL; - } else { - if (nc >= MAXWORDLEN) - return NULL; - } - int captype = 0; - int abbv = 0; - int wl = 0; - - // input conversion - RepList* rl = (pAMgr) ? pAMgr->get_iconvtable() : NULL; - int convstatus = rl ? rl->conv(word, wspace) : 0; - if (convstatus < 0) - return 0; - else if (convstatus > 0) - wl = cleanword2(cw, wspace, unicw, &nc, &captype, &abbv); - else - wl = cleanword2(cw, word, unicw, &nc, &captype, &abbv); - - if (wl == 0) - return NULL; - - char result[MAXLNLEN]; - char* st = NULL; - - *result = '\0'; - - switch (captype) { - case NOCAP: { - st = pSMgr->suggest_morph_for_spelling_error(cw); - if (st) { - mystrcat(result, st, MAXLNLEN); - free(st); - } - if (abbv) { - memcpy(wspace, cw, wl); - *(wspace + wl) = '.'; - *(wspace + wl + 1) = '\0'; - st = pSMgr->suggest_morph_for_spelling_error(wspace); - if (st) { - if (*result) - mystrcat(result, "\n", MAXLNLEN); - mystrcat(result, st, MAXLNLEN); - free(st); - } - } - break; - } - case INITCAP: { - memcpy(wspace, cw, (wl + 1)); - mkallsmall2(wspace, unicw, nc); - st = pSMgr->suggest_morph_for_spelling_error(wspace); - if (st) { - mystrcat(result, st, MAXLNLEN); - free(st); - } - st = pSMgr->suggest_morph_for_spelling_error(cw); - if (st) { - if (*result) - mystrcat(result, "\n", MAXLNLEN); - mystrcat(result, st, MAXLNLEN); - free(st); - } - if (abbv) { - memcpy(wspace, cw, wl); - *(wspace + wl) = '.'; - *(wspace + wl + 1) = '\0'; - mkallsmall2(wspace, unicw, nc); - st = pSMgr->suggest_morph_for_spelling_error(wspace); - if (st) { - if (*result) - mystrcat(result, "\n", MAXLNLEN); - mystrcat(result, st, MAXLNLEN); - free(st); - } - mkinitcap(wspace); - st = pSMgr->suggest_morph_for_spelling_error(wspace); - if (st) { - if (*result) - mystrcat(result, "\n", MAXLNLEN); - mystrcat(result, st, MAXLNLEN); - free(st); - } - } - break; - } - case HUHCAP: { - st = pSMgr->suggest_morph_for_spelling_error(cw); - if (st) { - mystrcat(result, st, MAXLNLEN); - free(st); - } - memcpy(wspace, cw, (wl + 1)); - mkallsmall2(wspace, unicw, nc); - st = pSMgr->suggest_morph_for_spelling_error(wspace); - if (st) { - if (*result) - mystrcat(result, "\n", MAXLNLEN); - mystrcat(result, st, MAXLNLEN); - free(st); - } - break; - } - case ALLCAP: { - memcpy(wspace, cw, (wl + 1)); - st = pSMgr->suggest_morph_for_spelling_error(wspace); - if (st) { - mystrcat(result, st, MAXLNLEN); - free(st); - } - mkallsmall2(wspace, unicw, nc); - st = pSMgr->suggest_morph_for_spelling_error(wspace); - if (st) { - if (*result) - mystrcat(result, "\n", MAXLNLEN); - mystrcat(result, st, MAXLNLEN); - free(st); - } - mkinitcap(wspace); - st = pSMgr->suggest_morph_for_spelling_error(wspace); - if (st) { - if (*result) - mystrcat(result, "\n", MAXLNLEN); - mystrcat(result, st, MAXLNLEN); - free(st); - } - if (abbv) { - memcpy(wspace, cw, (wl + 1)); - *(wspace + wl) = '.'; - *(wspace + wl + 1) = '\0'; - if (*result) - mystrcat(result, "\n", MAXLNLEN); - st = pSMgr->suggest_morph_for_spelling_error(wspace); - if (st) { - mystrcat(result, st, MAXLNLEN); - free(st); - } - mkallsmall2(wspace, unicw, nc); - st = pSMgr->suggest_morph_for_spelling_error(wspace); - if (st) { - if (*result) - mystrcat(result, "\n", MAXLNLEN); - mystrcat(result, st, MAXLNLEN); - free(st); - } - mkinitcap(wspace); - st = pSMgr->suggest_morph_for_spelling_error(wspace); - if (st) { - if (*result) - mystrcat(result, "\n", MAXLNLEN); - mystrcat(result, st, MAXLNLEN); - free(st); - } - } - break; - } - } - - if (*result) - return mystrdup(result); - return NULL; -} - -#endif // END OF HUNSPELL_EXPERIMENTAL CODE - Hunhandle* Hunspell_create(const char* affpath, const char* dpath) { return (Hunhandle*)(new Hunspell(affpath, dpath)); } @@ -2333,10 +1862,9 @@ int Hunspell::suffix_suggest(char*** slst, const char* root_word) { if (ignoredchars != NULL) { w2.assign(root_word); if (utf8) { - int ignoredchars_utf16_len; - unsigned short* ignoredchars_utf16 = - pAMgr->get_ignore_utf16(&ignoredchars_utf16_len); - remove_ignored_chars_utf(w2, ignoredchars_utf16, ignoredchars_utf16_len); + const std::vector<w_char>& ignoredchars_utf16 = + pAMgr->get_ignore_utf16(); + remove_ignored_chars_utf(w2, ignoredchars_utf16); } else { remove_ignored_chars(w2, ignoredchars); } diff --git a/libs/hunspell/src/hunspell.hxx b/libs/hunspell/src/hunspell.hxx index 259d44f86c..3bcf75e39c 100644 --- a/libs/hunspell/src/hunspell.hxx +++ b/libs/hunspell/src/hunspell.hxx @@ -77,6 +77,7 @@ #include "affixmgr.hxx" #include "suggestmgr.hxx" #include "langnum.hxx" +#include <vector> #define SPELL_XML "<?xml?>" @@ -215,7 +216,7 @@ class LIBHUNSPELL_DLL_EXPORTED Hunspell { /* get extra word characters definied in affix file for tokenization */ const char* get_wordchars(); - unsigned short* get_wordchars_utf16(int* len); + const std::vector<w_char>& get_wordchars_utf16(); struct cs_info* get_csconv(); const char* get_version(); @@ -229,45 +230,32 @@ class LIBHUNSPELL_DLL_EXPORTED Hunspell { { return pAMgr->get_try_string(); } -/* experimental and deprecated functions */ - -#ifdef HUNSPELL_EXPERIMENTAL - /* suffix is an affix flag string, similarly in dictionary files */ - int put_word_suffix(const char* word, const char* suffix); - char* morph_with_correction(const char* word); - - /* spec. suggestions */ - int suggest_auto(char*** slst, const char* word); - int suggest_pos_stems(char*** slst, const char* word); -#endif private: - int cleanword(char*, const char*, int* pcaptype, int* pabbrev); - int cleanword2(char*, - const char*, - w_char*, - int* w_len, - int* pcaptype, - int* pabbrev); - void mkinitcap(char*); - int mkinitcap2(char* p, w_char* u, int nc); - int mkinitsmall2(char* p, w_char* u, int nc); - void mkallcap(char*); - int mkallcap2(char* p, w_char* u, int nc); - void mkallsmall(char*); - int mkallsmall2(char* p, w_char* u, int nc); + void cleanword(std::string& dest, const char*, int* pcaptype, int* pabbrev); + size_t cleanword2(std::string& dest, + std::vector<w_char>& dest_u, + const char*, + int* w_len, + int* pcaptype, + size_t* pabbrev); + void mkinitcap(std::string& u8); + int mkinitcap2(std::string& u8, std::vector<w_char>& u16); + int mkinitsmall2(std::string& u8, std::vector<w_char>& u16); + void mkallcap(std::string& u8); + int mkallsmall2(std::string& u8, std::vector<w_char>& u16); struct hentry* checkword(const char*, int* info, char** root); - char* sharps_u8_l1(char* dest, char* source); + std::string sharps_u8_l1(const std::string& source); hentry* - spellsharps(char* base, char*, int, int, char* tmp, int* info, char** root); + spellsharps(std::string& base, size_t start_pos, int, int, int* info, char** root); int is_keepcase(const hentry* rv); - int insert_sug(char*** slst, char* word, int ns); - void cat_result(char* result, char* st); + int insert_sug(char*** slst, const char* word, int ns); + void cat_result(std::string& result, char* st); char* stem_description(const char* desc); int spellml(char*** slst, const char* word); - int get_xml_par(char* dest, const char* par, int maxl); + std::string get_xml_par(const char* par); const char* get_xml_pos(const char* s, const char* attr); - int get_xml_list(char*** slst, char* list, const char* tag); + int get_xml_list(char*** slst, const char* list, const char* tag); int check_xml_par(const char* q, const char* attr, const char* value); }; diff --git a/libs/hunspell/src/phonet.c++ b/libs/hunspell/src/phonet.c++ index 2b4d2ae504..17350e74a7 100644 --- a/libs/hunspell/src/phonet.c++ +++ b/libs/hunspell/src/phonet.c++ @@ -66,33 +66,30 @@ static int myisalpha(char ch) { return 1; } +/* Do phonetic transformation. */ /* phonetic transcription algorithm */ /* see: http://aspell.net/man-html/Phonetic-Code.html */ /* convert string to uppercase before this call */ -int phonet(const char* inword, char* target, int len, phonetable& parms) { - /** Do phonetic transformation. **/ - /** "len" = length of "inword" incl. '\0'. **/ +std::string phonet(const std::string& inword, phonetable& parms) { - /** result: >= 0: length of "target" **/ - /** otherwise: error **/ - - int i, j, k = 0, n, p, z; + int i, k = 0, p, z; int k0, n0, p0 = -333, z0; - char c, c0; + char c; const char* s; typedef unsigned char uchar; - char word[MAXPHONETUTF8LEN + 1]; - if (len == -1) - len = strlen(inword); + + size_t len = inword.size(); if (len > MAXPHONETUTF8LEN) - return 0; - strncpy(word, inword, MAXPHONETUTF8LEN); + return std::string(); + char word[MAXPHONETUTF8LEN + 1]; + strncpy(word, inword.c_str(), MAXPHONETUTF8LEN); word[MAXPHONETUTF8LEN] = '\0'; + std::string target; /** check word **/ - i = j = z = 0; + i = z = 0; while ((c = word[i]) != '\0') { - n = parms.hash[(uchar)c]; + int n = parms.hash[(uchar)c]; z0 = 0; if (n >= 0) { @@ -141,7 +138,7 @@ int phonet(const char* inword, char* target, int len, phonetable& parms) { (!myisalpha(word[i + k0])))) { /** search for followup rules, if: **/ /** parms.followup and k > 1 and NO '-' in searchstring **/ - c0 = word[i + k - 1]; + char c0 = word[i + k - 1]; n0 = parms.hash[(uchar)c0]; // if (parms.followup && k > 1 && n0 >= 0 @@ -216,9 +213,9 @@ int phonet(const char* inword, char* target, int len, phonetable& parms) { : 0; if (p0 == 1 && z == 0) { /** rule with '<' is used **/ - if (j > 0 && *s != '\0' && - (target[j - 1] == c || target[j - 1] == *s)) { - j--; + if (!target.empty() && *s != '\0' && + (target[target.size()-1] == c || target[target.size()-1] == *s)) { + target.erase(target.size() - 1); } z0 = 1; z = 1; @@ -236,10 +233,9 @@ int phonet(const char* inword, char* target, int len, phonetable& parms) { } else { /** no '<' rule used **/ i += k - 1; z = 0; - while (*s != '\0' && *(s + 1) != '\0' && j < len) { - if (j == 0 || target[j - 1] != *s) { - target[j] = *s; - j++; + while (*s != '\0' && *(s + 1) != '\0' && target.size() < len) { + if (target.empty() || target[target.size()-1] != *s) { + target.push_back(*s); } s++; } @@ -248,8 +244,7 @@ int phonet(const char* inword, char* target, int len, phonetable& parms) { if (parms.rules[n][0] != '\0' && strstr(parms.rules[n] + 1, "^^") != NULL) { if (c != '\0') { - target[j] = c; - j++; + target.push_back(c); } strmove(&word[0], &word[0] + i + 1); i = 0; @@ -262,15 +257,11 @@ int phonet(const char* inword, char* target, int len, phonetable& parms) { } /** end of while (parms.rules[n][0] == c) **/ } /** end of if (n >= 0) **/ if (z0 == 0) { - // if (k && (assert(p0!=-333),!p0) && j < len && c != '\0' - // && (!parms.collapse_result || j == 0 || target[j-1] != - // c)){ - if (k && !p0 && j < len && c != '\0' && - (1 || j == 0 || target[j - 1] != c)) { + if (k && !p0 && target.size() < len && c != '\0' && + (1 || target.empty() || target[target.size()-1] != c)) { /** condense only double letters **/ - target[j] = c; + target.push_back(c); /// printf("\n setting \n"); - j++; } i++; @@ -279,7 +270,5 @@ int phonet(const char* inword, char* target, int len, phonetable& parms) { } } /** end of while ((c = word[i]) != '\0') **/ - target[j] = '\0'; - return (j); - + return target; } /** end of function "phonet" **/ diff --git a/libs/hunspell/src/phonet.hxx b/libs/hunspell/src/phonet.hxx index cb0dbed3fb..eb9fd0c628 100644 --- a/libs/hunspell/src/phonet.hxx +++ b/libs/hunspell/src/phonet.hxx @@ -46,9 +46,7 @@ struct phonetable { LIBHUNSPELL_DLL_EXPORTED void init_phonet_hash(phonetable& parms); -LIBHUNSPELL_DLL_EXPORTED int phonet(const char* inword, - char* target, - int len, - phonetable& phone); +LIBHUNSPELL_DLL_EXPORTED std::string phonet(const std::string& inword, + phonetable& phone); #endif diff --git a/libs/hunspell/src/replist.c++ b/libs/hunspell/src/replist.c++ index ace6c4aaf8..b3e6b37d20 100644 --- a/libs/hunspell/src/replist.c++ +++ b/libs/hunspell/src/replist.c++ @@ -151,7 +151,7 @@ int RepList::add(char* pat1, char* pat2) { } int RepList::conv(const char* word, char* dest, size_t destsize) { - int stl = 0; + size_t stl = 0; int change = 0; for (size_t i = 0; i < strlen(word); i++) { int n = near(word + i); @@ -173,3 +173,21 @@ int RepList::conv(const char* word, char* dest, size_t destsize) { dest[stl] = '\0'; return change; } + +bool RepList::conv(const char* word, std::string& dest) { + dest.clear(); + + bool change = false; + for (size_t i = 0; i < strlen(word); i++) { + int n = near(word + i); + int l = match(word + i, n); + if (l) { + dest.append(dat[n]->pattern2); + i += l - 1; + change = true; + } else { + dest.push_back(word[i]); + } + } + return change; +} diff --git a/libs/hunspell/src/replist.hxx b/libs/hunspell/src/replist.hxx index 319eb03fb0..59366e9e02 100644 --- a/libs/hunspell/src/replist.hxx +++ b/libs/hunspell/src/replist.hxx @@ -79,6 +79,9 @@ #include "w_char.hxx" +#include <string> +#include <vector> + class LIBHUNSPELL_DLL_EXPORTED RepList { private: RepList(const RepList&); @@ -100,5 +103,6 @@ class LIBHUNSPELL_DLL_EXPORTED RepList { int near(const char* word); int match(const char* word, int n); int conv(const char* word, char* dest, size_t destsize); + bool conv(const char* word, std::string& dest); }; #endif diff --git a/libs/hunspell/src/suggestmgr.c++ b/libs/hunspell/src/suggestmgr.c++ index 4269a1181a..17becd7582 100644 --- a/libs/hunspell/src/suggestmgr.c++ +++ b/libs/hunspell/src/suggestmgr.c++ @@ -125,11 +125,11 @@ SuggestMgr::SuggestMgr(const char* tryme, int maxn, AffixMgr* aptr) { if (ckey) { if (utf8) { - w_char t[MAXSWL]; - ckeyl = u8_u16(t, MAXSWL, ckey); + std::vector<w_char> t; + ckeyl = u8_u16(t, ckey); ckey_utf = (w_char*)malloc(ckeyl * sizeof(w_char)); if (ckey_utf) - memcpy(ckey_utf, t, ckeyl * sizeof(w_char)); + memcpy(ckey_utf, &t[0], ckeyl * sizeof(w_char)); else ckeyl = 0; } else { @@ -142,11 +142,11 @@ SuggestMgr::SuggestMgr(const char* tryme, int maxn, AffixMgr* aptr) { if (ctry) ctryl = strlen(ctry); if (ctry && utf8) { - w_char t[MAXSWL]; - ctryl = u8_u16(t, MAXSWL, tryme); + std::vector<w_char> t; + ctryl = u8_u16(t, tryme); ctry_utf = (w_char*)malloc(ctryl * sizeof(w_char)); if (ctry_utf) - memcpy(ctry_utf, t, ctryl * sizeof(w_char)); + memcpy(ctry_utf, &t[0], ctryl * sizeof(w_char)); else ctryl = 0; } @@ -213,7 +213,7 @@ int SuggestMgr::suggest(char*** slst, int* onlycompoundsug) { int nocompoundtwowords = 0; char** wlst; - w_char word_utf[MAXSWL]; + std::vector<w_char> word_utf; int wl = 0; int nsugorig = nsug; std::string w2; @@ -242,7 +242,7 @@ int SuggestMgr::suggest(char*** slst, } if (utf8) { - wl = u8_u16(word_utf, MAXSWL, word); + wl = u8_u16(word_utf, word); if (wl == -1) { *slst = wlst; return nsug; @@ -257,7 +257,7 @@ int SuggestMgr::suggest(char*** slst, // suggestions for an uppercase word (html -> HTML) if ((nsug < maxSug) && (nsug > -1)) { - nsug = (utf8) ? capchars_utf(wlst, word_utf, wl, nsug, cpdsuggest) + nsug = (utf8) ? capchars_utf(wlst, &word_utf[0], wl, nsug, cpdsuggest) : capchars(wlst, word, nsug, cpdsuggest); } @@ -280,56 +280,56 @@ int SuggestMgr::suggest(char*** slst, // did we swap the order of chars by mistake if ((nsug < maxSug) && (nsug > -1) && (!cpdsuggest || (nsug < oldSug + maxcpdsugs))) { - nsug = (utf8) ? swapchar_utf(wlst, word_utf, wl, nsug, cpdsuggest) + nsug = (utf8) ? swapchar_utf(wlst, &word_utf[0], wl, nsug, cpdsuggest) : swapchar(wlst, word, nsug, cpdsuggest); } // did we swap the order of non adjacent chars by mistake if ((nsug < maxSug) && (nsug > -1) && (!cpdsuggest || (nsug < oldSug + maxcpdsugs))) { - nsug = (utf8) ? longswapchar_utf(wlst, word_utf, wl, nsug, cpdsuggest) + nsug = (utf8) ? longswapchar_utf(wlst, &word_utf[0], wl, nsug, cpdsuggest) : longswapchar(wlst, word, nsug, cpdsuggest); } // did we just hit the wrong key in place of a good char (case and keyboard) if ((nsug < maxSug) && (nsug > -1) && (!cpdsuggest || (nsug < oldSug + maxcpdsugs))) { - nsug = (utf8) ? badcharkey_utf(wlst, word_utf, wl, nsug, cpdsuggest) + nsug = (utf8) ? badcharkey_utf(wlst, &word_utf[0], wl, nsug, cpdsuggest) : badcharkey(wlst, word, nsug, cpdsuggest); } // did we add a char that should not be there if ((nsug < maxSug) && (nsug > -1) && (!cpdsuggest || (nsug < oldSug + maxcpdsugs))) { - nsug = (utf8) ? extrachar_utf(wlst, word_utf, wl, nsug, cpdsuggest) + nsug = (utf8) ? extrachar_utf(wlst, &word_utf[0], wl, nsug, cpdsuggest) : extrachar(wlst, word, nsug, cpdsuggest); } // did we forgot a char if ((nsug < maxSug) && (nsug > -1) && (!cpdsuggest || (nsug < oldSug + maxcpdsugs))) { - nsug = (utf8) ? forgotchar_utf(wlst, word_utf, wl, nsug, cpdsuggest) + nsug = (utf8) ? forgotchar_utf(wlst, &word_utf[0], wl, nsug, cpdsuggest) : forgotchar(wlst, word, nsug, cpdsuggest); } // did we move a char if ((nsug < maxSug) && (nsug > -1) && (!cpdsuggest || (nsug < oldSug + maxcpdsugs))) { - nsug = (utf8) ? movechar_utf(wlst, word_utf, wl, nsug, cpdsuggest) + nsug = (utf8) ? movechar_utf(wlst, &word_utf[0], wl, nsug, cpdsuggest) : movechar(wlst, word, nsug, cpdsuggest); } // did we just hit the wrong key in place of a good char if ((nsug < maxSug) && (nsug > -1) && (!cpdsuggest || (nsug < oldSug + maxcpdsugs))) { - nsug = (utf8) ? badchar_utf(wlst, word_utf, wl, nsug, cpdsuggest) + nsug = (utf8) ? badchar_utf(wlst, &word_utf[0], wl, nsug, cpdsuggest) : badchar(wlst, word, nsug, cpdsuggest); } // did we double two characters if ((nsug < maxSug) && (nsug > -1) && (!cpdsuggest || (nsug < oldSug + maxcpdsugs))) { - nsug = (utf8) ? doubletwochars_utf(wlst, word_utf, wl, nsug, cpdsuggest) + nsug = (utf8) ? doubletwochars_utf(wlst, &word_utf[0], wl, nsug, cpdsuggest) : doubletwochars(wlst, word, nsug, cpdsuggest); } @@ -357,88 +357,17 @@ int SuggestMgr::suggest(char*** slst, return nsug; } -// generate suggestions for a word with typical mistake -// pass in address of array of char * pointers -#ifdef HUNSPELL_EXPERIMENTAL -int SuggestMgr::suggest_auto(char*** slst, const char* w, int nsug) { - int nocompoundtwowords = 0; - char** wlst; - int oldSug; - - char w2[MAXWORDUTF8LEN]; - const char* word = w; - - // word reversing wrapper for complex prefixes - if (complexprefixes) { - strcpy(w2, w); - if (utf8) - reverseword_utf(w2); - else - reverseword(w2); - word = w2; - } - - if (*slst) { - wlst = *slst; - } else { - wlst = (char**)malloc(maxSug * sizeof(char*)); - if (wlst == NULL) - return -1; - } - - for (int cpdsuggest = 0; (cpdsuggest < 2) && (nocompoundtwowords == 0); - cpdsuggest++) { - // limit compound suggestion - if (cpdsuggest > 0) - oldSug = nsug; - - // perhaps we made a typical fault of spelling - if ((nsug < maxSug) && (nsug > -1)) - nsug = replchars(wlst, word, nsug, cpdsuggest); - - // perhaps we made chose the wrong char from a related set - if ((nsug < maxSug) && (nsug > -1) && - (!cpdsuggest || (nsug < oldSug + maxcpdsugs))) - nsug = mapchars(wlst, word, nsug, cpdsuggest); - - if ((cpdsuggest == 0) && (nsug > 0)) - nocompoundtwowords = 1; - - // perhaps we forgot to hit space and two words ran together - - if ((nsug < maxSug) && (nsug > -1) && - (!cpdsuggest || (nsug < oldSug + maxcpdsugs)) && - check_forbidden(word, strlen(word))) { - nsug = twowords(wlst, word, nsug, cpdsuggest); - } - - } // repeating ``for'' statement compounding support - - if (nsug < 0) { - for (int i = 0; i < maxSug; i++) - if (wlst[i] != NULL) - free(wlst[i]); - free(wlst); - return -1; - } - - *slst = wlst; - return nsug; -} -#endif // END OF HUNSPELL_EXPERIMENTAL CODE - // suggestions for an uppercase word (html -> HTML) int SuggestMgr::capchars_utf(char** wlst, const w_char* word, int wl, int ns, int cpdsuggest) { - char candidate[MAXSWUTF8L]; - w_char candidate_utf[MAXSWL]; - memcpy(candidate_utf, word, wl * sizeof(w_char)); - mkallcap_utf(candidate_utf, wl, langnum); - u16_u8(candidate, MAXSWUTF8L, candidate_utf, wl); - return testsug(wlst, candidate, strlen(candidate), ns, cpdsuggest, NULL, + std::vector<w_char> candidate_utf(word, word + wl); + mkallcap_utf(candidate_utf, langnum); + std::string candidate; + u16_u8(candidate, candidate_utf); + return testsug(wlst, candidate.c_str(), candidate.size(), ns, cpdsuggest, NULL, NULL); } @@ -449,7 +378,7 @@ int SuggestMgr::capchars(char** wlst, int cpdsuggest) { std::string candidate(word); mkallcap(candidate, csconv); - return testsug(wlst, candidate.data(), candidate.size(), ns, cpdsuggest, NULL, + return testsug(wlst, candidate.c_str(), candidate.size(), ns, cpdsuggest, NULL, NULL); } @@ -458,10 +387,9 @@ int SuggestMgr::mapchars(char** wlst, const char* word, int ns, int cpdsuggest) { - char candidate[MAXSWUTF8L]; + std::string candidate; clock_t timelimit; int timer; - candidate[0] = '\0'; int wl = strlen(word); if (wl < 2 || !pAMgr) @@ -474,14 +402,13 @@ int SuggestMgr::mapchars(char** wlst, timelimit = clock(); timer = MINTIMER; - return map_related(word, (char*)&candidate, 0, 0, wlst, cpdsuggest, ns, + return map_related(word, candidate, 0, wlst, cpdsuggest, ns, maptable, nummap, &timer, &timelimit); } int SuggestMgr::map_related(const char* word, - char* candidate, + std::string& candidate, int wn, - int cn, char** wlst, int cpdsuggest, int ns, @@ -491,17 +418,15 @@ int SuggestMgr::map_related(const char* word, clock_t* timelimit) { if (*(word + wn) == '\0') { int cwrd = 1; - *(candidate + cn) = '\0'; - int wl = strlen(candidate); for (int m = 0; m < ns; m++) { - if (strcmp(candidate, wlst[m]) == 0) { + if (candidate == wlst[m]) { cwrd = 0; break; } } - if ((cwrd) && checkword(candidate, wl, cpdsuggest, timer, timelimit)) { + if ((cwrd) && checkword(candidate.c_str(), candidate.size(), cpdsuggest, timer, timelimit)) { if (ns < maxSug) { - wlst[ns] = mystrdup(candidate); + wlst[ns] = mystrdup(candidate.c_str()); if (wlst[ns] == NULL) return -1; ns++; @@ -515,9 +440,11 @@ int SuggestMgr::map_related(const char* word, int len = strlen(maptable[j].set[k]); if (strncmp(maptable[j].set[k], word + wn, len) == 0) { in_map = 1; + size_t cn = candidate.size(); for (int l = 0; l < maptable[j].len; l++) { - strcpy(candidate + cn, maptable[j].set[l]); - ns = map_related(word, candidate, wn + len, strlen(candidate), wlst, + candidate.resize(cn); + candidate.append(maptable[j].set[l]); + ns = map_related(word, candidate, wn + len, wlst, cpdsuggest, ns, maptable, nummap, timer, timelimit); if (!(*timer)) return ns; @@ -526,8 +453,8 @@ int SuggestMgr::map_related(const char* word, } } if (!in_map) { - *(candidate + cn) = *(word + wn); - ns = map_related(word, candidate, wn + 1, cn + 1, wlst, cpdsuggest, ns, + candidate.push_back(*(word + wn)); + ns = map_related(word, candidate, wn + 1, wlst, cpdsuggest, ns, maptable, nummap, timer, timelimit); } return ns; @@ -539,9 +466,7 @@ int SuggestMgr::replchars(char** wlst, const char* word, int ns, int cpdsuggest) { - char candidate[MAXSWUTF8L]; - const char* r; - int lenr, lenp; + std::string candidate; int wl = strlen(word); if (wl < 2 || !pAMgr) return ns; @@ -550,45 +475,42 @@ int SuggestMgr::replchars(char** wlst, if (reptable == NULL) return ns; for (int i = 0; i < numrep; i++) { - r = word; - lenr = strlen(reptable[i].pattern2); - lenp = strlen(reptable[i].pattern); + const char* r = word; // search every occurence of the pattern in the word while ((r = strstr(r, reptable[i].pattern)) != NULL && (!reptable[i].end || strlen(r) == strlen(reptable[i].pattern)) && (!reptable[i].start || r == word)) { - strcpy(candidate, word); - if (r - word + lenr + strlen(r + lenp) >= MAXSWUTF8L) - break; - strcpy(candidate + (r - word), reptable[i].pattern2); - strcpy(candidate + (r - word) + lenr, r + lenp); - ns = testsug(wlst, candidate, wl - lenp + lenr, ns, cpdsuggest, NULL, + candidate.assign(word); + candidate.resize(r - word); + candidate.append(reptable[i].pattern2); + int lenp = strlen(reptable[i].pattern); + candidate.append(r + lenp); + ns = testsug(wlst, candidate.c_str(), candidate.size(), ns, cpdsuggest, NULL, NULL); if (ns == -1) return -1; // check REP suggestions with space - char* sp = strchr(candidate, ' '); - if (sp) { - char* prev = candidate; - while (sp) { - *sp = '\0'; - if (checkword(prev, strlen(prev), 0, NULL, NULL)) { + size_t sp = candidate.find(' '); + if (sp != std::string::npos) { + size_t prev = 0; + while (sp != std::string::npos) { + std::string prev_chunk = candidate.substr(prev, sp - prev); + if (checkword(prev_chunk.c_str(), prev_chunk.size(), 0, NULL, NULL)) { int oldns = ns; - *sp = ' '; - ns = testsug(wlst, sp + 1, strlen(sp + 1), ns, cpdsuggest, NULL, + std::string post_chunk = candidate.substr(sp + 1); + ns = testsug(wlst, post_chunk.c_str(), post_chunk.size(), ns, cpdsuggest, NULL, NULL); if (ns == -1) return -1; if (oldns < ns) { free(wlst[ns - 1]); - wlst[ns - 1] = mystrdup(candidate); + wlst[ns - 1] = mystrdup(candidate.c_str()); if (!wlst[ns - 1]) return -1; } } - *sp = ' '; prev = sp + 1; - sp = strchr(prev, ' '); + sp = candidate.find(' ', prev); } } r++; // search for the next letter @@ -603,7 +525,6 @@ int SuggestMgr::doubletwochars(char** wlst, const char* word, int ns, int cpdsuggest) { - char candidate[MAXSWUTF8L]; int state = 0; int wl = strlen(word); if (wl < 5 || !pAMgr) @@ -612,9 +533,9 @@ int SuggestMgr::doubletwochars(char** wlst, if (word[i] == word[i - 2]) { state++; if (state == 3) { - strcpy(candidate, word); - strcpy(candidate + i - 1, word + i + 1); - ns = testsug(wlst, candidate, wl - 2, ns, cpdsuggest, NULL, NULL); + std::string candidate(word, word + i - 1); + candidate.insert(candidate.end(), word + i + 1, word + wl); + ns = testsug(wlst, candidate.c_str(), candidate.size(), ns, cpdsuggest, NULL, NULL); if (ns == -1) return -1; state = 0; @@ -633,20 +554,18 @@ int SuggestMgr::doubletwochars_utf(char** wlst, int wl, int ns, int cpdsuggest) { - w_char candidate_utf[MAXSWL]; - char candidate[MAXSWUTF8L]; int state = 0; if (wl < 5 || !pAMgr) return ns; for (int i = 2; i < wl; i++) { - if (w_char_eq(word[i], word[i - 2])) { + if (word[i] == word[i - 2]) { state++; if (state == 3) { - memcpy(candidate_utf, word, (i - 1) * sizeof(w_char)); - memcpy(candidate_utf + i - 1, word + i + 1, - (wl - i - 1) * sizeof(w_char)); - u16_u8(candidate, MAXSWUTF8L, candidate_utf, wl - 2); - ns = testsug(wlst, candidate, strlen(candidate), ns, cpdsuggest, NULL, + std::vector<w_char> candidate_utf(word, word + i - 1); + candidate_utf.insert(candidate_utf.end(), word + i + 1, word + wl); + std::string candidate; + u16_u8(candidate, candidate_utf); + ns = testsug(wlst, candidate.c_str(), candidate.size(), ns, cpdsuggest, NULL, NULL); if (ns == -1) return -1; @@ -665,19 +584,16 @@ int SuggestMgr::badcharkey(char** wlst, const char* word, int ns, int cpdsuggest) { - char tmpc; - char candidate[MAXSWUTF8L]; - int wl = strlen(word); - strcpy(candidate, word); + std::string candidate(word); + // swap out each char one by one and try uppercase and neighbor // keyboard chars in its place to see if that makes a good word - - for (int i = 0; i < wl; i++) { - tmpc = candidate[i]; + for (size_t i = 0; i < candidate.size(); ++i) { + char tmpc = candidate[i]; // check with uppercase letters candidate[i] = csconv[((unsigned char)tmpc)].cupper; if (tmpc != candidate[i]) { - ns = testsug(wlst, candidate, wl, ns, cpdsuggest, NULL, NULL); + ns = testsug(wlst, candidate.c_str(), candidate.size(), ns, cpdsuggest, NULL, NULL); if (ns == -1) return -1; candidate[i] = tmpc; @@ -689,13 +605,13 @@ int SuggestMgr::badcharkey(char** wlst, while (loc) { if ((loc > ckey) && (*(loc - 1) != '|')) { candidate[i] = *(loc - 1); - ns = testsug(wlst, candidate, wl, ns, cpdsuggest, NULL, NULL); + ns = testsug(wlst, candidate.c_str(), candidate.size(), ns, cpdsuggest, NULL, NULL); if (ns == -1) return -1; } if ((*(loc + 1) != '|') && (*(loc + 1) != '\0')) { candidate[i] = *(loc + 1); - ns = testsug(wlst, candidate, wl, ns, cpdsuggest, NULL, NULL); + ns = testsug(wlst, candidate.c_str(), candidate.size(), ns, cpdsuggest, NULL, NULL); if (ns == -1) return -1; } @@ -713,19 +629,17 @@ int SuggestMgr::badcharkey_utf(char** wlst, int wl, int ns, int cpdsuggest) { - w_char tmpc; - w_char candidate_utf[MAXSWL]; - char candidate[MAXSWUTF8L]; - memcpy(candidate_utf, word, wl * sizeof(w_char)); + std::string candidate; + std::vector<w_char> candidate_utf(word, word + wl); // swap out each char one by one and try all the tryme // chars in its place to see if that makes a good word for (int i = 0; i < wl; i++) { - tmpc = candidate_utf[i]; + w_char tmpc = candidate_utf[i]; // check with uppercase letters - mkallcap_utf(candidate_utf + i, 1, langnum); - if (!w_char_eq(tmpc, candidate_utf[i])) { - u16_u8(candidate, MAXSWUTF8L, candidate_utf, wl); - ns = testsug(wlst, candidate, strlen(candidate), ns, cpdsuggest, NULL, + candidate_utf[i] = upper_utf(candidate_utf[i], 1); + if (tmpc != candidate_utf[i]) { + u16_u8(candidate, candidate_utf); + ns = testsug(wlst, candidate.c_str(), candidate.size(), ns, cpdsuggest, NULL, NULL); if (ns == -1) return -1; @@ -735,28 +649,28 @@ int SuggestMgr::badcharkey_utf(char** wlst, if (!ckey) continue; w_char* loc = ckey_utf; - while ((loc < (ckey_utf + ckeyl)) && !w_char_eq(*loc, tmpc)) + while ((loc < (ckey_utf + ckeyl)) && *loc != tmpc) loc++; while (loc < (ckey_utf + ckeyl)) { - if ((loc > ckey_utf) && !w_char_eq(*(loc - 1), W_VLINE)) { + if ((loc > ckey_utf) && *(loc - 1) != W_VLINE) { candidate_utf[i] = *(loc - 1); - u16_u8(candidate, MAXSWUTF8L, candidate_utf, wl); - ns = testsug(wlst, candidate, strlen(candidate), ns, cpdsuggest, NULL, + u16_u8(candidate, candidate_utf); + ns = testsug(wlst, candidate.c_str(), candidate.size(), ns, cpdsuggest, NULL, NULL); if (ns == -1) return -1; } - if (((loc + 1) < (ckey_utf + ckeyl)) && !w_char_eq(*(loc + 1), W_VLINE)) { + if (((loc + 1) < (ckey_utf + ckeyl)) && (*(loc + 1) != W_VLINE)) { candidate_utf[i] = *(loc + 1); - u16_u8(candidate, MAXSWUTF8L, candidate_utf, wl); - ns = testsug(wlst, candidate, strlen(candidate), ns, cpdsuggest, NULL, + u16_u8(candidate, candidate_utf); + ns = testsug(wlst, candidate.c_str(), candidate.size(), ns, cpdsuggest, NULL, NULL); if (ns == -1) return -1; } do { loc++; - } while ((loc < (ckey_utf + ckeyl)) && !w_char_eq(*loc, tmpc)); + } while ((loc < (ckey_utf + ckeyl)) && *loc != tmpc); } candidate_utf[i] = tmpc; } @@ -765,26 +679,23 @@ int SuggestMgr::badcharkey_utf(char** wlst, // error is wrong char in place of correct one int SuggestMgr::badchar(char** wlst, const char* word, int ns, int cpdsuggest) { - char tmpc; - char candidate[MAXSWUTF8L]; + std::string candidate(word); clock_t timelimit = clock(); int timer = MINTIMER; - int wl = strlen(word); - strcpy(candidate, word); // swap out each char one by one and try all the tryme // chars in its place to see if that makes a good word for (int j = 0; j < ctryl; j++) { - for (int i = wl - 1; i >= 0; i--) { - tmpc = candidate[i]; + for (std::string::reverse_iterator aI = candidate.rbegin(), aEnd = candidate.rend(); aI != aEnd; ++aI) { + char tmpc = *aI; if (ctry[j] == tmpc) continue; - candidate[i] = ctry[j]; - ns = testsug(wlst, candidate, wl, ns, cpdsuggest, &timer, &timelimit); + *aI = ctry[j]; + ns = testsug(wlst, candidate.c_str(), candidate.size(), ns, cpdsuggest, &timer, &timelimit); if (ns == -1) return -1; if (!timer) return ns; - candidate[i] = tmpc; + *aI = tmpc; } } return ns; @@ -796,22 +707,20 @@ int SuggestMgr::badchar_utf(char** wlst, int wl, int ns, int cpdsuggest) { - w_char tmpc; - w_char candidate_utf[MAXSWL]; - char candidate[MAXSWUTF8L]; + std::vector<w_char> candidate_utf(word, word + wl); + std::string candidate; clock_t timelimit = clock(); int timer = MINTIMER; - memcpy(candidate_utf, word, wl * sizeof(w_char)); // swap out each char one by one and try all the tryme // chars in its place to see if that makes a good word for (int j = 0; j < ctryl; j++) { for (int i = wl - 1; i >= 0; i--) { - tmpc = candidate_utf[i]; - if (w_char_eq(tmpc, ctry_utf[j])) + w_char tmpc = candidate_utf[i]; + if (tmpc == ctry_utf[j]) continue; candidate_utf[i] = ctry_utf[j]; - u16_u8(candidate, MAXSWUTF8L, candidate_utf, wl); - ns = testsug(wlst, candidate, strlen(candidate), ns, cpdsuggest, &timer, + u16_u8(candidate, candidate_utf); + ns = testsug(wlst, candidate.c_str(), candidate.size(), ns, cpdsuggest, &timer, &timelimit); if (ns == -1) return -1; @@ -829,24 +738,20 @@ int SuggestMgr::extrachar_utf(char** wlst, int wl, int ns, int cpdsuggest) { - char candidate[MAXSWUTF8L]; - w_char candidate_utf[MAXSWL]; - w_char* p; - w_char tmpc = W_VLINE; // not used value, only for VCC warning message - if (wl < 2) + std::vector<w_char> candidate_utf(word, word + wl); + if (candidate_utf.size() < 2) return ns; // try omitting one char of word at a time - memcpy(candidate_utf, word, wl * sizeof(w_char)); - for (p = candidate_utf + wl - 1; p >= candidate_utf; p--) { - w_char tmpc2 = *p; - if (p < candidate_utf + wl - 1) - *p = tmpc; - u16_u8(candidate, MAXSWUTF8L, candidate_utf, wl - 1); - ns = - testsug(wlst, candidate, strlen(candidate), ns, cpdsuggest, NULL, NULL); + for (size_t i = 0; i < candidate_utf.size(); ++i) { + size_t index = candidate_utf.size() - 1 - i; + w_char tmpc = candidate_utf[index]; + candidate_utf.erase(candidate_utf.begin() + index); + std::string candidate; + u16_u8(candidate, candidate_utf); + ns = testsug(wlst, candidate.c_str(), candidate.size(), ns, cpdsuggest, NULL, NULL); if (ns == -1) return -1; - tmpc = tmpc2; + candidate_utf.insert(candidate_utf.begin() + index, tmpc); } return ns; } @@ -856,21 +761,18 @@ int SuggestMgr::extrachar(char** wlst, const char* word, int ns, int cpdsuggest) { - char tmpc = '\0'; - char candidate[MAXSWUTF8L]; - char* p; - int wl = strlen(word); - if (wl < 2) + std::string candidate(word); + if (candidate.size() < 2) return ns; // try omitting one char of word at a time - strcpy(candidate, word); - for (p = candidate + wl - 1; p >= candidate; p--) { - char tmpc2 = *p; - *p = tmpc; - ns = testsug(wlst, candidate, wl - 1, ns, cpdsuggest, NULL, NULL); + for (size_t i = 0; i < candidate.size(); ++i) { + size_t index = candidate.size() - 1 - i; + char tmpc = candidate[index]; + candidate.erase(candidate.begin() + index); + ns = testsug(wlst, candidate.c_str(), candidate.size(), ns, cpdsuggest, NULL, NULL); if (ns == -1) return -1; - tmpc = tmpc2; + candidate.insert(candidate.begin() + index, tmpc); } return ns; } @@ -880,23 +782,22 @@ int SuggestMgr::forgotchar(char** wlst, const char* word, int ns, int cpdsuggest) { - char candidate[MAXSWUTF8L + 4]; - char* p; + std::string candidate(word); clock_t timelimit = clock(); int timer = MINTIMER; - int wl = strlen(word); + // try inserting a tryme character before every letter (and the null // terminator) - for (int i = 0; i < ctryl; i++) { - strcpy(candidate, word); - for (p = candidate + wl; p >= candidate; p--) { - *(p + 1) = *p; - *p = ctry[i]; - ns = testsug(wlst, candidate, wl + 1, ns, cpdsuggest, &timer, &timelimit); + for (int k = 0; k < ctryl; ++k) { + for (size_t i = 0; i <= candidate.size(); ++i) { + size_t index = candidate.size() - i; + candidate.insert(candidate.begin() + index, ctry[k]); + ns = testsug(wlst, candidate.c_str(), candidate.size(), ns, cpdsuggest, &timer, &timelimit); if (ns == -1) return -1; if (!timer) return ns; + candidate.erase(candidate.begin() + index); } } return ns; @@ -908,25 +809,25 @@ int SuggestMgr::forgotchar_utf(char** wlst, int wl, int ns, int cpdsuggest) { - w_char candidate_utf[MAXSWL + 1]; - char candidate[MAXSWUTF8L + 4]; - w_char* p; + std::vector<w_char> candidate_utf(word, word + wl); clock_t timelimit = clock(); int timer = MINTIMER; + // try inserting a tryme character at the end of the word and before every // letter - for (int i = 0; i < ctryl; i++) { - memcpy(candidate_utf, word, wl * sizeof(w_char)); - for (p = candidate_utf + wl; p >= candidate_utf; p--) { - *(p + 1) = *p; - *p = ctry_utf[i]; - u16_u8(candidate, MAXSWUTF8L, candidate_utf, wl + 1); - ns = testsug(wlst, candidate, strlen(candidate), ns, cpdsuggest, &timer, + for (int k = 0; k < ctryl; ++k) { + for (size_t i = 0; i <= candidate_utf.size(); ++i) { + size_t index = candidate_utf.size() - i; + candidate_utf.insert(candidate_utf.begin() + index, ctry_utf[k]); + std::string candidate; + u16_u8(candidate, candidate_utf); + ns = testsug(wlst, candidate.c_str(), candidate.size(), ns, cpdsuggest, &timer, &timelimit); if (ns == -1) return -1; if (!timer) return ns; + candidate_utf.erase(candidate_utf.begin() + index); } } return ns; @@ -937,8 +838,6 @@ int SuggestMgr::twowords(char** wlst, const char* word, int ns, int cpdsuggest) { - char candidate[MAXSWUTF8L]; - char* p; int c1, c2; int forbidden = 0; int cwrd; @@ -950,10 +849,12 @@ int SuggestMgr::twowords(char** wlst, if (langnum == LANG_hu) forbidden = check_forbidden(word, wl); + char* candidate = (char*)malloc(wl + 2); strcpy(candidate + 1, word); + // split the string into two pieces after every char // if both pieces are good words make them a suggestion - for (p = candidate + 1; p[1] != '\0'; p++) { + for (char* p = candidate + 1; p[1] != '\0'; p++) { p[-1] = *p; // go to end of the UTF-8 character while (utf8 && ((p[1] & 0xc0) == 0x80)) { @@ -988,12 +889,16 @@ int SuggestMgr::twowords(char** wlst, if (ns < maxSug) { if (cwrd) { wlst[ns] = mystrdup(candidate); - if (wlst[ns] == NULL) + if (wlst[ns] == NULL) { + free(candidate); return -1; + } ns++; } - } else + } else { + free(candidate); return ns; + } // add two word suggestion with dash, if TRY string contains // "a" or "-" // NOTE: cwrd doesn't modified for REP twoword sugg. @@ -1009,16 +914,21 @@ int SuggestMgr::twowords(char** wlst, if (ns < maxSug) { if (cwrd) { wlst[ns] = mystrdup(candidate); - if (wlst[ns] == NULL) + if (wlst[ns] == NULL) { + free(candidate); return -1; + } ns++; } - } else + } else { + free(candidate); return ns; + } } } } } + free(candidate); return ns; } @@ -1027,42 +937,40 @@ int SuggestMgr::swapchar(char** wlst, const char* word, int ns, int cpdsuggest) { - char candidate[MAXSWUTF8L]; - char* p; - char tmpc; - int wl = strlen(word); + std::string candidate(word); + if (candidate.size() < 2) + return ns; + // try swapping adjacent chars one by one - strcpy(candidate, word); - for (p = candidate; p[1] != 0; p++) { - tmpc = *p; - *p = p[1]; - p[1] = tmpc; - ns = testsug(wlst, candidate, wl, ns, cpdsuggest, NULL, NULL); + for (size_t i = 0; i < candidate.size() - 1; ++i) { + std::swap(candidate[i], candidate[i+1]); + ns = testsug(wlst, candidate.c_str(), candidate.size(), ns, cpdsuggest, NULL, NULL); if (ns == -1) return -1; - p[1] = *p; - *p = tmpc; + std::swap(candidate[i], candidate[i+1]); } + // try double swaps for short words // ahev -> have, owudl -> would - if (wl == 4 || wl == 5) { + if (candidate.size() == 4 || candidate.size() == 5) { candidate[0] = word[1]; candidate[1] = word[0]; candidate[2] = word[2]; - candidate[wl - 2] = word[wl - 1]; - candidate[wl - 1] = word[wl - 2]; - ns = testsug(wlst, candidate, wl, ns, cpdsuggest, NULL, NULL); + candidate[candidate.size() - 2] = word[candidate.size() - 1]; + candidate[candidate.size() - 1] = word[candidate.size() - 2]; + ns = testsug(wlst, candidate.c_str(), candidate.size(), ns, cpdsuggest, NULL, NULL); if (ns == -1) return -1; - if (wl == 5) { + if (candidate.size() == 5) { candidate[0] = word[0]; candidate[1] = word[2]; candidate[2] = word[1]; - ns = testsug(wlst, candidate, wl, ns, cpdsuggest, NULL, NULL); + ns = testsug(wlst, candidate.c_str(), candidate.size(), ns, cpdsuggest, NULL, NULL); if (ns == -1) return -1; } } + return ns; } @@ -1072,44 +980,39 @@ int SuggestMgr::swapchar_utf(char** wlst, int wl, int ns, int cpdsuggest) { - w_char candidate_utf[MAXSWL]; - char candidate[MAXSWUTF8L]; - w_char* p; - w_char tmpc; - int len = 0; + std::vector<w_char> candidate_utf(word, word + wl); + if (candidate_utf.size() < 2) + return ns; + + std::string candidate; // try swapping adjacent chars one by one - memcpy(candidate_utf, word, wl * sizeof(w_char)); - for (p = candidate_utf; p < (candidate_utf + wl - 1); p++) { - tmpc = *p; - *p = p[1]; - p[1] = tmpc; - u16_u8(candidate, MAXSWUTF8L, candidate_utf, wl); - if (len == 0) - len = strlen(candidate); - ns = testsug(wlst, candidate, len, ns, cpdsuggest, NULL, NULL); + for (size_t i = 0; i < candidate_utf.size() - 1; ++i) { + std::swap(candidate_utf[i], candidate_utf[i+1]); + u16_u8(candidate, candidate_utf); + ns = testsug(wlst, candidate.c_str(), candidate.size(), ns, cpdsuggest, NULL, NULL); if (ns == -1) return -1; - p[1] = *p; - *p = tmpc; + std::swap(candidate_utf[i], candidate_utf[i+1]); } + // try double swaps for short words // ahev -> have, owudl -> would, suodn -> sound - if (wl == 4 || wl == 5) { + if (candidate_utf.size() == 4 || candidate_utf.size() == 5) { candidate_utf[0] = word[1]; candidate_utf[1] = word[0]; candidate_utf[2] = word[2]; - candidate_utf[wl - 2] = word[wl - 1]; - candidate_utf[wl - 1] = word[wl - 2]; - u16_u8(candidate, MAXSWUTF8L, candidate_utf, wl); - ns = testsug(wlst, candidate, len, ns, cpdsuggest, NULL, NULL); + candidate_utf[candidate_utf.size() - 2] = word[candidate_utf.size() - 1]; + candidate_utf[candidate_utf.size() - 1] = word[candidate_utf.size() - 2]; + u16_u8(candidate, candidate_utf); + ns = testsug(wlst, candidate.c_str(), candidate.size(), ns, cpdsuggest, NULL, NULL); if (ns == -1) return -1; - if (wl == 5) { + if (candidate_utf.size() == 5) { candidate_utf[0] = word[0]; candidate_utf[1] = word[2]; candidate_utf[2] = word[1]; - u16_u8(candidate, MAXSWUTF8L, candidate_utf, wl); - ns = testsug(wlst, candidate, len, ns, cpdsuggest, NULL, NULL); + u16_u8(candidate, candidate_utf); + ns = testsug(wlst, candidate.c_str(), candidate.size(), ns, cpdsuggest, NULL, NULL); if (ns == -1) return -1; } @@ -1122,24 +1025,16 @@ int SuggestMgr::longswapchar(char** wlst, const char* word, int ns, int cpdsuggest) { - char candidate[MAXSWUTF8L]; - char* p; - char* q; - char tmpc; - int wl = strlen(word); + std::string candidate(word); // try swapping not adjacent chars one by one - strcpy(candidate, word); - for (p = candidate; *p != 0; p++) { - for (q = candidate; *q != 0; q++) { - if (abs((int)(p - q)) > 1) { - tmpc = *p; - *p = *q; - *q = tmpc; - ns = testsug(wlst, candidate, wl, ns, cpdsuggest, NULL, NULL); + for (std::string::iterator p = candidate.begin(); p < candidate.end(); ++p) { + for (std::string::iterator q = candidate.begin(); q < candidate.end(); ++q) { + if (abs(std::distance(q, p)) > 1) { + std::swap(*p, *q); + ns = testsug(wlst, candidate.c_str(), candidate.size(), ns, cpdsuggest, NULL, NULL); if (ns == -1) return -1; - *q = *p; - *p = tmpc; + std::swap(*p, *q); } } } @@ -1152,26 +1047,19 @@ int SuggestMgr::longswapchar_utf(char** wlst, int wl, int ns, int cpdsuggest) { - w_char candidate_utf[MAXSWL]; - char candidate[MAXSWUTF8L]; - w_char* p; - w_char* q; - w_char tmpc; + std::vector<w_char> candidate_utf(word, word + wl); // try swapping not adjacent chars - memcpy(candidate_utf, word, wl * sizeof(w_char)); - for (p = candidate_utf; p < (candidate_utf + wl); p++) { - for (q = candidate_utf; q < (candidate_utf + wl); q++) { - if (abs((int)(p - q)) > 1) { - tmpc = *p; - *p = *q; - *q = tmpc; - u16_u8(candidate, MAXSWUTF8L, candidate_utf, wl); - ns = testsug(wlst, candidate, strlen(candidate), ns, cpdsuggest, NULL, + for (std::vector<w_char>::iterator p = candidate_utf.begin(); p < candidate_utf.end(); ++p) { + for (std::vector<w_char>::iterator q = candidate_utf.begin(); q < candidate_utf.end(); ++q) { + if (abs(std::distance(q, p)) > 1) { + std::swap(*p, *q); + std::string candidate; + u16_u8(candidate, candidate_utf); + ns = testsug(wlst, candidate.c_str(), candidate.size(), ns, cpdsuggest, NULL, NULL); if (ns == -1) return -1; - *q = *p; - *p = tmpc; + std::swap(*p, *q); } } } @@ -1183,40 +1071,35 @@ int SuggestMgr::movechar(char** wlst, const char* word, int ns, int cpdsuggest) { - char candidate[MAXSWUTF8L]; - char* p; - char* q; - char tmpc; + std::string candidate(word); + if (candidate.size() < 2) + return ns; - int wl = strlen(word); // try moving a char - strcpy(candidate, word); - for (p = candidate; *p != 0; p++) { - for (q = p + 1; (*q != 0) && ((q - p) < 10); q++) { - tmpc = *(q - 1); - *(q - 1) = *q; - *q = tmpc; - if ((q - p) < 2) + for (std::string::iterator p = candidate.begin(); p < candidate.end(); ++p) { + for (std::string::iterator q = p + 1; q < candidate.end() && std::distance(p, q) < 10; ++q) { + std::swap(*q, *(q - 1)); + if (std::distance(p, q) < 2) continue; // omit swap char - ns = testsug(wlst, candidate, wl, ns, cpdsuggest, NULL, NULL); + ns = testsug(wlst, candidate.c_str(), candidate.size(), ns, cpdsuggest, NULL, NULL); if (ns == -1) return -1; } - strcpy(candidate, word); + std::copy(word, word + candidate.size(), candidate.begin()); } - for (p = candidate + wl - 1; p > candidate; p--) { - for (q = p - 1; (q >= candidate) && ((p - q) < 10); q--) { - tmpc = *(q + 1); - *(q + 1) = *q; - *q = tmpc; - if ((p - q) < 2) + + for (std::string::reverse_iterator p = candidate.rbegin(), pEnd = candidate.rend() - 1; p != pEnd; ++p) { + for (std::string::reverse_iterator q = p + 1, qEnd = candidate.rend(); q != qEnd && std::distance(p, q) < 10; ++q) { + std::swap(*q, *(q - 1)); + if (std::distance(p, q) < 2) continue; // omit swap char - ns = testsug(wlst, candidate, wl, ns, cpdsuggest, NULL, NULL); + ns = testsug(wlst, candidate.c_str(), candidate.size(), ns, cpdsuggest, NULL, NULL); if (ns == -1) return -1; } - strcpy(candidate, word); + std::copy(word, word + candidate.size(), candidate.begin()); } + return ns; } @@ -1226,49 +1109,47 @@ int SuggestMgr::movechar_utf(char** wlst, int wl, int ns, int cpdsuggest) { - w_char candidate_utf[MAXSWL]; - char candidate[MAXSWUTF8L]; - w_char* p; - w_char* q; - w_char tmpc; + std::vector<w_char> candidate_utf(word, word + wl); + if (candidate_utf.size() < 2) + return ns; + // try moving a char - memcpy(candidate_utf, word, wl * sizeof(w_char)); - for (p = candidate_utf; p < (candidate_utf + wl); p++) { - for (q = p + 1; (q < (candidate_utf + wl)) && ((q - p) < 10); q++) { - tmpc = *(q - 1); - *(q - 1) = *q; - *q = tmpc; - if ((q - p) < 2) + for (std::vector<w_char>::iterator p = candidate_utf.begin(); p < candidate_utf.end(); ++p) { + for (std::vector<w_char>::iterator q = p + 1; q < candidate_utf.end() && std::distance(p, q) < 10; ++q) { + std::swap(*q, *(q - 1)); + if (std::distance(p, q) < 2) continue; // omit swap char - u16_u8(candidate, MAXSWUTF8L, candidate_utf, wl); - ns = testsug(wlst, candidate, strlen(candidate), ns, cpdsuggest, NULL, + std::string candidate; + u16_u8(candidate, candidate_utf); + ns = testsug(wlst, candidate.c_str(), candidate.size(), ns, cpdsuggest, NULL, NULL); if (ns == -1) return -1; } - memcpy(candidate_utf, word, wl * sizeof(w_char)); + std::copy(word, word + candidate_utf.size(), candidate_utf.begin()); } - for (p = candidate_utf + wl - 1; p > candidate_utf; p--) { - for (q = p - 1; (q >= candidate_utf) && ((p - q) < 10); q--) { - tmpc = *(q + 1); - *(q + 1) = *q; - *q = tmpc; - if ((p - q) < 2) + + for (std::vector<w_char>::iterator p = candidate_utf.begin() + candidate_utf.size() - 1; p > candidate_utf.begin(); --p) { + for (std::vector<w_char>::iterator q = p - 1; q >= candidate_utf.begin() && std::distance(q, p) < 10; --q) { + std::swap(*q, *(q + 1)); + if (std::distance(q, p) < 2) continue; // omit swap char - u16_u8(candidate, MAXSWUTF8L, candidate_utf, wl); - ns = testsug(wlst, candidate, strlen(candidate), ns, cpdsuggest, NULL, + std::string candidate; + u16_u8(candidate, candidate_utf); + ns = testsug(wlst, candidate.c_str(), candidate.size(), ns, cpdsuggest, NULL, NULL); if (ns == -1) return -1; } - memcpy(candidate_utf, word, wl * sizeof(w_char)); + std::copy(word, word + candidate_utf.size(), candidate_utf.begin()); } + return ns; } // generate a set of suggestions for very poorly spelled words int SuggestMgr::ngsuggest(char** wlst, - char* w, + const char* w, int ns, HashMgr** pHMgr, int md) { @@ -1295,7 +1176,6 @@ int SuggestMgr::ngsuggest(char** wlst, int low = NGRAM_LOWERING; std::string w2; - char f[MAXSWUTF8L]; const char* word = w; // word reversing wrapper for complex prefixes @@ -1308,10 +1188,9 @@ int SuggestMgr::ngsuggest(char** wlst, word = w2.c_str(); } - char mw[MAXSWUTF8L]; - w_char u8[MAXSWL]; + std::vector<w_char> u8; int nc = strlen(word); - int n = (utf8) ? u8_u16(u8, MAXSWL, word) : nc; + int n = (utf8) ? u8_u16(u8, word) : nc; // set character based ngram suggestion for words with non-BMP Unicode // characters @@ -1325,21 +1204,20 @@ int SuggestMgr::ngsuggest(char** wlst, struct hentry* hp = NULL; int col = -1; phonetable* ph = (pAMgr) ? pAMgr->get_phonetable() : NULL; - char target[MAXSWUTF8L]; + std::string target; std::string candidate; if (ph) { if (utf8) { std::vector<w_char> _w; - int _wl = u8_u16(_w, word); - mkallcap_utf(_w, _wl, langnum); + u8_u16(_w, word); + mkallcap_utf(_w, langnum); u16_u8(candidate, _w); } else { candidate.assign(word); if (!nonbmp) mkallcap(candidate, csconv); } - phonet(candidate.c_str(), target, nc, - *ph); // XXX phonet() is 8-bit (nc, not n) + target = phonet(candidate, *ph); // XXX phonet() is 8-bit (nc, not n) } FLAG forbiddenword = pAMgr ? pAMgr->get_forbiddenword() : FLAG_NULL; @@ -1361,27 +1239,27 @@ int SuggestMgr::ngsuggest(char** wlst, leftcommonsubstring(word, HENTRY_WORD(hp)); // check special pronounciation + std::string f; if ((hp->var & H_OPT_PHON) && copy_field(f, HENTRY_DATA(hp), MORPH_PHON)) { int sc2 = ngram(3, word, f, NGRAM_LONGER_WORSE + low) + - +leftcommonsubstring(word, f); + +leftcommonsubstring(word, f.c_str()); if (sc2 > sc) sc = sc2; } int scphon = -20000; if (ph && (sc > 2) && (abs(n - (int)hp->clen) <= 3)) { - char target2[MAXSWUTF8L]; if (utf8) { std::vector<w_char> _w; - int _wl = u8_u16(_w, HENTRY_WORD(hp)); - mkallcap_utf(_w, _wl, langnum); + u8_u16(_w, HENTRY_WORD(hp)); + mkallcap_utf(_w, langnum); u16_u8(candidate, _w); } else { candidate.assign(HENTRY_WORD(hp)); mkallcap(candidate, csconv); } - phonet(candidate.c_str(), target2, -1, *ph); + std::string target2 = phonet(candidate, *ph); scphon = 2 * ngram(3, target, target2, NGRAM_LONGER_WORSE); } @@ -1415,14 +1293,17 @@ int SuggestMgr::ngsuggest(char** wlst, int thresh = 0; for (int sp = 1; sp < 4; sp++) { if (utf8) { - for (int k = sp; k < n; k += 4) - *((unsigned short*)u8 + k) = '*'; - u16_u8(mw, MAXSWUTF8L, u8, n); + for (int k = sp; k < n; k += 4) { + u8[k].l = '*'; + u8[k].h = 0; + } + std::string mw; + u16_u8(mw, u8); thresh = thresh + ngram(n, word, mw, NGRAM_ANY_MISMATCH + low); } else { - strcpy(mw, word); + std::string mw(word); for (int k = sp; k < n; k += 4) - *(mw + k) = '*'; + mw[k] = '*'; thresh = thresh + ngram(n, word, mw, NGRAM_ANY_MISMATCH + low); } } @@ -1454,11 +1335,14 @@ int SuggestMgr::ngsuggest(char** wlst, for (i = 0; i < MAX_ROOTS; i++) { if (roots[i]) { struct hentry* rp = roots[i]; + + std::string f; + const char *field = NULL; + if ((rp->var & H_OPT_PHON) && copy_field(f, HENTRY_DATA(rp), MORPH_PHON)) + field = f.c_str(); int nw = pAMgr->expand_rootword( glst, MAX_WORDS, HENTRY_WORD(rp), rp->blen, rp->astr, rp->alen, word, - nc, - ((rp->var & H_OPT_PHON) ? copy_field(f, HENTRY_DATA(rp), MORPH_PHON) - : NULL)); + nc, field); for (int k = 0; k < nw; k++) { sc = ngram(n, word, glst[k].word, NGRAM_ANY_MISMATCH + low) + @@ -1524,7 +1408,7 @@ int SuggestMgr::ngsuggest(char** wlst, if (utf8) { std::vector<w_char> _w; len = u8_u16(_w, guess[i]); - mkallsmall_utf(_w, len, langnum); + mkallsmall_utf(_w, langnum); u16_u8(gl, _w); } else { gl.assign(guess[i]); @@ -1578,7 +1462,7 @@ int SuggestMgr::ngsuggest(char** wlst, if (utf8) { std::vector<w_char> _w; len = u8_u16(_w, rootsphon[i]); - mkallsmall_utf(_w, len, langnum); + mkallsmall_utf(_w, langnum); u16_u8(gl, _w); } else { gl.assign(rootsphon[i]); @@ -1707,7 +1591,8 @@ int SuggestMgr::checkword(const char* word, if (pAMgr) { if (cpdsuggest == 1) { if (pAMgr->get_compound()) { - rv = pAMgr->compound_check(word, len, 0, 0, 100, 0, NULL, 0, 1, + struct hentry* rwords[100]; // buffer for COMPOUND pattern checking + rv = pAMgr->compound_check(word, len, 0, 0, 100, 0, NULL, (hentry**)&rwords, 0, 1, 0); // EXT if (rv && (!(rv2 = pAMgr->lookup(word)) || !rv2->astr || @@ -1790,51 +1675,6 @@ int SuggestMgr::check_forbidden(const char* word, int len) { return 0; } -#ifdef HUNSPELL_EXPERIMENTAL -// suggest possible stems -int SuggestMgr::suggest_pos_stems(char*** slst, const char* w, int nsug) { - char** wlst; - - struct hentry* rv = NULL; - - char w2[MAXSWUTF8L]; - const char* word = w; - - // word reversing wrapper for complex prefixes - if (complexprefixes) { - strcpy(w2, w); - if (utf8) - reverseword_utf(w2); - else - reverseword(w2); - word = w2; - } - - int wl = strlen(word); - - if (*slst) { - wlst = *slst; - } else { - wlst = (char**)calloc(maxSug, sizeof(char*)); - if (wlst == NULL) - return -1; - } - - rv = pAMgr->suffix_check(word, wl, 0, NULL, wlst, maxSug, &nsug); - - // delete dash from end of word - if (nsug > 0) { - for (int j = 0; j < nsug; j++) { - if (wlst[j][strlen(wlst[j]) - 1] == '-') - wlst[j][strlen(wlst[j]) - 1] = '\0'; - } - } - - *slst = wlst; - return nsug; -} -#endif // END OF HUNSPELL_EXPERIMENTAL CODE - char* SuggestMgr::suggest_morph(const char* w) { char result[MAXLNLEN]; char* r = (char*)result; @@ -1887,33 +1727,15 @@ char* SuggestMgr::suggest_morph(const char* w) { free(st); } - if (pAMgr->get_compound() && (*result == '\0')) - pAMgr->compound_check_morph(word, strlen(word), 0, 0, 100, 0, NULL, 0, &r, + if (pAMgr->get_compound() && (*result == '\0')) { + struct hentry* rwords[100]; // buffer for COMPOUND pattern checking + pAMgr->compound_check_morph(word, strlen(word), 0, 0, 100, 0, NULL, (hentry**)&rwords, 0, &r, NULL); + } return (*result) ? mystrdup(line_uniq(result, MSEP_REC)) : NULL; } -#ifdef HUNSPELL_EXPERIMENTAL -char* SuggestMgr::suggest_morph_for_spelling_error(const char* word) { - char* p = NULL; - char** wlst = (char**)calloc(maxSug, sizeof(char*)); - if (!**wlst) - return NULL; - // we will use only the first suggestion - for (int i = 0; i < maxSug - 1; i++) - wlst[i] = ""; - int ns = suggest(&wlst, word, maxSug - 1, NULL); - if (ns == maxSug) { - p = suggest_morph(wlst[maxSug - 1]); - free(wlst[maxSug - 1]); - } - if (wlst) - free(wlst); - return p; -} -#endif // END OF HUNSPELL_EXPERIMENTAL CODE - /* affixation */ char* SuggestMgr::suggest_hentry_gen(hentry* rv, const char* pattern) { char result[MAXLNLEN]; @@ -1973,23 +1795,24 @@ char* SuggestMgr::suggest_gen(char** desc, int n, const char* pattern) { if (n == 0 || !pAMgr) return NULL; - char result[MAXLNLEN]; - char result2[MAXLNLEN]; + std::string result2; std::string newpattern; - *result2 = '\0'; struct hentry* rv = NULL; // search affixed forms with and without derivational suffixes while (1) { for (int k = 0; k < n; k++) { - *result = '\0'; + std::string result; + // add compound word parts (except the last one) char* s = (char*)desc[k]; char* part = strstr(s, MORPH_PART); if (part) { char* nextpart = strstr(part + 1, MORPH_PART); while (nextpart) { - copy_field(result + strlen(result), part, MORPH_PART); + std::string field; + copy_field(field, part, MORPH_PART); + result.append(field); part = nextpart; nextpart = strstr(part + 1, MORPH_PART); } @@ -2030,16 +1853,14 @@ char* SuggestMgr::suggest_gen(char** desc, int n, const char* pattern) { free(sg); sg = NULL; for (int j = 0; j < genl; j++) { + result2.push_back(MSEP_REC); + result2.append(result); if (strstr(pl[i], MORPH_SURF_PFX)) { - int r2l = strlen(result2); - result2[r2l] = MSEP_REC; - strcpy(result2 + r2l + 1, result); - copy_field(result2 + strlen(result2), pl[i], MORPH_SURF_PFX); - mystrcat(result2, gen[j], MAXLNLEN); - } else { - sprintf(result2 + strlen(result2), "%c%s%s", MSEP_REC, result, - gen[j]); + std::string field; + copy_field(field, pl[i], MORPH_SURF_PFX); + result2.append(field); } + result2.append(gen[j]); } freelist(&gen, genl); } @@ -2050,14 +1871,14 @@ char* SuggestMgr::suggest_gen(char** desc, int n, const char* pattern) { freelist(&pl, pln); } - if (*result2 || !strstr(pattern, MORPH_DERI_SFX)) + if (!result2.empty() || !strstr(pattern, MORPH_DERI_SFX)) break; newpattern.assign(pattern); mystrrep(newpattern, MORPH_DERI_SFX, MORPH_TERM_SFX); pattern = newpattern.c_str(); } - return (*result2 ? mystrdup(result2) : NULL); + return (!result2.empty() ? mystrdup(result2.c_str()) : NULL); } // generate an n-gram score comparing s1 and s2 @@ -2080,7 +1901,7 @@ int SuggestMgr::ngram(int n, return 0; // lowering dictionary word if (opt & NGRAM_LOWERING) - mkallsmall_utf(su2, l2, langnum); + mkallsmall_utf(su2, langnum); for (int j = 1; j <= n; j++) { ns = 0; for (int i = 0; i <= (l1 - j); i++) { @@ -2147,25 +1968,20 @@ int SuggestMgr::ngram(int n, // length of the left common substring of s1 and (decapitalised) s2 int SuggestMgr::leftcommonsubstring(const char* s1, const char* s2) { if (utf8) { - w_char su1[MAXSWL]; - w_char su2[MAXSWL]; - su1[0].l = su2[0].l = su1[0].h = su2[0].h = 0; + std::vector<w_char> su1; + std::vector<w_char> su2; + int l1 = u8_u16(su1, s1); + int l2 = u8_u16(su2, s2); // decapitalize dictionary word if (complexprefixes) { - int l1 = u8_u16(su1, MAXSWL, s1); - int l2 = u8_u16(su2, MAXSWL, s2); - if (*((short*)su1 + l1 - 1) == *((short*)su2 + l2 - 1)) + if (su1[l1 - 1] == su2[l2 - 1]) return 1; } else { - int i; - u8_u16(su1, 1, s1); - u8_u16(su2, 1, s2); - unsigned short idx = (su2->h << 8) + su2->l; - unsigned short otheridx = (su1->h << 8) + su1->l; + unsigned short idx = su2.empty() ? 0 : (su2[0].h << 8) + su2[0].l; + unsigned short otheridx = su1.empty() ? 0 : (su1[0].h << 8) + su1[0].l; if (otheridx != idx && (otheridx != unicodetolower(idx, langnum))) return 0; - int l1 = u8_u16(su1, MAXSWL, s1); - int l2 = u8_u16(su2, MAXSWL, s2); + int i; for (i = 1; (i < l1) && (i < l2) && (su1[i].l == su2[i].l) && (su1[i].h == su2[i].h); i++) @@ -2176,9 +1992,9 @@ int SuggestMgr::leftcommonsubstring(const char* s1, const char* s2) { if (complexprefixes) { int l1 = strlen(s1); int l2 = strlen(s2); - if (*(s2 + l1 - 1) == *(s2 + l2 - 1)) + if (l1 <= l2 && s2[l1 - 1] == s2[l2 - 1]) return 1; - } else { + } else if (csconv) { const char* olds = s1; // decapitalise dictionary word if ((*s1 != *s2) && (*s1 != csconv[((unsigned char)*s2)].clower)) @@ -2201,22 +2017,22 @@ int SuggestMgr::commoncharacterpositions(const char* s1, int diffpos[2]; *is_swap = 0; if (utf8) { - w_char su1[MAXSWL]; - w_char su2[MAXSWL]; - int l1 = u8_u16(su1, MAXSWL, s1); - int l2 = u8_u16(su2, MAXSWL, s2); + std::vector<w_char> su1; + std::vector<w_char> su2; + int l1 = u8_u16(su1, s1); + int l2 = u8_u16(su2, s2); if (l1 <= 0 || l2 <= 0) return 0; // decapitalize dictionary word if (complexprefixes) { - mkallsmall_utf(su2 + l2 - 1, 1, langnum); + su2[l2 - 1] = lower_utf(su2[l2 - 1], langnum); } else { - mkallsmall_utf(su2, 1, langnum); + su2[0] = lower_utf(su2[0], langnum); } for (int i = 0; (i < l1) && (i < l2); i++) { - if (((short*)su1)[i] == ((short*)su2)[i]) { + if (su1[i] == su2[i]) { num++; } else { if (diff < 2) @@ -2225,8 +2041,8 @@ int SuggestMgr::commoncharacterpositions(const char* s1, } } if ((diff == 2) && (l1 == l2) && - (((short*)su1)[diffpos[0]] == ((short*)su2)[diffpos[1]]) && - (((short*)su1)[diffpos[1]] == ((short*)su2)[diffpos[0]])) + (su1[diffpos[0]] == su2[diffpos[1]]) && + (su1[diffpos[1]] == su2[diffpos[0]])) *is_swap = 1; } else { size_t i; @@ -2257,8 +2073,8 @@ int SuggestMgr::commoncharacterpositions(const char* s1, int SuggestMgr::mystrlen(const char* word) { if (utf8) { - w_char w[MAXSWL]; - return u8_u16(w, MAXSWL, word); + std::vector<w_char> w; + return u8_u16(w, word); } else return strlen(word); } @@ -2297,15 +2113,15 @@ void SuggestMgr::lcs(const char* s, int* l2, char** result) { int n, m; - w_char su[MAXSWL]; - w_char su2[MAXSWL]; + std::vector<w_char> su; + std::vector<w_char> su2; char* b; char* c; int i; int j; if (utf8) { - m = u8_u16(su, MAXSWL, s); - n = u8_u16(su2, MAXSWL, s2); + m = u8_u16(su, s); + n = u8_u16(su2, s2); } else { m = strlen(s); n = strlen(s2); @@ -2326,8 +2142,8 @@ void SuggestMgr::lcs(const char* s, c[j] = 0; for (i = 1; i <= m; i++) { for (j = 1; j <= n; j++) { - if (((utf8) && (*((short*)su + i - 1) == *((short*)su2 + j - 1))) || - ((!utf8) && ((*(s + i - 1)) == (*(s2 + j - 1))))) { + if (((utf8) && (su[i - 1] == su2[j - 1])) || + ((!utf8) && (s[i - 1] == s2[j - 1]))) { c[i * (n + 1) + j] = c[(i - 1) * (n + 1) + j - 1] + 1; b[i * (n + 1) + j] = LCS_UPLEFT; } else if (c[(i - 1) * (n + 1) + j] >= c[i * (n + 1) + j - 1]) { diff --git a/libs/hunspell/src/suggestmgr.hxx b/libs/hunspell/src/suggestmgr.hxx index c8762f81ef..675d98eb8f 100644 --- a/libs/hunspell/src/suggestmgr.hxx +++ b/libs/hunspell/src/suggestmgr.hxx @@ -74,8 +74,6 @@ #ifndef _SUGGESTMGR_HXX_ #define _SUGGESTMGR_HXX_ -#define MAXSWL 100 -#define MAXSWUTF8L (MAXSWL * 4) #define MAX_ROOTS 100 #define MAX_WORDS 100 #define MAX_GUESS 200 @@ -132,7 +130,7 @@ class LIBHUNSPELL_DLL_EXPORTED SuggestMgr { ~SuggestMgr(); int suggest(char*** slst, const char* word, int nsug, int* onlycmpdsug); - int ngsuggest(char** wlst, char* word, int ns, HashMgr** pHMgr, int md); + int ngsuggest(char** wlst, const char* word, int ns, HashMgr** pHMgr, int md); int suggest_auto(char*** slst, const char* word, int nsug); int suggest_stems(char*** slst, const char* word, int nsug); int suggest_pos_stems(char*** slst, const char* word, int nsug); @@ -177,8 +175,7 @@ class LIBHUNSPELL_DLL_EXPORTED SuggestMgr { int mapchars(char**, const char*, int, int); int map_related(const char*, - char*, - int, + std::string&, int, char** wlst, int, diff --git a/libs/hunspell/src/w_char.hxx b/libs/hunspell/src/w_char.hxx index 9de7989f4f..336c454f79 100644 --- a/libs/hunspell/src/w_char.hxx +++ b/libs/hunspell/src/w_char.hxx @@ -42,13 +42,27 @@ #define __WCHARHXX__ #ifndef GCC -typedef struct { +struct w_char { #else -typedef struct __attribute__((packed)) { +struct __attribute__((packed)) w_char { #endif unsigned char l; unsigned char h; -} w_char; + + friend bool operator<(const w_char a, const w_char b) { + unsigned short a_idx = (a.h << 8) + a.l; + unsigned short b_idx = (b.h << 8) + b.l; + return a_idx < b_idx; + } + + friend bool operator==(const w_char a, const w_char b) { + return (((a).l == (b).l) && ((a).h == (b).h)); + } + + friend bool operator!=(const w_char a, const w_char b) { + return !(a == b);; + } +}; // two character arrays struct replentry { diff --git a/plugins/SpellChecker/src/Version.h b/plugins/SpellChecker/src/Version.h index 237d49359b..cf5c959244 100644 --- a/plugins/SpellChecker/src/Version.h +++ b/plugins/SpellChecker/src/Version.h @@ -1,14 +1,14 @@ #define __MAJOR_VERSION 0
#define __MINOR_VERSION 2
#define __RELEASE_NUM 6
-#define __BUILD_NUM 3
+#define __BUILD_NUM 4
#include <stdver.h>
#define __PLUGIN_NAME "Spell checker"
#define __FILENAME "SpellChecker.dll"
#define __DESCRIPTION "Spell checker for the message windows. Uses Hunspell to do the checking."
-#define __AUTHOR "Ricardo Pescuma Domenecci, FREAK_THEMIGHTY"
+#define __AUTHOR "Ricardo Pescuma Domenecci, Wishmaster"
#define __AUTHOREMAIL "pescuma@miranda-im.org"
#define __AUTHORWEB "http://miranda-ng.org/p/SpellChecker/"
#define __COPYRIGHT "© 2006-2010 Ricardo Pescuma Domenecci"
diff --git a/plugins/SpellChecker/src/dictionary.cpp b/plugins/SpellChecker/src/dictionary.cpp index c726a9b92a..371b934d01 100644 --- a/plugins/SpellChecker/src/dictionary.cpp +++ b/plugins/SpellChecker/src/dictionary.cpp @@ -534,9 +534,8 @@ public: if (mir_strcmp(dic_enc, "UTF-8") == 0) { codePage = CP_UTF8; - - int wcs_len; - hwordchars = fromHunspell((char *)hunspell->get_wordchars_utf16(&wcs_len)); + const std::vector<w_char> wordchars_utf16 = hunspell->get_wordchars_utf16(); + hwordchars = fromHunspell((char *)&wordchars_utf16[0]); } else { |