int String::find (char ch, int start, int options) const
{
if (start < 0) start += _len (buf);
if (start < 0) start = 0;
if (start > _len (buf)) start = _len (buf);
if (options & FIND_CASE_INSENSITIVE)
ch = tolower ((unsigned char) ch);
if (options & FIND_REVERSE)
{
for (int i = start - 1; i >= 0; i--)
{
if ((options & FIND_CASE_INSENSITIVE ? tolower ((unsigned char) buf[i]) == ch : buf[i] == ch) &&
((options & FIND_WHOLE_WORD) == 0 || (
isWordBoundary (i) && isWordBoundary (i + 1))))
return i;
}
}
else
{
for (int i = start; i < _len (buf); i++)
{
if ((options & FIND_CASE_INSENSITIVE ? tolower ((unsigned char) buf[i]) == ch : buf[i] == ch) &&
((options & FIND_WHOLE_WORD) == 0 || (
isWordBoundary (i) && isWordBoundary (i + 1))))
return i;
}
}
return -1;
}
void c_SentenceEntry::add(const c_LabelEntry & l)
{
if (l.phon[l.phon.size()-1] != '-' || (l.phon[l.phon.size()-2] == '#' && l.phon[l.phon.size()-1] == '-')) {
if(isWordBoundary(l.phon)) {
if(l.pros.size())
type = isSentenceDelimiter(l.pros);
if(start) {
word.add(l);
word.setPros(l.pros);
word.setFirst(l.first);
start = false;
}
else {
word.setLast(l.first);
word.finish(l);
words.push_back(word);
word = c_WordEntry(sb_sym);
word.setPros(l.pros);
word.setFirst(l.first);
word.add(l);
}
}
else {
word.add(l);
}
}
}
bool LexicalAnalyser::readWord(std::string *word) {
std::string returnWord;
char ch;
bool startedWord = false;
while (readNext(&ch)) {
if (!isWordBoundary(ch)) {
// Append the character
returnWord += ch;
startedWord = true;
} else {
// If the word has started, return the word, else wait for the word to start
if (startedWord) {
*word = returnWord;
return true;
}
}
}
if (startedWord && returnWord.length() != 0) {
*word = returnWord;
return true;
}
return false;
}
bool CRegExp::checkMetaSymbol(EMetaSymbols symb, int &toParse)
{
const String &pattern = *global_pattern;
switch(symb){
case ReAnyChr:
if (toParse >= end) return false;
if (!singleLine && (pattern[toParse] == 0x0A || pattern[toParse] == 0x0B ||
pattern[toParse] == 0x0C || pattern[toParse] == 0x0D ||
pattern[toParse] == 0x85 ||
pattern[toParse] == 0x2028 ||
pattern[toParse] == 0x2029)) return false;
toParse++;
return true;
case ReSoL:
if (multiLine){
bool ok = false;
if (toParse && (pattern[toParse-1] == 0x0A || pattern[toParse-1] == 0x0B ||
pattern[toParse-1] == 0x0C || pattern[toParse-1] == 0x0D ||
pattern[toParse-1] == 0x85 ||
pattern[toParse-1] == 0x2028 ||
pattern[toParse-1] == 0x2029)) ok = true;
return (toParse == 0 || ok);
};
return (toParse == 0);
case ReEoL:
if (multiLine){
bool ok = false; // ???check
if (toParse && toParse < end &&
(pattern[toParse-1] == 0x0A || pattern[toParse-1] == 0x0B ||
pattern[toParse-1] == 0x0C || pattern[toParse-1] == 0x0D ||
pattern[toParse-1] == 0x85 ||
pattern[toParse-1] == 0x2028 ||
pattern[toParse-1] == 0x2029)) ok = true;
return (toParse == end || ok);
};
return (end == toParse);
case ReDigit:
if (toParse >= end || !Character::isDigit(pattern[toParse])) return false;
toParse++;
return true;
case ReNDigit:
if (toParse >= end || Character::isDigit(pattern[toParse])) return false;
toParse++;
return true;
case ReWordSymb:
if (toParse >= end || !(Character::isLetterOrDigit(pattern[toParse])
|| pattern[toParse] == '_')) return false;
toParse++;
return true;
case ReNWordSymb:
if (toParse >= end || Character::isLetterOrDigit(pattern[toParse])
|| pattern[toParse] == '_') return false;
toParse++;
return true;
case ReWSpace:
if (toParse >= end || !Character::isWhitespace(pattern[toParse])) return false;
toParse++;
return true;
case ReNWSpace:
if (toParse >= end || Character::isWhitespace(pattern[toParse])) return false;
toParse++;
return true;
case ReUCase:
if (toParse >= end || !Character::isUpperCase(pattern[toParse])) return false;
toParse++;
return true;
case ReNUCase:
if (toParse >= end || !Character::isLowerCase(pattern[toParse])) return false;
toParse++;
return true;
case ReWBound:
return isWordBoundary(toParse);
case ReNWBound:
return isNWordBoundary(toParse);
case RePreNW:
if (toParse >= end) return true;
return toParse == 0 || !Character::isLetter(pattern[toParse-1]);
#ifdef COLORERMODE
case ReSoScheme:
return (schemeStart == toParse);
case ReStart:
matches->s[0] = toParse;
startChange = true;
return true;
case ReEnd:
matches->e[0] = toParse;
endChange = true;
return true;
#endif
default:
return false;
};
}
bool CRegExp::isNWordBoundary(int &toParse)
{
return !isWordBoundary(toParse);
};
int String::find (char const* str, int start, int options) const
{
static int kmpbuf[256];
#define comp(a,b) (options&FIND_CASE_INSENSITIVE?tolower((unsigned char)a)==tolower((unsigned char)b):(a)==(b))
if (start < 0) start += _len (buf);
if (start < 0) start = 0;
if (start > _len (buf)) start = _len (buf);
if (*str == 0) return start;
int len = (int) strlen (str);
int* kmp = kmpbuf;
if (len > 256)
kmp = new int[len];
kmp[0] = 0;
int result = -1;
if (options & FIND_REVERSE)
{
for (int i = 1; i < len; i++)
{
kmp[i] = kmp[i - 1];
while (kmp[i] && !comp (str[len - 1 - kmp[i]], str[len - 1 - i]))
kmp[i] = kmp[kmp[i] - 1];
if (comp (str[len - 1 - kmp[i]], str[len - 1 - i]))
kmp[i]++;
}
int cur = 0;
for (int i = start - 1; i >= 0; i--)
{
while (cur && !comp (str[len - 1 - cur], buf[_len (buf) - 1 - i]))
cur = kmp[cur - 1];
if (comp (str[len - 1 - cur], buf[_len (buf) - 1 - i]))
cur++;
if (cur == len)
{
if ((options & FIND_WHOLE_WORD) == 0 || (
isWordBoundary (i) && isWordBoundary (i + len)))
{
result = i;
break;
}
}
}
}
else
{
for (int i = 1; i < len; i++)
{
kmp[i] = kmp[i - 1];
while (kmp[i] && !comp (str[kmp[i]], str[i]))
kmp[i] = kmp[kmp[i] - 1];
if (comp (str[kmp[i]], str[i]))
kmp[i]++;
}
int cur = 0;
for (int i = start; i < _len (buf); i++)
{
while (cur && !comp (str[cur], buf[i]))
cur = kmp[cur - 1];
if (comp (str[cur], buf[i]))
cur++;
if (cur == len)
{
if ((options & FIND_WHOLE_WORD) == 0 || (
isWordBoundary (i - len + 1) && isWordBoundary (i + 1)))
{
result = i - len + 1;
break;
}
}
}
}
if (kmp != kmpbuf)
delete[] kmp;
return result;
}
