bool Tokenizer::getNumber(const QString &str, int i, QString &numb)
{
// Examples of valid numbers are:
// 1234
// +1234
// -1234
// -1234.99
// .99
// +.99
// -1234.
numb.clear();
bool first = true;
bool dotFound = false;
for (; i < str.size(); ++i) {
if (first && isSign(str, i)) {
numb.append(str[i]);
} else if (isDigit(str, i)) {
numb.append(str[i]);
} else if (!dotFound && isDot(str, i)) {
numb.append(str[i]);
dotFound = true;
} else {
break;
}
first = false;
}
return numb.size() > 0;
}
void parseNumber() {
skipWhiteSpace();
while (!eof() && !isWhiteSpace() &&
(isSign() || isDot() || isDigit())) {
put();
next();
}
}
bool Number::IsValid(char character) {
switch ((States) _currentState) {
case States::Start:
return isMinus(character) || isDigit(character);
case States::Negation:
return isDigit(character);
case States::RationalPercent:
return isDot(character) || isExponentialSymbol(character);
case States::Number:
return isDigit(character) || isDot(character) || isExponentialSymbol(character);
case States::Decimal:
return isDigit(character) || isExponentialSymbol(character);
case States::Exponential_1:
return isDigit(character) || isPlus(character) || isMinus(character);
case States::Exponential_2:
return isDigit(character);
case States::Closing:
return false;
}
}
/**
* Searches in the "name" directory for the "find" file,
* if file is found and the user wanted to find that filetype
* the path is added to a list that will be printed at the end.
* Also puts all directories found in a list for it to be
* searched.
*/
bool searchForFile(char* name, char *find, int t, struct Node* list) {
DIR *dir;
struct dirent *ent;
struct stat f_info;
bool foundFile = false;
if (openDir(&dir, name, &ent)) {
do {
if ((lstat(ent->d_name, &f_info)) < 0) {
fprintf(stderr, "lstat error: ");
perror(ent->d_name);
} else {
if (t == 0 && compareName(ent->d_name, find)) {
foundFile = true;
} else if (checkDir(t, f_info, ent->d_name, find)) {
foundFile = true;
} else if (checkReg(t, f_info, ent->d_name, find)) {
foundFile = true;
} else if (checkLink(t, f_info, ent->d_name, find)) {
foundFile = true;
}
if (checkDirAndRights(f_info) && !isDot(ent->d_name)) {
int l = strlen(name) + strlen(ent->d_name);
char *str = calloc(1, sizeof(char[l + 2]));
strcpy(str, name);
strcat(str, "/");
strcat(str, ent->d_name);
insert(list, str);
}
}
} while ((ent = readdir(dir)));
}
closedir(dir);
return foundFile;
}
int main(void){
FILE *f, *g, *h, *out;
char c;
char keys[NUM][4];
char word[4];
int keyarr[NUM];
int len = 0, n, spam, a, b, dot, i;
int msglen, wordnum, numnum, weirdnum, spel;
double r;
f = fopen("data.txt", "r");
g = fopen("data2.txt", "r");
h = fopen("keywords.txt", "r");
out = fopen("out.txt", "w");
// ucitaj rjecnik keyworda
while (fscanf(h, "%c%c%c%c %d %d %lg ", keys[len], keys[len]+1, keys[len]+2,
keys[len]+3, &a, &b, &r) == 7) len++;
// napisi header u out.txt
fprintf(out,
"spam broj_rijeci duljina nepismenost udio_brojeva udio_nealfanumerickih ");
for (i = 0; i < len; i++){
fprintf(out, "%c%c%c%c ", keys[i][0], keys[i][1], keys[i][2], keys[i][3]);
}
fprintf(out, "\n");
for (n = 0; n < LINES; n++){
// prvo iz neobradenog fajla izvadi broj charactera itd..
fscanf(f, "%c", &c);
if (c == 'h') {
// zapamti jeli poruka ham ...
spam = 0;
}
else if (c == 's'){
// ... ili spam
spam = 1;
}
else printf("nesto nevalja: linija %d\n", n);
// ako na pocetku linije ne pise ni spam ni ham, nesto je krivo
while (isAlpha(c)) fscanf(f, "%c", &c);
// ucitaj cijelu rijec (spam/ham) do kraja
dot = 0;
msglen = 0;
numnum = 0;
weirdnum = 0;
spel = 0;
wordnum = 0;
while (1){
fscanf(f, "%c", &c);
if (c == '\n') break;
msglen++; // broj znakova
if (isNum(c)) numnum++; // broji udio brojeva
if (isWeird(c)) weirdnum++; // broji udio nealfanumerickih
if ((isNum(c) || isAlpha(c)) && (dot == 1)) spel = 1; // nepismenost
if (isDot(c)) dot = 1;
else dot = 0;
}
// zatim iz obradenog izvadi kljucne rijeci
fscanf(g, "%d", &a); // ucitaj jeli spam ili ham
if (a != spam) printf("nisu konzistentni na liniji: %d\n", n);
fscanf(g, "%c", &c); // ucitaj razmak
for (i = 0; i < len; i++){
keyarr[i] = 0; // ocisti niz koji cuva postojanje keyworda
}
while (c != '\n'){
for (i = 0; i < 4; i++){
fscanf(g, "%c", word+i);
}
wordnum++; // broj rijeci u poruci
b = exsist(word, keys, len); // kljucna rijec
if (b >= 0) keyarr[b] = 1;
fscanf(g, "%c", &c); // ucitaj razmak
}
// zapisi feature vektor u out.txt
fprintf(out, "%d %d %d %d %lg %lg ", spam, wordnum, msglen, spel,
(double)numnum/msglen, (double)weirdnum/msglen);
for (i = 0; i < len; i++){
fprintf(out, "%d ", keyarr[i]);
}
fprintf(out, "\n");
}
return 0;
}
short CRConData::CheckProgressInConsole(UINT nCursorLine)
{
if (!isValid(true, nWidth*(nCursorLine+1)))
return -1;
const wchar_t* pszCurLine = pConChar + (nWidth * nCursorLine);
// Обработка прогресса NeroCMD и пр. консольных программ (если курсор находится в видимой области)
//Плагин Update
//"Downloading Far 99%"
//NeroCMD
//"012% ########.................................................................."
//ChkDsk
//"Completed: 25%"
//Rar
// ... Vista x86\Vista x86.7z 6%
//aria2c
//[#1 SIZE:0B/9.1MiB(0%) CN:1 SPD:1.2KiBs ETA:2h1m11s]
int nIdx = 0;
bool bAllowDot = false;
short nProgress = -1;
const wchar_t *szPercentEng = L" percent";
const wchar_t *szComplEng = L"Completed:";
static wchar_t szPercentRus[16] = {}, szComplRus[16] = {};
static int nPercentEngLen = lstrlen(szPercentEng), nComplEngLen = lstrlen(szComplEng);
static int nPercentRusLen, nComplRusLen;
if (szPercentRus[0] == 0)
{
szPercentRus[0] = L' ';
TODO("Хорошо бы и другие национальные названия обрабатывать, брать из настройки");
lstrcpy(szPercentRus,L"процент");
lstrcpy(szComplRus,L"Завершено:");
nPercentRusLen = lstrlen(szPercentRus);
nComplRusLen = lstrlen(szComplEng);
}
// Сначала проверим, может цифры идут в начале строки (лидирующие пробелы)?
if (pszCurLine[nIdx] == L' ' && isDigit(pszCurLine[nIdx+1]))
nIdx++; // один лидирующий пробел перед цифрой
else if (pszCurLine[nIdx] == L' ' && pszCurLine[nIdx+1] == L' ' && isDigit(pszCurLine[nIdx+2]))
nIdx += 2; // два лидирующих пробела перед цифрой
else if (!isDigit(pszCurLine[nIdx]))
{
// Строка начинается НЕ с цифры. Может начинается одним из известных префиксов (ChkDsk)?
if (!wcsncmp(pszCurLine, szComplRus, nComplRusLen))
{
nIdx += nComplRusLen;
if (pszCurLine[nIdx] == L' ') nIdx++;
if (pszCurLine[nIdx] == L' ') nIdx++;
bAllowDot = true;
}
else if (!wcsncmp(pszCurLine, szComplEng, nComplEngLen))
{
nIdx += nComplEngLen;
if (pszCurLine[nIdx] == L' ') nIdx++;
if (pszCurLine[nIdx] == L' ') nIdx++;
bAllowDot = true;
}
else if (!wcsncmp(pszCurLine, L"[#", 2))
{
const wchar_t* p = wcsstr(pszCurLine, L"%) ");
while ((p > pszCurLine) && (p[-1] != L'('))
p--;
if (p > pszCurLine)
nIdx = p - pszCurLine;
}
// Известных префиксов не найдено, проверяем, может процент есть в конце строки?
if (!nIdx)
{
//TODO("Не работает с одной цифрой");
// Creating archive T:\From_Work\VMWare\VMWare.part006.rar
// ... Vista x86\Vista x86.7z 6%
int i = nWidth - 1;
// Откусить trailing spaces
while ((i > 3) && (pszCurLine[i] == L' '))
i--;
// Теперь, если дошли до '%' и перед ним - цифра
if (i >= 3 && pszCurLine[i] == L'%' && isDigit(pszCurLine[i-1]))
{
//i -= 2;
i--;
int j = i, k = -1;
while (j > 0 && isDigit(pszCurLine[j-1]))
j--;
// Может быть что-то типа "Progress 25.15%"
if (((i - j) <= 2) && (j >= 2) && isDot(pszCurLine[j-1]))
{
k = j - 1;
while (k > 0 && isDigit(pszCurLine[k-1]))
k--;
}
if (k >= 0)
{
if (((j - k) <= 3) // 2 цифры + точка
|| (((j - k) <= 4) && (pszCurLine[k] == L'1'))) // "100.0%"
{
nIdx = i = k;
bAllowDot = true;
}
}
else
{
if (((j - i) <= 2) // 2 цифры + точка
|| (((j - i) <= 3) && (pszCurLine[j] == L'1'))) // "100%"
{
nIdx = i = j;
}
}
#if 0
// Две цифры перед '%'?
if (isDigit(pszCurLine[i-1]))
i--;
// Три цифры допускается только для '100%'
if (pszCurLine[i-1] == L'1' && !isDigit(pszCurLine[i-2]))
{
nIdx = i - 1;
}
// final check
else if (!isDigit(pszCurLine[i-1]))
{
nIdx = i;
}
#endif
// Может ошибочно детектировать прогресс, если его ввести в prompt
// Допустим, что если в строке есть символ '>' - то это не прогресс
while (i>=0)
{
if (pszCurLine[i] == L'>')
{
nIdx = 0;
break;
}
i--;
}
}
}
}
// Менять nProgress только если нашли проценты в строке с курсором
if (isDigit(pszCurLine[nIdx]))
{
if (isDigit(pszCurLine[nIdx+1]) && isDigit(pszCurLine[nIdx+2])
&& (pszCurLine[nIdx+3]==L'%' || (bAllowDot && isDot(pszCurLine[nIdx+3]))
|| !wcsncmp(pszCurLine+nIdx+3,szPercentEng,nPercentEngLen)
|| !wcsncmp(pszCurLine+nIdx+3,szPercentRus,nPercentRusLen)))
{
nProgress = 100*(pszCurLine[nIdx] - L'0') + 10*(pszCurLine[nIdx+1] - L'0') + (pszCurLine[nIdx+2] - L'0');
}
else if (isDigit(pszCurLine[nIdx+1])
&& (pszCurLine[nIdx+2]==L'%' || (bAllowDot && isDot(pszCurLine[nIdx+2]))
|| !wcsncmp(pszCurLine+nIdx+2,szPercentEng,nPercentEngLen)
|| !wcsncmp(pszCurLine+nIdx+2,szPercentRus,nPercentRusLen)))
{
nProgress = 10*(pszCurLine[nIdx] - L'0') + (pszCurLine[nIdx+1] - L'0');
}
else if (pszCurLine[nIdx+1]==L'%' || (bAllowDot && isDot(pszCurLine[nIdx+1]))
|| !wcsncmp(pszCurLine+nIdx+1,szPercentEng,nPercentEngLen)
|| !wcsncmp(pszCurLine+nIdx+1,szPercentRus,nPercentRusLen))
{
nProgress = (pszCurLine[nIdx] - L'0');
}
}
if (nProgress != -1)
{
mp_RCon->setLastConsoleProgress(nProgress, true); // его обновляем всегда
}
else
{
DWORD nDelta = GetTickCount() - mp_RCon->m_Progress.LastConProgrTick;
if (nDelta < CONSOLEPROGRESSTIMEOUT) // Если таймаут предыдущего значения еще не наступил
nProgress = mp_RCon->m_Progress.ConsoleProgress; // возъмем предыдущее значение
mp_RCon->setLastConsoleProgress(-1, false); // его обновляем всегда
}
return nProgress;
}
bool Number::CoreValidate(char character) {
if (_lexeme.length() == 0)
_lexeme.reserve(20);
switch ((States) _currentState) {
case States::Start: {
_lexeme.append(1, character);
if (isMinus(character)) _currentState = (int) States::Negation;
if (isZero(character)) _currentState = (int) States::RationalPercent;
if (isDigitOneToNine(character)) _currentState = (int) States::Number;
break;
}
case States::Negation: {
_lexeme.append(1, character);
if (isZero(character)) _currentState = (int) States::RationalPercent;
if (isDigitOneToNine(character)) _currentState = (int) States::Number;
break;
}
case States::RationalPercent: {
_lexeme.append(1, character);
if (isDot(character)) _currentState = (int) States::Decimal;
if (isExponentialSymbol(character)) _currentState = (int) States::Exponential_1;
break;
}
case States::Number: {
_lexeme.append(1, character);
if (isDigit(character)) _currentState = (int) States::Number;
if (isDot(character)) _currentState = (int) States::Decimal;
if (isExponentialSymbol(character)) _currentState = (int) States::Exponential_1;
break;
}
case States::Decimal: {
_lexeme.append(1, character);
if (isDigit(character)) _currentState = (int) States::Decimal;
if (isExponentialSymbol(character)) _currentState = (int) States::Exponential_1;
break;
}
case States::Exponential_1: {
_lexeme.append(1, character);
_currentState = (int) States::Exponential_2;
break;
}
case States::Exponential_2: {
_lexeme.append(1, character);
_currentState = (int) States::Exponential_2;
break;
}
case States::Closing:
return false;
}
return true;
}
bool Number::BeginWithCharacter(char character) {
return isMinus(character) || isDigit(character) || isDot(character);
}
void Tokenizer::shift()
{
mpCurrent.reset();
if (mBlockComment)
{
skipEOL();
}
else
{
skipWhiteSpaces();
if (mpInput->eof())
return;
}
mpCurrent.reset(new Token);
mpCurrent->setLine(line());
mpCurrent->setBeginColumn(column());
if (mBlockComment)
{
consumeCStyleBlockComment();
return;
}
auto ch = mpInput->get();
if (isLetter(ch) || isUnderscore(ch))
{
consumeIdentifier(ch);
}
else if (isBiwiseOperatorSymbol(ch))
{
mpCurrent->setType(Token::TYPE_BITWISE_OPERATOR);
absorbed(ch);
mpCurrent->addChar(ch);
mpCurrent->setEndColumn(column());
}
else if (isDot(ch) && consumeDot(ch))
{
// nothing
}
else if ((isDecimalDigit(ch) || isDot(ch) || isSign(ch)) && consumeNumber(ch))
{
// nothing
}
else if (isQuotationMark(ch))
{
if (!consumeString(ch))
shift();
}
else if (isCStyleInitialCommentChar(ch))
{
consumeComment(ch);
}
else if (isArrowSymbol(ch) && consumeArrow(ch))
{
// nothing
}
else if (isBracket(ch))
{
mpCurrent->setType(Token::TYPE_BRACKET);
absorbed(ch);
mpCurrent->addChar(ch);
mpCurrent->setEndColumn(column());
}
else if (isAngleBracket(ch))
{
mpCurrent->setType(Token::TYPE_ANGLE_BRACKET);
absorbed(ch);
mpCurrent->addChar(ch);
mpCurrent->setEndColumn(column());
}
else if (isDelimiter(ch))
{
mpCurrent->setType(Token::TYPE_DELIMITER);
absorbed(ch);
mpCurrent->addChar(ch);
mpCurrent->setEndColumn(column());
}
else if (isOperator(ch) && consumeEqualOperator(ch))
{
// nothing
}
else if (isOperator(ch))
{
mpCurrent->setType(Token::TYPE_OPERATOR);
absorbed(ch);
mpCurrent->addChar(ch);
mpCurrent->setEndColumn(column());
}
else if (isAsterisk(ch))
{
mpCurrent->setType(Token::TYPE_ASTERISK);
absorbed(ch);
mpCurrent->addChar(ch);
mpCurrent->setEndColumn(column());
}
else
{
mpCurrent.reset();
}
}
bool Tokenizer::consumeNumber(std::streambuf::int_type ch)
{
int count = 0;
if (isSign(ch))
{
bool number = false;
while (!eof())
{
++count;
auto nch = mpInput->get();
if (isWhitespace(nch))
continue;
number = isDecimalDigit(nch);
break;
}
if (count)
mpInput->clear();
for (int i = 0; i < count; ++i)
mpInput->unget();
if (!number)
return false;
}
absorbed(ch);
mpCurrent->addChar(ch);
for (int i = 0; i < count; ++i)
{
auto nch = mpInput->get();
absorbed(nch);
mpCurrent->addChar(nch);
}
bool bDot = false;
if (isDot(ch))
{
bDot = true;
mpCurrent->setType(Token::TYPE_REAL_LITERAL);
}
else
{
mpCurrent->setType(Token::TYPE_INTEGER_LITERAL);
if (!eof() && isZero(ch))
{
ch = mpInput->get();
if (isHexIndicator(ch))
{
if (eof())
{
mpMessageCollector->addMessage(
Message::SEVERITY_ERROR, Message::t_invalidIntegerLiteral,
mpSourceId, mpCurrent->line(), mpCurrent->beginColumn());
mpCurrent.reset();
return false;
}
mpCurrent->addChar(ch);
absorbed(ch);
while (!eof())
{
ch = mpInput->get();
if (isNonHexicalLetter(ch) || isUnderscore(ch) || isDot(ch))
{
mpMessageCollector->addMessage(Message::SEVERITY_ERROR,
Message::t_invalidIntegerLiteral,
mpSourceId, mpCurrent->line(), mpCurrent->beginColumn());
mpCurrent.reset();
return false;
}
if (!isHexicalDigit(ch))
{
mpInput->clear();
mpInput->unget();
break;
}
absorbed(ch);
mpCurrent->addChar(ch);
}
mpCurrent->setEndColumn(column());
return true;
}
for (;;)
{
if ( isLetter(ch)
|| isDot(ch)
|| isNonOctalDecimalDigit(ch)
|| isUnderscore(ch))
{
mpMessageCollector->addMessage(
Message::SEVERITY_ERROR, Message::t_invalidIntegerLiteral,
mpSourceId, mpCurrent->line(), mpCurrent->beginColumn());
mpCurrent.reset();
return false;
}
if (!isOctalDigit(ch))
{
mpInput->clear();
mpInput->unget();
break;
}
absorbed(ch);
mpCurrent->addChar(ch);
if (eof())
{
break;
}
ch = mpInput->get();
}
mpCurrent->setEndColumn(column());
return true;
}
}
bool bExponent = false;
while (!eof())
{
ch = mpInput->get();
if (isExponent(ch))
{
if (bExponent)
{
mpMessageCollector->addMessage(
Message::SEVERITY_ERROR, Message::t_invalidIntegerLiteral,
mpSourceId, mpCurrent->line(), mpCurrent->beginColumn());
mpCurrent.reset();
return false;
}
bExponent = true;
mpCurrent->setType(Token::TYPE_REAL_LITERAL);
absorbed(ch);
mpCurrent->addChar(ch);
ch = mpInput->get();
if (isSign(ch))
{
mpCurrent->setType(Token::TYPE_REAL_LITERAL);
absorbed(ch);
mpCurrent->addChar(ch);
}
else
{
mpInput->clear();
mpInput->unget();
}
continue;
}
if (isLetter(ch) || isUnderscore(ch))
{
mpMessageCollector->addMessage(
Message::SEVERITY_ERROR, Message::t_invalidIntegerLiteral,
mpSourceId, mpCurrent->line(), mpCurrent->beginColumn());
mpCurrent.reset();
return false;
}
if (isDot(ch))
{
if (bExponent || bDot)
{
mpMessageCollector->addMessage(
Message::SEVERITY_ERROR, Message::t_invalidIntegerLiteral,
mpSourceId, mpCurrent->line(), mpCurrent->beginColumn());
mpCurrent.reset();
return false;
}
bDot = true;
mpCurrent->setType(Token::TYPE_REAL_LITERAL);
absorbed(ch);
mpCurrent->addChar(ch);
continue;
}
if (!isDecimalDigit(ch))
{
mpInput->clear();
mpInput->unget();
break;
}
absorbed(ch);
mpCurrent->addChar(ch);
}
mpCurrent->setEndColumn(column());
return true;
}
bool isAlNumDot(char c){
return isAlpha(c) || isNum(c) || isDot(c);
}
