readCSVRow.cpp

readCSVRow.cpp

std::vector<QStringRef> readCSVRow(const QString& line, const QStringList& separator, const QStringList& escape) {
	std::vector<QStringRef> part;
	
	// newState = delta[currentState][event]

	// rows are states

	// columns are events:
	// 0 = read separator
	// 1 = read escape
	// 2 = read new line
	// 3 = read "normal" character
	// 4 = reached the end of line (eof if line is a file content)
	static const int delta[][5] = {
	    //  ,    "   \n    ?  eof
	    {1, 2, -1, 0, -1}, // 0: parsing (store char)
	    {1, 2, -1, 0, -1}, // 1: parsing (store column)
	    {3, 4, 3, 3, -2},  // 2: quote entered (no-op)
	    {3, 4, 3, 3, -2},  // 3: parsing inside quotes (store char)
	    {1, 3, -1, 0, -1}, // 4: quote exited (no-op)
	                       // -1: end of row, store column, success
	                       // -2: eof inside quotes
	};
	part.clear(); //clear del contenitore dei pezzi in cui scomporre la riga
	if (line.isEmpty()) {
		return part;
	}

	int actualState = 0;
	int pos         = 0;

	// initial invalid value
	int                  event             = -1;
	int                  currentBlockStart = -1;
	int                  blockEnd          = 0;
	static const QString newline           = "\n";
	static const QString empty;

	while (actualState >= 0) {
		if (pos >= line.length())
			event = 4;
		else {
			QStringRef ch = line.midRef(pos, 1);
			pos++;
			if (separator.contains(ch, Qt::CaseInsensitive)){
				event = 0;
			} else if (escape.contains(ch, Qt::CaseInsensitive)) {
				event = 1;
			} else if (ch == newline){
				event = 2;
			} else{
				event = 3;
			}
		}

		actualState = delta[actualState][event];

		switch (actualState) {
		case 4:
			blockEnd = pos - 1;
			break;
		case 0:
		case 3:
			if (currentBlockStart == -1) {
				currentBlockStart = pos - 1;
			}

			break;
		case -1:
		case 1:

			if (!blockEnd) {
				blockEnd = pos - 1;
			}
			if (currentBlockStart == -1) { //a new block has never started, we have two separator in a row
				part.push_back(empty.midRef(0, 0));
				blockEnd = 0;
			} else {
				QStringRef v = line.midRef(currentBlockStart, (blockEnd - currentBlockStart));
				blockEnd     = 0;
				part.push_back(v);
				currentBlockStart = -1;
				//curentBlock.clear();
			}
			break;
		}
	}
	if (actualState == -2){
		throw std::runtime_error("End-of-file found while inside quotes.");
	}

	return part;
}