// UnirefParser.cpp


#include <cstring>
#include "UnirefParser.hpp"


namespace NGBW {


// UnirefParser


bool UnirefParser::Parse(FastaRecord &record)
{
	record.clear();

	if (m_line == NULL)
		ReadNextLine();

	while (1) {
		if (m_input.Eof())
			return false;

		if (m_line[0] == '>') {
			ParseHeader(record);

			break;
		}

		ReadNextLine();
	}

	while (1) {
		if (m_input.Eof() || m_line[0] == '>')
			break;

		record.sequence.append(m_line);

		ReadNextLine();
	}

	return true;
}

void UnirefParser::ParseHeader(FastaRecord &record)
{
	size_t first = FindFirstOf('_', 1);

	if (first == std::string::npos)
		first = 1;
	else
		first += 1;

	size_t length = std::strlen(m_line);
	size_t last = FindFirstSpace(first + 1);

	if (last == std::string::npos)
		last = length;

	record.identity.assign(m_line + first, last - first);

	if (last < length) {
		first = last + 1;

		const char *sub_string = std::strstr(m_line + first, " n=");

		if (sub_string != NULL) {
			last = sub_string - m_line;

			record.description.assign(m_line + first, last - first);

			first = last + 3;

			sub_string = std::strstr(m_line + first, " Tax=");

			if (sub_string != NULL) {
				first = (sub_string - m_line) + 5;

				sub_string = std::strstr(m_line + first, " RepID=");

				if (sub_string != NULL)
					last = sub_string - m_line;
				else
					last = length;

				record.organism.assign(m_line + first, last - first);
			}
		}
	}

	ReadNextLine();
}


} // namespace NGBW