LineReader.h

Go to the documentation of this file.

/*
 * This program is free software; you can redistribute it and/or modify
 * it under the terms of the GNU General Public License as published by
 * the Free Software Foundation; either version 3 of the License, or
 * (at your option) any later version.
 *
 * Written (W) 2013 Evgeniy Andreev (gsomix)
 */

#ifndef __LINE_READER_H__
#define __LINE_READER_H__

#include <shogun/lib/config.h>

#include <shogun/lib/SGVector.h>
#include <shogun/base/SGObject.h>

namespace shogun
{
class CCircularBuffer;
class CTokenizer;

class CLineReader : public CSGObject
{
public:
    CLineReader();

    CLineReader(FILE* stream, CTokenizer* tokenizer);

    CLineReader(int32_t max_string_length, FILE* stream, CTokenizer* tokenizer);

    virtual ~CLineReader();

    virtual bool has_next();

    virtual void skip_line();

    virtual SGVector<char> read_line();

    void reset();

    void set_tokenizer(CTokenizer* tokenizer);

    virtual const char* get_name() const { return "LineReader"; }

private:
    void init();

    int32_t read(int32_t& bytes_to_skip);

    SGVector<char> read_token(int32_t line_len);

private:
    CCircularBuffer* m_buffer;

    CTokenizer* m_tokenizer;

    FILE* m_stream;

    int32_t m_max_token_length;

    int32_t m_next_token_length;
};

}

#endif /* __FILE_READER_H__ */