ConcCommon.h

Go to the documentation of this file.

#ifndef __ConcCommon_H_
#define __ConcCommon_H_

#include "../common/utilit.h"
#include "list"
#include "limits.h"


#include "../GraphanLib/GraphmatFile.h"
#include "../LemmatizerLib/Lemmatizers.h"

#include "../AgramtabLib/EngGramTab.h"
#include "../AgramtabLib/RusGramTab.h"
#include "../AgramtabLib/GerGramTab.h"
#include "../common/DDC_common.h"
#include "../tinyxml/tinyxml.h"




const char globalFieldDelimeter = '\t';
const string PredefinedTableLineTag = "l";
const string ChunkIndexName = "chunk";
const string LeftBigramsIndexName = "left";
const string RightBigramsIndexName = "right";
const string PredefinedFileBreakName = "file";
const string PredefinedTextAreaBreakName = "textarea";

typedef DWORD CTokenNo;

const size_t MaxShortOccurCacheSize = 1000000;



class CShortOccurCache 
{
        struct CDataReference 
        {
                // the start index of subvector in m_OccurrencesBody
                size_t m_VectorStartOffset;
                // the length of subvector in m_OccurrencesBody
                size_t m_VectorLength;


                CDataReference (size_t VectorStartOffset, size_t VectorLength)
                {
                        m_VectorStartOffset = VectorStartOffset;
                        m_VectorLength = VectorLength;
                };
        };
        //map<CTokenNo, CDataReference> m_IndexItemNo2Occurrences;
        vector <CDataReference> m_IndexItemNo2Occurrences;

        vector<CTokenNo>                        m_OccurrencesBody;
public:
        void                            Clear();
        size_t                          AddNewIndexItemNoToCache(const CTokenNo* pStart, const CTokenNo* pEnd );
        const CTokenNo*         GetOccurrencesFromCache(const int CacheId, DWORD& Length) const;
        bool                            CouldContainMore() const;

};

enum HitSortEnum {

        NoSort = 0,
        LessByDate = 1,
        GreaterByDate = 2,
        LessBySize = 3,
        GreaterBySize = 4,
        LessByFreeBiblField = 5,
        GreaterByFreeBiblField = 6,
        LessByRank = 7,
        GreaterByRank = 8,
        LessByLeftContext = 9,
        LessByRightContext = 10,
        HitSortsCount = 11,
};


struct CDDCFilterWithBounds 
{
        HitSortEnum             m_FilterType;
        string                  m_FreeBiblAttribName;

        bool                    m_bRegExp;
        int                             m_LevelStart;
        int                             m_LevelEnd;
        set<int>                m_SatisfiedValues;  

        CDDCFilterWithBounds() 
        {
                m_bRegExp = false;
                m_FilterType = HitSortsCount;
                m_LevelStart = INT_MIN;
                m_LevelEnd = INT_MAX; 
        }

};

struct CHit 
{
        DWORD   m_BreakNo;
        DWORD   m_HighlightOccurrenceEnd;
        int             m_OrderId;
        DWORD   m_FileNo;
        size_t  m_DebugRankNo;

        CHit(DWORD BreakNo) : m_BreakNo(BreakNo), m_OrderId(0) {};

        bool operator < (const CHit& X) const 
        {
                return  m_OrderId < X.m_OrderId;
        };
};



typedef map<size_t, vector<DWORD> > PeriodsDivisionMap;

typedef map<string,CShortOccurCache> CShortOccurCacheMap;

typedef vector<CTokenNo> COccurrBuffer;

const string MorphAnnotationsDelim = "#";
const string MorphAnnotationsDelimRegExp = "[^#]*";

extern bool InitConcordDicts();
extern void FreeConcordDicts();
extern const CLemmatizer* GetLemmatizerByLanguage (MorphLanguageEnum Langua);
extern const CAgramtab* GetGramtabByLanguage (MorphLanguageEnum Langua);
extern void concord_daemon_log(const string&  t);
extern string GetDDCErrorString(DDCErrorEnum ErrorCode);


enum BigramDirectionEnum{
     bdDontUseBigrams,
     bdLeftBigram,
     bdRightBigram
};

#endif