seqdb_src.cpp

ncbi源码

/*
 * ===========================================================================
 * PRODUCTION $Log: seqdb_src.cpp,v $
 * PRODUCTION Revision 1000.1  2004/06/01 18:06:22  gouriano
 * PRODUCTION PRODUCTION: UPGRADED [GCC34_MSVC7] Dev-tree R1.11
 * PRODUCTION
 * ===========================================================================
 */

/*  $Id: seqdb_src.cpp,v 1000.1 2004/06/01 18:06:22 gouriano Exp $
* ===========================================================================
*
*                            PUBLIC DOMAIN NOTICE
*               National Center for Biotechnology Information
*
*  This software/database is a "United States Government Work" under the
*  terms of the United States Copyright Act.  It was written as part of
*  the author's official duties as a United States Government employee and
*  thus cannot be copyrighted.  This software/database is freely available
*  to the public for use. The National Library of Medicine and the U.S.
*  Government have not placed any restriction on its use or reproduction.
*
*  Although all reasonable efforts have been taken to ensure the accuracy
*  and reliability of the software and data, the NLM and the U.S.
*  Government do not and cannot warrant the performance or results that
*  may be obtained by using this software or data. The NLM and the U.S.
*  Government disclaim all warranties, express or implied, including
*  warranties of performance, merchantability or fitness for any particular
*  purpose.
*
*  Please cite the author in any work or product based on this material.
*
* ===========================================================================
*
* Author:  Ilya Dondoshansky
*
*/

/// @file seqdb_src.cpp
/// Implementation of the BlastSeqSrc interface for a C++ BLAST databases API

#include <ncbi_pch.hpp>
#include <objects/seqloc/Seq_id.hpp>
#include <objmgr/util/sequence.hpp>
#include <algo/blast/api/seqdb_src.hpp>
#include <algo/blast/core/blast_util.h>
#include <objtools/readers/seqdb/seqdb.hpp>
#include "blast_setup.hpp"

/** @addtogroup AlgoBlast
 *
 * @{
 */

USING_NCBI_SCOPE;
USING_SCOPE(objects);
USING_SCOPE(blast);

extern "C" {

/* Function prototypes */
static Int4 SeqDbGetMaxLength(void* seqdb_handle, void* ignoreme);
static Int4 SeqDbGetNumSeqs(void* seqdb_handle, void* ignoreme);
static Int8 SeqDbGetTotLen(void* seqdb_handle, void* ignoreme);
static Int4 SeqDbGetAvgLength(void* seqdb_handle, void* ignoreme);
static char* SeqDbGetName(void* seqdb_handle, void* ignoreme);
static char* SeqDbGetDefinition(void* seqdb_handle, void* ignoreme);
static char* SeqDbGetDate(void* seqdb_handle, void* ignoreme);
static Boolean SeqDbGetIsProt(void* seqdb_handle, void* ignoreme);
static ListNode* SeqDbGetSeqLoc(void* seqdb_handle, void* args);
static ListNode* SeqDbGetError(void* seqdb_handle, void* args);

/** Retrieves the length of the longest sequence in the BlastSeqSrc.
 * @param seqdb_handle Pointer to initialized CSeqDB object [in]
 * @param ignoreme Unused by this implementation [in]
 */
static Int4 SeqDbGetMaxLength(void* seqdb_handle, void*)
{
    CSeqDB* seqdb = (CSeqDB*) seqdb_handle;
    return seqdb->GetMaxLength();
}

/** Retrieves the number of sequences in the BlastSeqSrc.
 * @param seqdb_handle Pointer to initialized CSeqDB object [in]
 * @param ignoreme Unused by this implementation [in]
 */
static Int4 SeqDbGetNumSeqs(void* seqdb_handle, void*)
{
    CSeqDB* seqdb = (CSeqDB*) seqdb_handle;
    return seqdb->GetNumSeqs();
}

/** Retrieves the total length of all sequences in the BlastSeqSrc.
 * @param seqdb_handle Pointer to initialized CSeqDB object [in]
 * @param ignoreme Unused by this implementation [in]
 */
static Int8 SeqDbGetTotLen(void* seqdb_handle, void*)
{
    CSeqDB* seqdb = (CSeqDB*) seqdb_handle;
    return seqdb->GetTotalLength();
}

/** Retrieves the average length of sequences in the BlastSeqSrc.
 * @param seqdb_handle Pointer to initialized CSeqDB object [in]
 * @param ignoreme Unused by this implementation [in]
 */
static Int4 SeqDbGetAvgLength(void* seqdb_handle, void* ignoreme)
{
   Int8 total_length = SeqDbGetTotLen(seqdb_handle, ignoreme);
   Int4 num_seqs = MAX(1, SeqDbGetNumSeqs(seqdb_handle, ignoreme));

   return (Int4) (total_length/num_seqs);
}

/** Retrieves the name of the BLAST database.
 * FIXME: RIGHT NOW THERE IS NO GETTER in CSeqDB for database name
 * @param seqdb_handle Pointer to initialized CSeqDB object [in]
 * @param ignoreme Unused by this implementation [in]
 */
static char* SeqDbGetName(void* /*seqdb_handle*/, void*)
{
#if 0 // FIXME: GetName method not implemented in CSeqDb!!!
    CSeqDB* seqdb = (CSeqDB*) seqdb_handle;
    return strdup(seqdb->GetName().c_str());
#else
    return NULL;
#endif
}

/** Retrieves the definition (title) of the BLAST database.
 * @param seqdb_handle Pointer to initialized CSeqDB object [in]
 * @param ignoreme Unused by this implementation [in]
 */
static char* SeqDbGetDefinition(void* seqdb_handle, void*)
{
    CSeqDB* seqdb = (CSeqDB*) seqdb_handle;
    return strdup(seqdb->GetTitle().c_str());
}

/** Retrieves the date of the BLAST database.
 * @param seqdb_handle Pointer to initialized CSeqDB object [in]
 * @param ignoreme Unused by this implementation [in]
 */
static char* SeqDbGetDate(void* seqdb_handle, void*)
{
    CSeqDB* seqdb = (CSeqDB*) seqdb_handle;
    return strdup(seqdb->GetDate().c_str());
}

/** Retrieves the date of the BLAST database.
 * @param seqdb_handle Pointer to initialized CSeqDB object [in]
 * @param ignoreme Unused by this implementation [in]
 */
static Boolean SeqDbGetIsProt(void* seqdb_handle, void*)
{
    CSeqDB* seqdb = (CSeqDB*) seqdb_handle;

    return (seqdb->GetSeqType() == 'p');
}

/** Retrieves the sequence meeting the criteria defined by its second argument.
 * @param seqdb_handle Pointer to initialized CSeqDB object [in]
 * @param args Pointer to GetSeqArg structure [in]
 * @return return codes defined in blast_seqsrc.h
 */
static Int2 SeqDbGetSequence(void* seqdb_handle, void* args)
{
    CSeqDB* seqdb = (CSeqDB*) seqdb_handle;
    GetSeqArg* seqdb_args = (GetSeqArg*) args;
    Int4 oid = -1, len = 0;
    Boolean has_sentinel_byte;
    Boolean buffer_allocated;

    if (!seqdb || !seqdb_args)
        return BLAST_SEQSRC_ERROR;

    oid = seqdb_args->oid;
    Uint1 encoding = seqdb_args->encoding;
    has_sentinel_byte = (encoding == BLASTNA_ENCODING);

    buffer_allocated = 
       (encoding == BLASTNA_ENCODING || encoding == NCBI4NA_ENCODING);

    /* free buffers if necessary */
    if (seqdb_args->seq)
        BlastSequenceBlkClean(seqdb_args->seq);

    const char *buf;
    if (!buffer_allocated) {
        len = seqdb->GetSequence(oid, &buf);
    } else {
        len = seqdb->GetAmbigSeq(oid, &buf, has_sentinel_byte);
    }

    if (len <= 0)
        return BLAST_SEQSRC_ERROR;

    BlastSetUp_SeqBlkNew((Uint1*)buf, len, 0, &seqdb_args->seq, 
                         buffer_allocated);
    
    /* If there is no sentinel byte, and buffer is allocated, i.e. this is
       the traceback stage of a translated search, set "sequence" to the same 
       position as "sequence_start". */
    if (buffer_allocated && !has_sentinel_byte)
       seqdb_args->seq->sequence = seqdb_args->seq->sequence_start;

    seqdb_args->seq->oid = oid;

    return BLAST_SEQSRC_SUCCESS;
}

/** Returns the memory allocated for the sequence buffer to the CSeqDB 
 * interface.
 * @param seqdb_handle Pointer to initialized CSeqDB object [in]
 * @param args Pointer to the GetSeqArgs structure, containing sequence block
 *             with the buffer that needs to be deallocated. [in]
 * @return return codes defined in blast_seqsrc.h
 */
static Int2 SeqDbRetSequence(void* seqdb_handle, void* args)
{
    CSeqDB* seqdb = (CSeqDB*) seqdb_handle;
    GetSeqArg* seqdb_args = (GetSeqArg*) args;

    if (!seqdb || !seqdb_args)
        return BLAST_SEQSRC_ERROR;

    if (seqdb_args->seq->sequence_start_allocated) {
        seqdb->RetSequence((const char**)&seqdb_args->seq->sequence_start);
        seqdb_args->seq->sequence_start_allocated = FALSE;
        seqdb_args->seq->sequence_start = NULL;
    }
    return 0;
}

/** Retrieves the sequence identifier meeting the criteria defined by its 
 * second argument. Currently it is an ordinal id (integer value).
 * Client code is responsible for deallocating the return value. 
 * @param seqdb_handle Pointer to initialized CSeqDB object [in]
 * @param args Pointer to integer indicating ordinal id [in]
 * @return Sequence id structure generated from ASN.1 spec, 
 *         cast to a void pointer.
 */
static ListNode* SeqDbGetSeqId(void* seqdb_handle, void* args)
{
    CSeqDB* seqdb = (CSeqDB*) seqdb_handle;
    Uint4* oid = (Uint4*) args;
    ListNode* seqid_wrap;

    if (!seqdb || !oid)
        return NULL;

    list< CRef<CSeq_id> > seqid_list;
    seqid_list = seqdb->GetSeqIDs(*oid);

    CRef<CSeq_id>* seqid_ref = new CRef<CSeq_id>(seqid_list.front());
    seqid_wrap = ListNodeAddPointer(NULL, BLAST_SEQSRC_CPP_SEQID_REF, 
                                    (void*) seqid_ref);

    return seqid_wrap;
}