gbseq_formatter.cpp

来自「ncbi源码」· C++ 代码 · 共 632 行 · 第 1/2 页

/*
 * ===========================================================================
 * PRODUCTION $Log: gbseq_formatter.cpp,v $
 * PRODUCTION Revision 1000.1  2004/06/01 19:44:34  gouriano
 * PRODUCTION PRODUCTION: UPGRADED [GCC34_MSVC7] Dev-tree R1.4
 * PRODUCTION
 * ===========================================================================
 */

/*  $Id: gbseq_formatter.cpp,v 1000.1 2004/06/01 19:44:34 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:  Aaron Ucko, NCBI
*          Mati Shomrat
*
* File Description:
*   GBseq formatting        
*/
#include <ncbi_pch.hpp>
#include <corelib/ncbistd.hpp>
#include <serial/objostr.hpp>

#include <objects/gbseq/GBSet.hpp>
#include <objects/gbseq/GBSeq.hpp>
#include <objects/gbseq/GBReference.hpp>
#include <objects/gbseq/GBKeyword.hpp>
#include <objects/gbseq/GBSeqid.hpp>
#include <objects/gbseq/GBFeature.hpp>
#include <objects/gbseq/GBInterval.hpp>
#include <objects/gbseq/GBQualifier.hpp>
#include <objects/seq/Seqdesc.hpp>
#include <objmgr/scope.hpp>
#include <objmgr/impl/scope_info.hpp>
#include <objmgr/seqdesc_ci.hpp>
#include <objmgr/util/sequence.hpp>
#include <objmgr/impl/synonyms.hpp>

#include <objtools/format/text_ostream.hpp>
#include <objtools/format/gbseq_formatter.hpp>
#include <objtools/format/items/locus_item.hpp>
#include <objtools/format/items/defline_item.hpp>
#include <objtools/format/items/accession_item.hpp>
#include <objtools/format/items/version_item.hpp>
#include <objtools/format/items/keywords_item.hpp>
#include <objtools/format/items/source_item.hpp>
#include <objtools/format/items/reference_item.hpp>
#include <objtools/format/items/comment_item.hpp>
#include <objtools/format/items/feature_item.hpp>
#include <objtools/format/items/sequence_item.hpp>
#include <objtools/format/items/segment_item.hpp>
#include <objtools/format/items/contig_item.hpp>
#include "utils.hpp"


BEGIN_NCBI_SCOPE
BEGIN_SCOPE(objects)

/////////////////////////////////////////////////////////////////////////////
// static functions

static void s_GBSeqStringCleanup(string& str, bool location = false)
{
    list<string> l;
    NStr::Split(str, " \n\r\t\b", l);
    str = NStr::Join(l, " ");
    if ( location ) {
        str = NStr::Replace(str, ", ", ",");
    }
    str = NStr::TruncateSpaces(str);
}


static void s_GBSeqQualCleanup(string& val)
{
    
    val = NStr::Replace(val, "\"", " ");
    s_GBSeqStringCleanup(val);
    /*
    if ( NStr::EndsWith(val, ".") ) {
        val.erase(val.length() - 1);
    }
    */
}

/////////////////////////////////////////////////////////////////////////////
// Public


// constructor
CGBSeqFormatter::CGBSeqFormatter(void)
{
}

// detructor
CGBSeqFormatter::~CGBSeqFormatter(void) 
{
}


void CGBSeqFormatter::Start(IFlatTextOStream& text_os)
{
    x_WriteFileHeader(text_os);
        
    x_StartWriteGBSet(text_os);
}


void CGBSeqFormatter::StartSection(const CStartSectionItem&, IFlatTextOStream&)
{
    m_GBSeq.Reset(new CGBSeq);
    _ASSERT(m_GBSeq);
}


void CGBSeqFormatter::EndSection(const CEndSectionItem&, IFlatTextOStream& text_os)
{
    x_WriteGBSeq(text_os);

    m_GBSeq.Reset();
    _ASSERT(!m_GBSeq);
}


void CGBSeqFormatter::End(IFlatTextOStream& text_os)
{
    x_EndWriteGBSet(text_os);
}


///////////////////////////////////////////////////////////////////////////
//
// Locus
//


CGBSeq::TStrandedness s_GBSeqStrandedness(CSeq_inst::TStrand strand)
{
    switch ( strand ) {
    case CSeq_inst::eStrand_ss:
        return CGBSeq::eStrandedness_single_stranded;
    case CSeq_inst::eStrand_ds:
        return CGBSeq::eStrandedness_double_stranded;
    case CSeq_inst::eStrand_mixed:
        return CGBSeq::eStrandedness_mixed_stranded;
    case CSeq_inst::eStrand_other:
    case CSeq_inst::eStrand_not_set:
    default:
        break;
    }

    return CGBSeq::eStrandedness_not_set;
}


CGBSeq::TMoltype s_GBSeqMoltype(CMolInfo::TBiomol biomol)
{
    switch ( biomol ) {
    case CMolInfo::eBiomol_unknown:
        return CGBSeq::eMoltype_nucleic_acid;
    case CMolInfo::eBiomol_genomic:
    case CMolInfo::eBiomol_other_genetic:
    case CMolInfo::eBiomol_genomic_mRNA:
        return CGBSeq::eMoltype_dna;
    case CMolInfo::eBiomol_pre_RNA:
    case CMolInfo::eBiomol_cRNA:
    case CMolInfo::eBiomol_transcribed_RNA:
        return CGBSeq::eMoltype_rna;
    case CMolInfo::eBiomol_mRNA:
        return CGBSeq::eMoltype_mrna;
    case CMolInfo::eBiomol_rRNA:
        return CGBSeq::eMoltype_rrna;
    case CMolInfo::eBiomol_tRNA:
        return CGBSeq::eMoltype_trna;
    case CMolInfo::eBiomol_snRNA:
        return CGBSeq::eMoltype_urna;
    case CMolInfo::eBiomol_scRNA:
        return CGBSeq::eMoltype_snrna;
    case CMolInfo::eBiomol_peptide:
        return CGBSeq::eMoltype_peptide;
    case CMolInfo::eBiomol_snoRNA:
        return CGBSeq::eMoltype_snorna;
    default:
        break;
    }
    return CGBSeq::eMoltype_nucleic_acid;
}


CGBSeq::TTopology s_GBSeqTopology(CSeq_inst::TTopology topology)
{
    if ( topology == CSeq_inst::eTopology_circular ) {
        return CGBSeq::eTopology_circular;
    }
    return CGBSeq::eTopology_linear;
}


string s_GetDate(const CBioseq_Handle& bsh, CSeqdesc::E_Choice choice)
{
    _ASSERT(choice == CSeqdesc::e_Update_date  ||
            choice == CSeqdesc::e_Create_date);
    CSeqdesc_CI desc(bsh, choice);
    if ( desc ) {
        string result;
        if ( desc->IsUpdate_date() ) {
            DateToString(desc->GetUpdate_date(), result);
        } else {
            DateToString(desc->GetCreate_date(), result);
        }
        return result;
    }

    return "01-JAN-1900";
}


void CGBSeqFormatter::FormatLocus
(const CLocusItem& locus, 
 IFlatTextOStream&)
{
    _ASSERT(m_GBSeq);
    CBioseqContext& ctx = *locus.GetContext();

    m_GBSeq->SetLocus(locus.GetName());
    m_GBSeq->SetLength(locus.GetLength());
    m_GBSeq->SetStrandedness(s_GBSeqStrandedness(locus.GetStrand()));
    m_GBSeq->SetMoltype(s_GBSeqMoltype(locus.GetBiomol()));
    m_GBSeq->SetTopology(s_GBSeqTopology(locus.GetTopology()));
    m_GBSeq->SetDivision(locus.GetDivision());
    m_GBSeq->SetUpdate_date(s_GetDate(ctx.GetHandle(), CSeqdesc::e_Update_date));
    m_GBSeq->SetCreate_date(s_GetDate(ctx.GetHandle(), CSeqdesc::e_Create_date));
    ITERATE (CBioseq::TId, it, ctx.GetBioseqIds()) {
        m_GBSeq->SetOther_seqids().push_back(CGBSeqid((*it)->AsFastaString()));
    }
}


///////////////////////////////////////////////////////////////////////////
//
// Definition

void CGBSeqFormatter::FormatDefline
(const CDeflineItem& defline,
 IFlatTextOStream&)
{
    _ASSERT(m_GBSeq);
    m_GBSeq->SetDefinition(defline.GetDefline());
    if ( NStr::EndsWith(m_GBSeq->GetDefinition(), ".") ) {
        m_GBSeq->SetDefinition().resize(m_GBSeq->GetDefinition().length() - 1);
    }
}


///////////////////////////////////////////////////////////////////////////
//
// Accession

void CGBSeqFormatter::FormatAccession
(const CAccessionItem& acc, 
 IFlatTextOStream&)
{
    m_GBSeq->SetPrimary_accession(acc.GetAccession());
    ITERATE (CAccessionItem::TExtra_accessions, it, acc.GetExtraAccessions()) {
        m_GBSeq->SetSecondary_accessions().push_back(CGBSecondary_accn(*it));
    }
}


///////////////////////////////////////////////////////////////////////////
//
// Version

void CGBSeqFormatter::FormatVersion
(const CVersionItem& version,
 IFlatTextOStream&)
{
    m_GBSeq->SetAccession_version(version.GetAccession());
}


///////////////////////////////////////////////////////////////////////////
//
// Segment

void CGBSeqFormatter::FormatSegment
(const CSegmentItem& seg,
 IFlatTextOStream&)
{
    CNcbiOstrstream segment_line;

    segment_line << seg.GetNum() << " of " << seg.GetCount();

    m_GBSeq->SetSegment(CNcbiOstrstreamToString(segment_line));
}