validatorp.cpp

来自「ncbi源码」· C++ 代码 · 共 2,068 行 · 第 1/5 页

/*
 * ===========================================================================
 * PRODUCTION $Log: validatorp.cpp,v $
 * PRODUCTION Revision 1000.3  2004/06/01 19:47:39  gouriano
 * PRODUCTION PRODUCTION: UPGRADED [GCC34_MSVC7] Dev-tree R1.52
 * PRODUCTION
 * ===========================================================================
 */

/*  $Id: validatorp.cpp,v 1000.3 2004/06/01 19:47:39 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:  Jonathan Kans, Clifford Clausen, Aaron Ucko, Mati Shomrat, ....
 *
 * File Description:
 *   Implementation of private parts of the validator
 *   .......
 *
 */
#include <ncbi_pch.hpp>
#include <corelib/ncbistd.hpp>
#include <corelib/ncbistr.hpp>
#include <objmgr/object_manager.hpp>

#include "validatorp.hpp"
#include "utilities.hpp"

#include <serial/iterator.hpp>
#include <serial/enumvalues.hpp>

#include <objects/general/Dbtag.hpp>
#include <objects/general/Person_id.hpp>
#include <objects/general/Name_std.hpp>

#include <objects/seqalign/Seq_align.hpp>

#include <objects/seqset/Bioseq_set.hpp>
#include <objects/seqset/Seq_entry.hpp>

#include <objects/seq/Bioseq.hpp>
#include <objects/seq/Seq_annot.hpp>
#include <objects/seq/Seqdesc.hpp>
#include <objects/seq/Seq_descr.hpp>
#include <objects/seq/Pubdesc.hpp>

#include <objects/seqfeat/BioSource.hpp>
#include <objects/seqfeat/OrgMod.hpp>
#include <objects/seqfeat/OrgName.hpp>
#include <objects/seqfeat/Org_ref.hpp>
#include <objects/seqfeat/Seq_feat.hpp>
#include <objects/seqfeat/SubSource.hpp>

#include <objects/seqloc/Seq_loc.hpp>
#include <objects/seqloc/Seq_interval.hpp>
#include <objects/seqloc/Seq_point.hpp>
#include <objects/seqloc/Textseq_id.hpp>

#include <objects/seqres/Seq_graph.hpp>

#include <objects/submit/Seq_submit.hpp>
#include <objects/submit/Submit_block.hpp>

#include <objmgr/util/feature.hpp>
#include <objmgr/util/sequence.hpp>

#include <objmgr/feat_ci.hpp>
#include <objmgr/seq_vector.hpp>
#include <objmgr/scope.hpp>

#include <objects/pub/Pub.hpp>
#include <objects/pub/Pub_equiv.hpp>

#include <objects/biblio/Author.hpp>
#include <objects/biblio/Auth_list.hpp>
#include <objects/biblio/Cit_art.hpp>
#include <objects/biblio/Cit_book.hpp>
#include <objects/biblio/Cit_gen.hpp>
#include <objects/biblio/Cit_jour.hpp>
#include <objects/biblio/Cit_let.hpp>
#include <objects/biblio/Cit_proc.hpp>
#include <objects/biblio/Cit_sub.hpp>
#include <objects/biblio/PubMedId.hpp>
#include <objects/biblio/PubStatus.hpp>
#include <objects/biblio/Title.hpp>
#include <objects/biblio/Imprint.hpp>
#include <objects/biblio/Affil.hpp>

#include <algorithm>

BEGIN_NCBI_SCOPE
BEGIN_SCOPE(objects)
BEGIN_SCOPE(validator)
using namespace sequence;


auto_ptr<CTextFsa> CValidError_imp::m_SourceQualTags;


// =============================================================================
//                            CValidError_imp Public
// =============================================================================

// Constructor
CValidError_imp::CValidError_imp
(CObjectManager& objmgr, 
 CValidError*       errs,
 Uint4              options) :
      m_ObjMgr(&objmgr),
      m_Scope(0),
      m_TSE(0),
      m_ErrRepository(errs),
      m_NonASCII((options & CValidator::eVal_non_ascii) != 0),
      m_SuppressContext((options & CValidator::eVal_no_context) != 0),
      m_ValidateAlignments((options & CValidator::eVal_val_align) != 0),
      m_ValidateExons((options & CValidator::eVal_val_exons) != 0),
      m_SpliceErr((options & CValidator::eVal_splice_err) != 0),
      m_OvlPepErr((options & CValidator::eVal_ovl_pep_err) != 0),
      m_RequireTaxonID((options & CValidator::eVal_need_taxid) != 0),
      m_RequireISOJTA((options & CValidator::eVal_need_isojta) != 0),
      m_ValidateIdSet((options & CValidator::eVal_validate_id_set) != 0),
      m_RemoteFetch((options & CValidator::eVal_remote_fetch) != 0),
      m_PerfBottlenecks((options & CValidator::eVal_perf_bottlenecks) != 0),
      m_IsStandaloneAnnot(false),
      m_NoPubs(false),
      m_NoBioSource(false),
      m_IsGPS(false),
      m_IsGED(false),
      m_IsPDB(false),
      m_IsTPA(false),
      m_IsPatent(false),
      m_IsRefSeq(false),
      m_IsNC(false),
      m_IsNG(false),
      m_IsNM(false),
      m_IsNP(false),
      m_IsNR(false),
      m_IsNS(false),
      m_IsNT(false),
      m_IsNW(false),
      m_IsXR(false),
      m_IsGI(false),
      m_PrgCallback(0),
      m_NumAlign(0),
      m_NumAnnot(0),
      m_NumBioseq(0),
      m_NumBioseq_set(0),
      m_NumDesc(0),
      m_NumDescr(0),
      m_NumFeat(0),
      m_NumGraph(0)
{
    if ( m_SourceQualTags.get() == 0 ) {
        InitializeSourceQualTags();
    }
}


// Destructor
CValidError_imp::~CValidError_imp()
{
}


// Error post methods
void CValidError_imp::PostErr
(EDiagSev sv,
 EErrType et,
 const string&  msg,
 const CSerialObject& obj)
{
    const CSeqdesc* desc = dynamic_cast < const CSeqdesc* > (&obj);
    if (desc != 0) {
        PostErr (sv, et, msg, *desc);
        return;
    }
    const CSeq_feat* feat = dynamic_cast < const CSeq_feat* > (&obj);
    if (feat != 0) {
        PostErr (sv, et, msg, *feat);
        return;
    }
    const CBioseq* seq = dynamic_cast < const CBioseq* > (&obj);
    if (seq != 0) {
        PostErr (sv, et, msg, *seq);
        return;
    }
    const CBioseq_set* set = dynamic_cast < const CBioseq_set* > (&obj);
    if (set != 0) {
        PostErr (sv, et, msg, *set);
        return;
    }
    const CSeq_annot* annot = dynamic_cast < const CSeq_annot* > (&obj);
    if (annot != 0) {
        PostErr (sv, et, msg, *annot);
        return;
    }
    const CSeq_graph* graph = dynamic_cast < const CSeq_graph* > (&obj);
    if (graph != 0) {
        PostErr (sv, et, msg, *graph);
        return;
    }
    const CSeq_align* align = dynamic_cast < const CSeq_align* > (&obj);
    if (align != 0) {
        PostErr (sv, et, msg, *align);
        return;
    }
    const CSeq_entry* entry = dynamic_cast < const CSeq_entry* > (&obj);
    if (entry != 0) {
        PostErr (sv, et, msg, *entry);
        return;
    }
}


void CValidError_imp::PostErr
(EDiagSev sv,
 EErrType et,
 const string&   message,
 TDesc    ds)
{
    // Append Descriptor label
    string msg(message + " DESCRIPTOR: ");
    ds.GetLabel (&msg, CSeqdesc::eBoth);

    m_ErrRepository->AddValidErrItem(new CValidErrItem(sv, et, msg, ds));
}


void CValidError_imp::PostErr
(EDiagSev sv,
 EErrType et,
 const string&   message,
 TFeat    ft)
{
    // Add feature part of label
    string msg(message + " FEATURE: ");
    feature::GetLabel(ft, &msg, feature::eBoth, m_Scope);

    // Add feature location part of label
    string loc_label;
    if (m_SuppressContext) {
        CSeq_loc loc;
        loc.Assign(ft.GetLocation());
        ChangeSeqLocId(&loc, false, m_Scope.GetPointer());
        loc.GetLabel(&loc_label);
    } else {
        ft.GetLocation().GetLabel(&loc_label);
    }
    if (loc_label.size() > 800) {
        loc_label = loc_label.substr(0, 797) + "...";
    }
    if (!loc_label.empty()) {
        loc_label = string("[") + loc_label + "]";
        msg += loc_label;
    }

    // Append label for bioseq of feature location
    if (!m_SuppressContext) {
        try {
            CBioseq_Handle hnd = m_Scope->GetBioseqHandle(ft.GetLocation());
            if( hnd ) {
                CBioseq_Handle::TBioseqCore bc = hnd.GetBioseqCore();
                msg += "[";
                bc->GetLabel(&msg, CBioseq::eBoth);
                msg += "]";
            }
        } catch (...){
        };
    }

    // Append label for product of feature
    loc_label.erase();
    if (ft.IsSetProduct()) {
        if (m_SuppressContext) {
            CSeq_loc loc;
            loc.Assign(ft.GetProduct());
            ChangeSeqLocId(&loc, false, m_Scope.GetPointer());
            loc.GetLabel(&loc_label);
        } else {
            ft.GetProduct().GetLabel(&loc_label);
        }
        if (loc_label.size() > 800) {
            loc_label = loc_label.substr(0, 797) + "...";
        }
        if (!loc_label.empty()) {
            loc_label = string("[") + loc_label + "]";
            msg += loc_label;
        }
    }
    m_ErrRepository->AddValidErrItem(new CValidErrItem(sv, et, msg, ft));
}


void CValidError_imp::PostErr
(EDiagSev sv,
 EErrType et,
 const string&   message,
 TBioseq  sq)
{
    // Append bioseq label
    string msg(message + " BIOSEQ: ");
    if (m_SuppressContext) {
        sq.GetLabel(&msg, CBioseq::eContent, true);
    } else {
        sq.GetLabel(&msg, CBioseq::eBoth, false);
    }
    m_ErrRepository->AddValidErrItem(new CValidErrItem(sv, et, msg, sq));
}


void CValidError_imp::PostErr
(EDiagSev sv,
 EErrType et,
 const string&   message,
 TBioseq  sq,
 TDesc    ds)
{
    // Append Descriptor label
    string msg(message + " DESCRIPTOR: ");
    ds.GetLabel(&msg, CSeqdesc::eBoth);
    PostErr(sv, et, msg, sq);
}


void CValidError_imp::PostErr
(EDiagSev      sv,
 EErrType      et,
 const string& message,
 TSet          set)
{
    // Append Bioseq_set label
    string msg(message + " BIOSEQ-SET: ");
    if (m_SuppressContext) {
        set.GetLabel(&msg, CBioseq_set::eContent);
    } else {
        set.GetLabel(&msg, CBioseq_set::eBoth);
    }
    m_ErrRepository->AddValidErrItem(new CValidErrItem(sv, et, msg, set));
}


void CValidError_imp::PostErr
(EDiagSev        sv,
 EErrType        et,
 const string&   message,
 TSet            set,
 TDesc           ds)
{
    // Append Descriptor label
    string msg(message + " DESCRIPTOR: ");
    ds.GetLabel(&msg, CSeqdesc::eBoth);
    PostErr(sv, et, msg, set);
}


void CValidError_imp::PostErr
(EDiagSev sv,
 EErrType et,
 const string&   message,
 TAnnot    an)
{
    // Append Annotation label
    string msg(message + " ANNOTATION: ");

    // !!! need to decide on the message

    m_ErrRepository->AddValidErrItem(new CValidErrItem(sv, et, msg, an));
}


void CValidError_imp::PostErr
(EDiagSev sv,
 EErrType et,
 const string& message,
 TGraph graph)
{
    // Append Graph label
    string msg(message + " GRAPH: ");
    if ( graph.IsSetTitle() ) {
        msg += graph.GetTitle();
    } else {
        msg += "Not Named";
    }
    msg += " ";
    graph.GetLoc().GetLabel(&msg);

    m_ErrRepository->AddValidErrItem(new CValidErrItem(sv, et, msg, graph));
}


void CValidError_imp::PostErr
(EDiagSev sv,
 EErrType et,