genbank_loader.cpp

来自「ncbi源码」· C++ 代码 · 共 386 行

/*
 * ===========================================================================
 * PRODUCTION $Log: genbank_loader.cpp,v $
 * PRODUCTION Revision 1000.5  2004/06/01 20:57:58  gouriano
 * PRODUCTION PRODUCTION: UPGRADED [GCC34_MSVC7] Dev-tree R1.38
 * PRODUCTION
 * ===========================================================================
 */

/*  $Id: genbank_loader.cpp,v 1000.5 2004/06/01 20:57:58 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.
 *
 * ===========================================================================
 *
 * Authors:  Mike DiCuccio
 *
 * File Description:
 *    CDataPlugin_GenbankLoader - load sequence information form Genbank.
 */

#include <ncbi_pch.hpp>
#include "genbank_loader.hpp"

#include <gui/core/doc_exception.hpp>
#include <gui/core/doc_manager.hpp>
#include <gui/core/idocument.hpp>
#include <gui/core/version.hpp>
#include <gui/core/plugin_utils.hpp>
#include <gui/plugin/PluginCommandSet.hpp>
#include <gui/plugin/PluginInfo.hpp>
#include <gui/plugin/PluginValue.hpp>
#include <gui/utils/message_box.hpp>
#include <gui/objutils/label.hpp>
#include <objmgr/scope.hpp>
#include <objmgr/util/sequence.hpp>

#include "loader_utils.hpp"

#include "gb_manage_dlg.hpp"


BEGIN_NCBI_SCOPE
USING_SCOPE(objects);



//
// class CDataPlugin_GenbankLoaderException defines some internal exception types used in
// processing files.
//
// This class is used internally to avoid an ad-hoc exception mechanism;
// currently, it reports only errors concerning invalid format types.
//
class CDataPlugin_GenbankLoaderException : EXCEPTION_VIRTUAL_BASE public CException
{
public:
    // Enumerated list of exception types
    enum EErrCode {
        eInvalidId
    };

    // Convert an enuerated exception to a human-readable string representation
    // of this exception.
    virtual const char* GetErrCodeString(void) const
    {
        switch (GetErrCode()) {
        case eInvalidId:        return "eInvalidId";
        default:                    return CException::GetErrCodeString();
        }
    }

    // constructor boilerplate
    NCBI_EXCEPTION_DEFAULT(CDataPlugin_GenbankLoaderException, CException);
};


void CDataPlugin_GenbankLoader::GetInfo(CPluginInfo& info)
{
    info.Reset();

    // version info macro
    info.SetInfo(CPluginVersion::eMajor, CPluginVersion::eMinor, 0,
                 string(__DATE__) + " " + string(__TIME__),
                 "CDataPlugin_GenbankLoader", "Genbank Accession",
                 "Open a Genbank accession",
                 "");

    // command info
    CPluginCommandSet& cmds    = info.SetCommands();
    CPluginCommand& load_args  = cmds.AddDataCommand(eDataCommand_load);
    load_args.AddArgument("acc", "Accessions to load",
                          CPluginArg::eString,
                          CPluginArg::TData::e_Array);
}


CDataPlugin_GenbankLoader::CDataPlugin_GenbankLoader()
{
}


CDataPlugin_GenbankLoader::~CDataPlugin_GenbankLoader()
{
}


//
// load a record from Genbank into a fully-prepared document
//
void CDataPlugin_GenbankLoader::Load(CPluginMessage& msg)
{
    const CPluginCommand& args = msg.GetRequest().GetCommand();
    CPluginReply& reply = msg.SetReply();
    vector<string> str_vec = CPluginUtils::ArgToStringVec(args["acc"]);

    reply.SetStatus(eMessageStatus_failed);
    size_t load_success = 0;
    ITERATE (vector<string>, iter, str_vec) {
        //
        // now, go about loading our entry, if we can
        // check to make sure the Accession is (moderately) valid
        // if the user puts in just a number, assume its a gi
        //
        string acc = NStr::TruncateSpaces(*iter);
        if (acc.empty()) {
            continue;
        }

        CLoaderStrFormatter::AddGIPrefix(&acc);

        try {
            CConstRef<CSeq_id> id(new CSeq_id(acc));
            if (id->Which() == CSeq_id::e_not_set) {
                NCBI_THROW(CDataPlugin_GenbankLoaderException, eInvalidId,
                           "Accession " + acc + " is not valid");
            }
            CRef<CScope> scope(new CScope(CDocManager::GetObjectManager()));
            scope->AddDefaults();
            CBioseq_Handle handle = scope->GetBioseqHandle(*id);
            if ( !handle ) {
                LOG_POST(Error
                         << "failed to load sequence for accession " << acc);
                continue;
            }

            // marshal our document with this scope
            IDocument* doc = CDocManager::CreateDocument(*scope, *id);
            reply.AddObject(*doc);
            ++load_success;
            LOG_POST(Info <<
                     "CDataPlugin_GenbankLoader: loaded accession " << acc);


            // Needs to be added to MRU list

            // Get label for the MRU list
            string label("GenBank: ");
            CLabel::GetLabel(*handle.GetSeqId(), &label,
                             CLabel::eDefault, &handle.GetScope());

            // Create a new Plugin Message containing only this entry
            //  for the MRU list
            CRef<CPluginMessage> new_msg(new CPluginMessage());
            new_msg->Assign(msg);
            CPluginCommand& new_args = new_msg->SetRequest().SetCommand();
            CPluginArg& acc_arg = new_args["acc"];
            acc_arg.SetString(acc);
            acc_arg.SetList();

            // add to MRU list
            CDocManager::AddMRUEntry(*new_msg, label);
        }
        catch (CException& e) {
            string msg("Failed to load accession '");
            msg += acc;
            msg += "':\n";
            msg += e.GetMsg();
            NcbiMessageBox(msg);
        }
        catch (std::exception& e) {
            string msg("Failed to load accession '");
            msg += acc;
            msg += "':\n";
            msg += e.what();
            NcbiMessageBox(msg);
        }
#ifndef _DEBUG
        catch (...) {
            string msg("Failed to load accession '");
            msg += acc;
            msg += "':\nUnknown error";
            NcbiMessageBox(msg);
        }
#endif
    }

    if (str_vec.size() != load_success) {
        reply.SetStatus(eMessageStatus_failed);
        if (load_success) {
            NcbiMessageBox("Failed to load some accessions.");
        } else {
            NcbiMessageBox("Failed to load any accessions.");
        }
        return;
    }

    if (load_success != 0) {
        LOG_POST(Info << "CDataPlugin_GenbankLoader: all accessions loaded successfully");
    }
    reply.SetStatus(eMessageStatus_success);
}


END_NCBI_SCOPE


/*
 * ===========================================================================
 * $Log: genbank_loader.cpp,v $
 * Revision 1000.5  2004/06/01 20:57:58  gouriano
 * PRODUCTION: UPGRADED [GCC34_MSVC7] Dev-tree R1.38
 *
 * Revision 1.38  2004/05/25 17:21:59  dicuccio
 * Modified class names.  Fonts to 12 point
 *
 * Revision 1.37  2004/05/21 22:27:48  gorelenk
 * Added PCH ncbi_pch.hpp
 *
 * Revision 1.36  2004/05/18 11:22:34  friedman
 * Add recently loaded documents to the MRU list.
 *
 * Revision 1.35  2004/01/06 20:48:17  dicuccio
 * Moved CFltkCursorGuard into CPlugin::Execute() - broadly applied to all plugins
 *
 * Revision 1.34  2004/01/06 20:15:54  dicuccio
 * Use CFltkCursorGuard on load / import / search
 *
 * Revision 1.33  2004/01/05 18:50:28  vasilche
 * Fixed path to include files.
 *
 * Revision 1.32  2003/12/10 15:15:44  dicuccio
 * Added message box to warn when accessions fail to load.
 *
 * Revision 1.31  2003/12/05 13:07:53  dicuccio
 * Split management interface into a separate plugin.  Fixed linker error
 * introduced with status bar
 *
 * Revision 1.30  2003/11/24 15:45:36  dicuccio
 * Renamed CVersion to CPluginVersion
 *
 * Revision 1.29  2003/11/18 17:49:01  dicuccio
 * Added standard processing of return values
 *
 * Revision 1.28  2003/11/06 20:12:15  dicuccio
 * Cleaned up handling of USING_SCOPE - removed from all headers
 *
 * Revision 1.27  2003/11/04 17:49:24  dicuccio
 * Changed calling parameters for plugins - pass CPluginMessage instead of paired
 * CPluginCommand/CPluginReply
 *
 * Revision 1.26  2003/10/10 17:19:00  dicuccio
 * Added Manage() interface.
 *
 * Revision 1.25  2003/10/07 13:47:05  dicuccio
 * Renamed CPluginURL* to CPluginValue*
 *
 * Revision 1.24  2003/09/04 14:51:24  dicuccio
 * Use IDocument instead of CDocument
 *
 * Revision 1.23  2003/08/05 17:16:25  kuznets
 * Changes to improve code reuse between loaders + CLDS_Database passed as the
 * construction parameter to the lds search dialog
 *
 * Revision 1.22  2003/07/14 11:16:52  shomrat
 * Plugin messageing system related changes
 *
 * Revision 1.21  2003/06/25 17:02:58  dicuccio
 * Split CPluginHandle into a handle (pointer-to-implementation) and
 * implementation file.  Lots of #include file clean-ups.
 *
 * Revision 1.20  2003/06/20 14:52:57  dicuccio
 * Revised plugin registration - moved GetInfo() into the plugin handler
 *
 * Revision 1.19  2003/06/02 16:06:22  dicuccio
 * Rearranged src/objects/ subtree.  This includes the following shifts:
 *     - src/objects/asn2asn --> arc/app/asn2asn
 *     - src/objects/testmedline --> src/objects/ncbimime/test
 *     - src/objects/objmgr --> src/objmgr
 *     - src/objects/util --> src/objmgr/util
 *     - src/objects/alnmgr --> src/objtools/alnmgr
 *     - src/objects/flat --> src/objtools/flat
 *     - src/objects/validator --> src/objtools/validator
 *     - src/objects/cddalignview --> src/objtools/cddalignview
 * In addition, libseq now includes six of the objects/seq... libs, and libmmdb
 * replaces the three libmmdb? libs.
 *
 * Revision 1.18  2003/05/30 14:15:42  dicuccio
 * Renamed MessageBox to NcbiMessageBox because brain-dead MSVC thinks this is
 * ::MessageBox and rewrites the symbol as MessageBoxA, which results in an
 * unresolved external and conflict with the Win32 API :(.
 *
 * Revision 1.17  2003/05/30 13:10:37  dicuccio
 * Use MessageBox() instead of fl_alter()
 *
 * Revision 1.16  2003/05/19 13:40:44  dicuccio
 * Moved gui/core/plugin/ -> gui/plugin/.  Merged core libraries into libgui_core.
 * Removed old, unused dialog box.
 *
 * Revision 1.15  2003/05/01 12:54:34  dicuccio
 * Changed a useful trace -> log post for failure to load accession
 *
 * Revision 1.14  2003/04/29 14:52:31  dicuccio
 * Deprecated old accession entry dialog.  Added correct plugin argument to specify accession names
 *
 * Revision 1.13  2003/04/24 16:39:08  dicuccio
 * Updated to reflect changes in plugin API.  Changed location of document
 * marshal from framework to plugin
 *
 * Revision 1.12  2003/02/26 14:26:56  dicuccio
 * Removed unnecessary _TRACE() statements
 *
 * Revision 1.11  2003/02/24 13:03:16  dicuccio
 * Renamed classes in plugin spec:
 *     CArgSeg --> CPluginArgSet
 *     CArgument --> CPluginArg
 *     CPluginArgs --> CPluginCommand
 *     CPluginCommands --> CPluginCommandSet
 *
 * Revision 1.10  2003/02/20 19:49:56  dicuccio
 * Created new plugin architecture, based on ASN.1 spec.  Moved GBENCH
 * frameowrk over to use new plugin architecture.
 *
 * Revision 1.9  2003/02/06 18:48:36  dicuccio
 * Made 'catch (...)' conditional for non-debug builds
 *
 * Revision 1.8  2003/01/15 21:08:34  dicuccio
 * Removed dead _TRACE() statements
 *
 * Revision 1.7  2003/01/13 13:10:07  dicuccio
 * Namespace clean-up.  Retired namespace gui -> converted all to namespace
 * ncbi.  Moved all FLUID-generated code into namespace ncbi.
 *
 * Revision 1.6  2003/01/03 17:25:50  dicuccio
 * Fixed (finally!) issues regarding input validation of seq-ids.
 *
 * Revision 1.5  2002/12/30 18:44:20  dicuccio
 * Fixed a number of buglets in accession dialog and in handling of empty
 * accessions.
 *
 * Revision 1.4  2002/12/30 17:48:28  dicuccio
 * Added mechanism for data loader plugins to announce supported modes of
 * operation (load, import, save currently)
 *
 * Revision 1.3  2002/12/20 19:17:58  dicuccio
 * Added better handling of sueqnece ID errors (try not to segfault...)
 *
 * Revision 1.2  2002/12/12 15:22:38  dicuccio
 * Minot tweak to avoid a compiler warning.
 *
 * Revision 1.1  2002/11/25 20:51:01  dicuccio
 * Initial revision.
 *
 * ===========================================================================
 */