id1_fetch.cpp

ncbi源码

/*
 * ===========================================================================
 * PRODUCTION $Log: id1_fetch.cpp,v $
 * PRODUCTION Revision 1000.2  2004/06/01 18:30:30  gouriano
 * PRODUCTION PRODUCTION: UPGRADED [GCC34_MSVC7] Dev-tree R1.50
 * PRODUCTION
 * ===========================================================================
 */

/*  $Id: id1_fetch.cpp,v 1000.2 2004/06/01 18:30:30 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:  Denis Vakatov, Aleksey Grichenko, Aaron Ucko
 *
 * File Description:
 *   New IDFETCH network client (get data from "ID1")
 */

#include <ncbi_pch.hpp>
#include <corelib/ncbiapp.hpp>
#include <corelib/ncbienv.hpp>
#include <corelib/ncbiargs.hpp>
#include <corelib/ncbireg.hpp>

#include <connect/ncbi_core_cxx.hpp>

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

#include <objects/entrez2/Entrez2_boolean_element.hpp>
#include <objects/entrez2/Entrez2_boolean_exp.hpp>
#include <objects/entrez2/Entrez2_boolean_reply.hpp>
#include <objects/entrez2/Entrez2_db_id.hpp>
#include <objects/entrez2/Entrez2_docsum.hpp>
#include <objects/entrez2/Entrez2_docsum_data.hpp>
#include <objects/entrez2/Entrez2_docsum_list.hpp>
#include <objects/entrez2/Entrez2_eval_boolean.hpp>
#include <objects/entrez2/Entrez2_id_list.hpp>
#include <objects/entrez2/entrez2_client.hpp>
#include <objects/general/Date.hpp>
#include <objects/general/Date_std.hpp>
#include <objects/general/Dbtag.hpp>
#include <objects/general/Object_id.hpp>
#include <objects/id1/ID1Seq_hist.hpp>
#include <objects/id1/ID1server_maxcomplex.hpp>
#include <objects/id1/id1_client.hpp>

#include <objmgr/graph_ci.hpp>
#include <objmgr/object_manager.hpp>
#include <objmgr/scope.hpp>
#include <objmgr/seq_vector.hpp>
#include <objtools/data_loaders/genbank/gbloader.hpp>

#include <objects/seq/Bioseq.hpp>
#include <objects/seq/Seq_descr.hpp>
#include <objects/seq/Seq_hist_rec.hpp>
#include <objects/seq/Seqdesc.hpp>
#include <objects/seqloc/Seq_id.hpp>
#include <objects/seqloc/Textseq_id.hpp>
#include <objects/seqres/Byte_graph.hpp>
#include <objects/seqres/Seq_graph.hpp>
#include <objects/seqset/Seq_entry.hpp>
#include <objects/seqset/Bioseq_set.hpp>
#if 1
#include <objtools/flat/flat_ncbi_formatter.hpp>
#else
#include <objmgr/util/genbank.hpp>
#endif
#include <objmgr/util/sequence.hpp>

#include <memory>
#include <algorithm>


BEGIN_NCBI_SCOPE
USING_SCOPE(NCBI_NS_NCBI::objects); // MSVC requires qualification (!)


/////////////////////////////////
//  CId1FetchApp::
//

class CId1FetchApp : public CNcbiApplication
{
    virtual void Init(void);
    virtual int  Run (void);
    virtual void Exit(void);

private:
    bool LookUpGI(int gi);
    int  LookUpFastaSeqID(const string& s);
    int  LookUpFlatSeqID(const string& s);

    void WriteFastaIDs     (const list< CRef< CSeq_id > >& ids);

    void WriteHistoryTable (const CID1server_back& id1_reply);
    void WriteQualityScores(CBioseq_Handle& handle);

    CNcbiOstream*        m_OutputFile;
    CID1Client           m_ID1Client;
    CEntrez2Client       m_E2Client;
    CRef<CObjectManager> m_ObjMgr;
    CRef<CScope>         m_Scope;
};


void CId1FetchApp::Init(void)
{
    // Prepare command line descriptions
    //

    // Create
    auto_ptr<CArgDescriptions> arg_desc(new CArgDescriptions);

    // GI
    arg_desc->AddOptionalKey
        ("gi", "SeqEntryID",
         "GI id of the Seq-Entry to fetch",
         CArgDescriptions::eInteger);
    arg_desc->SetConstraint
        ("gi", new CArgAllow_Integers(0, 99999999));

    // Output format
    arg_desc->AddDefaultKey
        ("fmt", "OutputFormat",
         "Format to dump the resulting data in",
         CArgDescriptions::eString, "asn");
    arg_desc->SetConstraint
        ("fmt", &(*new CArgAllow_Strings,
                  "asn", "asnb", "xml", "genbank", "genpept", "fasta",
                  "quality", "docsum"));

    // Output datafile
    arg_desc->AddDefaultKey
        ("out", "ResultFile",
         "File to dump the resulting data to",
         CArgDescriptions::eOutputFile, "-", CArgDescriptions::fBinary);

    // Log file
    arg_desc->AddOptionalKey
        ("log", "LogFile",
         "File to post errors and messages to",
         CArgDescriptions::eOutputFile,
         0);

    // Database to use
    arg_desc->AddOptionalKey
        ("db", "Database", // was "-d" in `idfetch'
         "Database to use",
         CArgDescriptions::eString);
    
    // Entity number
    arg_desc->AddOptionalKey
        ("ent", "EntityNumber", // was "-e" in `idfetch'
         "(Sub)entity number (retrieval number) to dump",
         CArgDescriptions::eInteger);
    arg_desc->SetConstraint
        ("ent", new CArgAllow_Integers(0, 99999999));

    // Type of lookup
    arg_desc->AddDefaultKey
        ("lt", "LookupType", // combination of "-i" (!) and "-n" in `idfetch'
         "Type of lookup",
         CArgDescriptions::eString, "entry");
    arg_desc->SetConstraint
        ("lt", &(*new CArgAllow_Strings,
                 "entry", "state", "ids", "history", "revisions", "none"));
    
    // File with list of stuff to dump
    arg_desc->AddOptionalKey
        ("in", "RequestFile", // was "-G" (!) in `idfetch'
         "File with list of GIs, (versioned) accessions, FASTA SeqIDs to dump",
         CArgDescriptions::eInputFile, CArgDescriptions::fPreOpen);
         
    // Maximum complexity
    arg_desc->AddDefaultKey
        ("maxplex", "MaxComplexity", // was "-c" in `idfetch'
         "Maximum complexity to return",
         CArgDescriptions::eString, "entry");
    arg_desc->SetConstraint
        ("maxplex", &(*new CArgAllow_Strings,
                      "entry", "bioseq", "bioseq-set", "nuc-prot",
                      "pub-set"));
    
    // Flattened SeqID
    arg_desc->AddOptionalKey
        ("flat", "FlatID", // was "-f" in `idfetch'
         "Flattened SeqID; format can be\n"
         "\t'type([name][,[accession][,[release][,version]]])'"
         " [e.g., '5(HUMHBB)'],\n"
         "\ttype=accession[.version], or type:number",
         CArgDescriptions::eString);
    
    // FASTA-style SeqID
    arg_desc->AddOptionalKey
        ("fasta", "FastaID", // was "-s" in `idfetch'
         "FASTA-style SeqID, in the form \"type|data\"; choices are\n"
         "\tlcl|int lcl|str bbs|int bbm|int gim|int gb|acc|loc emb|acc|loc\n"
         "\tpir|acc|name sp|acc|name pat|country|patent|seq ref|acc|name|rel\n"
         "\tgnl|db|id gi|int dbj|acc|loc prf|acc|name pdb|entry|chain\n"
         "\ttpg|acc|name tpe|acc|name tpd|acc|name",
         CArgDescriptions::eString);

    // Generate GI list by Entrez query
    arg_desc->AddOptionalKey
        ("query", "EntrezQueryString", // was "-q" in `idfetch'
         "Generate GI list by Entrez query given on command line",
         CArgDescriptions::eString);
    arg_desc->AddOptionalKey
        ("qf", "EntrezQueryFile", // was "-Q" in `idfetch'
         "Generate GI list by Entrez query in given file",
         CArgDescriptions::eInputFile, CArgDescriptions::fPreOpen);

    // Program description
    string prog_description =
        "Fetch SeqEntry from ID server by its GI ID, possibly obtained from\n"
        "its SeqID or an Entrez query";
    arg_desc->SetUsageContext(GetArguments().GetProgramBasename(),
                              prog_description, false);


    arg_desc->AddDefaultKey
        ("repeat", "repeat",
         "Repeat fetch number of times",
         CArgDescriptions::eInteger, "1");

    // Pass argument descriptions to the application
    //

    SetupArgDescriptions(arg_desc.release());
}


// Workaround for "replace_if"
inline bool s_IsControl(char c)
{
    return iscntrl(c) ? true : false;
}


int CId1FetchApp::Run(void)
{
    // Process command line args
    const CArgs& args = GetArgs();

    // Setup and tune logging facilities
    if ( args["log"] ) {
        SetDiagStream( &args["log"].AsOutputFile() );
    }
#ifdef _DEBUG
    // SetDiagTrace(eDT_Enable);
    SetDiagPostLevel(eDiag_Info);
    SetDiagPostFlag(eDPF_All);
#endif

    // Make sure the combination of arguments is valid
    {{
        int id_count = 0;
        const string& fmt = args["fmt"].AsString();

        if (args["gi"])     id_count++;
        if (args["in"])     id_count++;
        if (args["flat"])   id_count++;
        if (args["fasta"])  id_count++;
        if (args["query"])  id_count++;
        if (args["qf"])     id_count++;

        if (id_count != 1) {
            NCBI_THROW(CArgException,eNoArg,
                "You must supply exactly one argument"
                " indicating what to look up.");
        }
        if ((args["query"]  ||  args["qf"]  ||  fmt == "docsum")
            &&  !args["db"]) {
            ERR_POST("No Entrez database supplied.  Try -db Nucleotide or "
                     "-db Protein.");
            return -1;
        }
        if ((fmt == "genbank"  ||  fmt == "genpept"  ||  fmt == "quality")
            &&  args["lt"].AsString() != "entry") {
            ERR_POST("The output format '" << fmt
                     << "' is only available for Seq-Entries.");
            return -1;
        }
    }}

    // Setup application registry, error log, and MT-lock for CONNECT library
    CONNECT_Init(&GetConfig());

    m_E2Client.SetDefaultRequest().SetTool("id1_fetch");

    // Open output file
    m_OutputFile = &args["out"].AsOutputFile();

    // Set up object manager
    m_ObjMgr = new CObjectManager;
    m_Scope = new CScope(*m_ObjMgr);
    m_ObjMgr->RegisterDataLoader( *new CGBDataLoader("GENBANK"));
    m_Scope->AddDataLoader("GENBANK");

    int repeat = args["repeat"].AsInteger();