web_algo.cpp

来自「ncbi源码」· C++ 代码 · 共 365 行

/*
 * ===========================================================================
 * PRODUCTION $Log: web_algo.cpp,v $
 * PRODUCTION Revision 1000.2  2004/06/01 20:56:13  gouriano
 * PRODUCTION PRODUCTION: UPGRADED [GCC34_MSVC7] Dev-tree R1.7
 * PRODUCTION
 * ===========================================================================
 */

/*  $Id: web_algo.cpp,v 1000.2 2004/06/01 20:56:13 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:  Josh Cherry
 *
 * File Description:  Plugin to call remote procedure via cgi
 *
 */


#include <ncbi_pch.hpp>
#include <corelib/ncbistd.hpp>

// http stuff
#include <connect/ncbi_conn_stream.hpp>
#include <connect/ncbi_core_cxx.hpp>
#include <connect/ncbi_util.h>

#include "../basic/output_dlg.hpp"

#include <gui/core/idocument.hpp>
#include <gui/core/plugin_utils.hpp>
#include <gui/plugin/PluginRequest.hpp>
#include <gui/plugin/PluginInfo.hpp>
#include <gui/plugin/PluginCommandSet.hpp>
#include <gui/plugin/PluginValueConstraint.hpp>
#include <gui/core/version.hpp>
#include <gui/objutils/utils.hpp>
#include <gui/utils/browser_utils.hpp>

#include "web_algo.hpp"
#include "plugin_args_as_strs.hpp"

#include <objmgr/util/sequence.hpp>

#include <objmgr/scope.hpp>
#include <objmgr/seq_vector.hpp>
#include <objmgr/util/sequence.hpp>
#include <objtools/readers/readfeat.hpp>
#include <objects/seqloc/Seq_loc.hpp>
#include <objects/general/Object_id.hpp>
#include <connect/ncbi_pipe.hpp>

#include <serial/serial.hpp>
#include <serial/objostr.hpp>
#include <serial/objistr.hpp>

#include <cgi/ncbicgi.hpp>
#include <util/regexp.hpp>

BEGIN_NCBI_SCOPE
USING_SCOPE(objects);

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

    // version info macro
    info.SetInfo(CPluginVersion::eMajor, CPluginVersion::eMinor, 0,
                 string(__DATE__) + " " + string(__TIME__),
                 "CAlgoWebService", "", "", "");

    // command info
    CPluginCommandSet& cmds = info.SetCommands();
    CPluginCommand&    args = cmds.AddAlgoCommand(eAlgoCommand_run);
    args.AddArgument("locs", "Sequences to evaluate",
                     CSeq_loc::GetTypeInfo(),
                     CPluginArg::TData::e_Array);
    args.SetConstraint("locs",
                       *CPluginValueConstraint::CreateSeqSameMol());
    args.AddArgument("__url", "url",
                     CPluginArg::eString);
}


void CAlgoWebService::RunCommand(CPluginMessage& msg)
{
    const CPluginCommand& args = msg.GetRequest().GetCommand();
    CPluginReply& reply = msg.SetReply();
    _TRACE("CAlgoWebService::Run()");
    if ( !m_OutputDlg.get() ) {
        m_OutputDlg.reset(new COutputDlg());
    }
    m_OutputDlg->SetTitle("Results of calling web algorithm");

    const CPluginArgSet::Tdata& argset = args.GetArgs().Get();
    vector<string> keys, values;
    try {
        PluginArgsAsStrs(argset, keys, values);
    }
    catch (exception& e) {
        LOG_POST(Info << "CAlgoWebService: couldn't convert to string rep: "
                 << e.what());
        reply.SetStatus(eMessageStatus_failed);
        return;
    }
    catch (...) {
        LOG_POST(Info << "CAlgoWebService: couldn't convert to string rep");
        reply.SetStatus(eMessageStatus_failed);
        return;
    }
    keys.push_back("action");
    values.push_back("run");

    string query;
    for (unsigned int i = 0;  i < keys.size();  i++) {
        if (i > 0) {
            query += '&';
        }
        query += URL_EncodeString(keys[i]) + "=" + URL_EncodeString(values[i]);
    }
    
    string body;
    string headers;
    string base;

    if (args.HasArgument("__url")  &&
        CPluginUtils::IsValid(args["__url"])) {
        x_DoPost(args["__url"].AsString(), query,
                 body, headers);
        base = args["__url"].AsString();
    } else {
        throw runtime_error("CAlgoWebService::run:  no url");
    }

    // see if we have a Content-type; if so, act appropriately
    string content_type = x_GetContentType(headers);
    if (content_type == "chemical/ncbi-asn1-ascii" ||
        content_type == "chemical/ncbi-asn1-binary" ||
        content_type == "chemical/ncbi-xml" ||
        content_type == "chemical/ncbi-five-col") {

        // in these cases we'll try to make an annot
        CRef<CSeq_annot> annot;
        CNcbiIstrstream is(body.c_str(), body.size());
        if (content_type == "chemical/ncbi-asn1-ascii" ||
        content_type == "chemical/ncbi-asn1-binary" ||
        content_type == "chemical/ncbi-xml") {
            ESerialDataFormat serial_format;
            if (content_type == "chemical/ncbi-asn1-ascii") {
                serial_format = eSerial_AsnText;
            } else if (content_type == "chemical/ncbi-asn1-binary") {
                serial_format = eSerial_AsnBinary;
            } else if (content_type == "chemical/ncbi-xml") {
                serial_format = eSerial_Xml;
            }
            auto_ptr<CObjectIStream>
                istr(CObjectIStream::Open(serial_format, is));
            annot.Reset(new CSeq_annot);
            *istr >> *annot;
        } else if (content_type == "chemical/ncbi-five-col") {
            annot = CFeature_table_reader::ReadSequinFeatureTable(is);
        }
        try {
            x_DealWithAnnot(annot, args, reply);
        }
        catch (exception& e) {
            LOG_POST(Error << e.what());
            reply.SetStatus(eMessageStatus_failed);
            return;
        }
        reply.AddAction(CPluginReplyAction::e_Add_to_document);
    } else if (content_type == "text/html") {
        if (!base.empty()) {
            CBrowserUtils::AddBaseTag(body, base);
        }
        CBrowserUtils::SendToBrowser(body);
    } else if (content_type == "text/plain" ||
               content_type.empty()) {
        string str;
        str += "Output:\n\n";
        str += body + "\n\n";

        m_OutputDlg->SetText(str);
        m_OutputDlg->Show();
    } else {
        CBrowserUtils::SendToBrowser(body, content_type);
    }
    reply.SetStatus(eMessageStatus_success);
}


void CAlgoWebService::x_DealWithAnnot(CRef<CSeq_annot> annot,
                               const CPluginCommand& args,
                               CPluginReply& reply)
{
    // figure out which argument this annot belongs to
    CSeq_annot::TData::TFtable& ftable = annot->SetData().SetFtable();
    const CSeq_id& orig_id = sequence::GetId(ftable.front()->GetLocation());
    const string& orig_id_str = orig_id.GetLocal().GetStr();

    const CSeq_loc *loc;
    const IDocument *doc;

    // if it has the form name[integer], interpret it as array member
    CRegexp re("(.*)\\[ *([0-9]+) *\\]");
    if (!re.GetMatch(orig_id_str.c_str(), 0).empty()) {
        const int *name_range = re.GetResults(1);
        string name = orig_id_str.substr(name_range[0], name_range[1]);
        const int *index_range = re.GetResults(2);
        unsigned int index = 
            NStr::StringToInt(orig_id_str
                              .substr(index_range[0],
                                      index_range[1] - index_range[0]));
        plugin_args::TLocList loc_list;
        GetArgValue(args[name], loc_list);
        if (index > loc_list.size() - 1) {
            throw runtime_error(orig_id_str + " does not exist; "
                                + name + " has only " + 
                                NStr::IntToString(loc_list.size())
                                + " elements");
        }
        plugin_args::TLocList::iterator iter = loc_list.begin();
        advance(iter, index);
        loc = iter->second;
        doc = iter->first;
    } else {
        const CPluginArg& arg = args[orig_id_str];
        loc = dynamic_cast<const CSeq_loc*> (&arg.AsObject());
        doc = arg.GetDocument();
    }

    // now set the ids and remap

    NON_CONST_ITERATE (list<CRef<CSeq_feat> >, feat, ftable) {
        CSeq_loc& feat_loc = (*feat)->SetLocation();
        feat_loc.SetId(sequence::GetId(*loc));
        (*feat)->SetLocation(*CSeqUtils::RemapChildToParent(*loc, feat_loc));
    }
    
    reply.AddObject(*doc, *annot);

    m_OutputDlg->SetText("Annotations added to " + doc->GetShortTitle());
    m_OutputDlg->Show();

}


string CAlgoWebService::x_GetContentType(const string& headers)
{
    list<string> lines;
    string content_type;
    NStr::Split(headers, "\n", lines);
    ITERATE (list<string>, line, lines) {
        if (NStr::StartsWith(*line, "content-type:", NStr::eNocase)) {
            content_type = line->substr(string("content-type:").size());
            content_type = NStr::TruncateSpaces(content_type);
            NStr::ToLower(content_type);
            break;
        }
    }
    return content_type;
}


void CAlgoWebService::x_DoPost(const string& url, const string& query,
                           string& body, string& headers)
{

    body.erase();
    headers.erase();

    SConnNetInfo* net_info = ConnNetInfo_Create(0);
    if (!net_info) {
        throw runtime_error("ConnNetInfo_Create(0) returned null");
    }
    ConnNetInfo_ParseURL(net_info, url.c_str());
    net_info->req_method = eReqMethod_Post;

    // pretend to be a 'real' web browser; some servers
    // care about this
    string user_header =
        "User-Agent: Mozilla/4.0 (compatible; MSIE 5.01; "
        "Windows NT 5.0; (R1 1.1))";
    user_header = "Content-type: application/x-www-form-urlencoded\n";
    CConn_HttpStream http(net_info, user_header,
                          fHCC_AutoReconnect | fHCC_KeepHeader);
    ConnNetInfo_Destroy(net_info);

    // write the query string to server
    http << query;

    // now read
    char buf[1024];
    string line;

    while (getline(http, line)) {
        if (line.find_first_not_of(" \t\n\r\f\v") == string::npos) {
            break;
        }
        headers += line + '\n';
    }

    while (!http.fail()) {
        http.read(buf, 1024);
        body.append(buf, http.gcount());
    }

}


END_NCBI_SCOPE


/*
 * ===========================================================================
 * $Log: web_algo.cpp,v $
 * Revision 1000.2  2004/06/01 20:56:13  gouriano
 * PRODUCTION: UPGRADED [GCC34_MSVC7] Dev-tree R1.7
 *
 * Revision 1.7  2004/05/21 22:27:47  gorelenk
 * Added PCH ncbi_pch.hpp
 *
 * Revision 1.6  2004/05/03 13:05:42  dicuccio
 * gui/utils --> gui/objutils where needed
 *
 * Revision 1.5  2004/01/27 18:45:25  dicuccio
 * Added missing header files
 *
 * Revision 1.4  2003/12/19 18:38:18  jcherry
 * Alert the user when annotations are added
 *
 * Revision 1.3  2003/12/01 23:22:00  jcherry
 * Added registry file format for plugin info
 *
 * Revision 1.2  2003/11/26 17:13:08  dicuccio
 * Lots of code clean-up.  CHanged names of algorithms to CAlgoWebServices{Init}
 *
 * Revision 1.1  2003/11/25 19:08:50  jcherry
 * Initial version
 *
 * ===========================================================================
 */