TRestDataSet.h

/*************************************************************************
 * This file is part of the REST software framework.                     *
 *                                                                       *
 * Copyright (C) 2016 GIFNA/TREX (University of Zaragoza)                *
 * For more information see https://gifna.unizar.es/trex                 *
 *                                                                       *
 * REST is free software: you can redistribute it and/or modify          *
 * it under the terms of the GNU General Public License as published by  *
 * the Free Software Foundation, either version 3 of the License, or     *
 * (at your option) any later version.                                   *
 *                                                                       *
 * REST is distributed in the hope that it will be useful,               *
 * but WITHOUT ANY WARRANTY; without even the implied warranty of        *
 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the          *
 * GNU General Public License for more details.                          *
 *                                                                       *
 * You should have a copy of the GNU General Public License along with   *
 * REST in $REST_PATH/LICENSE.                                           *
 * If not, see https://www.gnu.org/licenses/.                            *
 * For the list of contributors see $REST_PATH/CREDITS.                  *
 *************************************************************************/
 
#ifndef REST_TRestDataSet
#define REST_TRestDataSet
 
#include <TTimeStamp.h>
 
#include <ROOT/RDataFrame.hxx>
 
#include "TRestCut.h"
#include "TRestMetadata.h"
 
class TRestDataSet : public TRestMetadata {
   public:
    struct RelevantQuantity {
        std::string metadata;
 
        std::string strategy;
 
        std::string description;
 
        std::string value;
    };
 
   private:
    std::string fFilterStartTime = "2000/01/01";  //<
 
    std::string fFilterEndTime = "3000/12/31";  //<
 
    std::string fFilePattern = "";  //<
 
    std::vector<std::string> fObservablesList;  //<
 
    std::vector<std::string> fFilterMetadata;  //<
 
    std::vector<std::string> fFilterContains;  //<
 
    std::vector<Double_t> fFilterGreaterThan;  //<
 
    std::vector<Double_t> fFilterLowerThan;  //<
 
    std::vector<Double_t> fFilterEqualsTo;  //<
 
    std::map<std::string, RelevantQuantity> fQuantity;  //<
 
    TRestCut* fCut = nullptr;  //<
 
    Double_t fTotalDuration = 0;  //<
 
    std::vector<std::string> fFileSelection;  //<
 
    Double_t fStartTime = REST_StringHelper::StringToTimeStamp(fFilterEndTime);  //<
 
    Double_t fEndTime = REST_StringHelper::StringToTimeStamp(fFilterStartTime);  //<
 
    Bool_t fMergedDataset = false;  //<
 
    std::vector<std::string> fImportedFiles;  //<
 
    std::vector<std::pair<std::string, std::string>> fColumnNameExpressions;  //<
 
    Bool_t fMT = false;  //<
 
    // If the dataframe was defined externally it will be true
    Bool_t fExternal = false;  //<
 
    ROOT::RDF::RNode fDataFrame = ROOT::RDataFrame(0);  
 
    TChain* fTree = nullptr;  
 
    void InitFromConfigFile() override;
 
   protected:
    virtual std::vector<std::string> FileSelection();
 
    void RegenerateTree(std::vector<std::string> finalList = {});
 
   public:
    ROOT::RDF::RNode GetDataFrame() const {
        if (!fExternal && fTree == nullptr)
            RESTWarning << "DataFrame has not been yet initialized" << RESTendl;
        return fDataFrame;
    }
 
    void EnableMultiThreading(Bool_t enable = true) { fMT = enable; }
 
    TTree* GetTree() const {
        if (fTree == nullptr && fExternal) {
            RESTInfo << "The tree is not accessible. Only GetDataFrame can be used in an externally "
                        "generated dataset"
                     << RESTendl;
            RESTInfo << "You may write a tree using GetDataFrame()->Snapshot(\"MyTree\", \"output.root\");"
                     << RESTendl;
            return fTree;
        }
 
        if (fTree == nullptr) {
            RESTError << "Tree has not been yet initialized" << RESTendl;
            RESTError << "You should invoke TRestDataSet::GenerateDataSet() or " << RESTendl;
            RESTError << "TRestDataSet::Import( fname ) before trying to access the tree" << RESTendl;
        }
        return fTree;
    }
 
    size_t GetNumberOfColumns() { return fDataFrame.GetColumnNames().size(); }
 
    size_t GetNumberOfBranches() { return GetNumberOfColumns(); }
 
    std::vector<std::string> GetFileSelection() { return fFileSelection; }
 
    Double_t GetTotalTimeInSeconds() const { return fTotalDuration; }
 
    inline auto GetFilterStartTime() const { return fFilterStartTime; }
    inline auto GetFilterEndTime() const { return fFilterEndTime; }
    inline auto GetStartTime() const { return fStartTime; }
    inline auto GetEndTime() const { return fEndTime; }
    inline auto GetFilePattern() const { return fFilePattern; }
    inline auto GetObservablesList() const { return fObservablesList; }
    inline auto GetFileSelection() const { return fFileSelection; }
    inline auto GetFilterMetadata() const { return fFilterMetadata; }
    inline auto GetFilterContains() const { return fFilterContains; }
    inline auto GetFilterGreaterThan() const { return fFilterGreaterThan; }
    inline auto GetFilterLowerThan() const { return fFilterLowerThan; }
    inline auto GetFilterEqualsTo() const { return fFilterEqualsTo; }
    inline auto GetQuantity() const { return fQuantity; }
    inline auto GetAddedColumns() const { return fColumnNameExpressions; }
    inline auto GetCut() const { return fCut; }
    inline auto IsMergedDataSet() const { return fMergedDataset; }
 
    inline void SetObservablesList(const std::vector<std::string>& obsList) { fObservablesList = obsList; }
    inline void SetFilePattern(const std::string& pattern) { fFilePattern = pattern; }
    inline void SetQuantity(const std::map<std::string, RelevantQuantity>& quantity) { fQuantity = quantity; }
 
    void SetTotalTimeInSeconds(Double_t seconds) { fTotalDuration = seconds; }
    void SetDataFrame(const ROOT::RDF::RNode& dS) {
        fDataFrame = dS;
        fExternal = true;
    }
 
    TRestDataSet& operator=(TRestDataSet& dS);
    Bool_t Merge(const TRestDataSet& dS);
    void Import(const std::string& fileName);
    void Import(std::vector<std::string> fileNames);
    void Export(const std::string& filename, std::vector<std::string> excludeColumns = {});
 
    ROOT::RDF::RNode MakeCut(const TRestCut* cut);
    ROOT::RDF::RNode ApplyRange(size_t from, size_t to);
    ROOT::RDF::RNode Range(size_t from, size_t to);
    ROOT::RDF::RNode DefineColumn(const std::string& columnName, const std::string& formula);
 
    size_t GetEntries();
 
    void PrintMetadata() override;
    void Initialize() override;
 
    void GenerateDataSet();
 
    TRestDataSet();
    TRestDataSet(const char* cfgFileName, const std::string& name = "");
    ~TRestDataSet();
 
    ClassDefOverride(TRestDataSet, 9);
};
#endif